ORIGINAL RESEARCH

Enhancement of an Auto-Injector Device for Self-Administration of Etanercept in Patients WithRheumatoid Arthritis Confers Emotionaland Functional Benefits

Naceur Rekaya . Steven M. Vicik . Bors T. Hulesch .

Laura L. McDonald

Received: April 7, 2020 / Published online: June 4, 2020� The Author(s) 2020

ABSTRACT

Introduction: Etanercept is effective in themanagement of rheumatoid arthritis (RA) andcan be self-administered via an auto-injector.While these devices are generally well accepted,some patients are not comfortable with theprocess of self-administration; this has beencited as a reason for discontinuation of biologictreatment. Alternative routes of administration(e.g., infusion) are more resource intensive. Theaim of this analysis was to explore the attributesof auto-injection devices that impact patientconfidence and ability to self-administer.Methods: Patients with RA (n = 168) andhealthcare providers (n = 82) in Belgium, Ger-many, Japan, Spain, and the UK were inter-viewed (n = 250 overall). Mock injectionprocedures were carried out using an auto-in-jector device with the addition of a sleeve with a

wider rubber grip. Importance of and perfor-mance of the device against a range attributeswere captured using a Likert scale (1–7). Diseaseseverity was captured using the Cochin handfunction scale.Results: Device attributes reported by patientsto be most important were ‘use without assis-tance’ ‘ease of administration’, ‘ease of opera-tion’, and ‘ease of grip’. The device withadditional sleeve performed strongly againstthese attributes, scoring 6.9 (out of 7), 6.8, 6.8,and 6.6, respectively with no differenceobserved between countries. Nurses and physi-cians reported similar responses. Qualitatively,patients reported that stability and grip pro-vided a sense of control and reduced anxiety.Similar overall ‘ease of operation’ was reportedbetween patients with mild (n = 89) or moder-ate/severe (n = 71) disease (score 6.4 vs. 6.5,respectively).Conclusions: The auto-injector plus sleeve per-formed strongly against key attributes even inpatients with moderate/severe RA and patientswith reduced grip strength. The robust gripimproved patient confidence and reducedinjection-related anxiety. This may be beneficialin patients who are anxious about self-admin-istration, those new to self-administration, andpotentially in patients with reduced hand dex-terity as a result of either advanced disease or apainful day.

Digital Features To view digital features for this articlego to: https://doi.org/10.6084/m9.figshare.12382679.

Electronic Supplementary Material The onlineversion of this article (https://doi.org/10.1007/s40744-020-00216-5) contains supplementary material, which isavailable to authorized users.

N. Rekaya � S. M. VicikPfizer, Walton Oaks, Surrey, UK

B. T. Hulesch (&) � L. L. McDonaldBrains and Cheek, London, UKe-mail: bors@brainsandcheek.com

Rheumatol Ther (2020) 7:537–552

https://doi.org/10.1007/s40744-020-00216-5

PLAIN LANGUAGE SUMMARY

Etanercept is a medicine used to treat rheuma-toid arthritis that is given by injection. It can bedelivered by patients themselves, in their ownhomes, using an auto-injection device thatlooks like a pen. Some patients are not com-fortable with the process of self-injecting. Otherpatients have severe arthritis in their hands thatstops them from carrying out self-injection.One auto-injection device (MyClic) has beenenhanced by the addition of a sleeve that fitsover the top of the pen. This was shown to 168patients with rheumatoid arthritis, who alreadyuse the MyClic ‘pen’ and 82 doctors and nursesin five countries (Belgium, Germany, Japan,Spain, and the UK). Patients, nurses, and doc-tors said what they believed to be the mostimportant features of a self-injection device.Generally, patients, nurses, and doctors fromthe different countries had similar views. Next,the participants scored the modified auto-in-jection device against those features. The deviceplus sleeve scored highly, meaning thatpatients, nurses, and doctors believed it wouldwork well for the patient population tested.Many patients said that with the added sleeve,the self-injection device was more stable andthey were able to grip it better. This helped tomake patients feel in control and reduced anyanxiety or fear they were feeling about their self-injection. This suggests that the sleeve is a use-ful addition and may be particularly useful forpatients who are nervous about self-injection orhave difficulty gripping self-injection devicesbecause of their arthritis.

Keywords: Auto-injector; Ergonomic device;Etanercept; Rheumatoid arthritis

Key Summary Points

Why carry out this study?

Many treatments in rheumatoid arthritisare delivered via self-injection, but somepatients are not comfortable with thisprocess, or have severe arthritis in theirhands that stops them from carrying outthe self-injection.

In this study, an auto-injector with anadditional sleeve was demonstrated tousers of the device and to healthcareprofessionals in order to explore whetherthe added sleeve improved the self-injection experience in any specificpatient groups.

We elicited the key features in a self-injection device, and then askedrespondents to score the performance ofthe device ? sleeve solution on these keyattributes. We assumed that the sleevewould provide a modified experience, ascompared with the auto-injector alone.

What was learned from the study?

