bmjopen.bmj.com€¦ · For peer review only 113 Randomisation 114 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 115 investigator without patient

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available. When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to. The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript. BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com). If you have any questions on BMJ Open’s open peer review process please email

[email protected]

on October 16, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2018-023583 on 20 S

eptember 2019. D

ownloaded from

http://bmjopen.bmj.com/

[email protected]


For peer review only

Feasibility of a Transition Diabetes Team to discharge patients with type 2 diabetes starting injectable therapies

Journal: BMJ Open

Manuscript ID bmjopen-2018-023583

Article Type: Research

Date Submitted by the Author: 04-May-2018

Complete List of Authors: Pyrlis, Felicity; Austin Health, Endocrinology Ogrin, Rajna; University of Melbourne, Medicine; Bolton Clarke Research Institute Arthur, Sonja; University of Melbourne, Medicine Zhai, Cathy; University of Melbourne, Medicine Churilov, Leonid; University of Melbourne, Medicine Baqar, Sara; Austin Health, Endocrinology Zajac, Jeffrey; Austin Health, Endocrinology; University of Melbourne,

Medicine Ekinci , Elif ; Austin Health, Endocrinology; University of Melbourne, Medicine

Keywords: diabetes, insulin, inpatient, injectable, discharge

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml

BMJ Open on O

ctober 16, 2020 by guest. Protected by copyright.

http://bmjopen.bm

j.com/

BM

J Open: first published as 10.1136/bm

jopen-2018-023583 on 20 Septem

ber 2019. Dow

nloaded from



Feasibility of a Transition Diabetes Team to discharge patients with type 2 diabetes

starting injectable therapies

Pyrlis F1, Ogrin R

2,3,, Arthur S³, Zhai B

3, Churilov L³, Baqar S

1, Zajac JD

1,3, Ekinci E I

1,3

1 Endocrinology Department, Austin Health, Heidelberg, Australia

2 Bolton Clarke Research Institute, Melbourne, Australia

3Department of Medicine Austin Health, University of Melbourne, Melbourne, Australia

Corresponding Author:

Dr Felicity Pyrlis

Endocrinology Department, Austin Health

300 Waterdale Road, Heidelberg, VIC 3081

Australia

Email: [email protected]

Word count: 2496

Keywords: diabetes, inpatient, insulin, injectable, discharge

Ethical Approval: HREC Austin Health, Victoria, Australia

Page 1 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Abstract

Objectives: This study aimed to investigate if the use of a transition team was feasible for

patients with diabetes being discharged from hospital on injectable diabetes therapies.

Design, Setting, Participants: This pilot, randomised controlled trial was conducted between

2014 and 2016 conjointly by a tertiary referral hospital and a community healthcare provider. Hospital inpatients (n=105) on new injectable diabetes therapies were randomised 1:1 to

transition team or standard care.

Interventions: The transition team received in-home diabetes education 24-48 hours post-

discharge, with endocrinologist review 2-4 weeks and 16 weeks post-discharge.

Main outcome measures: The primary outcome was feasibility, defined by percentage of

patients successfully receiving the intervention. Secondary outcomes included safety, defined

by hospital readmission and emergency department presentations within 16 weeks post-

randomization, and treatment satisfaction, measured using Diabetes Treatment Satisfaction

Questionnaire (DTSQ). Exploratory outcomes included length of stay (LOS), and change in

HbA1c throughout the study.

Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in

safety between groups was detected. No difference in change in HbA1c between groups was

detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team

median HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a significant improvement

in patient satisfaction in the transition team (standard care median 10.5 (IQR 8.5, 16);

transition team DTSQc median 15 (IQR 10, 17.5), p=0.047), although interpretability is

limited by missing data.

Conclusion: This study demonstrated that the use of a novel transition diabetes team is a

feasible alternative model of care.

Page 2 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Article Summary: Strengths and Limitations in this Study

• A strength of this study was the successful collaboration of two large organisations

providing support for the transition of care from hospital to home. Daily challenges

in coordinating team members across organisations to achieve the same goals was evident throughout the trial. Despite this, the current study recruited a large number

of patients.

• We acknowledge the limitations of the study. Eight patients in each group did not

have follow-up HbA1c measurements, despite active encouragement.

• The number of missing DTSQc questionnaires may have limited interpretability of

DTSQc results. The follow-up DTSQc was completed by 40 (16.7% missing) of the

standard care patients and 36 (34.5% missing) of the transition team patients, despite the best efforts of investigators to obtain completed DTSQc questionnaires. We

believe the non-random nature of missing data was largely due to factors related to

the transition team group, particularly difficulties involved in attending appointments at hospital. However, it may be that patients failing to complete the DTSQc were less

satisfied with treatment, suggesting that this was more likely in the transition team

group. A significantly higher number of male participants failed to complete the

DTSQc, and missing data were higher in those patients with hypertension and

dyslipidaemia. The difficulty in obtaining this data raises concerns about the

feasibility of using this method to assess treatment satisfaction in patients recently

hospitalised with acute illness.

• Withdrawal of participants from the transition team in hospital prior to the

intervention may reflect reluctance in hospitalized patients to accept health providers

entering their home, which may be a limitation of this model.

Page 3 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Introduction

Despite relative stability or a slight decline in diabetes-related mortality (1), global prevalence

of diabetes is rising (2). Consequently, health care costs related to diabetes continue to

increase over time (3).

It has been demonstrated that 34% of hospital inpatients aged over 54 attending a tertiary hospital have diabetes mellitus (4). Many studies have demonstrated that inpatients with type

2 diabetes have longer hospital length of stay and higher mortality rates compared to those

without (5, 6). Factors such as stress hyperglycaemia, medications, and inadequate glycaemic control at the time of hospital admission, often result in the need for intensification with

injectable diabetes therapies (7). Furthermore, the hospitalisation period provides an

opportunity to identify those with poor glycaemic control and optimise diabetes management

(8). However, commencement of injectable therapies can be difficult in the context of

concurrent acute illness.

Guidelines recommend that patients are required to demonstrate self-management with

injectable therapies prior to hospital discharge (9). Diabetes education is crucial in enabling

patients to effectively self-manage, and assists in optimising glycaemic control post-discharge

(10). However, diabetes education in the hospital setting is subject to a number of limitations

including acute illness, pain and a sense of being overwhelmed (11). Home-based diabetes education may prove more effective based on understanding a patient’s life context and

allowing adaptation of self-management routines such as timing of injection, sharps disposal

and medication storage to suit the patient’s home environment (12).

This pilot study developed and evaluated the use of a transition team comprising in-home

diabetes education by a credentialed diabetes educator (CDE), and early post-discharge

assessment by an endocrinologist. We hypothesized that the proposed intervention would be

feasible and would not negatively affect patient satisfaction when compared to standard care.

Page 4 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Methods

Design

This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in

metropolitan Melbourne and a community-based healthcare provider. Study participants were

recruited during inpatient admissions between March 2014 and November 2015 and follow-up continued until March 2016. SQUIRE Reporting guidelines adhered to (13).

Participants

Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, were screened for the study, and randomised to receive the intervention or standard care after

providing informed consent.

Participant inclusion criteria:

• Type 2 diabetes,

• Age greater than 18 years,

• Need to start or change insulin or other injectable therapy, therefore requiring a

credentialed diabetes educator (CDE) to provide education prior to discharge,

• Ability to be contactable by telephone,

• Medically stable and awaiting diabetes education,

• Reside within a 30-minute travel radius of the hospital. This initial criterion was later

abandoned and the decision to provide home care in the patient’s residence was

placed at discretion of the CDE,

• Ability to attend hospital for outpatient follow-up,

• Stable glycaemia defined as blood glucose levels between 5-15 mmol/L in the 24

hours prior to randomisation.

Participant exclusion criteria:

• Participants who did not fulfil inclusion criteria,

• Participants unable to provide informed consent.

Randomisation

Participants were randomised in a 1:1 ratio using permuted block randomisation, by an

investigator without patient contact. Group allocations were concealed by writing allocations

on a card, and placing in sealed, unlabelled envelopes, with each consecutive participant given their allocation after informed consent was obtained. Due to the difference in treatment

protocols, the study was open label to the participants and investigators.

Patient and Public Involvement

Changes in diabetes management, including initiation of new injectable medications, are often

required during hospitalisation, and good glycaemic control is important during admission and post-discharge. Patients must be actively involved in changes to diabetes management,

and require education regarding therapy changes and administration of injectable medications.

Effective education of patients can be difficult in the hospital setting, particularly in the

setting of concurrent acute illness. It is subject to a number of limitations including acute

Page 5 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



illness, pain and a sense of being overwhelmed in hospital. These clinical observations

contributed to the formulation of our research questions, however there was no direct patient

involvment in this.

Patients were not involved in the original study design, and involvement of patients in

recruitment was impractical as patients were hospitalised at the time of recruitment. However, during participation in the trial some intervention group patients expressed

difficulty keeping their hospital appointment for endocrinologist follow-up. In response, a

protocol for following up these patients using “telehealth” was established towards the end of the study. In this respect, patients had an influence on the study design and conduct of the

study. A qualitative analysis of these patients’ experiences and their perceptions of the

intervention was performed, and this will be reported separately.

Following publication of our findings, the paper outlining study results will be sent to the

study participants.

Interventions

Standard Care

Participants randomised to standard care were educated by hospital credentialed diabetes

educators (CDEs) prior to discharge. This comprised education regarding injectable therapy, storage, injection technique, and sharps disposal, and provision of additional resources when

required. Additional resources included National Diabetes Services Scheme (NDSS)

registration, supply of glucometer if required, written patient information regarding diabetes,

and outpatient follow-up. Participants were discharged when medically appropriate and the

inpatient team were satisfied that the participant could safely administer the injectable

therapy. Prior to discharge, appropriate follow-up was organised. General Practitioners (GPs)

were notified that participants had commenced or changed treatment.

Transition Team (intervention)

The transition team group participants received in-home education to start injectable therapy by the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided

with an appropriate glucometer in addition to education regarding medication, storage,

injection technique, sharps disposal, NDSS registration, an education package in the relevant language, and CDE contact details.

Further contact with participants was based on CDE evaluation of the participant’s capacity to

self-manage injectable therapy. Once the CDE ascertained that participants were able to self-manage without further intervention, the endocrinologist was notified. Participants were then

linked with community CDE services, if necessary, for ongoing monitoring of self-

management.

Follow-up with the same endocrinologist was provided within four weeks and at 16 weeks

post-randomization. HbA1c was assessed at baseline and 16 weeks. The endocrinologist

liaised with the participant’s GP regarding changes to management and plans for ongoing

follow-up after the 16-week visit.

Data collection

Baseline demographic, medication and medical data were collected and participants

completed the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at

enrolment. Follow-up data were collected at 16 weeks post-randomization, including rates of hospital readmissions and emergency presentations, length of hospital stay (LOS), glycaemic

control as measured by HbA1c and treatment satisfaction using the DTSQ change version

(DTSQc). Initial attempts to collect DTSQc questionnaires in the standard care group was by

reply-paid mail, however after a limited response rate using this method, patients were

Page 6 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



contacted by phone. Study participation ceased at the 16-week endocrinology appointment,

and final data collection occurred. Further follow-up for ongoing diabetes management was

arranged at conclusion of the study.

Ethics

Ethics approval was obtained from the Austin hopsital human ethics committee and the

community healthcare provider ethics committee. Each participant provided written informed

consent.

Outcomes

The primary outcome measured was feasibility (proportion of participants in the transition

team group completing the intervention as per protocol). Secondary outcomes were safety, as

defined by hospital readmission and emergency department presentations within 16 weeks

post-randomization, and patient satisfaction with care (measured by DTSQ). Exploratory

outcomes were change in HbA1c and length of hospital stay.

Sample size determination

Due to the pilot nature of the study, the sample size estimation was based on precision arguments: assuming the feasibility of transition team intervention being 0.9 (i.e. that 90% of

participants randomized into transition team group would be able to complete the intervention

as per protocol), the sample of 55 participants randomized to the transition team group

provides the precision (desired half-width of the 95% confidence interval) of 0.08.

The same number of participants was to be randomized to the standard care group, thus

ensuring 80% power to detect potential medium-to-large effects of transition team

intervention compared to the standard care (Cohen’s d=0.55) assuming the settings of

alpha=0.05. Thus, the total sample size for this study was proposed as 110 participants.

Statistical methods

The demographic and clinical characteristics of participants were summarized as medians

(interquartile ranges, IQRs) for continuous variables and counts (proportions) for categorical variables.

The feasibility of the intervention was estimated as a proportion of participants in the

transition team group completing the intervention as per protocol with corresponding 95% confidence interval (95% CI).

The difference in safety profiles (diabetes related hospital presentation or admission) between

two groups was investigated using Fisher's exact test.

DTSQ outcomes were analyzed using Wilcoxon-Mann-Whitney test and a median regression

model with the DTSQ score at 16 weeks post-randomization as an output and treatment group

and baseline DTSQ score as inputs. Sensitivity analysis was conducted by including the

auxillary variables demonstrating significant association with the DTSQ data being missing,

into the median regression model.

Differences in change in HbA1c and LOS between groups were investigated using Wilcoxon rank-sum test.

Statistical analysis was performed using STATA software (StataCorp, College Station, TX,

USA).

All statistical tests were two sided and were performed at a significance level of α = 0.05.

Page 7 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Statistical analysis was performed on both intention to treat and per protocol bases. Per

protocol analysis was deemed necessary to account for the patients who were initially

assigned to transition team but withdrew prior to intervention; in this situation they received

standard care.

Page 8 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Results

One hundred and five participants were randomised to transition team or standard care.

Following initial drop outs, 103 participants received the intervention or standard care (Figure

1). 55 participants randomised to the transition team and 48 participants randomised to

receive standard care were included in the per protocol analysis (Figure 1). Participants randomised to the transition team withdrew for a variety of reasons such as decisions to

change treatment, change of discharge destination, and changes to clinical status. For the

purposes of the per protocol analysis, these participants crossed over to the standard care group.

Five participants withdrew following the home visit by the CDE but prior to completing the

endocrinologist component of the intervention as they did not wish to participate.

One participant in the transition team group was unable to be followed up as they were being

actively palliated for terminal malignancy, and one died before completion of the trial, for

reasons unrelated to diabetes. One participant withdrew from the standard care group and one

died before final data analysis in this group.

Baseline characteristics of participants are outlined in Table 1. No differences between groups

at baseline were identified

Feasibility

Forty seven out of 55 participants in the TDT group completed the study as per protocol

(85%, 95%CI: 73%, 94%).

Safety

There was one hospital presentation in each group (Table 2), one for hypoglycaemia (standard

care) and one for inability to cope with insulin management at home due to change in social

circumstances (transition team). Neither participant was admitted.

Patient satisfaction

A significant improvement in satisfaction with diabetes treatment was demonstrated (DTSQc transition team median 15 (IQR 10.0, 17.5), standard care median 10.5 (IQR 8.5,16.0)

Wilcoxon-Mann-Whitney, p=0.047), Figure 2, Table 2. On analysis adjusted for the baseline

DTSQs value, the transition team median DTSQc value was 4 points higher than the standard care median (95%CI: 0.25, 7.75; p=0.037). The follow-up DTSQc was completed by 40

(16.7% missing) of the standard care patients and 36 (34.5% missing) of the transition team

patients, thus the “missingness” of the data was not likely to be random. On the sensitivity analysis adjusted for the variables significantly associated with the missing DTSQc data at 16

weeks, the results remained qualitatively similar.

HbA1c

No statistically significant difference in change in HbA1c (standard care median HbA1c -

1.5% (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -

0.2%), p = 0.83) was observed, Figure 3, Table 2.

Length of Stay

There was a trend towards a reduced length of stay in the transition team group when

analysed on a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team median 6 (IQR 3.0-12.0), p=0.06), Figure 4, Table 3.

Page 9 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Discussion

Key findings

The most important finding in this trial was that a transition team to initiate injectable diabetes therapies following discharge is a feasible model of care. These data suggest that a

transition team is safe and acceptable with a trend towards reduced length of hospital stay.

