Changing What’s Possible ® The Longevity after Injury Project Subjective and Objective Environmental Factors’ Influence and Spinal Cord Injury Yue Cao,

Changing What’s Possible®The Longevity after Injury Project

Subjective and Objective Environmental Factors’ Influence

and Spinal Cord Injury

Yue Cao, PhD, MSPHMedical University of South Carolina

Changing What’s Possible®

Acknowledgement• The contents of this presentation were developed

under a grant from the Department of Education, NIDRR grant numbers H133B090005, and grant SCIRF 11-006 and 09-001 from the South Carolina Spinal Cord Injury Research Fund (SCSCIRF). However, those contents do not necessarily represent the policy of the Department of Education or the SCSCIRF, and you should not assume endorsement by the Federal Government or the state of South Carolina.

longevityafterinjury.com


Accreditation• The Medical University of South Carolina is accredited by the Accreditation Council for

Continuing Medical Education (ACCME) to provide continuing medical education for physicians. The Medical University of South Carolina designates this live activity for a maximum of 1.0 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

• In accordance with the ACCME Essentials &Standards, anyone involved in planning or presenting this educational activity will be required to disclose any relevant financial relationships with commercial interests in the healthcare industry. This information is listed below. Speakers who incorporate information about off-label or investigational use of drugs or devices will be asked to disclose that information at the beginning of their presentation.

• The Center for Professional Development is an approved provider of the continuing nursing education by the South Carolina Nurses Association an accredited approver by the American Nurses Credentialing Center’s Commission on Accreditation



Continuing Education


The Center for Education and Best Practice is an approved provider of continuing nursing education by the South Carolina Nurses Association, an accredited approver by the American Nurses Credentialing Center’s Commission on Accreditation.

Only RNs are eligible to receive nursing contact hours

Each participant will receive two forms for CNE- Verification of attendance - Individual evaluation form

For all CNE sessions, in order to receive full contact hour credit for the CNE activities, you must:

- Be present no later than five minutes after starting time- Remain until the scheduled ending time- Complete and return the evaluation form at the end of the session

Continuing Nursing Education (CNE) credit:

A conflict of interest occurs when an individual has an opportunity to affect educational content about health care products or services of a commercial interest with which she/he has a financial relationship.

The planners and presenters of this CNE activity have disclosed relevant financial relationships with any commercial interests pertaining to this activity. A list of event sponsors and vendors may be found in your handouts or disclosure slide.

The Center for Education and Best Practice has conflict of interest disclosures on file for all presenters and planners.

Non-endorsement of ProductsProvision of this education activity by the Center for Education and Best Practice does not imply endorsement by the Center or SCNA of any commercial products displayed in conjunction with this activity. Commercial support does not influence the design and scientific objectivity of any Center educational activity.

Conflict of Interest


DisclosuresYue Cao has no financial or non-financial interest to disclose. Commercial Support was not received for this activity.



Learning Objectives1.Identify the strategies and mechanisms to measure objective environmental factors by linking Census Data with research projects.

2.Identify the relationship between SCI incidence and objective neighborhood disadvantage.

3.Identify the relationship between health outcomes after SCI and objective neighborhood disadvantage.

4.Identify the prevalence of self-perceived environmental barriers and their impacts on subjective health in persons with SCI.



Background



WHO International Classifications of Functioning, Disability and Health (ICF) model


Health Condition

Body Functions & Structures

Activity Participation

Environmental Factors Personal Factors

Contextual Factors

(Impairments) (Limitations)

(Disorder or Disease)

(Restrictions)


Background• Guided by ICF model, previous research indicates five environmental factors

that have impacts on people’s health: 1. The natural environment;

2. The environments of where one lives, works, and plays;

3. The availability of public and private services to support people’s daily living;

4. The socio-cultural perspective of local areas;

5. An area’s reputation for its esteem, and quality of material infrastructure.



Background• For example, many studies have found the places where people live were

associated with some health problems in general population: obesity, cardiovascular disease, diabetes, and depression.