The device plus sleeve scored highly,meaning that patients, nurses, anddoctors believed it would work well for thepatient population tested. Many patientssaid that with the added sleeve, the self-injection device was more stable and theywere able to grip it better. This helped tomake patients feel in control and reducedany anxiety or fear they were feeling abouttheir self-injection.

This suggests that the sleeve is a usefuladdition and may be particularly usefulfor patients who are nervous about self-injection or have difficulty gripping self-injection devices because of their arthritis.

538 Rheumatol Ther (2020) 7:537–552

INTRODUCTION

Etanercept is an anti-tumor necrosis factor(TNF)-a biologic treatment, effective in themanagement of rheumatoid arthritis (RA) [1, 2],with a safety profile somewhat superior to thatof alternative anti- TNF-a treatments [2–4].Etanercept can be self-administered via auto-injector devices and this has shown to be apreferred method of administering the treat-ment [4–6] with alternative routes of adminis-tration being less convenient [4] and moreresource intensive [6]. Self-injection is associ-ated with a wide range of benefits compared toalternative methods, including increased flexi-bility in the time and place of injectionadministration, reduced cost to both the patientand healthcare system, reduced travel time, andreduced caregiver burden [6].

While these devices are generally wellaccepted, self-injection can be associated with anumber of challenges. Some patients are lesscomfortable with the process of self-adminis-tration; in a study of 250 patients who hadrecently discontinued a range of biologic treat-ments over the previous 12 months, injectionexperience was cited as the second most com-mon unprompted primary reason to discon-tinue (18% of discontinuations, second only tolack of efficacy) and the most commonunprompted secondary reason (34%) [7]. Over-all, 41% patients reported the injection experi-ence as either a primary or secondary reason fordiscontinuation [7].

Specific challenges associated with injectionexperience can include: needle phobia, fear andanxiety, concerns about pain, stinging, andother injection site reactions, the fear of incor-rectly administering the drug, and the struggleto use a self-injection device while sufferingfrom arthritic pain, and swelling of the hands[8–10]. Many of these factors stem from, orcontribute to, patients’ lack of confidence,which can negatively influence adherence [11].However, many of these factors can be addres-sed or minimized through careful attention tothe design of the device [8]. In particular, designfeatures that can increase a patient’s confidencein their ability to correctly deliver treatment

may have a positive impact [11]. Self-injectiondevice design can help overcome some of thesefunctional and emotional challenges associatedwith self-injection [10]. It is suggested that theresulting improvements in patient preference,satisfaction, and injection experience may havea positive effect on adherence to therapy [8].

Subgroups of patients who may struggle withauto-inject procedures include those with handdexterity issues either in association withadvanced disease or due to periodic flare-ups [8].Research into the design of pre-filled syringeshas previously shown that ergonomicallydesigned features can help to decrease the forcesneeded during an injection. This helps over-come some of the challenges stemming fromproblems with hand dexterity [12].

The focus of this investigation was to explorepatient acceptance of an etanercept auto-injec-tor (MyClic�, Pfizer) that had been enhanced bythe addition of an ergonomic sleeve (MyClicauto-injector with added sleeve), with the pur-pose of improving patient usability and self-in-jection experience. The specific objectives ofthis analysis were to explore which attributes ofthe enhanced MyClic auto-injector with addedsleeve most impacted patient confidence, andtheir ability to self-inject. A further objectivewas to identify the principle benefit of theenhanced device and to identify specific patienttypes that could most benefit from it.

METHODS

Study Design

This study consisted of a survey, structuredinterview, and practical demonstration of anenhanced injection experience using the Enbrel(Etanercept) MyClic auto-injector (Pfizer, Sur-rey, UK) with a reusable and washable sleeve,(Genia Medical, Inc., Harrisburg, PA, USA)designed to enhance the ergonomics and ease ofthe self-injection process (Fig. 1). Specifically,the sleeve was designed to increase the width ofthe pen to make it easier to grip, was composedof non-slip material, and had a lip to allow thehand to rest comfortably during the adminis-tration process. The size of the pen was

Rheumatol Ther (2020) 7:537–552 539

increased by 59% by addition of the sleeve.Design of the sleeve ensured that prominentand beneficial features of the auto-injector werenot impaired, such as the audible clicks (a signalto confirm successful initiation and completionof the injection) and the viewing window (usedby the patient to confirm successful completionof the injection). The MyClic auto-injector isoperated by a button located at one end of thepen. This button was purposefully obscured bythe end of the sleeve. During the self-injectionprocess, the presence of the sleeve over theauto-injector, automatically triggers the button,meaning that the patient has only to applypressure to initiate the injection, without theneed to physically press the button themselves.

Participants included adult patients with RA.All patients were current MyClic auto-injectorusers (without the additional sleeve) at thepoint of enrolment into the study. Participantsalso included nurses and physicians who wereknowledgeable about the care of patients withRA and who had a current caseload of RApatients. The study was conducted in fivecountries: Belgium, Germany, Japan, Spain, andthe UK. Moderators from each individual

country were contracted to conduct the surveysin the local language. Moderators were trainedand supervised by members of the core teamthroughout the project.