Moreover, patients randomized to the transition team group had greater treatment satisfaction as demonstrated by a greater difference in DTSQc score.

Relationship with previous studies

The results pertaining to the quality of this intervention are supported by results from other

studies examining home-based care in diabetes. The quality markers of this intervention

include feasibility and objective measures of medical indices, including readmission and

emergency presentation rates, change in HbA1c from baseline and length of stay, and patient-

reported outcomes.

We demonstrated a trend towards (p=0.06) reduced length of stay in the transition team when

analysed on a per protocol basis. Future studies with greater numbers may demonstrate

statistically significant reductions in length of hospital stay.

We demonstrated significant improvements in treatment satisfaction in the transition team

group. The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical

trials and validated in several languages. The status version (DTSQs) evaluates baseline

satisfaction with diabetes treatment and the change version (DTSQc) evaluates the impact of an intervention on satisfaction with treatment (14, 15, 16). Interpretability of this parameter is

limited by missing data, discussed below.

Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with

HbA1c reduction of over 2% in both groups. Importantly, there was no difference detected

between the reduction in HbA1c in the transition team and the standard care group.

Other quality outcomes assessed included emergency department presentations and hospital

readmissions. There were no significant differences in our study. This suggests that patient

safety is unlikely to be compromised by delivery of home-based education. However, given

the limited literature in this field, further studies with greater numbers would be necessary to

validate these findings.

Strengths and limitations of this study:

A strength of this study was the successful collaboration of two large organisations providing support for the transition of care from hospital to home. Daily challenges in coordinating

team members across organisations to achieve the same goals was evident throughout the

trial. Despite this, the current study recruited a large number of patients.

We acknowledge the limitations of the study. Eight patients in each group did not have

follow-up HbA1c measurements, despite active encouragement.

The number of missing DTSQc questionnaires may have limited interpretability of DTSQc

results. The follow-up DTSQc was completed by 40 (16.7% missing) of the standard care

Page 10 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



patients and 36 (34.5% missing) of the transition team patients, despite the best efforts of

investigators to obtain completed DTSQc questionnaires. We believe the non-random nature

of missing data was largely due to factors related to the transition team group, particularly

difficulties involved in attending appointments at hospital. However, it may be that patients

failing to complete the DTSQc were less satisfied with treatment, suggesting that this was more likely in the transition team group. A significantly higher number of male participants

failed to complete the DTSQc, and missing data were higher in those patients with

hypertension and dyslipidaemia. The difficulty in obtaining this data raises concerns about the feasibility of using this method to assess treatment satisfaction in patients recently

hospitalised with acute illness.

Withdrawal of participants from the transition team in hospital prior to the intervention may

reflect reluctance in hospitalized patients to accept health providers entering their home,

which may be a limitation of this model.

Study implications

Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to

improve outcomes over the course of disease. Escalating rates of diabetes necessitates the

development of feasible alternative models of care, with the potential to improve clinical

outcomes and health care costs. This study has investigated one such option, and has

demonstrated feasibility, improved treatment satisfaction, and a trend to reduced length of

stay, with no safety concerns detected.

Conclusion

The results of this novel pilot study suggest that use of a transition team provides a feasible

alternative model of care for patients with type 2 diabetes requiring initiation of injectable therapies. More research is necessary to validate these findings in larger populations, and to

ascertain whether it may lead to reduced length of stay and healthcare costs.

Funding

The Lord Mayor’s Charitable Foundation and the Estate of the Late Glen W A Griffiths who

funded this project. A/Prof Ekinci was supported by Australian National Health and Medical

Research Council (NHMRC) Early Career Fellowship, Viertel Clinical Investigatorship,

Royal Australasian College of Physicians (RACP) Fellowship and Sir Edward Weary Dunlop

Medical Research Foundation research grant.

Acknowledgements

The Diabetes Treatment Satisfaction Questionnaire (DTSQ) used in this publication is owned by Prof Clare Bradley) and sourced from HPR Ltd.

The team gratefully acknowledges Mr Paul Steel, CDE, for providing the in-home diabetes education for the intervention group.

Page 11 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Competing Interests statement

We have read and understood BMJ policy on declaration of interests and declare that we have

no competing interests.

Author Contributions

F Pyrlis - involved in development of trial, clinical management of participants, performed

final write up

R Ogrin - involved in development of protocol, supervision of trial, reviewed final write up

S Arthur - recruitment of participants and management of logistics of trial (research assistant),

data collection and analysis

B Zhai - logistics of trial (research assistant), final data collection and statistical analysis

S Baqar - recruitment of participants, data collection and analysis

L Churilov - statistical analysis

JD Zajac - supervisory/ advisory role

EI Ekinci - development of protocol, supervisory and advisory role during trial, assisted with

statistical analysis and final write up

Data Statement: Dataset is restricted but can be provided upon reasonable request.

Page 12 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



References

1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010.

http://www.aihw.gov.au/diabetes-indicators/deaths)

2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes

prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 103:137-149

3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children

and young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. No. CVD 53; Diabetes Series No 21,)

http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359

4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient

HbA1c testing: a prospective observational study. BMJ open diabetes research &

care. 2015;3(1):e000113

5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for

coronary artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9

6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and

new hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3

7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of

diabetes self-management education in the hospital setting. Primary Care Diabetes.

2014;8(3):187-94.

8. Schafer I, Pawels M, Kuver C, Pohontsch NJ, Scherer M, Bussche Hv et al.

Strategies for Improving Participation in Diabetes Education. A Qualitative Study.

PLoS One. 2014; 9(4)

9. Joint British Diabetes Societies for Inpatient Care. Discharge planning for adult

inpatients with diabetes. October 2015

10. Wexler DJ, Beauharnais CC, Regan S, Nathan DM, Cagliero E, Larkin ME. Impact

of inpatient diabetes management, education, and improved discharge transition on

glycemic control 12 months after discharge. Diabetes Research and Clinical Practice.

2012;98:249-56

11. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of

diabetes self-management education in the hospital setting. Primary Care Diabetes.

2014;8(3):187-94.

12. de Carvalho Torres H, dos Santos LM, de Souza Cordeiro PMC. Home visit: an

educational health strategy for self-care in diabetes. Visita domiciliária: estratégia

educativa em saúde para o autocuidado em diabetes. 2014;27(1):23

13. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0

(Standards for QUality Improvement Reporting Excellence): revised publication

guidelines from a detailed consensus process

14. Bradley C, Lewis KS. Measures of psychological well-being and treatment

satisfaction developed from the responses of people with tablet-treated diabetes.

Diabetic Medicine. 1990; 7:445-451.

15. Bradley C, Speight J. Patient perceptions of diabetes and diabetes therapy: assessing quality of life. Diabetes Metabolism Research and Reviews. 2002; 18: S64-S69

16. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change

version for use alongside status version provides appropriate solution where ceiling

effects occur. Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J,

Valentine J and Witthaus E (2007) The Diabetes Treatment Satisfaction

Questionnaire change version (DTSQc) evaluated in insulin glargine trials shows greater responsiveness to improvements than the original DTSQ. Health and Quality

of Life Outcomes. 1999; 5 (5) 57

Page 13 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 1 : Recruitment and participation flowchart

Page 14 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Table 1: Comparison of baseline clinical and biochemical characteristics between in the

control and TDT groups by initial randomization/ intention to treat)

Characteristic Control group

(n=48)

Transitions group

(n=55)

Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (34.4) 22 (61.1)

Median type 2 diabetes (SD) 7.48 (7.78) 10.18 (9.77)

Mean HbA1c baseline (%, SD)

mmol/mol (mmol/mol, SD)

9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)

*CALD = culturally and linguistically diverse

# two-sample Wilcoxon rank-sum (Mann-Whitney) test used for continuous variables, and

Fisher Exact tests used for binary variables

Table 2: Results by intention to treat analysis. Differences in hospital readmission rates,

change in HbA1c, length of hospital stay, & treatment satisfaction according to original randomization.

Control group (n=48) Transitions group

(n=55)

P-value

Readmission (count (%)) 1/48 (2%) 1/55 (2%) >0.99

DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047

Change HbA1c (median,

IQR)

-1.5% (-3.7, -0.2%) -1.9% (-3.8, -

0.2%)

0.83

Length of stay (days)

(median, IQR)

8 (5.5,11.5) 7 (3,12) 0.26

Page 15 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Table 3: Results by per Protocol analysis. Differences in hospital readmission rates, change in HbA1c, length of hospital stay, & treatment satisfaction according to management received.

Control group (n=56) Transitions group

(n=47)

P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047

Change HbA1c (median,

IQR)

-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85

Length of stay (days)

(median, IQR)

8 (5.5, 12) 6 (3, 12) 0.06

Figure 2: Change in treatment satisfaction (DTSQ)

Figure 3: A. Change in HbA1c (%) intention to treat analysis and B. Change in HbA1c

(%) per protocol analysis

Figure 4: A. Length of hospital stay (days) by intention to treat analysis and B. Length

of hospital stay (days) per protocol analysis

Page 16 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 17 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Reporting checklist for quality improvement study.

Based on the SQUIRE guidelines.

Instructions to authors

Complete this checklist by entering the page numbers from your manuscript where readers will find

each of the items listed below.

Your article may not currently address all the items on the checklist. Please modify your text to

include the missing information. If you are certain that an item does not apply, please write "n/a" and

provide a short explanation.

Upload your completed checklist as an extra file when you submit to a journal.

In your methods section, say that you used the SQUIRE reporting guidelines, and cite them as:

Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for

QUality Improvement Reporting Excellence): revised publication guidelines from a detailed

consensus process

Reporting Item

Page

Number

#1 Indicate that the manuscript concerns an initiative to improve

healthcare (broadly defined to include the quality, safety,

effectiveness, patientcenteredness, timeliness, cost, efficiency,

and equity of healthcare)

1

#02a Provide adequate information to aid in searching and indexing 2

#02b Summarize all key information from various sections of the text

using the abstract format of the intended publication or a

structured summary such as: background, local problem,

methods, interventions, results, conclusions

2

Problem

description

#3 Nature and significance of the local problem 4

Available

knowledge

#4 Summary of what is currently known about the problem,

including relevant previous studies

4

Rationale #5 Informal or formal frameworks, models, concepts, and / or

theories used to explain the problem, any reasons or

4

Page 18 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



assumptions that were used to develop the intervention(s), and

reasons why the intervention(s) was expected to work

Specific aims #6 Purpose of the project and of this report 4

Context #7 Contextual elements considered important at the outset of

introducing the intervention(s)

5

Intervention(s) #08a Description of the intervention(s) in sufficient detail that others

could reproduce it

5-6

#08b Specifics of the team involved in the work 5-6

Study of the

Intervention(s)

#09a Approach chosen for assessing the impact of the intervention(s) 6-7

#09b Approach used to establish whether the observed outcomes

were due to the intervention(s)

6-7

Measures #10a Measures chosen for studying processes and outcomes of the

intervention(s), including rationale for choosing them, their

operational definitions, and their validity and reliability

6-7

#10b Description of the approach to the ongoing assessment of

contextual elements that contributed to the success, failure,

efficiency, and cost

7

#10c Methods employed for assessing completeness and accuracy

of data

7

Analysis #11a Qualitative and quantitative methods used to draw inferences

from the data

7

#11b Methods for understanding variation within the data, including

the effects of time as a variable

7

Ethical

considerations

#12 Ethical aspects of implementing and studying the intervention(s)

and how they were addressed, including, but not limited to,

formal ethics review and potential conflict(s) of interest

6

#13a Initial steps of the intervention(s) and their evolution over time

(e.g., time-line diagram, flow chart, or table), including

modifications made to the intervention during the project

8.10

#13b Details of the process measures and outcome 8.10

Page 19 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



#13c Contextual elements that interacted with the intervention(s) 10

#13d Observed associations between outcomes, interventions, and

relevant contextual elements

10

#13e Unintended consequences such as unexpected benefits,

problems, failures, or costs associated with the intervention(s).

10

#13f Details about missing data 9

Summary #14a Key findings, including relevance to the rationale and specific

aims

9

#14b Particular strengths of the project 9

Interpretation #15a Nature of the association between the intervention(s) and the

outcomes

9

#15b Comparison of results with findings from other publications 9

#15c Impact of the project on people and systems 10

#15d Reasons for any differences between observed and anticipated

outcomes, including the influence of context

9-10

#15e Costs and strategic trade-offs, including opportunity costs 10

Limitations #16a Limits to the generalizability of the work 9-10

#16b Factors that might have limited internal validity such as

confounding, bias, or imprecision in the design, methods,

measurement, or analysis

9-10

#16c Efforts made to minimize and adjust for limitations 9-10

Conclusion #17a Usefulness of the work 10

#17b Sustainability 10

#17c Potential for spread to other contexts 10

#17d Implications for practice and for further study in the field 10

#17e Suggested next steps 10

Funding #18 Sources of funding that supported this work. Role, if any, of the

funding organization in the design, implementation,

10

Page 20 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



interpretation, and reporting

The SQUIRE 2.0 checklist is distributed under the terms of the Creative Commons Attribution License

CC BY-NC 4.0. This checklist was completed on 12. April 2018 using http://www.goodreports.org/, a

tool made by the EQUATOR Network in collaboration with Penelope.ai

Page 21 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 1 : Recruitment and participation flowchart

373x341mm (150 x 150 DPI)

Page 22 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




401x555mm (96 x 96 DPI)

Page 23 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 3: A. Change in HbA1c (%) intention to treat analysis and B. Change in HbA1c (%) per protocol analysis

725x555mm (96 x 96 DPI)

Page 24 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 4: A. Length of hospital stay (days) by intention to treat analysis and B. Length of hospital stay (days) per protocol analysis

794x555mm (96 x 96 DPI)

Page 25 of 25


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from


For peer review onlyFeasibility of using a transition diabetes team to commence injectable therapies post discharge from a tertiary hospital

Journal: BMJ Open

Manuscript ID bmjopen-2018-023583.R1

Article Type: Original research

Date Submitted by the Author: 24-Oct-2018

Complete List of Authors: Pyrlis, Felicity; Austin Health, EndocrinologyOgrin, Rajna; University of Melbourne, Medicine; Bolton Clarke Research InstituteArthur, Sonja; University of Melbourne, MedicineZhai, Cathy; University of Melbourne, MedicineChurilov, Leonid; University of Melbourne, MedicineBaqar, Sara; Austin Health, EndocrinologyZajac, Jeffrey; Austin Health, Endocrinology; University of Melbourne, MedicineEkinci , Elif ; Austin Health, Endocrinology; University of Melbourne, Medicine

<b>Primary Subject Heading</b>: Diabetes and endocrinology

Secondary Subject Heading: Patient-centred medicine



BMJ Open on O


http://bmjopen.bm

j.com/

BM



ber 2019. Dow

nloaded from



1 Feasibility of using a transition diabetes team to commence injectable therapies post discharge 2 from a tertiary hospital

3

4 Pyrlis, F1, Ogrin, R2,3, Arthur, S³, Zhai, C3, Churilov, L³, Baqar, S1, Zajac, J D1,3, Ekinci, E. I. 1,3

5 1 Endocrinology Department, Austin Health, Heidelberg, Australia

6 2 Royal District Nursing Service, Melbourne, Australia

7 3Department of Medicine Austin Health, University of Melbourne, Melbourne, Australia

8

9 Corresponding Author: 10 Dr Felicity Pyrlis11 Endocrinology Department, Austin Health12 300 Waterdale Road, Heidelberg, VIC 308113 Australia 14 Email: [email protected]

15

16 Word count: 2426

17 Keywords: diabetes, inpatient, insulin, injectable, discharge

18 Ethical Approval: HREC Austin Health, Victoria, Australia (LNR_13_Austin_179)

1920

Page 1 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



21 Article Summary:

2223 This study aimed to investigate if the use of a transition team was feasible for patients 24 with diabetes being discharged from hospital on injectable diabetes therapies.25 This pilot, randomised controlled trial was conducted between 2014 and 2016 conjointly 26 by a tertiary referral hospital and a community healthcare provider.27 Hospital inpatients (n=105) on new injectable diabetes therapies were randomised to 28 transition team (receiving in-home diabetes education 24-48 hours post-discharge, with 29 endocrinologist review 2-4 weeks and 16 weeks post-discharge) or standard care.30 The primary outcome was feasibility, defined by percentage of patients successfully 31 receiving the intervention.32 This study demonstrated that the use of a novel transition diabetes team is a feasible 33 alternative model of care.