• Comparing with general population, people with SCI may have more health problems, some secondary health conditions such as pressure ulcers and urinary tract infection (UTI).

• For those living with a long-term disability, the limitations and restrictions imposed by their environment might have greater impacts on their health.

• It is highly possible that environmental factors might have great impacts on the health of people with SCI.



Background• There are few studies that have investigated the differences in

SCI occurrence and the survivors’ health outcomes due to the places where people live.

• We have initiated three research projects focusing on the environmental factors’ impacts on SCI.



STUDY 1

Neighborhood Disadvantage and

SCI Incidence



Definition of neighborhood based on Census geographic entities


Legal/Administrative• States• Counties• Minor civil divisions• Incorporated places• Congressional

districts• School districts

Statistical• Census tracts• Metropolitan/Micropolitan

statistical areas• Urban areas• Census designated places



Blocks• Smallest units for 100% data tabulation• Cover entire nation since 1990 Census, and about 11

million Census blocks for Census 2010.• Nest within all other types of geographic areas• Generally bounded by visible features or boundaries

Block Groups• Smallest units for sample data tabulation • Population ranges from 600 to 3,000• Nest within Census tracts




Census Tracts• Small, relatively permanent and consistent

statistical subdivisions of a county• Size: around 4,000 people • About 74,000 census tracts for Census 2010• Defined nationwide for the first time in the

Census 2000




Zip Code Tabulation Areas (ZCTAs)• Calculated geographic entities based on USPS ZIP code

distribution

• Composed of whole Census blocks

• The most frequently occurring ZIP Code in a block used as the ZCTA code for this block

• Some addresses will end up with a ZCTA code different from their ZIP Code, but in most instances, the ZCTA code is the same as the ZIP Code



Definition of neighborhood in our study

In our studies, we chose census tracts as the neighborhood definition.

• Small, relatively permanent and consistent.

• Homogenous areas with similar economic, demographic characteristics and living conditions.

• Smallest unit that has all socioeconomic information available from Census Bureau.



Census Geocodes

All Census entities can be identified by a unique FIPS (Federal Information Processing Standard) code.

Every census block has a unique 15-digit value with four components:

SSCCCTTTTTTBBBB.

For example, the following value 450190006001013 represents the census block (1013) containing our university in SC (45), within census tract 6 of the county of Charleston (019).

How to get the Geocode for an address?a. Manually: American FactFinder (http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml )

b. GIS software.


http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml



Census Data SetsDecennial Census• SF1 and SF2 contain data on the questions asked of all households. • SF3 and SF4 have data from questions asked of a sample of households

only. --------Not available in 2010 Census• Public Use Microdata Sample (PUMS, 1% or 5% subsample)

American Community Survey• 1-year estimate• 3-year estimate• 5-year estimate• PUMS

Economic Census

Current Population Survey

State and Local Government data



Access to Census DataAmerican FactFinder: Gateway to Census statisticshttp://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml

American FactFinder download centerhttp://factfinder2.census.gov/faces/nav/jsf/pages/download_center.xhtml

TheDataWeb: An internet based distributed data network and provides access to over 828 datasets with new files added monthly. You can works with microdata records, longitudinal linked records, time series records, and aggregate records.http://thedataweb.rm.census.gov/index.html

Census FTP site: http://www2.census.gov/



http://factfinder2.census.gov/faces/nav/jsf/pages/download_center.xhtml

http://thedataweb.rm.census.gov/index.html

http://www2.census.gov/


Linking Census data with South Carolina SCI Surveillance System Registry (SCSCISR)

SCSCISR is a population-based registry of SCI in South Carolina. All non-federal hospitals in South Carolina are required to report discharge data on all hospitalizations involving SCI to the State Budget and Control Board through the uniform billing discharge data (UB-04).

Duplicate admissions were eliminated using personal identifiers. Out of state residents were excluded.



Linking Census data with South Carolina SCI Surveillance System Registry (SCSCISR)

Data in the SCSCISR are validated through verification of randomly selected medical charts and have been shown to be 99% accurate.