Research instrument design, participantrecruitment, compensation, consent, and datahandling were handled according to the guid-ance provided by the British Healthcare Busi-ness Intelligence Association (BHBIA),specifically the Legal and Ethical guidelines ofthe BHBIA relating to patient research [13].Participants consented to participate in ademonstration of the product, and they alsoconsented to having their data analyzed forpublication in a scholarly journal. Healthcareprofessionals have also consented to the above.All participants have received compensation fortheir participation, within the ‘fair marketvalue’ rules set out by Pfizer.

Mock injection procedures were carried outonto a pad using a demonstration MyClic auto-injector (with no needle and no injectable liq-uid) with the addition of the sleeve. Participantswere shown how to apply the sleeve to the auto-injector and shown how to administer the

Fig. 1 Auto-inject device with additional sleeve

540 Rheumatol Ther (2020) 7:537–552

injection onto a pad. Participants were thenasked to carry out the procedure themselves.

Data Capture and Analysis

Disease duration and characteristics were cap-tured. Hand involvement (i.e., the extent towhich the patients’ hands were impacted bytheir RA during the day of the survey) wascaptured using the Cochin Hand Function Scale[14, 15]. This is a functional disability ques-tionnaire containing 18 questions about apatient’s daily activities related to hand func-tion and the degree of difficulty they experiencein completing those activities (Supplementarymaterial; Table 1). This tool is presented as asimple-to-use questionnaire, with proven relia-bility, validity, and responsiveness demon-strated in RA [14, 15]. Questions were scoredfrom 0 (least disability) to 5 (representinggreatest disability). The Cochin score for anindividual patient was calculated as the sum ofthe responses for all 18 questions; the scaleranges from 0 to a maximum of 90 where thegreater the score the greater the disability. Toanalyze the impact of hand disability on theresponse to the enhanced self-inject device,patients were categorized into those with no/mild hand disability (Cochin score of 0–20) or amoderate/severe degree of hand disability(scores[ 21).

Prior to the demonstration of the MyClicauto-injector with added sleeve, all participantswere asked how important they felt a rangeattributes were, in a self-inject device. Impor-tance of these attributes was captured using aLikert scale from 0 to 7, where 0 indicated thatthe attribute was ‘‘completely unimportant’’and 7 indicated that the attribute was ‘‘ex-tremely important’’. Following the demonstra-tion of the MyClic auto-injector with addedsleeve, the performance of the MyClic auto-in-jector with added sleeve was captured; partici-pants were asked how well they felt the auto-injector had performed against each of thosesame attributes. A similar Likert scale was usedfrom 0 to 7, where 0 indicated that the device(?sleeve) had ‘‘very poor performance’’ and 7indicated that the auto-injector had excellent

performance. The attributes explored are shownin the Supplementary material, Table 2, and inFig. 2. This approach has been used elsewhere[16, 17].

Each participant also underwent a structuredinterview. All responses were captured andqualitative analysis of participant interviewswas carried out by comparison of patientresponses to each individual question. Emer-gent themes in response to individual questionswere identified and tallies were used to quantifythe proportion of respondents expressing thesethemes.

The study sample size was determinedaccording to feasibility; a minimum of approx-imately 20 patients from each of the fivecountries were required to allow sub-analysis.Data analysis was carried out via calculation ofmean values. For statistical analysis, Chi-squared tests were carried out using SPSS11 andp values of less than 0.05 were considered to beof statistical significance.

RESULTS

Details of Participants

Overall, 239 participants were recruited includ-ing patients (n = 168) who were all current usersof the MyClic auto-injector, as well as physi-cians (n = 43) and nurses (n = 39) who repre-sented the healthcare professionals caring forpatients with RA.

Patient characteristics are shown in Table 1.The 168 patients enrolled were from a range ofcountries including Belgium (n = 30), Germany(n = 41), Japan (n = 16), Spain (n = 40), and theUK (n = 41). Patients’ average duration of RAwas 12.73 years and the average time thatpatients had been receiving their current med-ication (etanercept delivered via MyClic auto-injector) was 5.39 years. Three-quarters ofpatients (75%) reported that their RA hadimpacted their lives to a moderate or significantextent. Also, 55.62% of patients reported no ormild hand involvement (Cochin scores B 20);the remaining 44.38% reported moderate orsevere degree of hand disability (scores[20).

Rheumatol Ther (2020) 7:537–552 541

Physician and nurse characteristics areshown in Table 2. The 82 healthcare practi-tioners enrolled were from a range of countriesincluding Belgium (n = 10), Germany, Spain,the UK (n = 20 per country), and Japan (n = 12).The principle place of work was captured;overall, most physicians and nurses were hos-pital based (23/43, 53.49%; 18/39, 46.15%,respectively) with fewer working in an office orclinic setting, in teaching or university hospi-tals, or in the community. All nurses were spe-ciality rheumatology nurses. All physicians werespecialist rheumatologists except for oneorthopedic surgeon (Japan) and one neurologist(Belgium).