34

35 Strengths and limitations of the study:

36 A strength of this study was the successful collaboration of two large organisations providing 37 support for the transition of care from hospital to home.

38 We acknowledge the limitations of the study.

39 Eight patients in each group did not have follow-up HbA1c measurements, despite active 40 encouragement.

41 The follow-up DTSQc was completed by 40 (16.7% missing) of the standard care patients and 36 42 (34.5% missing) of the transition team patients,and the number of missing DTSQc questionnaires 43 may have limited interpretability of DTSQc results. .

44 Withdrawal of participants from the transition team in hospital prior to the intervention may 45 reflect reluctance in hospitalized patients to accept health providers entering their home, which 46 may be a limitation of this model.

47

48

49

50

Page 2 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



52 Abstract

53 Objectives: This study aimed to investigate if the use of a transition team was feasible for patients 54 with diabetes being discharged from hospital on injectable diabetes therapies.

55

56 Design, Setting, Participants: This pilot, randomised controlled trial was conducted between 2014 57 and 2016 conjointly by a tertiary referral hospital and a community healthcare provider. Hospital 58 inpatients (n=105) on new injectable diabetes therapies were randomised 1:1 to transition team or 59 standard care. The transition team received in-home diabetes education 24-48 hours post-60 discharge, with endocrinologist review 2-4 weeks and 16 weeks post-discharge.

61

62 Main outcome measures: The primary outcome was feasibility, defined by percentage of patients 63 successfully receiving the intervention. Secondary outcomes included safety, defined by hospital 64 readmission and emergency department presentations within 16 weeks post-randomization, and 65 treatment satisfaction, measured using Diabetes Treatment Satisfaction Questionnaire (DTSQ). 66 Exploratory outcomes included length of stay (LOS), and change in HbA1c throughout the study.67

68 Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in 69 safety between groups was detected. No difference in change in HbA1c between groups was 70 detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team median 71 HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a significant improvement in patient 72 satisfaction in the transition team (standard care median 10.5 (IQR 8.5, 16); transition team 73 DTSQc median 15 (IQR 10, 17.5), p=0.047), although interpretability is limited by missing data.

74 Conclusion: This study demonstrated that the use of a novel transition diabetes team is a feasible 75 alternative model of care.

76

77

78

79

Page 3 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



80 Introduction

81 Despite stability in diabetes-related mortality (1), global prevalence of diabetes is rising (2). 82 Consequently, health care costs related to diabetes continue to increase over time (3).

83 It has been demonstrated that 34% of hospital inpatients aged over 54 attending a tertiary hospital 84 have diabetes mellitus (4). Many studies have demonstrated that inpatients with type 2 diabetes 85 have longer hospital length of stay and higher mortality rates compared to those without (5, 6). 86 Factors such as stress hyperglycaemia, medications, and inadequate glycaemic control at the time 87 of hospital admission, often result in the need for intensification with injectable diabetes therapies 88 (7). Furthermore, the hospitalisation period provides an opportunity to identify those with poor 89 glycaemic control and optimise diabetes management (8). However, commencement of 90 injectable therapies can be difficult in the context of concurrent acute illness.

91 Guidelines recommend that patients are required to demonstrate self-management with injectable 92 therapies prior to hospital discharge (9). Diabetes education is crucial in enabling patients to 93 effectively self-manage, and assists in optimising glycaemic control post-discharge (10). 94 However, diabetes education in the hospital setting is subject to a number of limitations including 95 acute illness, pain and a sense of being overwhelmed (11). Home-based diabetes education may 96 prove more effective based on understanding a patient’s life context and allowing adaptation of 97 self-management routines such as timing of injection, sharps disposal and medication storage to 98 suit the patient’s home environment (12).

99 This pilot study developed and evaluated the use of a transition team comprising in-home 100 diabetes education by a credentialed diabetes educator (CDE), and early post-discharge 101 assessment by an endocrinologist. We hypothesized that the proposed intervention would be 102 feasible and would not negatively affect patient satisfaction when compared to standard care.

103

104

105 Methods

106 Design

107 This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in 108 metropolitan Melbourne and a community-based healthcare provider. Study participants were 109 recruited during inpatient admissions between March 2014 and November 2015 and follow-up 110 continued until March 2016. The trial was stopped after funding was exhausted and sufficient 111 participants were recruited.

112 Participants

113 Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, 114 were screened for the study, and randomised to receive the intervention or standard care after 115 providing informed consent. Participant inclusion criteria were the ability to provide informed 116 consent, the presence of type 2 diabetes, age > 18 years, requirement to commence or change 117 injectable therapies, therefore requiring a credentialed diabetes educator (CDE) to provide 118 education prior to discharge, medically stable and awaiting diabetes education, reside within a 119 30-minute travel radius of the hospital, ability to attend hospital for outpatient follow-up and 120 stable glycaemia defined as blood glucose levels between 5-15 mmol/L in the 24 hours prior to 121 randomisation. 122

Page 4 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



123 Randomisation

124 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 125 investigator without patient contact. Group allocations were concealed by writing allocations on 126 a card, and placing in sealed, unlabelled envelopes, with each consecutive participant given their 127 allocation by a research assistant after informed consent was obtained. Due to the difference in 128 treatment protocols, the study was open label to the participants and investigators.

129 Intervention

130 Standard Care

131 Participants randomised to standard care were educated by hospital credentialed diabetes 132 educators (CDEs) prior to discharge. Diabetes education regarding injectable therapies in our 133 institution complies with guidelines of the Australian Diabetes Educators Association (13). This 134 comprised education regarding injectable therapy, storage, injection technique, and sharps 135 disposal, and provision of additional resources when required. Additional resources included 136 National Diabetes Services Scheme (NDSS) registration, supply of glucometer if required, 137 written patient information regarding diabetes, and outpatient follow-up. Participants were 138 discharged when medically appropriate and the inpatient team were satisfied that the participant 139 could safely administer the injectable therapy. Prior to discharge, appropriate follow-up was 140 organised. General Practitioners (GPs) were notified that participants had commenced or changed 141 treatment.

142 Transition Team (intervention)

143 The transition team group participants received in-home education to start injectable therapy by 144 the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided with 145 an appropriate glucometer in addition to education regarding medication, storage, injection 146 technique, sharps disposal, NDSS registration, an education package in the relevant language, and 147 CDE contact details.

148 Further contact with participants was based on CDE evaluation of the participant’s capacity to 149 self-manage injectable therapy. Once the CDE ascertained that participants were able to self-150 manage without further intervention, the endocrinologist was notified. Participants were then 151 linked with community CDE services, if necessary, for ongoing monitoring of self-management.

152 Follow-up with the same endocrinologist was provided within four weeks, and at 16 weeks post-153 randomization. HbA1c was assessed at baseline and 16 weeks. The endocrinologist liaised with 154 the participant’s GP regarding changes to management and plans for ongoing follow-up after the 155 16-week visit.

156 Data collection

157 Baseline demographic, medication and medical data were collected and participants completed 158 the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at enrolment. Whether 159 the patient was from a culturally and linguistically diverse (CALD) background was recorded. 160 Follow-up data were collected at 16 weeks post-randomization, including rates of hospital 161 readmissions and emergency presentations, length of hospital stay (LOS), glycaemic control as 162 measured by HbA1c and treatment satisfaction using the DTSQ change version (DTSQc). Initial 163 attempts to collect DTSQc questionnaires in the standard care group was by reply-paid mail, 164 however after a limited response rate using this method, patients were contacted by phone. 165 Study participation ceased at the 16-week endocrinology appointment, and final data collection 166 occurred. Further follow-up for ongoing diabetes management was arranged at conclusion of the 167 study.

Page 5 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



168

169 Ethics

170 Ethics approval was obtained from the hospital human ethics committee and the community 171 healthcare provider ethics committee. Each participant provided written informed consent.

172

173 Outcomes

174 The primary outcome measured was feasibility (proportion of participants in the transition team 175 group completing the intervention as per protocol). Secondary outcomes were safety, as defined 176 by hospital readmission and emergency department presentations within 16 weeks post-177 randomization, and patient satisfaction with care (measured by DTSQ). Exploratory outcomes 178 were change in HbA1c and length of hospital stay (days).

179

180 Sample size determination

181 Due to the pilot nature of the study, the sample size estimation was based on precision arguments: 182 assuming the feasibility of transition team intervention being 0.9 (i.e. that 90% of participants 183 randomized into transition team group would be able to complete the intervention as per 184 protocol), the sample of 55 participants randomized to the transition team group provides the 185 precision (desired half-width of the 95% confidence interval) of 0.08.

186 The same number of participants was to be randomized to the standard care group, thus ensuring 187 80% power to detect potential medium-to-large effects of transition team intervention compared 188 to the standard care (Cohen’s d=0.55) assuming the settings of alpha=0.05. Thus, the total sample 189 size for this study was proposed as 110 participants.

190

191 Statistical methods

192 The demographic and clinical characteristics of participants were summarized as medians 193 (interquartile ranges, IQRs) for continuous variables and counts (proportions) for categorical 194 variables.

195 The feasibility of the intervention was estimated as a proportion of participants in the transition 196 team group completing the intervention as per protocol with corresponding 95% confidence 197 interval (95% CI).

198 The difference in safety profiles (diabetes related hospital presentation or admission) between two 199 groups was investigated using Fisher's exact test.

200 DTSQ outcomes were analyzed using Wilcoxon-Mann-Whitney test and a median regression 201 model with the DTSQ score at 16 weeks post-randomization as an output and treatment group 202 and baseline DTSQ score as inputs. Sensitivity analysis was conducted by including the auxillary 203 variables demonstrating significant association with the DTSQ data being missing, into the 204 median regression model.

205 Differences in change in HbA1c and LOS between groups were investigated using Wilcoxon 206 rank-sum test.

207 Statistical analysis was performed using STATA software (StataCorp, College Station, TX, 208 USA).

209 All statistical tests were two sided and were performed at a significance level of α = 0.05.

Page 6 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



210 Statistical analysis was performed on both intention to treat and per protocol bases. Per protocol 211 analysis was deemed necessary to account for the patients who were initially assigned to 212 transition team but withdrew prior to intervention; in this situation they received standard care.

213 Patient and public involvement

214 Patients must be actively involved in changes to diabetes management, and require education 215 regarding therapy changes and administration of injectable medications. Effective education of 216 patients can be difficult in the hospital setting, particularly in the setting of concurrent acute 217 illness. It is subject to a number of limitations including acute illness, pain and a sense of being 218 overwhelmed in hospital. These clinical observations contributed to the formulation of our 219 research questions, however there was no direct patient involvment in this.

220 Patients were not involved in the original study design, and involvement of patients in 221 recruitment was impractical as patients were hospitalised at the time of recruitment. A qualitative 222 analysis of a subgroup of patients’ experiences and their perceptions of the intervention was 223 performed, and this will be reported separately.

224 Following publication of our findings, the paper outlining study results will be sent to the study 225 participants.

226

227

Page 7 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



229 Results

230 One hundred and five participants were randomised to transition team or standard care. Following 231 initial drop outs, 103 participants received the intervention or standard care (Figure 1). 55 232 participants randomised to the transition team and 48 participants randomised to receive standard 233 care were included in the per protocol analysis (figure 1). Participants randomised to the 234 transition team withdrew for a variety of reasons such as decisions to change treatment, change of 235 discharge destination, and changes to clinical status. Of these, five participants withdrew 236 following the home visit by the CDE but prior to completing the endocrinologist component of 237 the intervention. For the purposes of the per protocol analysis, these participants crossed over to 238 the standard care group.

239 One participant in the transition team group was unable to be followed up as they were being 240 actively palliated for terminal malignancy, and one died before completion of the trial, for reasons 241 unrelated to diabetes. One participant withdrew from the standard care group and one died before 242 final data analysis in this group.

243 Baseline characteristics of participants are outlined in Table 1. No differences between groups at 244 baseline were identified.

245

246 Feasibility

247 Forty seven out of 55 participants in the transition team group completed the study as per protocol 248 (85%, 95%CI: 73%, 94%).

249 Safety

250 There was one hospital presentation in each group (Table 2), one for hypoglycaemia (standard 251 care) and one for inability to cope with insulin management at home due to change in social 252 circumstances (transition team). Neither participant was admitted.

253 Patient satisfaction

254 A significant improvement in satisfaction with diabetes treatment was demonstrated (DTSQc 255 transition team median 15 (IQR 10.0, 17.5), standard care median 10.5 (IQR 8.5,16.0) Wilcoxon-256 Mann-Whitney, p=0.047), Figure 2, Table 2. On analysis adjusted for the baseline DTSQs value, 257 the transition team median DTSQc value was 4 points higher than the standard care median 258 (95%CI: 0.25, 7.75; p=0.037). The follow-up DTSQc was completed by 40 (16.7% missing) of 259 the standard care patients and 36 (34.5% missing) of the transition team patients, thus the 260 “missingness” of the data was not likely to be random. On the sensitivity analysis adjusted for the 261 variables significantly associated with the missing DTSQc data at 16 weeks, the results remained 262 qualitatively similar.

263 HbA1c

264 No statistically significant difference in change in HbA1c (standard care median HbA1c -1.5% 265 (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -0.2%), p = 266 0.83) was observed, Figure 3, Table 2. An equal number of patients (8) in each group did not 267 have an HbA1c measurement at 16 weeks.

268 Length of Stay

269 There was a trend towards a reduced length of stay in the transition team group when analysed on 270 a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team median 6 (IQR 271 3.0-12.0), p=0.06), Figure 4, Table 3.

272

Page 8 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



273

274

275 Discussion

276

277 Key findings

278 The most important finding in this trial was that a transition team to initiate injectable diabetes 279 therapies following discharge is a feasible model of care. These data suggest that a transition 280 team is safe and acceptable with a trend towards reduced length of hospital stay. Moreover, 281 patients randomized to the transition team group had greater treatment satisfaction as 282 demonstrated by a greater difference in DTSQc score.

283

284

285 Relationship with previous studies

286287 The results pertaining to the quality of this intervention are supported by results from other 288 studies examining home-based care in diabetes. The quality markers of this intervention include 289 feasibility and objective measures of medical indices, including readmission and emergency 290 presentation rates, change in HbA1c from baseline and length of stay, and patient-reported 291 outcomes. 292293 We demonstrated a trend towards (p=0.06) reduced length of stay in the transition team when 294 analysed on a per protocol basis. Future studies with greater numbers may demonstrate 295 statistically significant reductions in length of hospital stay.296

297 We demonstrated significant improvements in treatment satisfaction in the transition team group. 298 The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical trials and 299 validated in several languages. The status version (DTSQs) evaluates baseline satisfaction with 300 diabetes treatment and the change version (DTSQc) evaluates the impact of an intervention on 301 satisfaction with treatment (14, 15, 16). Interpretability of this parameter is limited by missing 302 data, and there is potential for bias due to some data being obtained over the phone by a research 303 assistant, rather than in person. 304305 Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with 306 HbA1c reduction approaching 2% in both groups. Importantly, there was no statistical difference 307 detected between the reduction in HbA1c in the transition team and the standard care group. 308309 Other quality outcomes assessed included emergency department presentations and hospital 310 readmissions. There were no significant differences in our study. This suggests that patient safety 311 is unlikely to be compromised by delivery of home-based education. However, given the limited 312 literature in this field, further studies with greater numbers would be necessary to validate these 313 findings.

314

315 Study implications

Page 9 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



316 Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to improve 317 outcomes over the course of disease. Escalating rates of diabetes necessitates the development of 318 feasible alternative models of care, with the potential to improve clinical outcomes and health 319 care costs. This study has investigated one such option, and has demonstrated feasibility, 320 improved treatment satisfaction, and a trend to reduced length of stay, with no safety concerns 321 detected.