During 1998 to 2011, there were a total of 3,524 SCI injuries. Among them, 2,583 (73%) patients had been identified with valid census tracts geocodes, which were used to link with census tract level summary data of American Community Survey 2009 5-year estimates.



Measurement of neighborhood disadvantage

We downloaded the total civilian population, percentage of households with incomes below the Federal poverty threshold, and the percentage of female-headed households with children for each census tract.

We used the index of objective neighborhood disadvantage. The index summarized the prevalence of poverty and of mother-only households. We calculated the index by dividing each percentage by ten and then taking the mean of the two. One unit increase of the index represented an average of ten percent increases in the poverty and mother-only household prevalence.



Annual cumulative SCI incidence 1998-2011

Our 2,583 SCI cases were distributed in 711 census tracts (a total of 867 census tracts in SC). The annual cumulative SCI incidence between 1998 and 2011 at the census tract level range from 0 to 323 per million.


Neighborhood disadvantage index

Number of census tracts

Average SCI incidence

Overall SCI incidence

<=0.5 65 36.31 29.95

0.5-1 222 39.75 38.20

1-1.5 232 48.10 46.07

1.5-2 165 53.10 50.00

2-2.5 91 60.71 61.70

>2.5 92 67.21 65.26


STUDY 2:

Neighborhood Disadvantage and

Health Outcomes after SCI



Neighborhood disadvantage and health outcomes after SCI

Participants:

They were recruited from SCSCISR. Participants were eligible if they: (1) were > 18 years(2) were > one year post injury (3) had traumatic SCI with residual effects

Data were collected through mail-in survey between 2010 and 2013.

We surveyed a total of 833 participants. Among them, we found 488 participants (59%) having valid census geocode.



Self-Report Physical Health Questions1. Now thinking about your physical health, which includes physical

illness and injury, for how many days in the past 30 days was your physical health not good?

2. In the past year (12 months), how many times (if any) have you gone to a doctor for treatment of a medical problem of any kind (DO NOT INCLUDE ROUTINE CHECKUPS)?

3. In the past year (12 months), how many times (if any) have you been hospitalized for any reason?

4. In the past year (12 months), approximately how many days have you spent in the hospital?

5. In the past year (12 months), how many TOTAL times (if any) did you go to the emergency room because of illness or injury?



Self-Report Mental Health Questions

1. Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30

days was your mental health not good?

2. During the past 30 days, for about how many days have you felt sad, blue, or depressed?

3. During the past 30 days, for about how many of the days have you felt worried, tense, or anxious?



Other Measurements• Age at Injury• Years Post Injury• Racial/Ethnic Groups• Gender• Injury Severity• Household Income• Education


Table 2. Pearson correlation between neighborhood disadvantage index and health measures

Pearson Coefficients p-value

Days of physical health not good in the past month0.06

0.20

Times of doctor visits in the past year -0.07 0.14

Times of hospitalization in the past year 0.09 <.05

Days of hospitalization in the past year 0.08 0.09

Times of emergency room visit in the past year0.08

0.06

Days of mental health not good in the past month0.09

0.05

Days of feeling sad, blue, or depressed in the past month0.14

<.01

Days of feeling worried, tense, or anxious in the past month0.10

<.05

Table 3. Multivariate regression coefficients of neighborhood disadvantage

Before controlling for individual SES1

After controlling for individual SES2

b p-value b p-value

Days of physical health not good in the past month 0.98 0.12 0.46 0.47

Times of doctor visits in the past year -0.12 0.58 -0.07 0.76

Times of hospitalization in the past year 0.23 <0.05 0.17 0.13

Days of hospitalization in the past year 1.40 <0.05 0.01 0.99

Times of emergency room visit in the past year 0.17 0.24 0.10 0.50

Days of mental health not good in the past month 1.84 <0.01 1.34 <0.05

Days of feeling sad, blue, or depressed in the past month 2.19 <0.01 1.61 <0.05

Days of feeling worried, tense, or anxious in the past month 2.37 <0.01 1.96 <0.01

1. Controlling for race, gender, injury severity, age at injury, years post injury.

2. Two additional Individual SES indictors were added into the model: household income, and education degree.


Conclusion• The more disadvantaged a neighborhood, the

higher the SCI incidence.