Device Attributes

The overall importance of specific attributes ofan auto-injector are shown in Fig. 2 ranked bythe perceived importance of each attribute.Attributes considered to be most important bypatients were ‘use without assistance’ (scoring

6.9 out of a maximum of 7), ‘ease of adminis-tration’ (6.8), and ‘ease of operation overall’(scoring 6.8). Other attributes perceived to behighly important (with average scores over 6 inall groups) included that the device be ‘easy togrip’ (6.6), ‘easy to learn how to use’ (6.6), and‘easy to prepare’ (6.5) (Fig. 2). Differences inperceived importance of device attributesbetween patients, nurses, and physicians wereexplored. There was a very similar pattern ofresponses between different groups of partici-pants meaning that patients, nurses, andphysicians identified similar attributes to be themost important (refer to Fig. 2 for values).Patients and nurses also rated as important(scores[ 6) that they received sound feedbackafter injection (in this case, an audible click) aswell as visual feedback after injection, that theinjection window was easy to see and that theneedle was concealed. For all of these attributes,physicians rated the importance of slightly lessimportant with scores just below 6 (not shown).The attribute perceived to be least important for

Fig. 2 Importance of device attributes. Participants scoredthe importance of specific attributes of an auto-injectordevice using a Likert scale from 1 to 7, where 1 indicatedthat the attribute was ‘‘not at all important’’ and 7indicated that the attribute was extremely important.

Mean scores for patients (n = 160) are shown. Attributesare ranked in order according to the patient’s mostimportant attribute shown at the top of the figure and thenin decreasing order of importance

542 Rheumatol Ther (2020) 7:537–552

patients was the ‘attractive design of the device’(scoring 3.4); nurses and physicians also scoredthis attribute the lowest (3.8 and 4.0, respec-tively). In general, physicians tended to giveslightly lower scores, but the relative order ofimportance was similar. Patients and nursestended to give higher scores and had very sim-ilar perceptions of the relative importance ofthe device attributes; mean scores were moresimilar between nurses and patients thanbetween physicians and patients.

A separate analysis was undertaken toexplore the differences in perceived importanceof device attributes between patients in differ-ent countries (not shown). With regards to thethree attributes identified as most important(‘use without assistance’, ‘ease of administra-tion’, and ‘ease of operation’), there were noapparent differences between countries; meanscores for each individual country were simi-larly high for these attributes, all scoring above6.7 (out of a maximum of 7). Differencesbetween patients from different countries were

Table 1 Patient characteristics

Country [n (%)]

Belgium 30 (17.86)

Germany 41 (24.40)

Japan 16 (9.52)

Spain 40 (23.81)

UK 41 (24.40)

Total 168

Disease characteristics

Time since diagnosis of RA, mean years (SD) 12.73 (9.67)

Time on current medication, years (SD) 5.39 (4.73)

To what extent has RA impacted patients’ life [n (%)]a

No effect 5 (3.01)

Slight effect 36 (21.95)

Moderate effect 68 (41.46)

Significant effect 55 (33.54)

Cochin score [n (%)]

0 (mild) 15 (9.37)

1–20 74 (46.25)

21–40 41 (25.63)

41–60 25 (15.63)

60 ? (severe) 5 (3.13)

Total 160a (100)

RA rheumatoid arthritis, SD standard deviationa Data not provided for a minority of patients

Rheumatol Ther (2020) 7:537–552 543

apparent with some of the other attributes. Forexample, ‘ease of grip’, ‘intuitive’ design of thedevice, and having a ‘button-free device’, wereless important to Japanese patients (scoring 5.7,5.5, and 4.3, respectively), compared with par-ticipants from other countries (scores rangingfrom 6.4 to 6.9, 6.0 to 6.9, and 4.8 to 6.0,respectively). The ‘size’ and ‘weight’ and the‘attractive design’ of the device were all moreimportant to participants from Belgium and to alesser extent for respondents from Spain, thanfrom the other countries. Participants fromBelgium tended to score consistently morehighly than other countries. In general, noother obvious inter-country differences wereobserved.

After patients, nurses, and physicians hadbeen taken through a demonstration of theenhanced MyClic auto-injector with additionalsleeve, they were asked to score how well theenhanced auto-injector performed against thoseattributes. Patients reported that the auto-in-jector with additional sleeve performed wellagainst all the attributes (Fig. 3a); in particular,

it performed strongly against those attributesranked as most important including ‘use with-out assistance (scoring 6.8 out of a maximumscore of 7), ‘ease of administration’ (scoring 6.6)and ‘ease of operation overall’(scoring 6.5).Nurse and physicians also scored the perfor-mance of the enhanced auto-injector highlyagainst the listed attributes. Responses for thetop 6 most important attributes are shown inFig. 3b. No meaningful difference was observedbetween patients, nurses, or physicians in theseattributes.

Importantly, mean scores in all categoriesand for all categories of participants weregreater than 5, the majority being greater than6, on average, indicating a high degree of per-formance of the auto-injector with addedsleeve.