322

323

324

325 Conclusion

326 The results of this novel pilot study suggest that use of a transition team provides a feasible 327 alternative model of care for patients with type 2 diabetes requiring initiation of injectable 328 therapies. More research is necessary to validate these findings in larger populations, and to 329 ascertain whether it may lead to reduced length of stay and healthcare costs.

330

331 Funding

332 The Lord Mayor’s Charitable Foundation and the Estate of the Late Glen W A Griffiths who 333 funded this project. A/Prof Ekinci was supported by Australian National Health and Medical 334 Research Council (NHMRC) Early Career Fellowship, Viertel Clinical Investigatorship, Royal 335 Australasian College of Physicians (RACP) Fellowship and Sir Edward Weary Dunlop Medical 336 Research Foundation research grant.

337

338 Acknowledgements

339 The Diabetes Treatment Satisfaction Questionnaire (DTSQ) used in this publication is owned by 340 Prof Clare Bradley) and sourced from HPR Ltd.341 The team gratefully acknowledges Mr Paul Steel, CDE, for providing the in-home diabetes 342 education for the intervention group.

343

344 Competing Interests statement

345 We have read and understood BMJ policy on declaration of interests and declare that we have no 346 competing interests.

347

348

349 Authors Statement

350 F Pyrlis - involved in development of trial, clinical management of participants, performed final 351 write up

352 R Ogrin - involved in development of protocol, supervision of trial, reviewed final write up

353 S Arthur - logistics of trial (research assistant), data collection and statistical analysis

354 B Zhai - logistics of trial (research assistant), final data collection and statistical analysis

Page 10 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



355 S Baqar - recruitment of participants and management of logistics of trial (research assistant), 356 data collection and analysis

357 L Churilov - statistical analysis

358 J Zajac - supervisory/ advisory role

359 E Ekinci - development of protocol, supervisory and advisory role during trial, assisted with 360 statistical analysis and final write up

361

362 Data Statement: Dataset is restricted but can be provided upon reasonable request

363

364 Figure Legend

365 Figure 1 – Recruitment and participation flowchart

366 Figure 2 – Change in treatment satisfaction (DTSQ)

367 Figure 3 – a) Change in HbA1c (%) intention to treat analysis b) Change in HbA1c (%) per 368 protocol analysis

369 Figure 4 – a) Length of hospital stay (days) by intention to treat analysis b) Length of hospital 370 stay (days) by per protocol analysis

371 Table 1 - Comparison of baseline clinical and biochemical characteristics between in the control 372 and transition team groups by initial randomization/ intention to treat

373 Table 2 - Results by intention to treat analysis. Differences in hospital readmission rates, change 374 in HbA1c, length of hospital stay, & treatment satisfaction according to original randomization.

375 Table 3 - Results by per Protocol analysis. Differences in hospital readmission rates, change in 376 HbA1c, length of hospital stay, & treatment satisfaction according to management received.377

378

Page 11 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



379 References380 1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010. 381 http://www.aihw.gov.au/diabetes-indicators/deaths)382 2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes prevalence 383 for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 103:137-149384 3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children and 385 young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. No. 386 CVD 53; Diabetes Series No 21,) 387 http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359388 4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient 389 HbA1c testing: a prospective observational study. BMJ open diabetes research & care. 390 2015;3(1):e000113391 5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for coronary 392 artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9393 6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and new 394 hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3395 7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of diabetes 396 self-management education in the hospital setting. Primary Care Diabetes. 397 2014;8(3):187-94.

398 8. Schafer I, Pawels M, Kuver C, Pohontsch NJ, Scherer M, Bussche Hv et al. Strategies 399 for Improving Participation in Diabetes Education. A Qualitative Study. PLoS One. 400 2014; 9(4)

401 9. Joint British Diabetes Societies for Inpatient Care. Discharge planning for adult inpatients 402 with diabetes. October 2015 403 10. Wexler DJ, Beauharnais CC, Regan S, Nathan DM, Cagliero E, Larkin ME. Impact of 404 inpatient diabetes management, education, and improved discharge transition on 405 glycemic control 12 months after discharge. Diabetes Research and Clinical Practice. 406 2012;98:249-56

407 11. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of diabetes 408 self-management education in the hospital setting. Primary Care Diabetes. 409 2014;8(3):187-94.

410 12. de Carvalho Torres H, dos Santos LM, de Souza Cordeiro PMC. Home visit: an 411 educational health strategy for self-care in diabetes. Visita domiciliária: estratégia 412 educativa em saúde para o autocuidado em diabetes. 2014;27(1):23

413 13. https://www.adea.com.au/wp-content/uploads/2015/11/Injection-Technique-Final-414 digital-version2.pdf

415 14. Bradley C, Lewis KS. Measures of psychological well-being and treatment satisfaction 416 developed from the responses of people with tablet-treated diabetes. Diabetic Medicine. 417 1990; 7:445-451.

418 15. Bradley C, Speight J. Patient perceptions of diabetes and diabetes therapy: assessing 419 quality of life. Diabetes Metabolism Research and Reviews. 2002; 18: S64-S69

420 16. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change version 421 for use alongside status version provides appropriate solution where ceiling effects occur. 422 Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J, Valentine J and Witthaus

Page 12 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from

http://www.aihw.gov.au/diabetes-indicators/deaths


https://www.adea.com.au/wp-content/uploads/2015/11/Injection-Technique-Final-digital-version2.pdf




423 E (2007) The Diabetes Treatment Satisfaction Questionnaire change version (DTSQc) 424 evaluated in insulin glargine trials shows greater responsiveness to improvements than 425 the original DTSQ. Health and Quality of Life Outcomes. 1999; 5 (5) 57426427428429

430

Page 13 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Table 1 Comparison of baseline clinical and biochemical characteristics between in the control and transition team groups by initial randomization/ intention to treat.

Characteristic Control group (n=48)

Transitions group (n=55)

Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (33.3%) 22 (40%)

Median type 2 diabetes (SD) 7.48 (7.78) 10.18 (9.77)



9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)


# two-sample Wilcoxon rank-sum (Mann-Whitney) test used for continuous variables, and Fisher Exact tests used for binary variables

Table 2. Results by intention to treat analysis. Differences in hospital readmission rates, change in HbA1c, length of hospital stay, & treatment satisfaction according to original randomization.

Control group (n=48) Transitions group (n=55)

P-value


DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047

Change HbA1c (median, IQR)

-1.5% (-3.7, -0.2%) -1.9% (-3.8, -0.2%)

0.83

Length of stay (days) (median, IQR)

8 (5.5,11.5) 7 (3,12) 0.26

Page 14 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from


For peer review onlyTable 3. Results by per Protocol analysis. Differences in hospital readmission rates, change in HbA1c, length of hospital stay, & treatment satisfaction according to management received.


P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047


-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85


8 (5.5, 12) 6 (3, 12) 0.06

Page 15 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 16 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 1: Recruitment and participation flowchart

Inpatients commencing injectable diabetes therapies

Intention to treat analysis Analysed (n=55)

6 patients did not have follow-up HbA1c 19 patients did not complete DTSQc

Transition team care (n=55) • Received allocated intervention (n=47) • Did not receive allocated intervention (n= 8)


Standard care (n= 50) • Received allocated care (n=48) • Withdrew prior to baseline data collection due

to changes in care during hospitalisation, so could not be included in analysis (n=2)


Randomised (n=105)

Per protocol analysis Analysed (n=47)


Participants who did not receive transition team intervention (n= 8)

Page 17 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




401x555mm (96 x 96 DPI)

Page 18 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




725x555mm (96 x 96 DPI)

Page 19 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 4: A. Length of hospital stay (days) by intention to treat analysis and B. Length of hospital stay (days) per protocol analysis

794x555mm (96 x 96 DPI)

Page 20 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



CONSORT 2010 checklist of information to include when reporting a pilot or feasibility trial*

Section/TopicItem No Checklist item

Reported on page No

Title and abstract1a Identification as a pilot or feasibility randomised trial in the title 11b Structured summary of pilot trial design, methods, results, and conclusions (for specific guidance see

CONSORT abstract extension for pilot trials)2,3

Introduction2a Scientific background and explanation of rationale for future definitive trial, and reasons for randomised pilot

trial4Background and

objectives2b Specific objectives or research questions for pilot trial 3, 4

Methods3a Description of pilot trial design (such as parallel, factorial) including allocation ratio 4Trial design3b Important changes to methods after pilot trial commencement (such as eligibility criteria), with reasons 54a Eligibility criteria for participants 4Participants4b Settings and locations where the data were collected 44c How participants were identified and consented 4

Interventions 5 The interventions for each group with sufficient details to allow replication, including how and when they were actually administered

5

6a Completely defined prespecified assessments or measurements to address each pilot trial objective specified in 2b, including how and when they were assessed

6Outcomes

6b Any changes to pilot trial assessments or measurements after the pilot trial commenced, with reasons n/a6c If applicable, prespecified criteria used to judge whether, or how, to proceed with future definitive trial n/a7a Rationale for numbers in the pilot trial 6Sample size7b When applicable, explanation of any interim analyses and stopping guidelines n/a

Randomisation:8a Method used to generate the random allocation sequence 5Sequence

generation 8b Type of randomisation(s); details of any restriction (such as blocking and block size) 5Allocationconcealmentmechanism

9 Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned

5

Implementation 10 Who generated the random allocation sequence, who enrolled participants, and who assigned participants to 5

Page 21 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



interventions11a If done, who was blinded after assignment to interventions (for example, participants, care providers, those

assessing outcomes) and hown/aBlinding

11b If relevant, description of the similarity of interventions 5Statistical methods 12 Methods used to address each pilot trial objective whether qualitative or quantitative 6

Results13a For each group, the numbers of participants who were approached and/or assessed for eligibility, randomly

assigned, received intended treatment, and were assessed for each objective8Participant flow (a

diagram is strongly recommended) 13b For each group, losses and exclusions after randomisation, together with reasons 8

14a Dates defining the periods of recruitment and follow-up 4Recruitment14b Why the pilot trial ended or was stopped 4

Baseline data 15 A table showing baseline demographic and clinical characteristics for each group 14Numbers analysed 16 For each objective, number of participants (denominator) included in each analysis. If relevant, these numbers

should be by randomised group8

Outcomes and estimation

17 For each objective, results including expressions of uncertainty (such as 95% confidence interval) for anyestimates. If relevant, these results should be by randomised group

8

Ancillary analyses 18 Results of any other analyses performed that could be used to inform the future definitive trial 8Harms 19 All important harms or unintended effects in each group (for specific guidance see CONSORT for harms) n/a

19a If relevant, other important unintended consequences n/a

DiscussionLimitations 20 Pilot trial limitations, addressing sources of potential bias and remaining uncertainty about feasibility 2, 9Generalisability 21 Generalisability (applicability) of pilot trial methods and findings to future definitive trial and other studies 9Interpretation 22 Interpretation consistent with pilot trial objectives and findings, balancing potential benefits and harms, and

considering other relevant evidence9

22a Implications for progression from pilot to future definitive trial, including any proposed amendments n/a

Other informationRegistration 23 Registration number for pilot trial and name of trial registry 1Protocol 24 Where the pilot trial protocol can be accessed, if available n/aFunding 25 Sources of funding and other support (such as supply of drugs), role of funders 10

26 Ethical approval or approval by research review committee, confirmed with reference number 1

Citation: Eldridge SM, Chan CL, Campbell MJ, Bond CM, Hopewell S, Thabane L, et al. CONSORT 2010 statement: extension to randomised pilot and feasibility trials. BMJ. 2016;355.

Page 22 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



*We strongly recommend reading this statement in conjunction with the CONSORT 2010, extension to randomised pilot and feasibility trials, Explanation and Elaboration for important clarifications on all the items. If relevant, we also recommend reading CONSORT extensions for cluster randomised trials, non-inferiority and equivalence trials, non-pharmacological treatments, herbal interventions, and pragmatic trials. Additional extensions are forthcoming: for those and for up to date references relevant to this checklist, see www.consort-statement.org.

Page 23 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from

http://www.consort-statement.org


For peer review onlyFeasibility of using a transition diabetes team to commence injectable therapies post discharge from a tertiary hospital

Journal: BMJ Open



Date Submitted by the Author: 20-Dec-2018






BMJ Open on O


http://bmjopen.bm

j.com/

BM



ber 2019. Dow

nloaded from



1 Feasibility of using a transition diabetes team to commence injectable therapies post 2 discharge from a tertiary hospital

3

4 Pyrlis, F1, Ogrin, R2,3, Arthur, S³, Zhai, C3, Churilov, L³, Baqar, S1, Zajac, J D1,3, Ekinci, E. I. 5 1,3




9


16

17 Word count: 2468



2021

Page 1 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



22 Article Summary:

2324 This study aimed to investigate if the use of a transition team was feasible for patients 25 with diabetes being discharged from hospital on injectable diabetes therapies.26 This pilot, randomised controlled trial was conducted between 2014 and 2016 27 conjointly by a tertiary referral hospital and a community healthcare provider.28 Hospital inpatients (n=105) on new injectable diabetes therapies were randomised to 29 transition team (receiving in-home diabetes education 24-48 hours post-discharge, 30 with endocrinologist review 2-4 weeks and 16 weeks post-discharge) or standard 31 care.32 The primary outcome was feasibility, defined by percentage of patients successfully 33 receiving the intervention.34 This study demonstrated that the use of a novel transition diabetes team is a feasible 35 alternative model of care.

36



40 We acknowledge the limitations of the study.

41 Eight patients in each group did not have follow-up HbA1c measurements, despite active 42 encouragement.

43 The follow-up diabetes treatment satisfaction questionnaire change version (DTSQc) was 44 completed by 40 (16.7% missing) of the standard care patients and 36 (34.5% missing) of the 45 transition team patients,and the number of missing DTSQc questionnaires may have limited 46 interpretability of DTSQc results. .

47 Withdrawal of participants from the transition team in hospital prior to the intervention may 48 reflect reluctance in hospitalized patients to accept health providers entering their home, 49 which may be a limitation of this model.

50

51

52

53

Page 2 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



55 Abstract

56 Objectives: This study aimed to investigate if the use of a transition team was feasible for 57 patients with diabetes being discharged from hospital on injectable diabetes therapies.

58

59 Design, Setting, Participants: This pilot, randomised controlled trial was conducted between 60 2014 and 2016 conjointly by a tertiary referral hospital and a community healthcare provider. 61 Hospital inpatients (n=105) on new injectable diabetes therapies were randomised 1:1 to 62 transition team or standard care. The transition team received in-home diabetes education 24-63 48 hours post-discharge, with endocrinologist review 2-4 weeks and 16 weeks post-discharge.

64

65 Main outcome measures: The primary outcome was feasibility, defined by percentage of 66 patients successfully receiving the intervention. Secondary outcomes included safety, defined 67 by hospital readmission and emergency department presentations within 16 weeks post-68 randomization, and treatment satisfaction, measured using Diabetes Treatment Satisfaction 69 Questionnaire (DTSQ). Exploratory outcomes included length of stay (LOS), and change in 70 Haemoglobin A1c (HbA1c) throughout the study.71

72 Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in 73 safety between groups was detected. No difference in change in HbA1c between groups was 74 detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team 75 median HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a trend towards reduced 76 length of stay in the transition team group (per protocol, standard care median LOS 8 (IQR 77 5.5-12); transition team median LOS 6 (IQR3-12), p=0.06). There was a significant 78 improvement in patient satisfaction in the transition team (standard care median 10.5 (IQR 79 8.5, 16); transition team DTSQc median 15 (IQR 10, 17.5), p=0.047), although 80 interpretability is limited by missing data.

81 Conclusion: This study demonstrated that the use of a novel transition diabetes team is a 82 feasible alternative model of care.