• The neighborhood disadvantages were significantly related to negative health outcomes after SCI.

• Although parts of the neighborhood’s effects were explained by individual SES factors, neighborhood

disadvantages still had significant independent impacts on our participants’ mental health.



Limitations1. Objective measure, but not completely

• Census data collected for purposes other than health research.

• Subject to survey data biases.



Limitations2. People and places are dynamicOur current research is cross-sectional and does not capture:• Individual-level change

‒ Exposure ‒ Migration

• Neighborhood-level change‒ Development/deterioration‒ Economic change‒ Major events



Limitations3. Selection effectsThe location of individuals in different places is not random.• Residential segregation

‒ Socio-economic‒ Race-ethnicity

• Accommodation needs after rehabilitation may further drive selection.



ContributionsOur work extends rehabilitation research by incorporating multiple data sources:

Survey Surveillance Census

BRFSS GIS Regional



ContributionsClinically

Target prevention and intervention efforts in the areas with the greatest risk.

Public PolicyImproving communities is relevant to decreasing

disability• social inclusion• health equality



STUDY 3:

Self-Perceived Environmental Barriers and Subjective Health in Persons with

SCI--A Cohort Study



ParticipantsData were collected through self-report at baseline (2007-2009) and at a 5-year follow-up.

Participants were recruited based on the following criteria:• Traumatic SCI • Minimum of 1-year post-injury• 18+ years of age• Residual deficits

A total of 1,635 individuals completed both baseline and follow-up assessment.



Measures• The outcome variable, subjective health status, was

measured at both time 1 and 2, by two health status items from the Behavioral Risk Factor Surveillance System Survey Questionnaire developed by Centers for Disease Control and Prevention (CDC):

a) The first health item asked how many days physical

health was not good within the past 30 days; b) The second health item asked how many days mental

health was not good within the past 30 days.



Measures• Environmental barriers were measured at time 1 by the Craig

Hospital Inventory of Environmental Factors-Short Form (CHIEF-SF), a well-validated and reliable 12-item scale to measure the frequency and magnitude of environmental barriers perceived by individuals.

• There are 5 subscales identified from CHIEF-SF: Policy Barriers, Physical/Structural Barriers, Work/School Barriers, Attitudes/Support Barriers, and Services/Assistance Barriers.



MeasuresOther controlling variables used in this study were measured at time 1.

• Demographic indicators: gender, race, age at injury.

• Injury related indicators: years since injury, and injury severity.



Analysis

• We will present the statistics for each of the 5 CHIEF subscales.

• We will also dichotomize each subscale by using the cut-point of 1. If a subscale > 1, we will consider the participants encounter at least one barrier within this subscale items.



Analysis• In the multivariate regression models, we will use all

variables measured at time 1 to predict the health outcomes measured at time 2.

• In order to estimate more specifically the unique power of predictors, we will use lagged-Y-regressor (or lagged dependent variable) analysis for all the regression models. This method adds the outcome variable measured at time 1 in the models as a controlling variable.



Table 1.

Characteristics of Respondents (n=1635)


Table 1. Characteristics of Respondents (n=1635) Mean Standard Deviation %

Days physical health not good measured at time 16 8.4

NA

Days physical health not good measured at time 27 8.6

NA

Days mental health not good measured at time 16 8.2

NA

Days mental health not good measured at time 26 8.0

NA

Age at injury33 13.8

NA

Years since injury at time 113 9.6

NA

Male NA NA 74.0

Race

Non-Hispanic White NA NA73.0

Non-Hispanic Black NA NA21.8

Others NA NA5.3

Injury Severity

Non-ambulatory C1-4 NA NA9.8

Non-ambulatory C5-8 NA NA25.0

Non-ambulatory non-cervical NA NA34.0

Ambulatory NA NA31.3


Table 2.