A separate sub-analysis was undertaken toexplore how well participants from differentcountries perceived the auto-injector to performagainst the listed attributes and whether anydifferences could be observed between coun-tries. Similarly high performance ratings were

Table 2 HCP characteristics

Nurses (n = 39) Physicians (n = 43) Total (n = 82)

Country

Belgium 5 5 10

Germany 10 10 20

Japan 6 6 12

Spain 11 9 20

UK 7 13 20

Professional experience

Principle setting, n (%)

Teaching/university hospital 8 9 17

Hospital 18 23 41

Community 2 0 2

Office/clinic setting 11 11 22

All nurses were speciality rheumatology nurses. All physicians were rheumatologists except for one orthopedic surgeon(Japan) and one neurologist (Belgium)HCP healthcare professional, RA rheumatoid arthritis, SD standard deviation

544 Rheumatol Ther (2020) 7:537–552

Fig. 3 Performance of the device with the sleeve againstkey attributes. Participants scored the performance of theauto-injector with the added sleeve, against a variety ofattributes using a Likert scale from 1 to 7, where 1 wouldindicate that it had ‘‘very poor performance’’ and 7 wouldindicate that the device had excellent performance. Meanscores for patients (n = 160) are shown in a. Responses of

physicians (n = 41) and nurses (n = 38) for the mostimportant attributes are shown in b. Mean patient scoresfor the most important attributes, divided by country areshown in c. Attributes are ranked in order according to thepatient’s most important attribute shown at the top of thefigure and then in decreasing order of importance

Rheumatol Ther (2020) 7:537–552 545

measured for these attributes in all countriesnotably for ‘easy to use without assistance’(mean scores ranging from 6.7 to 6.9/7) and‘easy to administer’ (mean scores ranging from6.4 to 7.0; Fig. 3c). For Belgian participants, theauto-injector with additional sleeve scored moststrongly for ‘easy to administer’ and ‘easy tooperate overall’ (both scoring 7/7). Mostpatients also reported that the device was easyto grip, with mean scores from Belgium, Ger-many, the UK, Spain, and Japan of 6.8, 6.8, 6.7,6.4, and 6.0, respectively). Patients also reportedthat the auto-injector with additional sleevescored highly against the attribute ‘button free’(mean scores per country ranging from 6.2 to6.9/7; not shown).

Nurses and physicians were asked how likelythey would be to recommend the auto-injectorwith added sleeve to their patients and to theircolleagues on a scale of 1–7 (1 = extremelyunlikely 7 = extremely likely). Nurses andphysicians both reported a high likelihood torecommend the auto-injector with added sleeveto their patients (mean scores of 6.6 and 6.2,

respectively, Fig. 4). Nurses were more likely torecommend the auto-injector with added sleeveto a colleague compared with physicians (6.5 vs.5.5, respectively). Patients were extremely likelyto recommend the auto-injector with addedsleeve to other patients (mean score of 6.3) andhighly likely to request it for themselves (5.8,Fig. 4).

Impact of disease severity on patientresponses

A sub-analysis was carried out to compareresponses for patients with moderate/severehand disability (patients with Cochin scores[21; n = 71) or no/mild hand disability (Cochinscore of 0–20; n = 89). This was to investigatewhether patients with impaired hand dexteritywould particularly benefit from the addition ofthe sleeve. Similar overall ‘ease of operation’was reported between patients with no/mild ormoderate/severe hand disability (scores of 6.4vs. 6.5, respectively, Fig. 5. This suggested thatpatients with moderate or severe RA were able

Fig. 3 continued

546 Rheumatol Ther (2020) 7:537–552

to use the enhanced auto-injector just as easilyas patients with less impairment. No differencein other auto-injector attributes were apparentbetween patients with no/mild or moder-ate/severe RA including ease of grip, ease ofpreparation, ease of administration, or ability to

use without assistance (Fig. 5). However,patients with moderate-to-high Cochin scoreswere significantly more likely to request thesleeve for themselves (6.2 out of a maximum of7) compared with patients with no/mild Cochinscores (5.3, p\ 0.05), suggesting that patients

Fig. 4 Likelihood of nurses and physicians to recommendthe auto-injector with added sleeve. Nurses, physicians, andpatients scored the likelihood of their recommending theauto-injector with the added sleeve using a Likert scale

from 1 to 7, where 1 indicated that they were extremelyunlikely and 7 extremely likely to recommend it. Meanscores for patients (n = 160), physicians (n = 41), andnurses (n = 38) are shown

Fig. 5 Effect of disease severity of performance of theauto-injector with added sleeve. Disease severity wasmeasured using the Cochin scale [15]. Participants scoredthe performance of the auto-injector with the added sleeveagainst a variety of attributes using a Likert scale from 1 to7, where 1 indicated that the device had ‘‘very poor

performance’’ and 7 indicated that the device had excellentperformance. Mean scores for patients with no/mild handdisability (Cochin score 0-20, n = 89), and those witheither moderate or severe hand disability (Cochin score 21or greater, n = 71) are shown

Rheumatol Ther (2020) 7:537–552 547

with hand disability as a result of their RA couldmore easily see the benefit of the additionalsleeve.

Themes Arising from Qualitative InterviewSeveral recurring themes relating to functionaland emotional benefits of the sleeve arose fromthe qualitative interview.