83

84

85

86

Page 3 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



87 Introduction


90 It has been demonstrated that 34% of hospital inpatients aged over 54 admitted to a tertiary 91 hospital have diabetes mellitus (4). Many studies have demonstrated that inpatients with type 92 2 diabetes have longer hospital length of stay and higher mortality rates compared to those 93 without (5, 6). Factors such as stress hyperglycaemia, medications, and inadequate glycaemic 94 control at the time of hospital admission, often result in the need for intensification with 95 injectable diabetes therapies (7). Furthermore, the hospitalisation period provides an 96 opportunity to identify those with poor glycaemic control and optimise diabetes management 97 (8). However, commencement of injectable therapies can be difficult in the context of 98 concurrent acute illness.

99 Guidelines recommend that patients are required to demonstrate self-management with 100 injectable therapies prior to hospital discharge (9). Diabetes education is crucial in enabling 101 patients to effectively self-manage, and assists in optimising glycaemic control post-discharge 102 (10). However, diabetes education in the hospital setting is subject to a number of limitations 103 including acute illness, pain and a sense of being overwhelmed (11). Home-based diabetes 104 education may prove more effective based on understanding a patient’s life context and 105 allowing adaptation of self-management routines such as timing of injection, sharps disposal 106 and medication storage to suit the patient’s home environment (12).


111

112

113 Methods

114 Design

115 This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in 116 metropolitan Melbourne and a community-based healthcare provider. Study participants were 117 recruited during inpatient admissions between March 2014 and November 2015 and follow-118 up continued until March 2016. The trial was stopped after funding was exhausted and 119 sufficient participants were recruited.

120 Participants

121 Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, 122 were screened for the study, and randomised to receive the intervention or standard care after 123 providing informed consent. Participant inclusion criteria were the ability to provide 124 informed consent, the presence of type 2 diabetes, age > 18 years, requirement to commence 125 or change injectable therapies therefore requiring a CDE to provide education prior to 126 discharge, medically stable and awaiting diabetes education, reside within a 30-minute travel 127 radius of the hospital, ability to attend hospital for outpatient follow-up and stable glycaemia 128 defined as blood glucose levels between 5-15 mmol/L in the 24 hours prior to randomisation. 129 Patients were excluded if they did not fulfil inclusion criteria130

131 Randomisation

132 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 133 investigator without patient contact. Group allocations were concealed by writing allocations

Page 4 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



134 on a card, and placing in sealed, unlabelled envelopes, with each consecutive participant 135 given their allocation by a research assistant after informed consent was obtained. Due to the 136 difference in treatment protocols, the study was open label to the participants and 137 investigators.

138 Intervention

139 Standard Care

140 Participants randomised to standard care were educated by hospital credentialed diabetes 141 educators (CDEs) prior to discharge. Diabetes education regarding injectable therapies in our 142 institution complies with guidelines of the Australian Diabetes Educators Association (13). 143 This comprised education regarding injectable therapy, storage, injection technique, and 144 sharps disposal, and provision of additional resources when required. Additional resources 145 included National Diabetes Services Scheme (NDSS) registration, supply of glucometer if 146 required, written patient information regarding diabetes, and outpatient follow-up. The NDSS 147 is a scheme administered by the Australian federal government which provides access for 148 people living with diabetes to education and equipment in order to enhance their ability to 149 effectively self-manage their diabetes. Participants were discharged when medically 150 appropriate and the inpatient team were satisfied that the participant could safely administer 151 the injectable therapy. Prior to discharge, appropriate follow-up was organised. General 152 Practitioners (GPs) were notified that participants had commenced or changed treatment.


154 The transition team group participants received in-home education to start injectable therapy 155 by the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided 156 with an appropriate glucometer in addition to education regarding medication, storage, 157 injection technique, sharps disposal, NDSS registration, an education package in the relevant 158 language, and CDE contact details.

159 Further contact with participants was based on CDE evaluation of the participant’s capacity to 160 self-manage injectable therapy. Once the CDE ascertained that participants were able to self-161 manage without further intervention, the endocrinologist was notified. Participants were then 162 linked with community CDE services, if necessary, for ongoing monitoring of self-163 management.

164 Follow-up with the same endocrinologist was provided within four weeks, and at 16 weeks 165 post-randomization. HbA1c was assessed at baseline and 16 weeks. The endocrinologist 166 liaised with the participant’s GP regarding changes to management and plans for ongoing 167 follow-up after the 16-week visit.

168 Data collection

169 Baseline demographic, medication and medical data were collected and participants 170 completed the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at 171 enrolment. Whether the patient was from a culturally and linguistically diverse (CALD) 172 background was recorded. Follow-up data were collected at 16 weeks post-randomization, 173 including rates of hospital readmissions and emergency presentations, length of hospital stay 174 (LOS), glycaemic control as measured by HbA1c and treatment satisfaction using the DTSQ 175 change version (DTSQc). Initial attempts to collect DTSQc questionnaires in the standard 176 care group was by reply-paid mail, however after a limited response rate using this method, 177 patients were contacted by phone. Study participation ceased at the 16-week endocrinology 178 appointment, and final data collection occurred. Further follow-up for ongoing diabetes 179 management was arranged at conclusion of the study.

180

181 Ethics

Page 5 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



182 Ethics approval was obtained from the hospital human ethics committee and the community 183 healthcare provider ethics committee. Each participant provided written informed consent.

184

185 Outcomes

186 The primary outcome measured was feasibility (proportion of participants in the transition 187 team group completing the intervention as per protocol). Secondary outcomes were safety, as 188 defined by hospital readmission and emergency department presentations within 16 weeks 189 post-randomization, and patient satisfaction with care (measured by DTSQ). Exploratory 190 outcomes were change in HbA1c and length of hospital stay (days).

191


193 Due to the pilot nature of the study, the sample size estimation was based on precision 194 arguments: assuming the feasibility of transition team intervention being 0.9 (i.e. that 90% of 195 participants randomized into transition team group would be able to complete the intervention 196 as per protocol), the sample of 55 participants randomized to the transition team group 197 provides the precision (desired half-width of the 95% confidence interval) of 0.08.

198 The same number of participants was to be randomized to the standard care group, thus 199 ensuring 80% power to detect potential medium-to-large effects of transition team 200 intervention compared to the standard care (Cohen’s d=0.55) assuming the settings of 201 alpha=0.05. Thus, the total sample size for this study was proposed as 110 participants.

202



207 The feasibility of the intervention was estimated as a proportion of participants in the 208 transition team group completing the intervention as per protocol with corresponding 95% 209 confidence interval (95% CI).

210 The difference in safety profiles (diabetes related hospital presentation or admission) between 211 two groups was investigated using Fisher's exact test.

212 DTSQ outcomes were analyzed using Wilcoxon-Mann-Whitney test and a median regression 213 model with the DTSQ score at 16 weeks post-randomization as an output and treatment group 214 and baseline DTSQ score as inputs. Sensitivity analysis was conducted by including the 215 auxillary variables demonstrating significant association with the DTSQ data being missing, 216 into the median regression model.




222 Statistical analysis was performed on both intention to treat and per protocol bases. Per 223 protocol analysis was deemed necessary to account for the patients who were initially 224 assigned to transition team but withdrew prior to intervention; in this situation they received 225 standard care.


Page 6 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



227 Patients must be actively involved in changes to diabetes management, and require education 228 regarding therapy changes and administration of injectable medications. Effective education 229 of patients can be difficult in the hospital setting, particularly in the setting of concurrent 230 acute illness. It is subject to a number of limitations including acute illness, pain and a sense 231 of being overwhelmed in hospital. These clinical observations contributed to the formulation 232 of our research questions, however there was no direct patient involvment in this.

233 Patients were not involved in the original study design, and involvement of patients in 234 recruitment was impractical as patients were hospitalised at the time of recruitment. A 235 qualitative analysis of a subgroup of patients’ experiences and their perceptions of the 236 intervention was performed, and this will be reported separately.

237 Following publication of our findings, the paper outlining study results will be sent to the 238 study participants.

239

240

Page 7 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



242 Results

243 One hundred and five participants were randomised to transition team or standard care. 244 Following initial drop outs, 103 participants received the intervention or standard care (Figure 245 1). 55 participants randomised to the transition team and 48 participants randomised to 246 receive standard care were included in the per protocol analysis (figure 1). Participants 247 randomised to the transition team withdrew for a variety of reasons such as decisions to 248 change treatment, change of discharge destination, and changes to clinical status. Of these, 249 five participants withdrew following the home visit by the CDE but prior to completing the 250 endocrinologist component of the intervention. For the purposes of the per protocol 251 analysis, these participants crossed over to the standard care group.

252 One participant in the transition team group was unable to be followed up as they were being 253 actively palliated for terminal malignancy, and one died before completion of the trial, for 254 reasons unrelated to diabetes. One participant withdrew from the standard care group and one 255 died before final data analysis in this group.

256 Baseline characteristics of participants are outlined in Table 1. No differences between groups 257 at baseline were identified.

258

259 Feasibility

260 Forty seven out of 55 participants in the transition team group completed the study as per 261 protocol (85%, 95%CI: 73%, 94%).

262 Safety



267 A significant improvement in satisfaction with diabetes treatment was demonstrated (DTSQc 268 transition team median 15 (IQR 10.0, 17.5), standard care median 10.5 (IQR 8.5,16.0) 269 Wilcoxon-Mann-Whitney, p=0.047), Figure 2, Table 2. On analysis adjusted for the baseline 270 DTSQs value, the transition team median DTSQc value was 4 points higher than the standard 271 care median (95%CI: 0.25, 7.75; p=0.037). The follow-up DTSQc was completed by 40 272 (16.7% missing) of the standard care patients and 36 (34.5% missing) of the transition team 273 patients, thus the “missingness” of the data was not likely to be random. On the sensitivity 274 analysis adjusted for the variables significantly associated with the missing DTSQc data at 16 275 weeks, the results remained qualitatively similar.

276 HbA1c

277 No statistically significant difference in change in HbA1c (standard care median HbA1c -278 1.5% (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -279 0.2%), p = 0.83) was observed, Figure 3, Table 2. An equal number of patients (8) in each 280 group did not have an HbA1c measurement at 16 weeks.

281 Length of Stay

282 There was a trend towards a reduced length of stay in the transition team group when 283 analysed on a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team 284 median 6 (IQR 3.0-12.0), p=0.06), Figure 4, Table 3.

285

286

287

Page 8 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



288 Discussion

289

290 Key findings

291 The most important finding in this trial was that a transition team to initiate injectable 292 diabetes therapies following discharge is a feasible model of care. These data suggest that a 293 transition team is safe and acceptable with a trend towards reduced length of hospital stay. 294 Moreover, patients randomized to the transition team group had greater treatment satisfaction 295 as demonstrated by a greater difference in DTSQc score.

296

297


299300 The results pertaining to the quality of this intervention are supported by results from other 301 studies examining home-based care in diabetes. The quality markers of this intervention 302 include feasibility and objective measures of medical indices, including readmission and 303 emergency presentation rates, change in HbA1c from baseline and length of stay, and patient-304 reported outcomes. 305306 We demonstrated a trend towards (p=0.06) reduced length of stay in the transition team when 307 analysed on a per protocol basis. Future studies with greater numbers may demonstrate 308 statistically significant reductions in length of hospital stay.309

310 We demonstrated significant improvements in treatment satisfaction in the transition team 311 group. The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical 312 trials and validated in several languages. The status version (DTSQs) evaluates baseline 313 satisfaction with diabetes treatment and the change version (DTSQc) evaluates the impact of 314 an intervention on satisfaction with treatment (14, 15, 16). Interpretability of this parameter is 315 limited by missing data, and there is potential for bias due to some data being obtained over 316 the phone by a research assistant, rather than in person. 317318 Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with 319 HbA1c reduction approaching 2% in both groups. Importantly, there was no statistical 320 difference detected between the reduction in HbA1c in the transition team and the standard 321 care group. 322323 Other quality outcomes assessed included emergency department presentations and hospital 324 readmissions. There were no significant differences in our study. This suggests that patient 325 safety is unlikely to be compromised by delivery of home-based education. However, given 326 the limited literature in this field, further studies with greater numbers would be necessary to 327 validate these findings.

328


330 Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to 331 improve outcomes over the course of disease. Escalating rates of diabetes necessitates the 332 development of feasible alternative models of care, with the potential to improve clinical 333 outcomes and health care costs. This study has investigated one such option, and has

Page 9 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



334 demonstrated feasibility, improved treatment satisfaction, and a trend to reduced length of 335 stay, with no safety concerns detected.

336

337

338

339 Conclusion


344

345 Funding

346 The Lord Mayor’s Charitable Foundation and the Estate of the Late Glen W A Griffiths who 347 funded this project. A/Prof Ekinci was supported by Australian National Health and Medical 348 Research Council (NHMRC) Early Career Fellowship, Viertel Clinical Investigatorship, 349 Royal Australasian College of Physicians (RACP) Fellowship and Sir Edward Weary Dunlop 350 Medical Research Foundation research grant.

351


353 The Diabetes Treatment Satisfaction Questionnaire (DTSQ) used in this publication is owned 354 by Prof Clare Bradley) and sourced from HPR Ltd.355 The team gratefully acknowledges Mr Paul Steel, CDE, for providing the in-home diabetes 356 education for the intervention group.

357


359 We have read and understood BMJ policy on declaration of interests and declare that we have 360 no competing interests.

361

362


364 F Pyrlis - involved in development of trial, clinical management of participants, performed 365 final write up








Page 10 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



375

376 Data Statement: Dataset is restricted but can be provided upon reasonable request

377

378 Figure Legend





385 Table 1 - Comparison of baseline clinical and biochemical characteristics between in the 386 control and transition team groups by initial randomization/ intention to treat

387 Table 2 - Results by intention to treat analysis. Differences in hospital readmission rates, 388 change in HbA1c, length of hospital stay, & treatment satisfaction according to original 389 randomization.

390 Table 3 - Results by per Protocol analysis. Differences in hospital readmission rates, change 391 in HbA1c, length of hospital stay, & treatment satisfaction according to management 392 received.393

394

Page 11 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



395 References396 1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010. 397 http://www.aihw.gov.au/diabetes-indicators/deaths)398 2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes 399 prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 400 103:137-149401 3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children 402 and young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. 403 No. CVD 53; Diabetes Series No 21,) 404 http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359405 4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient 406 HbA1c testing: a prospective observational study. BMJ open diabetes research & 407 care. 2015;3(1):e000113408 5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for 409 coronary artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9410 6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and 411 new hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3412 7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of 413 diabetes self-management education in the hospital setting. Primary Care Diabetes. 414 2014;8(3):187-94.

415 8. Schafer I, Pawels M, Kuver C, Pohontsch NJ, Scherer M, Bussche Hv et al. 416 Strategies for Improving Participation in Diabetes Education. A Qualitative Study. 417 PLoS One. 2014; 9(4)

418 9. Joint British Diabetes Societies for Inpatient Care. Discharge planning for adult 419 inpatients with diabetes. October 2015 420 10. Wexler DJ, Beauharnais CC, Regan S, Nathan DM, Cagliero E, Larkin ME. Impact of 421 inpatient diabetes management, education, and improved discharge transition on 422 glycemic control 12 months after discharge. Diabetes Research and Clinical Practice. 423 2012;98:249-56

424 11. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of diabetes 425 self-management education in the hospital setting. Primary Care Diabetes. 426 2014;8(3):187-94.

427 12. de Carvalho Torres H, dos Santos LM, de Souza Cordeiro PMC. Home visit: an 428 educational health strategy for self-care in diabetes. Visita domiciliária: estratégia 429 educativa em saúde para o autocuidado em diabetes. 2014;27(1):23


432 14. Bradley C, Lewis KS. Measures of psychological well-being and treatment 433 satisfaction developed from the responses of people with tablet-treated diabetes. 434 Diabetic Medicine. 1990; 7:445-451.