Scores on CHIEF-SF subscales measured at time 1


Table 2. Scores on CHIEF-SF subscales measured at time 1

Variable N Mean product score Standard Deviation %

Policies subscale 1328 0.6 1.1 NA

Physical/structural subscale 1409 1.5 1.5 NA

Work/school subscale 1043 0.5 1.0 NA

Attitudes and support subscale 1373 0.8 1.1 NA

Services and assistance subscale 1205 0.8 1.0 NA

Policies subscale >1 1328 NA NA 19.7

Physical/structural subscale >1 1409 NA NA 46.0

Work/school subscale >1 1043 NA NA 13.2

Attitudes and support subscale >1 1373 NA NA 22.4

Services and assistance subscale >1 1205 NA NA 26.0


Table 3.

OLS Regression Analysis: Predicting Days of Physical Health

Not Good at Time 2


Table 3. OLS regression analysis: Predicting days of physical health not good at time 2

Unstandardized Standardp-value

Coefficient Error

Intercept -0.32 1.10 0.77

Days of physical health not good at time1 0.37 0.03 <.01

Male -0.08 0.58 0.90

Race (Ref: Non-Hispanic White)

Non-Hispanic Black -0.02 0.65 0.98

Others -1.39 1.12 0.21

Injury Severity (Ref: Ambulatory)

Non-ambulatory C1-4 1.49 0.89 0.10

Non-ambulatory C5-8 -0.37 0.69 0.59

Non-ambulatory non-cervical -0.26 0.63 0.68

Age at injury 0.07 0.02 <.01

Years since injury at time 1 0.07 0.03 0.01

Policies subscale at time 1 -0.05 0.29 0.87

Physical/structural subscale at time 1 0.42 0.20 0.03

Attitudes and support subscale at time 1 -0.09 0.27 0.74

Services and assistance subscale at time 1 1.07 0.31 <.01


Table 4.

OLS Regression Analysis: Predicting Days of Physical Health

Not Good at Time 2


Table 4. OLS regression analysis: Predicting days of mental health not good at time 2

Unstandardized Standardp-value

Coefficient Error

Intercept 2.53 0.97 <.01

Days of mental health not good at time1 0.44 0.03 <.01

Male -0.31 0.51 0.54

Race (Ref: Non-Hispanic White)

Non-Hispanic Black -0.64 0.57 0.26

Others -0.75 0.98 0.44

Injury Severity (Ref: Ambulatory)



Non-ambulatory non-cervical 0.02 0.55 0.97

Age at injury 0.00 0.02 0.95

Years since injury at time 1 -0.04 0.02 0.08

Policies subscale at time 1 -0.07 0.25 0.77

Physical/structural subscale at time 1 0.45 0.17 <.01

Attitudes and support subscale at time 1 0.13 0.24 0.59

Services and assistance subscale at time 1 0.64 0.28 0.02


Conclusion• The environmental barriers are prevalent among people with chronic SCI.

• Two CHIEF subscales, physical/structural barriers and service/assistance barriers, showed significant relationship with subjective physical health and mental health. These two subscales mainly reflect people’s physical and material environment, while the other two subscales are mostly social aspects of people’s environment.

• CHIEF proved valuable as a tool for measuring the effect of environmental factors as barriers on the lives of those living with a disability.



Limitations• First, our participants were selected through a clinical site, rather

than the population-based.

• Although the CHIEF-SF is a valid tool to measure environmental barriers, it does not take into account the environmental factors acting as facilitators to health.

• Third, we have an attrition rate of 36%, which is respectable considering the 5 years follow-up period, but we should be aware of the possibility of selection bias resulting from non-random loss of respondents.



Thank you!

Dr. Yue CaoEmail: [email protected]:1.866.313.9963

Websiteswww.helpafterdisability.comwww.longevityafterinjury.com


mailto:[email protected]

http://www.helpafterdisability.com/

http://www.longevityafterinjury.com/

Documents

Changing What’s Possible ® The Longevity after Injury Project Subjective and Objective Environmental Factors’ Influence and Spinal Cord Injury Yue Cao,