In terms of functional benefits, patientshighlighted that the increased width and non-slip texture provided an enhanced grip on theMyClic auto-injector, resulting in greater con-fidence to hold and position the device forlonger durations. The non-slip texture was alsoseen as beneficial if an injection was to be per-formed in moist conditions such as could beexperienced in a bathroom. Use of the sleeveallowed displacement of pressure across thewhole hand during administration, putting lesspressure and stress on the joints, fingers, andwrists. In particular, patients could confidentlyoperate the MyClic auto-injector with theadded sleeve without the need to use thethumb, allowing a functional benefit forpatients with limited thumb dexterity, or forpatients who prefer not to have to press a but-ton to administer the injection. Patients alsoidentified that the sleeve enabled the auto-in-jector to be held at a 90� angle without anyrestriction. These benefits were all recognized tobe useful for patients with hand disabilities.

In terms of emotional benefits, patientsreported that stability and grip provided a senseof greater control and reduced anxiety. This wasequally true for patients with no/mild or mod-erate/severe hand disabilities. One aspect of thereduced anxiety was related to the button-freeformat of the auto-injector when the sleeve wasin position meaning that patients did not needto separately press the button to initiate theinjection, merely apply pressure. A recurringtheme was that this offered a release from theanxiety that patients had built up around theaction of pressing the button. Patients alsoreported that the addition of the sleeve allayedtheir fears that they were not carrying theinjection out correctly; in particular, the sleevewas reported to improve stability of placementand reassured patients that the auto-injectorwas at the correct angle.

Patients reflected that the sleeve may be ableto provide greater independence if their condi-tion worsened to such an extent that theywould otherwise struggle to self-administertheir treatment and that addition of the sleevemay minimize the future need for assistancewith administration. Even patients with goodhand dexterity on the day of the evaluationrecognized the benefit that the sleeve couldbring on a ‘bad’ day when their hands may bestiff and swollen. In some cases, where patientscurrently required assistance to administer theirmedication, they felt that the enhancementsprovided by the sleeve could allow them to self-administer and not have to rely on a caregiver.Autonomy was found to be an important factoramong the patients and one of the preferredbenefits of the enhanced MyClic auto-injectorwith the additional sleeve.

Many patients suggested that the sleeve wassomething that should be offered to patientsnew to self-injection, as they felt that it sim-plified and improved the process; 80% ofpatients agreed that they liked or preferred thedevice with the addition of the sleeve andwould accept it if offered to them. A further15% of the patient cohort felt that they neededthe sleeve to enable them to effectively self-ad-minister their medication.

DISCUSSION

The addition of a sleeve to the Enbrel MyClicauto-injector was well received by patients,physicians, and nurses. The MyClic auto-injec-tor enhanced by the addition of a sleeve per-formed very well against the attributes deemedby patients, nurses, and physicians to be mostimportant: ‘ease of use without assistance’, ‘easeof administration’, and ‘ease of operation over-all’. These most important attributes were allrelated to the use of the device; patient auton-omy in administration of medication is of greatimportance in this population. This was con-firmed during the qualitative interviews wherethe desire to remain autonomous and avoidingthe reliance on others for routine tasks such asadministration of medication was a commontheme. Many patients could see that the sleeve

548 Rheumatol Ther (2020) 7:537–552

could be easier for them to use during futureflare-ups when they may otherwise struggle toself-inject he easier grip and increased controlprovided by the sleeve. Improved autonomywas seen as a major benefit, not only forpatients with hand disabilities at the time of thestudy but also for patients worried about thepotential progression of their disease in thefuture.

We had hypothesized that patients withfunctional impairment and reduced hand dex-terity as a result of severe RA may report thegreatest benefit from the addition of the sleeve.Surprisingly, most patients reported benefittingfrom the sleeve, regardless of their level of handdisability; no difference was observed betweenpatients with moderate-to-severe hand disabil-ity (Cochin scores of 21 or above) and thosewith mild or no hand disability (Cochin scoresbetween 0 and 20). Reassuringly, patients withsevere hand disability as a result of their RAwere able to confidently use the MyClic auto-injector with additional sleeve, confirming thatthe sleeve is appropriate for use in this patientsubpopulation. A previous study by Hudry et al.[18] explored the impact of hand disability onthe ability of patients with RA to self-inject theirmethotrexate medication. In this study, nodifference was found between patients withCochin scores above or below 20. This is similarto the observation made in this study. Thissuggests that concerns around correct adminis-tration of the medication can often be emo-tional barriers rather than physical barriers (i.e.,inability to carry out the self-injection proce-dure). Steps taken to address these emotionalbarriers may therefore have a sizable impact onpatient satisfaction and confidence with self-injection procedures, not only in RA, butpotentially in other conditions.