437 16. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change 438 version for use alongside status version provides appropriate solution where ceiling 439 effects occur. Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J, 440 Valentine J and Witthaus E (2007) The Diabetes Treatment Satisfaction 441 Questionnaire change version (DTSQc) evaluated in insulin glargine trials shows

Page 12 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from







442 greater responsiveness to improvements than the original DTSQ. Health and Quality 443 of Life Outcomes. 1999; 5 (5) 57444445446447

448

Page 13 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




Characteristic Standard care (n=48)


Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (33.3%) 22 (40%)

Median duration type 2 diabetes (SD)

7.48 (7.78) 10.18 (9.77)



9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)




Standard care (n=48) Transitions group (n=55)

P-value


DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047


-1.5% (-3.7, -0.2%) -1.9% (-3.8, -0.2%)

0.83


8 (5.5,11.5) 7 (3,12) 0.26

Page 14 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Table 3. Results by per Protocol analysis. Differences in hospital readmission rates, change in HbA1c, length of hospital stay, & treatment satisfaction according to management received.


P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047


-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85


8 (5.5, 12) 6 (3, 12) 0.06

Page 15 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 16 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from












Randomised (n=105)




Page 17 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




114x180mm (150 x 150 DPI)

Page 18 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




292x227mm (150 x 150 DPI)

Page 19 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Figure 4: A. Length of hospital stay (days) by intention to treat analysis and B. Length of hospital stay (days) per protocol analysis.

233x185mm (150 x 150 DPI)

Page 20 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Reported on page No


CONSORT abstract extension for pilot trials)2,3



objectives2b Specific objectives or research questions for pilot trial 3, 4

Methods3a Description of pilot trial design (such as parallel, factorial) including allocation ratio 4Trial design3b Important changes to methods after pilot trial commencement (such as eligibility criteria), with reasons 54a Eligibility criteria for participants 4Participants4b Settings and locations where the data were collected 44c How participants were identified and consented 4


5


6Outcomes


Randomisation:8a Method used to generate the random allocation sequence 5Sequence

generation 8b Type of randomisation(s); details of any restriction (such as blocking and block size) 5Allocationconcealmentmechanism


5

Implementation 10 Who generated the random allocation sequence, who enrolled participants, and who assigned participants to 5

Page 21 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



interventions11a If done, who was blinded after assignment to interventions (for example, participants, care providers, those

assessing outcomes) and hown/aBlinding










8






Other informationRegistration 23 Registration number for pilot trial and name of trial registry 1Protocol 24 Where the pilot trial protocol can be accessed, if available n/aFunding 25 Sources of funding and other support (such as supply of drugs), role of funders 10

26 Ethical approval or approval by research review committee, confirmed with reference number 1

Citation: Eldridge SM, Chan CL, Campbell MJ, Bond CM, Hopewell S, Thabane L, et al. CONSORT 2010 statement: extension to randomised pilot and feasibility trials. BMJ. 2016;355.

Page 22 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



*We strongly recommend reading this statement in conjunction with the CONSORT 2010, extension to randomised pilot and feasibility trials, Explanation and Elaboration for important clarifications on all the items. If relevant, we also recommend reading CONSORT extensions for cluster randomised trials, non-inferiority and equivalence trials, non-pharmacological treatments, herbal interventions, and pragmatic trials. Additional extensions are forthcoming: for those and for up to date references relevant to this checklist, see www.consort-statement.org.

Page 23 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



For peer review onlyFeasibility of using a transition diabetes team to commence injectable therapies post discharge from a tertiary hospital:

a pilot, randomised controlled trial

Journal: BMJ Open



Date Submitted by the Author: 18-Mar-2019






BMJ Open on O


http://bmjopen.bm

j.com/

BM



ber 2019. Dow

nloaded from



1 Feasibility of using a transition diabetes team to commence injectable therapies post 2 discharge from a tertiary hospital: a pilot, randomised controlled trial

3

4 Pyrlis, F1, Ogrin, R2,3, Arthur, S³, Zhai, C3, Churilov, L³, Baqar, S1, Zajac, J D1,3, Ekinci, E. I. 5 1,3




9


16

17 Word count: 2456



2021

Page 1 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



22 Article Summary:

2324 This study aimed to investigate if the use of a transition team was feasible for patients 25 with diabetes being discharged from hospital on injectable diabetes therapies.26 This pilot, randomised controlled trial was conducted between 2014 and 2016 27 conjointly by a tertiary referral hospital and a community healthcare provider.28 Hospital inpatients (n=105) on new injectable diabetes therapies were randomised to 29 transition team (receiving in-home diabetes education 24-48 hours post-discharge, 30 with endocrinologist review 2-4 weeks and 16 weeks post-discharge) or standard 31 care.32 The primary outcome was feasibility, defined by percentage of patients successfully 33 receiving the intervention.34 This study demonstrated that the use of a novel transition diabetes team is a feasible 35 alternative model of care.

36

37

38

39

40

Page 2 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



42 Abstract


45

46 Design: Pilot, randomised controlled trial.

47

48 Setting: The trial was conducted between 2014 and 2016 conjointly by a tertiary referral 49 hospital and a community healthcare provider.

50

51 Participants: Hospital inpatients (n=105) on new injectable diabetes therapies were 52 randomised 1:1 to transition team or standard care. The transition team received in-home 53 diabetes education 24-48 hours post-discharge, with endocrinologist review 2-4 weeks and 16 54 weeks post-discharge.

55


63 Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in 64 safety between groups was detected. No difference in change in HbA1c between groups was 65 detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team 66 median HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a trend towards reduced 67 length of stay in the transition team group (per protocol, standard care median LOS 8 (IQR 68 5.5-12); transition team median LOS 6 (IQR3-12), p=0.06). There was a significant 69 improvement in patient satisfaction in the transition team (standard care median 10.5 (IQR 70 8.5, 16); transition team DTSQc median 15 (IQR 10, 17.5), p=0.047), although 71 interpretability is limited by missing data.


74

75

Page 3 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




77 A strength of this study was the successful collaboration of two large organisations providing 78 support for the transition of care from hospital to home. The feasibility of delivering this 79 patient-centred alternative model of care was established.

80 Withdrawal of participants from the transition team in hospital prior to the intervention may 81 reflect reluctance in patients recovering from acute illness to return to hospital or accept 82 health providers entering their home.

83 Eight patients in each group did not have follow-up HbA1c measurements, and the follow-up 84 diabetes treatment satisfaction questionnaire change version (DTSQc) was not completed by 85 16.7% of the standard care patients and 34.5% of the transition team patients. Missing data 86 may have limited interpretability of results.

87 Overall, this novel intervention was found to be feasible and acceptable to people living with 88 diabetes.

89

90

Page 4 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



91 Introduction


94 It has been demonstrated that 34% of hospital inpatients aged over 54 admitted to a tertiary 95 hospital have diabetes mellitus (4). Many studies have demonstrated that inpatients with type 96 2 diabetes have longer hospital length of stay and higher mortality rates compared to those 97 without (5, 6). Factors such as stress hyperglycaemia, medications, and inadequate glycaemia 98 management at the time of hospital admission, often result in the need for intensification with 99 injectable diabetes therapies (7). Furthermore, the hospitalisation period provides an

100 opportunity to identify those with poor glycaemic control and optimise diabetes management 101 (8). However, commencement of injectable therapies can be difficult in the context of 102 concurrent acute illness.



115

116

117 Methods

118 Design

119 This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in 120 metropolitan Melbourne and a community-based healthcare provider. Study participants were 121 recruited during inpatient admissions between March 2014 and November 2015 and follow-122 up continued until March 2016. The trial was stopped after funding was exhausted and 123 sufficient participants were recruited.

124 Participants

125 Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, 126 were screened for the study, and randomised to receive the intervention or standard care after 127 providing informed consent. Participant inclusion criteria were the ability to provide 128 informed consent, the presence of type 2 diabetes, age > 18 years, requirement to commence 129 or change injectable therapies therefore requiring a CDE to provide education prior to 130 discharge, medically stable and awaiting diabetes education, reside within a 30-minute travel 131 radius of the hospital, ability to attend hospital for outpatient follow-up and stable glycaemia 132 defined as blood glucose levels between 5-15 mmol/L in the 24 hours prior to randomisation. 133 Patients were excluded if they did not fulfil inclusion criteria.134

135 Randomisation

136 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 137 investigator without patient contact. Group allocations were concealed by writing allocations

Page 5 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



138 on a card, and placing in sealed, unlabelled envelopes, with each consecutive participant 139 given their allocation by a research assistant after informed consent was obtained. Due to the 140 difference in treatment protocols, the study was open label to the participants and 141 investigators.

142 Intervention

143 Standard Care

144 Participants randomised to standard care were educated by hospital credentialed diabetes 145 educators (CDEs) prior to discharge. Diabetes education regarding injectable therapies in our 146 institution complies with guidelines of the Australian Diabetes Educators Association (13). 147 This comprised education regarding the injectable therapy and its storage, and sharps 148 disposal. Additional resources were provided when required. Additional resources included 149 National Diabetes Services Scheme (NDSS) registration, supply of glucometer if required, 150 written patient information regarding diabetes, and outpatient follow-up. The NDSS is a 151 scheme administered by the Australian federal government which provides access for people 152 living with diabetes to education and equipment in order to enhance their ability to effectively 153 self-manage their diabetes. Participants were discharged when medically appropriate and the 154 inpatient team was satisfied that the participant could safely administer the injectable therapy. 155 Prior to discharge, appropriate follow-up was organised. General Practitioners (GPs) were 156 notified that participants had commenced or changed treatment.


158 The transition team group participants received in-home education to start injectable therapy 159 by the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided 160 with an appropriate glucometer in addition to education regarding medication, storage, 161 injection technique, sharps disposal, NDSS registration, an education package in the relevant 162 language, and CDE contact details.



172 Data collection

173 Baseline demographic, medication and medical data were collected and participants 174 completed the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at 175 enrolment. Whether the patient was from a culturally and linguistically diverse (CALD) 176 background was recorded. Follow-up data were collected at 16 weeks post-randomization, 177 including rates of hospital readmissions and emergency presentations, length of hospital stay 178 (LOS), glycaemic control as measured by HbA1c and treatment satisfaction using the DTSQ 179 change version (DTSQc). Initial attempts to collect DTSQc questionnaires in the standard 180 care group was by reply-paid mail, however after a limited response rate using this method, 181 patients were contacted by phone. Study participation ceased at the 16-week endocrinology 182 appointment, and final data collection occurred. Further follow-up for ongoing diabetes 183 management was arranged at conclusion of the study.

184

185 Ethics

Page 6 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



186 Ethics approval was obtained from Austin Health human research ethics committee (reference 187 LNR/13/Austin 179) and the community healthcare provider ethics committee. Each 188 participant provided written informed consent.

189

190 Outcomes

191 The primary outcome measured was feasibility (proportion of participants in the transition 192 team group completing the intervention as per protocol). Secondary outcomes were safety, as 193 defined by hospital readmission and emergency department presentations within 16 weeks 194 post-randomization, and patient satisfaction with care (measured by DTSQ). Exploratory 195 outcomes were change in HbA1c and length of hospital stay (days).

196


198 Due to the pilot nature of the study, the sample size estimation was based on precision 199 arguments: assuming the feasibility of transition team intervention being 0.9 (i.e. that 90% of 200 participants randomized into transition team group would be able to complete the intervention 201 as per protocol) (14), the sample of 55 participants randomized to the transition team group 202 provides the precision (desired half-width of the 95% confidence interval) of 0.08.

203 The same number of participants was to be randomized to the standard care group, thus 204 ensuring 80% power to detect potential medium-to-large effects of transition team 205 intervention compared to the standard care (Cohen’s d=0.55) assuming the settings of 206 alpha=0.05. Thus, the total sample size for this study was proposed as 110 participants.

207



212 The feasibility of the intervention was estimated as a proportion of participants in the 213 transition team group completing the intervention as per protocol with corresponding 95% 214 confidence interval (95% CI).






227 Statistical analysis was performed on both intention to treat and per protocol bases. Per 228 protocol analysis was deemed necessary to account for the patients who were initially 229 assigned to transition team but withdrew prior to intervention; in this situation they received 230 standard care.


Page 7 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



232 Patients must be actively involved in changes to diabetes management, and require education 233 regarding therapy changes and administration of injectable medications. Effective education 234 of patients can be difficult in the hospital setting, particularly in the setting of concurrent 235 acute illness. It is subject to a number of limitations including acute illness, pain and a sense 236 of being overwhelmed in hospital. These clinical observations contributed to the formulation 237 of our research questions, however there was no direct patient involvement in this.


242

243

Page 8 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



245 Results

246 One hundred and five participants were randomised to transition team or standard care. 247 Following initial drop outs, 103 participants received the intervention or standard care (Figure 248 1). 55 participants randomised to the transition team and 48 participants randomised to 249 receive standard care were included in the per protocol analysis (figure 1). Participants 250 randomised to the transition team withdrew for a variety of reasons such as decisions to 251 change treatment, change of discharge destination, and changes to clinical status. Of these, 252 five participants withdrew following the home visit by the CDE but prior to completing the 253 endocrinologist component of the intervention. For the purposes of the per protocol 254 analysis, these participants crossed over to the standard care group.



261

262 Feasibility


265 Safety




279 HbA1c

280 No statistically significant difference in change in HbA1c (standard care median HbA1c -281 1.5% (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -282 0.2%), p = 0.83) was observed, Figure 3, Table 2. An equal number of patients (8) in each 283 group did not have an HbA1c measurement at 16 weeks.

284 Length of Stay

285 There was a trend towards a reduced length of stay in the transition team group when 286 analysed on a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team 287 median 6 (IQR 3.0-12.0), p=0.06), Figure 4, Table 3.

288

289

290

Page 9 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



291 Discussion

292

293 Key findings

294 The most important finding in this trial was that a transition team to initiate injectable 295 diabetes therapies following discharge is a feasible model of care. These data suggest that a 296 transition team is safe and acceptable with a trend towards reduced length of hospital stay. 297 Moreover, patients randomized to the transition team group had greater treatment satisfaction 298 as demonstrated by a greater difference in DTSQc score.

299

300


302303 The results pertaining to the quality of this intervention are supported by results from other 304 studies examining home-based care in diabetes. The quality markers of this intervention 305 include feasibility and objective measures of medical indices, including readmission and 306 emergency presentation rates, change in HbA1c from baseline and length of stay, and patient-307 reported outcomes. 308309 We demonstrated a trend towards (p=0.06) reduced length of stay in the transition team when 310 analysed on a per protocol basis. Future studies with greater numbers may demonstrate 311 statistically significant reductions in length of hospital stay.312

313 We demonstrated significant improvements in treatment satisfaction in the transition team 314 group. The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical 315 trials and validated in several languages. The status version (DTSQs) evaluates baseline 316 satisfaction with diabetes treatment and the change version (DTSQc) evaluates the impact of 317 an intervention on satisfaction with treatment (15, 16, 17). Interpretability of this parameter is 318 limited by missing data, and there is potential for bias due to some data being obtained over 319 the phone by a research assistant, rather than in person. 320321 Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with 322 HbA1c reduction approaching 2% in both groups. Importantly, there was no statistical 323 difference detected between the reduction in HbA1c in the transition team and the standard 324 care group. 325326 Other quality outcomes assessed included emergency department presentations and hospital 327 readmissions. There were no significant differences in our study. This suggests that patient 328 safety is unlikely to be compromised by delivery of home-based education. However, given 329 the limited literature in this field, further studies with greater numbers would be necessary to 330 validate these findings.

331


333 Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to 334 improve outcomes over the course of disease. Escalating rates of diabetes necessitates the 335 development of feasible alternative models of care, with the potential to improve clinical 336 outcomes and health care costs. This study has investigated one such option, and has

Page 10 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



337 demonstrated feasibility, improved treatment satisfaction, and a trend to reduced length of 338 stay, with no safety concerns detected.

339

340

341

342 Conclusion


347

348 Funding

349 The Lord Mayor’s Charitable Foundation and the Estate of the Late Glen W A Griffiths who 350 funded this project. A/Prof Ekinci was supported by Australian National Health and Medical 351 Research Council (NHMRC) Early Career Fellowship, Viertel Clinical Investigatorship, 352 Royal Australasian College of Physicians (RACP) Fellowship and Sir Edward Weary Dunlop 353 Medical Research Foundation research grant.

354



360



364

365


367 F Pyrlis - involved in development of trial, clinical management of participants, performed 368 final write up








Page 11 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



378

379 Data Statement: Deindentified patient data is available upon reasonable request from the 380 corresponding author within 3 months up to 6 months following publication.

381

382 Figure Legend





389 Table 1 - Comparison of baseline clinical and biochemical characteristics between in the 390 control and transition team groups by initial randomization/ intention to treat


394 Table 3 - Results by per Protocol analysis. Differences in hospital readmission rates, change 395 in HbA1c, length of hospital stay, & treatment satisfaction according to management 396 received.397

398

Page 12 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



399 References400 1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010. 401 http://www.aihw.gov.au/diabetes-indicators/deaths)402 2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes 403 prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 404 103:137-149405 3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children 406 and young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. 407 No. CVD 53; Diabetes Series No 21,) 408 http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359409 4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient 410 HbA1c testing: a prospective observational study. BMJ open diabetes research & 411 care. 2015;3(1):e000113412 5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for 413 coronary artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9414 6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and 415 new hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3416 7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of 417 diabetes self-management education in the hospital setting. Primary Care Diabetes. 418 2014;8(3):187-94.

419 8. Schafer I, Pawels M, Kuver C, Pohontsch NJ, Scherer M, Bussche Hv et al. 420 Strategies for Improving Participation in Diabetes Education. A Qualitative Study. 421 PLoS One. 2014; 9(4)

422 9. Joint British Diabetes Societies for Inpatient Care. Discharge planning for adult 423 inpatients with diabetes. October 2015 424 10. Wexler DJ, Beauharnais CC, Regan S, Nathan DM, Cagliero E, Larkin ME. Impact 425 of inpatient diabetes management, education, and improved discharge transition on 426 glycemic control 12 months after discharge. Diabetes Research and Clinical Practice. 427 2012;98:249-56

428 11. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of 429 diabetes self-management education in the hospital setting. Primary Care Diabetes. 430 2014;8(3):187-94.

431 12. de Carvalho Torres H, dos Santos LM, de Souza Cordeiro PMC. Home visit: an 432 educational health strategy for self-care in diabetes. Visita domiciliária: estratégia 433 educativa em saúde para o autocuidado em diabetes. Acta Paulista de Enfermagem. 434 2014;27(1):23


437 14. Perez A, Reales P, Barahona MJ, Romero MG, Minambres I. Efficacy and feasibility 438 of basal-bolus insulin regimens and a discharge strategy in hospitalized patients with 439 type 2 diabetes – the HO.SMIDIA study. International Journal of Clinical Practice. 440 2014; 68, 10, 1264-71.

441 15. Bradley C, Lewis KS. Measures of psychological well-being and treatment 442 satisfaction developed from the responses of people with tablet-treated diabetes. 443 Diabetic Medicine. 1990; 7:445-451.


Page 13 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



https://www.researchgate.net/journal/0103-2100_Acta_Paulista_de_Enfermagem





446 17. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change 447 version for use alongside status version provides appropriate solution where ceiling 448 effects occur. Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J, 449 Valentine J and Witthaus E (2007) The Diabetes Treatment Satisfaction 450 Questionnaire change version (DTSQc) evaluated in insulin glargine trials shows 451 greater responsiveness to improvements than the original DTSQ. Health and Quality 452 of Life Outcomes. 1999; 5 (5) 57453454455456

457

Page 14 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from






Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (33.3%) 22 (40%)


7.48 (7.78) 10.18 (9.77)



9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)





P-value


DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047


-1.5% (-3.7, -0.2%) -1.9% (-3.8, -0.2%)

0.83


8 (5.5,11.5) 7 (3,12) 0.26

Page 15 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047


-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85


8 (5.5, 12) 6 (3, 12) 0.06

Page 16 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 17 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from












Randomised (n=105)




Page 18 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




114x180mm (150 x 150 DPI)

Page 19 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




292x227mm (150 x 150 DPI)

Page 20 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




233x185mm (150 x 150 DPI)

Page 21 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Reported on page No


CONSORT abstract extension for pilot trials)3



objectives2b Specific objectives or research questions for pilot trial 5

Methods3a Description of pilot trial design (such as parallel, factorial) including allocation ratio 5, 6Trial design3b Important changes to methods after pilot trial commencement (such as eligibility criteria), with reasons 64a Eligibility criteria for participants 5Participants4b Settings and locations where the data were collected 54c How participants were identified and consented 5, 6


6


6Outcomes


Randomisation:8a Method used to generate the random allocation sequence 5, 6Sequence

generation 8b Type of randomisation(s); details of any restriction (such as blocking and block size) 5, 6Allocationconcealmentmechanism


5, 6

Page 22 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Implementation 10 Who generated the random allocation sequence, who enrolled participants, and who assigned participants to interventions

5, 6

11a If done, who was blinded after assignment to interventions (for example, participants, care providers, those assessing outcomes) and how

n/aBlinding










9






Other informationRegistration 23 Registration number for pilot trial and name of trial registry 1, 7Protocol 24 Where the pilot trial protocol can be accessed, if available n/aFunding 25 Sources of funding and other support (such as supply of drugs), role of funders 11

26 Ethical approval or approval by research review committee, confirmed with reference number 1, 7

Page 23 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Citation: Eldridge SM, Chan CL, Campbell MJ, Bond CM, Hopewell S, Thabane L, et al. CONSORT 2010 statement: extension to randomised pilot and feasibility trials. BMJ. 2016;355.*We strongly recommend reading this statement in conjunction with the CONSORT 2010, extension to randomised pilot and feasibility trials, Explanation and Elaboration for important clarifications on all the items. If relevant, we also recommend reading CONSORT extensions for cluster randomised trials, non-inferiority and equivalence trials, non-pharmacological treatments, herbal interventions, and pragmatic trials. Additional extensions are forthcoming: for those and for up to date references relevant to this checklist, see www.consort-statement.org.

Page 24 of 24


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Journal: BMJ Open



Date Submitted by the Author: 11-May-2019






BMJ Open on O


http://bmjopen.bm

j.com/

BM



ber 2019. Dow

nloaded from



1 Feasibility of using a transition diabetes team to commence injectable therapies post discharge 2 from a tertiary hospital: a pilot, randomised controlled trial

3





8


15

16 Word count: 2481



1920

Page 1 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



21 Abstract


24


26


29

30 Participants: Hospital inpatients (n=105) on new injectable diabetes therapies were randomised 31 1:1 to transition team or standard care. The transition team received in-home diabetes education 32 24-48 hours post-discharge, with endocrinologist review 2-4 weeks and 16 weeks post-33 discharge.

34


42 Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in 43 safety between groups was detected. No difference in change in HbA1c between groups was 44 detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team 45 median HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a trend towards reduced 46 length of stay in the transition team group (per protocol, standard care median LOS 8 (IQR 5.5-47 12); transition team median LOS 6 (IQR3-12), p=0.06). There was a significant improvement 48 in patient satisfaction in the transition team (standard care median 10.5 (IQR 8.5, 16); transition 49 team DTSQc median 15 (IQR 10, 17.5), p=0.047), although interpretability is limited by 50 missing data.


53

54

Page 2 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





58 Withdrawal of participants from the transition team in hospital prior to the intervention may 59 reflect reluctance in patients recovering from acute illness to return to hospital or accept health 60 providers entering their home.

61 Missing data may have limited interpretability of results.

62 Overall, this novel, patient-centred intervention was found to be feasible and acceptable to 63 people living with diabetes.

64

65

Page 3 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



66 Introduction


69 It has been demonstrated that 34% of hospital inpatients aged over 54 admitted to a tertiary 70 hospital have diabetes mellitus (4). Many studies have demonstrated that inpatients with type 71 2 diabetes have longer hospital length of stay and higher mortality rates compared to those 72 without (5, 6). Factors such as stress hyperglycaemia, medications, and inadequate glycaemia 73 management at the time of hospital admission, often result in the need for intensification with 74 injectable diabetes therapies (7). Furthermore, the hospitalisation period provides an 75 opportunity to identify those with poor glycaemic control and optimise diabetes management 76 (8). However, commencement of injectable therapies can be difficult in the context of 77 concurrent acute illness.



90

91

92 Methods

93 Design

94 This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in 95 metropolitan Melbourne, Australia, and a community-based healthcare provider. Study 96 participants were recruited during inpatient admissions between March 2014 and November 97 2015 and follow-up continued until March 2016. The trial was stopped after funding was 98 exhausted and sufficient participants were recruited.

99 Participants

100 Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, 101 were screened for the study, and randomised to receive the intervention or standard care after 102 providing informed consent. Baseline characteristics of the participants are outlined in table 1. 103 Funding of participants’ healthcare is through the Australian government’s universal health 104 insurance system, Medicare.

105 Participant inclusion criteria were the ability to provide informed consent, the presence of type 106 2 diabetes, age > 18 years, requirement to commence or change injectable therapies therefore 107 requiring a CDE to provide education prior to discharge, medically stable and awaiting diabetes 108 education, reside within a 30-minute travel radius of the hospital, ability to attend hospital for 109 outpatient follow-up and stable glycaemia defined as blood glucose levels between 5-15 110 mmol/L in the 24 hours prior to randomisation. Patients were excluded if they did not fulfil 111 inclusion criteria.112

Page 4 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



113 Randomisation

114 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 115 investigator without patient contact. Group allocations were concealed by writing allocations 116 on a card, and placing in sealed, unlabelled envelopes, with each consecutive participant given 117 their allocation by a research assistant after informed consent was obtained. Due to the 118 difference in treatment protocols, the study was open label to the participants and investigators.

119 Intervention

120 Standard Care

121 Participants randomised to standard care were educated by hospital credentialed diabetes 122 educators (CDEs) prior to discharge. Diabetes education regarding injectable therapies in our 123 institution complies with guidelines of the Australian Diabetes Educators Association (13). 124 This comprised education regarding the injectable therapy and its storage, and sharps disposal. 125 Additional resources were provided when required. Additional resources included National 126 Diabetes Services Scheme (NDSS) registration, supply of glucometer if required, written 127 patient information regarding diabetes, and outpatient follow-up. The NDSS is a scheme 128 administered by the Australian federal government which provides access for people living 129 with diabetes to education and equipment in order to enhance their ability to effectively self-130 manage their diabetes. Participants were discharged when medically appropriate and the 131 inpatient team was satisfied that the participant could safely administer the injectable therapy. 132 Prior to discharge, appropriate follow-up was organised. General Practitioners (GPs) were 133 notified that participants had commenced or changed treatment.


135 The transition team group participants received in-home education to start injectable therapy 136 by the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided 137 with an appropriate glucometer in addition to education regarding medication, storage, injection 138 technique, sharps disposal, NDSS registration, an education package in the relevant language, 139 and CDE contact details.



149 Data collection

150 Baseline demographic, medication and medical data were collected and participants completed 151 the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at enrolment. 152 Whether the patient was from a culturally and linguistically diverse (CALD) background was 153 recorded. Follow-up data were collected at 16 weeks post-randomization, including rates of 154 hospital readmissions and emergency presentations, length of hospital stay (LOS), glycaemic 155 control as measured by HbA1c and treatment satisfaction using the DTSQ change version 156 (DTSQc). Initial attempts to collect DTSQc questionnaires in the standard care group was by 157 reply-paid mail, however after a limited response rate using this method, patients were 158 contacted by phone. Study participation ceased at the 16-week endocrinology appointment, 159 and final data collection occurred. Further follow-up for ongoing diabetes management was 160 arranged at conclusion of the study.

161

Page 5 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



162 Ethics

163 Ethics approval was obtained from Austin Health human research ethics committee (reference 164 LNR/13/Austin 179) and the community healthcare provider ethics committee. Each participant 165 provided written informed consent.

166

167 Outcomes


173



180 The same number of participants was to be randomized to the standard care group, thus ensuring 181 80% power to detect potential medium-to-large effects of transition team intervention 182 compared to the standard care (Cohen’s d=0.55) assuming the settings of alpha=0.05. Thus, the 183 total sample size for this study was proposed as 110 participants.

184











Page 6 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



208 Patients must be actively involved in changes to diabetes management, and require education 209 regarding therapy changes and administration of injectable medications. Effective education 210 of patients can be difficult in the hospital setting, particularly in the setting of concurrent acute 211 illness. It is subject to a number of limitations including acute illness, pain and a sense of being 212 overwhelmed in hospital. These clinical observations contributed to the formulation of our 213 research questions, however there was no direct patient involvement in this.


218

219

Page 7 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



221 Results

222 One hundred and five participants were randomised to transition team or standard care. 223 Following initial drop outs, 103 participants received the intervention or standard care (Figure 224 1). 55 participants randomised to the transition team and 48 participants randomised to receive 225 standard care were included in the per protocol analysis (figure 1). Participants randomised to 226 the transition team withdrew for a variety of reasons such as decisions to change treatment, 227 change of discharge destination, and changes to clinical status. Of these, five participants 228 withdrew following the home visit by the CDE but prior to completing the endocrinologist 229 component of the intervention. For the purposes of the per protocol analysis, these participants 230 crossed over to the standard care group.



237

238 Feasibility


241 Safety




255 HbA1c

256 No statistically significant difference in change in HbA1c (standard care median HbA1c -1.5% 257 (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -0.2%), p 258 = 0.83) was observed, Figure 3, Table 2. An equal number of patients (8) in each group did not 259 have an HbA1c measurement at 16 weeks.

260 Length of Stay

261 There was a trend towards a reduced length of stay in the transition team group when analysed 262 on a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team median 6 263 (IQR 3.0-12.0), p=0.06), Figure 4, Table 3.

264

265

266

Page 8 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



267 Discussion

268

269 Key findings


275

276



289 We demonstrated significant improvements in treatment satisfaction in the transition team 290 group. The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical 291 trials and validated in several languages. The status version (DTSQs) evaluates baseline 292 satisfaction with diabetes treatment and the change version (DTSQc) evaluates the impact of 293 an intervention on satisfaction with treatment (15, 16, 17). Interpretability of this parameter is 294 limited by missing data, and there is potential for bias due to some data being obtained over the 295 phone by a research assistant, rather than in person. 296297 Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with 298 HbA1c reduction approaching 2% in both groups. Importantly, there was no statistical 299 difference detected between the reduction in HbA1c in the transition team and the standard care 300 group. 301302 Other quality outcomes assessed included emergency department presentations and hospital 303 readmissions. There were no significant differences in our study. This suggests that patient 304 safety is unlikely to be compromised by delivery of home-based education. However, given 305 the limited literature in this field, further studies with greater numbers would be necessary to 306 validate these findings.

307


309 Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to improve 310 outcomes over the course of disease. Escalating rates of diabetes necessitates the development 311 of feasible alternative models of care, with the potential to improve clinical outcomes and health 312 care costs. This study has investigated one such option, and has demonstrated feasibility,

Page 9 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



313 improved treatment satisfaction, and a trend to reduced length of stay, with no safety concerns 314 detected.

315

316

317

318 Conclusion


323

324 Funding


330



336



340

341





347 C Zhai - logistics of trial (research assistant), final data collection and statistical analysis





Page 10 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



354


357

358 Figure Legend








373

Page 11 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



374 References375 1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010. 376 http://www.aihw.gov.au/diabetes-indicators/deaths)377 2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes prevalence 378 for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 103:137-149379 3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children and 380 young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. No. 381 CVD 53; Diabetes Series No 21,) 382 http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359383 4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient 384 HbA1c testing: a prospective observational study. BMJ open diabetes research & care. 385 2015;3(1):e000113386 5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for 387 coronary artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9388 6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and new 389 hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3390 7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of 391 diabetes self-management education in the hospital setting. Primary Care Diabetes. 392 2014;8(3):187-94.







415 15. Bradley C, Lewis KS. Measures of psychological well-being and treatment satisfaction 416 developed from the responses of people with tablet-treated diabetes. Diabetic 417 Medicine. 1990; 7:445-451.