The unforeseen benefits of the additionalsleeve, stemming from a number of differentareas, were related to the emotional benefitsthat the sleeve provided. Firstly, the sleeveenhanced the size, grip, and stability of theauto-injector, making patients feel confident intheir ability to carry out the injection procedurecorrectly. Secondly, the sleeve also relievedpatient’s anxiety that had built up around theself-injection procedure. Previous reports have

suggested that diminishing confidence in apatient’s own ability to successfully perform atask is linked to their lack of adherence totreatment. As such, increasing a patient’s beliefin their ability to correctly administer theirmedication may help increase the level ofadherence to that medication [11]. Devices canalso impact positively on the self-esteem ofelderly patients if they feel they have masteredthe device [19]. Addition of the sleeve maytherefore have a beneficial effect; the increasedconfidence and sense of control it gave topatients with resulting impact on confidence inthe auto-inject procedure, may boost patient’scompliance with treatment. Whether theenhancements described in this study result inimproved patient adherence to treatmentremains to be evaluated, but is plausible.

These observations suggest that the additionof the sleeve to the auto-inject device may be ofgreatest benefit to:

• patients who have decreased confidence intheir ability to self-inject

• patients with hand impairments due to theirRA

• Patients with reduced grip strength• patients with flaring RA• patients with any fear or anxiety around the

process of self-injection• patients whose continued autonomy is very

important to them• patients who are new to the process of auto-

injection.

Patients also have the choice to perform aninjection using the additional sleeve or to con-tinue using the device without the sleeve.

This study has some limitations; we did notexplore the impact of patient age on their per-ception of the additional sleeve or the ability touse it. The study is also a non-comparativestudy, which does not explore the relativepreference of patients of healthcare practition-ers to different auto-inject devices. Finally,because a demonstration (sham) device wasused, no information relating to the perfor-mance of the device in clinical practice could becaptured.

Rheumatol Ther (2020) 7:537–552 549

CONCLUSIONS

Enhancement of current auto-injectors mayprovide additional functional and emotionalbenefits for many patients. The MyClic auto-injector, enhanced through addition of anergonomic sleeve providing a robust grip, per-formed strongly against key attributes (notably‘ease of administration’, ‘ease of operationoverall’, and ‘ease of grip’) even in patients withmoderate/severe RA affecting their hand dex-terity. Qualitative analysis of patient interviewsclearly identified the importance of not onlyproviding a device that is physically easy to use(functional benefits) but also a device thatpatients feel confident in using (emotionalbenefits). The MyClic auto-injector with addi-tional sleeve may be of functional benefit inpatients with reduced hand dexterity as a resultof either advanced disease or those currentlyexperiencing a disease flare but may be ofemotional benefit in a far wider range ofpatients; the auto-injector with additionalsleeve was found to relieve much of the anxietyand increase patient confidence relating to theinjection process. This may be particularlybeneficial in patients who are already known tobe anxious about self-injection. The sleeve mayalso be of benefit to patients new to self-injec-tion because it enhances the injection experi-ence and may also make administration easier.Patients with decreased confidence in theirability to self-inject, including patients withhand disability as a result of their RA, may alsobenefit emotionally from using the sleeve,through its potential to increase self-confidencein the process. Future investigations shouldexplore the psychological and emotionalimpact of auto-injectors as well as the ability ofpatients to physically use them.

ACKNOWLEDGEMENTS

Funding. Sponsorship for this study andRapid Service Fee were funded by Pfizer (WaltonOaks, Surrey, UK). All authors had full access toall of the data in this study and take complete

responsibility for the integrity of the data andaccuracy of the data analysis.

Editorial Assistance. Editorial assistance inthe preparation of this article was provided byDr. Jenny Smith of JMS Medical Writing Ser-vices Ltd and was funded by Pfizer (WaltonOaks, Surrey, UK).

Authorship. All named authors meet theInternational Committee of Medical JournalEditors (ICMJE) criteria for authorship for thisarticle, take responsibility for the integrity ofthe work as a whole, and have given theirapproval for this version to be published.

Disclosures. Naceur Rekaya and Steven M.Vicik are both employees of Pfizer. Bors T.Hulesch and Laura L. McDonald are researchdirectors at Brains & Cheek (London) LLP, thecompany that has been commissioned to carryout the research and funded by Pfizer (WaltonOaks).

Compliance with Ethics Guidelines. Re-search instrument design, participant recruit-ment, compensation, consent, and datahandling were handled according to the guid-ance provided by the British Healthcare Busi-ness Intelligence Association (BHBIA),specifically the Legal and Ethical guidelines ofthe BHBIA relating to patient research [13].Participants consented to participate in ademonstration of the product, and they alsoconsented to having their data analyzed forpublication in a scholarly journal. Healthcareprofessionals have also consented to the above.All participants have received compensation fortheir participation, within the ‘fair marketvalue’ rules set out by Pfizer.

Data Availability. The datasets generatedduring and/or analyzed during the currentstudy are not publicly available at the presenttime but are available from the correspondingauthor on reasonable request.

Open Access. This article is licensed under aCreative Commons Attribution-NonCommer-cial 4.0 International License, which permits

550 Rheumatol Ther (2020) 7:537–552

any non-commercial use, sharing, adaptation,distribution and reproduction in any mediumor format, as long as you give appropriate creditto the original author(s) and the source, providea link to the Creative Commons licence, andindicate if changes were made. The images orother third party material in this article areincluded in the article’s Creative Commonslicence, unless indicated otherwise in a creditline to the material. If material is not includedin the article’s Creative Commons licence andyour intended use is not permitted by statutoryregulation or exceeds the permitted use, youwill need to obtain permission directly from thecopyright holder. To view a copy of this licence,visit http://creativecommons.org/licenses/by-nc/4.0/.