420 17. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change 421 version for use alongside status version provides appropriate solution where ceiling

Page 12 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from








422 effects occur. Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J, Valentine 423 J and Witthaus E (2007) The Diabetes Treatment Satisfaction Questionnaire change 424 version (DTSQc) evaluated in insulin glargine trials shows greater responsiveness to 425 improvements than the original DTSQ. Health and Quality of Life Outcomes. 1999; 5 426 (5) 57427428429430

431

Page 13 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from






Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (33.3%) 22 (40%)


7.48 (7.78) 10.18 (9.77)



9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)





P-value


DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047


-1.5% (-3.7, -0.2%) -1.9% (-3.8, -0.2%) 0.83


8 (5.5,11.5) 7 (3,12) 0.26

Page 14 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047


-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85


8 (5.5, 12) 6 (3, 12) 0.06

Page 15 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 16 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from












Randomised (n=105)




Page 17 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




114x180mm (150 x 150 DPI)

Page 18 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




292x227mm (150 x 150 DPI)

Page 19 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




233x185mm (150 x 150 DPI)

Page 20 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Reported on page No








6


6Outcomes





5, 6

Page 21 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




5, 6


n/aBlinding










9








Page 22 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




Page 23 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Journal: BMJ Open


Article Type: Research

Date Submitted by the Author: 05-Aug-2019






BMJ Open on O


http://bmjopen.bm

j.com/

BM



ber 2019. Dow

nloaded from



1 Feasibility of using a transition diabetes team to commence injectable therapies post discharge 2 from a tertiary hospital: a pilot, randomised controlled trial

3





8


15

16 Word count: 2542



1920

Page 1 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



21 Abstract


24


26


29

30 Participants: Hospital inpatients (n=105) on new injectable diabetes therapies were randomised 31 1:1 to transition team or standard care. The transition team received in-home diabetes education 32 24-48 hours post-discharge, with endocrinologist review 2-4 weeks and 16 weeks post-33 discharge.

34


42 Results: The intervention was deemed feasible (85%; (95% CI: 73%, 94%)). No difference in 43 safety between groups was detected. No difference in change in HbA1c between groups was 44 detected (standard care median HbA1c -1.5% (IQR-3.7% to -0.2%) versus transition team 45 median HbA1c -1.9% (IQR -3.8% to -0.2%), p = 0.83). There was a trend towards reduced 46 length of stay in the transition team group (per protocol, standard care median LOS 8 (IQR 5.5-47 12); transition team median LOS 6 (IQR3-12), p=0.06). There was a significant improvement 48 in patient satisfaction in the transition team (standard care median 10.5 (IQR 8.5, 16); transition 49 team DTSQc median 15 (IQR 10, 17.5), p=0.047), although interpretability is limited by 50 missing data.


53

54

Page 2 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





58 Withdrawal of participants from the transition team in hospital prior to the intervention may 59 reflect reluctance in patients recovering from acute illness to return to hospital or accept health 60 providers entering their home.

61 Missing data may have limited interpretability of results.

62 Overall, this novel, patient-centred intervention was found to be feasible and acceptable to 63 people living with diabetes.

64

65

Page 3 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



66 Introduction


69 It has been demonstrated that 34% of hospital inpatients aged over 54 admitted to a tertiary 70 hospital have diabetes mellitus (4). Many studies have demonstrated that inpatients with type 71 2 diabetes have longer hospital length of stay and higher mortality rates compared to those 72 without (5, 6). Factors such as stress hyperglycaemia, medications, and inadequate glycaemia 73 management at the time of hospital admission, often result in the need for intensification with 74 injectable diabetes therapies (7). Furthermore, the hospitalisation period provides an 75 opportunity to identify those with poor glycaemic control and optimise diabetes management 76 (8). However, commencement of injectable therapies can be difficult in the context of 77 concurrent acute illness.


86

87 Prior to assessing the effectiveness in a full scale Phase III randomised trial, the feasibility and 88 safety of such an intervention needs to be evaluated (13). This can be accomplished by 89 conducting a pilot, feasibility study that would assess both feasibility and safety of the 90 intervention in question, the recruitment potential, and would increase clinical experience with 91 the study intervention (13).

92

93 This pilot study therefore developed and evaluated the use of a transition team comprising in-94 home diabetes education by a credentialed diabetes educator (CDE), and early post-discharge 95 assessment by an endocrinologist. We hypothesized that the proposed intervention would be 96 feasible and would not negatively affect patient satisfaction when compared to standard care.

97

98

99 Methods

100 Design

101 This pilot, randomised controlled trial was conducted conjointly by a tertiary hospital in 102 metropolitan Melbourne, Australia, and a community-based healthcare provider. Study 103 participants were recruited during inpatient admissions between March 2014 and November 104 2015 and follow-up continued until March 2016. The trial was stopped after funding was 105 exhausted and sufficient participants were recruited.

106 Participants

107 Hospital inpatients with type 2 diabetes, commencing or altering injectable diabetes therapies, 108 were screened for the study, and randomised to receive the intervention or standard care after 109 providing informed consent. Baseline characteristics of the participants are outlined in table 1. 110 Funding of participants’ healthcare is through the Australian government’s universal health 111 insurance system, Medicare.

Page 4 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



112 Participant inclusion criteria were the ability to provide informed consent, the presence of type 113 2 diabetes, age > 18 years, requirement to commence or change injectable therapies therefore 114 requiring a CDE to provide education prior to discharge, medically stable and awaiting diabetes 115 education, reside within a 30-minute travel radius of the hospital, ability to attend hospital for 116 outpatient follow-up and stable glycaemia defined as blood glucose levels between 5-15 117 mmol/L in the 24 hours prior to randomisation. Patients were excluded if they did not fulfil 118 inclusion criteria.119

120 Randomisation

121 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 122 investigator without patient contact. Group allocations were concealed by writing allocations 123 on a card, and placing in sealed, unlabelled envelopes, with each consecutive participant given 124 their allocation by a research assistant after informed consent was obtained. Due to the 125 difference in treatment protocols, the study was open label to the participants and investigators.

126 Intervention

127 Standard Care

128 Participants randomised to standard care were educated by hospital credentialed diabetes 129 educators (CDEs) prior to discharge. Diabetes education regarding injectable therapies in our 130 institution complies with guidelines of the Australian Diabetes Educators Association (14). 131 This comprised education regarding the injectable therapy and its storage, and sharps disposal. 132 Additional resources were provided when required. Additional resources included National 133 Diabetes Services Scheme (NDSS) registration, supply of glucometer if required, written 134 patient information regarding diabetes, and outpatient follow-up. The NDSS is a scheme 135 administered by the Australian federal government which provides access for people living 136 with diabetes to education and equipment in order to enhance their ability to effectively self-137 manage their diabetes. Participants were discharged when medically appropriate and the 138 inpatient team was satisfied that the participant could safely administer the injectable therapy. 139 Prior to discharge, appropriate follow-up was organised. General Practitioners (GPs) were 140 notified that participants had commenced or changed treatment.


142 The transition team group participants received in-home education to start injectable therapy 143 by the CDE within 24-48 hours of discharge. At the initial visit, the participant was provided 144 with an appropriate glucometer in addition to education regarding medication, storage, injection 145 technique, sharps disposal, NDSS registration, an education package in the relevant language, 146 and CDE contact details.



156 Data collection

157 Baseline demographic, medication and medical data were collected and participants completed 158 the Diabetes Treatment Satisfaction Questionnaire status version (DTSQs) at enrolment. 159 Whether the patient was from a culturally and linguistically diverse (CALD) background was 160 recorded. Follow-up data were collected at 16 weeks post-randomization, including rates of

Page 5 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



161 hospital readmissions and emergency presentations, length of hospital stay (LOS), glycaemic 162 control as measured by HbA1c and treatment satisfaction using the DTSQ change version 163 (DTSQc). Initial attempts to collect DTSQc questionnaires in the standard care group was by 164 reply-paid mail, however after a limited response rate using this method, patients were 165 contacted by phone. Study participation ceased at the 16-week endocrinology appointment, 166 and final data collection occurred. Further follow-up for ongoing diabetes management was 167 arranged at conclusion of the study.

168

169 Ethics

170 Ethics approval was obtained from Austin Health human research ethics committee (reference 171 LNR/13/Austin 179) and the community healthcare provider ethics committee. Each participant 172 provided written informed consent.

173

174 Outcomes


180



187 The same number of participants was to be randomized to the standard care group, thus ensuring 188 80% power to detect potential medium-to-large effects of transition team intervention 189 compared to the standard care (Cohen’s d=0.55) assuming the settings of alpha=0.05. Thus, the 190 total sample size for this study was proposed as 110 participants.

191







Page 6 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from







215 Patients must be actively involved in changes to diabetes management, and require education 216 regarding therapy changes and administration of injectable medications. Effective education 217 of patients can be difficult in the hospital setting, particularly in the setting of concurrent acute 218 illness. It is subject to a number of limitations including acute illness, pain and a sense of being 219 overwhelmed in hospital. These clinical observations contributed to the formulation of our 220 research questions, however there was no direct patient involvement in this.


225

226

Page 7 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



228 Results

229 One hundred and five participants were randomised to transition team or standard care. 230 Following initial drop outs, 103 participants received the intervention or standard care (Figure 231 1). 55 participants randomised to the transition team and 48 participants randomised to receive 232 standard care were included in the per protocol analysis (figure 1). Participants randomised to 233 the transition team withdrew for a variety of reasons such as decisions to change treatment, 234 change of discharge destination, and changes to clinical status. Of these, five participants 235 withdrew following the home visit by the CDE but prior to completing the endocrinologist 236 component of the intervention. For the purposes of the per protocol analysis, these participants 237 crossed over to the standard care group.



244

245 Feasibility


248 Safety




262 HbA1c

263 No statistically significant difference in change in HbA1c (standard care median HbA1c -1.5% 264 (IQR-3.7%, -0.2%) versus transition team group median HbA1c -1.9% (IQR -3.8%, -0.2%), p 265 = 0.83) was observed, Figure 3, Table 2. An equal number of patients (8) in each group did not 266 have an HbA1c measurement at 16 weeks.

267 Length of Stay

268 There was a trend towards a reduced length of stay in the transition team group when analysed 269 on a per protocol basis (standard care median LOS 8 (IQR 5.5-12.0), transition team median 6 270 (IQR 3.0-12.0), p=0.06), Figure 4, Table 3.

271

272

273

Page 8 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



274 Discussion

275

276 Key findings


282

283



296 We demonstrated significant improvements in treatment satisfaction in the transition team 297 group. The diabetes treatment satisfaction questionnaire (DTSQ) is widely used in clinical 298 trials and validated in several languages. The status version (DTSQs) evaluates baseline 299 satisfaction with diabetes treatment and the change version (DTSQc) evaluates the impact of 300 an intervention on satisfaction with treatment (16, 17, 18). Interpretability of this parameter is 301 limited by missing data, and there is potential for bias due to some data being obtained over the 302 phone by a research assistant, rather than in person. 303304 Analysis of HbA1c at baseline and at 16 weeks revealed a significant treatment effect with 305 HbA1c reduction approaching 2% in both groups. Importantly, there was no statistical 306 difference detected between the reduction in HbA1c in the transition team and the standard care 307 group. 308309 Other quality outcomes assessed included emergency department presentations and hospital 310 readmissions. There were no significant differences in our study. This suggests that patient 311 safety is unlikely to be compromised by delivery of home-based education. However, given 312 the limited literature in this field, further studies with greater numbers would be necessary to 313 validate these findings.

314


316 Hospitalisation in patients with type 2 diabetes provides an opportunity to intervene to improve 317 outcomes over the course of disease. Escalating rates of diabetes necessitates the development 318 of feasible alternative models of care, with the potential to improve clinical outcomes and health 319 care costs. This study has investigated one such option, and has demonstrated feasibility,

Page 9 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



320 improved treatment satisfaction, and a trend to reduced length of stay, with no safety concerns 321 detected.

322

323

324

325 Conclusion


330

331 Funding


337



343



347

348





354 C Zhai - logistics of trial (research assistant), final data collection and statistical analysis





Page 10 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



361


364

365 Figure Legend








380

Page 11 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



381 References382 1. Australian Institute of Health and welfare. Diabetes deaths. Canberra: AIHW. 2010. 383 http://www.aihw.gov.au/diabetes-indicators/deaths)384 2. Guariguata L, Whiting DR, Hambleton I et al. Global estimates of diabetes prevalence 385 for 2013 and projections for 2035. Diabetes Res Clin Pract 2014; 103:137-149386 3. Australian Institute of Health and welfare. Type 2 Diabetes in Australia’s children and 387 young people:a working paper. Canberra: AIHW. 2014. Canberra: (AIHW Cat. No. 388 CVD 53; Diabetes Series No 21,) 389 http://www.aihw.gov.au/WorkArea.DowloadAsset.aspx?id=50129546359390 4. Nanayakkara N, Nguyen H, Churilov L, Kong A, Pang N, Hart GK, et al. Inpatient 391 HbA1c testing: a prospective observational study. BMJ open diabetes research & care. 392 2015;3(1):e000113393 5. Medhi, Marshall, Burke; HbA1c predicts length of stay in patients admitted for 394 coronary artery bypass surgery. Heart Dis 2001 Mar-Apr; 3(2):77-9395 6. Baker, S et al. Outcomes for general medical inpatients with diabetes mellitus and new 396 hyperglycaemia, MJA 2008 Mar 17; 188(6): 340-3397 7. Korytkowski MT, Koerbel GL, Kotagal L, Donihi A, DiNardo MM. Pilot trial of 398 diabetes self-management education in the hospital setting. Primary Care Diabetes. 399 2014;8(3):187-94.





416 13. Thabane L, Ma J, Chu R, Cheng J, Ismaila A, Rios LP, Robson R, Thabane M, 417 Giangregorio L, Goldsmith C. A tutorial on pilot studies: the what, why and how. BMC 418 Medical Research Methodology. 2010; 10(1).



425 16. Bradley C, Lewis KS. Measures of psychological well-being and treatment satisfaction 426 developed from the responses of people with tablet-treated diabetes. Diabetic 427 Medicine. 1990; 7:445-451.

Page 12 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from









430 18. Bradley C. The Diabetes Treatment Satisfaction Questionnaire (DTSQ): change 431 version for use alongside status version provides appropriate solution where ceiling 432 effects occur. Diabetes Care 22, 3,530-2. Bradley C, Plowright R, Stewart J, Valentine 433 J and Witthaus E (2007) The Diabetes Treatment Satisfaction Questionnaire change 434 version (DTSQc) evaluated in insulin glargine trials shows greater responsiveness to 435 improvements than the original DTSQ. Health and Quality of Life Outcomes. 1999; 5 436 (5) 57437438439440

441

Page 13 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from






Mean age (SD) 59.4 (10.92) 62.96 (16.31)

Number Male (%) 32(66.7%) 38 (69.1%)

Number CALD* (%) 16 (33.3%) 22 (40%)


7.48 (7.78) 10.18 (9.77)



9.81% (2.64)

83.7 (5.36)

10.06% (2.23)

86.4 (0.87)





P-value


DTSQc (median, IQR) 10.5 (8.5,16) 15 (10,17.5) 0.047


-1.5% (-3.7, -0.2%) -1.9% (-3.8, -0.2%) 0.83


8 (5.5,11.5) 7 (3,12) 0.26

Page 14 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





P-value


DTSQc (median, IQR) 10.5 (8.5, 16) 15 (10, 17.5) 0.047


-1.5% (-3.7%, -0.2%) -1.85% (-4, -0.2%) 0.85


8 (5.5, 12) 6 (3, 12) 0.06

Page 15 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Page 16 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from












Randomised (n=105)




Page 17 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




114x180mm (150 x 150 DPI)

Page 18 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




292x227mm (150 x 150 DPI)

Page 19 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




233x185mm (150 x 150 DPI)

Page 20 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from





Reported on page No








6


6Outcomes





5, 6

Page 21 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




5, 6


n/aBlinding










9








Page 22 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from




Page 23 of 23


BMJ Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



jopen.bmj.com

/B

MJ O


eptember 2019. D

ownloaded from



Documents

bmjopen.bmj.com€¦ · For peer review only 113 Randomisation 114 Participants were randomised in a 1:1 ratio using permuted block randomisation, by an 115 investigator without patient