REFERENCES

1. Tarp S, Furst DE, Dossing A, et al. Defining theoptimal biological monotherapy in rheumatoidarthritis: a systematic review and meta-analysis ofrandomised trials. Semin Arthritis Rheum.2017;46(6):699–708.

2. Singh JA, Hossain A, Tanjong Ghogomu E, MudanoAS, Tugwell P, Wells GA. Biologic or tofacitinibmonotherapy for rheumatoid arthritis in peoplewith traditional disease-modifying anti-rheumaticdrug (DMARD) failure: a Cochrane SystematicReview and network meta-analysis (NMA).Cochrane Database Syst Rev. 2016;11:Cd012437.

3. Michaud TL, Rho YH, Shamliyan T, Kuntz KM, ChoiHK. The comparative safety of tumor necrosis factorinhibitors in rheumatoid arthritis: a meta-analysisupdate of 44 trials. Am J Med. 2014;127(12):1208–32.

4. Navarro-Millan I, Herrinton LJ, Chen L, Harrold L,Liu L, Curtis JR. Comparative effectiveness ofetanercept and adalimumab in patient reportedoutcomes and injection-related tolerability. PLoSOne. 2016;11(3):e0149781.

5. Borras-Blasco J, Gracia-Perez A, Castera MD-E,Rosique-Robles J, Abad J. Educational session as atool to increase patient satisfaction of switchingetanercept from the prefilled syringe to theautoinjection pen. Expert Opin Biol Ther.2013;13(8):1103–8.

6. Borras-Blasco J, Gracia-Perez A, Rosique-Robles JD,Castera MD, Abad FJ. Acceptability of switching

adalimumab from a prefilled syringe to an autoin-jection pen. Expert Opin Biol Ther. 2010;10(3):301–7.

7. Bolge SC, Goren A, Tandon N. Reasons for discon-tinuation of subcutaneous biologic therapy in thetreatment of rheumatoid arthritis: a patient per-spective. Patient Prefer Adher. 2015;9:121–31.

8. van den Bemt BJF, Gettings L, Domanska B, Brug-graber R, Mountian I, Kristensen LE. A portfolio ofbiologic self-injection devices in rheumatology:how patient involvement in device design canimprove treatment experience. Drug Deliv.2019;26(1):384–92.

9. Schiff M, Saunderson S, Mountian I, Hartley P.Chronic disease and self-injection: ethnographicinvestigations into the patient experience duringtreatment. Rheumatol Ther. 2017;4(2):445–63.

10. Keininger D, Coteur G. Assessment of self-injectionexperience in patients with rheumatoid arthritis:psychometric validation of the Self-InjectionAssessment Questionnaire (SIAQ). Health Qual LifeOutcomes. 2011;9:2.

11. Nafradi L, Nakamoto K, Schulz PJ. Is patientempowerment the key to promote adherence? Asystematic review of the relationship between self-efficacy, health locus of control and medicationadherence. PLoS One. 2017;12(10):e0186458.

12. Sheikhzadeh A, Yoon J, Formosa D, Domanska B,Morgan D, Schiff M. The effect of a new syringedesign on the ability of rheumatoid arthritispatients to inject a biological medication. ApplErgonom. 2012;43(2):368–75.

13. BHBIA guidelines. https://www.bhbia.org.uk/guidelines-and-legislation/legal-and-ethical-guidelines. Accessed 8 July 2019.

14. Poiraudeau S, Lefevre-Colau MM, Fermanian J,Revel M. The ability of the Cochin rheumatoidarthritis hand functional scale to detect changeduring the course of disease. Arthritis Care Res.2000;13(5):296–303.

15. Duruoz M, Poiraudeau S. Development and valida-tion of a rheumatoid hand functional disabilityscale that assesses functional handicap. J Rheuma-tol. 1996;23(7):1167–72.

16. Thakur K, Biberger A, Handrich A, Rezk MF. Patientperceptions and preferences of two etanercept auto-injectors for rheumatoid arthritis: findings from apatient survey in Europe. Rheumatol Ther.2016;3(2):245–56.

17. Tischer B, Mehl A. Patients’ and nurses’ preferencesfor auto-injectors for rheumatoid arthritis: results of

Rheumatol Ther (2020) 7:537–552 551

a European survey. Patient Prefer Adher. 2018;12:1413–24.

18. Hudry C, Lebrun A, Moura B, Zinovieva E, BackersO, Herman-Demars H. Evaluation of usability andacceptance of a new auto-injector intended formethotrexate subcutaneous self-administration in

the management of rheumatoid arthritis. Rheuma-tol Ther. 2017;4(1):183–94.

19. Thomson R, Martin JL, Sharples S. The psychosocialimpact of home use medical devices on the lives ofolder people: a qualitative study. BMC health ServRes. 2013;13:467.

552 Rheumatol Ther (2020) 7:537–552