1. Changing What’s Possible ® The Longevity after Injury Project Subjective and Objective Environmental Factors’ Influence and Spinal Cord Injury Yue Cao, PhD, MSPH Medical University of South Carolina

2. 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

3. Changing What’s Possible ® 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 longevityafterinjury.com

4. Changing What’s Possible ® Continuing Education longevityafterinjury.com  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 Products Provision 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

5. Changing What’s Possible ® Disclosures Yue Cao has no financial or non-financial interest to disclose. Commercial Support was not received for this activity. longevityafterinjury.com

6. Changing What’s Possible ® Learning Objectives 1.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. longevityafterinjury.com

7. Changing What’s Possible ® Background longevityafterinjury.com

8. Changing What’s Possible ® WHO International Classifications of Functioning, Disability and Health (ICF) model longevityafterinjury.com Health Condition Body Functions & Structures ActivityParticipation Environmental Factors Personal Factors Contextual Factors (Impairments)(Limitations) (Disorder or Disease) (Restrictions)

9. Changing What’s Possible ® 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. longevityafterinjury.com

10. Changing What’s Possible ® 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. longevityafterinjury.com

11. Changing What’s Possible ® 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. longevityafterinjury.com

12. Changing What’s Possible ® STUDY 1 Neighborhood Disadvantage and SCI Incidence longevityafterinjury.com

13. Changing What’s Possible ® Definition of neighborhood based on Census geographic entities longevityafterinjury.com 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

15. Changing What’s Possible ® Definition of neighborhood based on Census geographic entities 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 longevityafterinjury.com

16. Changing What’s Possible ® Definition of neighborhood based on Census geographic entities 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 longevityafterinjury.com

17. Changing What’s Possible ® Definition of neighborhood based on Census geographic entities 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 longevityafterinjury.com

18. Changing What’s Possible ® 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. longevityafterinjury.com

19. Changing What’s Possible ® 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 ) http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml b. GIS software. longevityafterinjury.com

21. Changing What’s Possible ® Census Data Sets Decennial 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 longevityafterinjury.com

22. Changing What’s Possible ® Access to Census Data American FactFinder: Gateway to Census statistics http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml American FactFinder download center http://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://www2.census.gov/ longevityafterinjury.com

23. Changing What’s Possible ® 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. longevityafterinjury.com

24. Changing What’s Possible ® 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. longevityafterinjury.com

25. Changing What’s Possible ® 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. longevityafterinjury.com

26. Changing What’s Possible ® 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. longevityafterinjury.com Neighborhood disadvantage index Number of census tracts Average SCI incidence Overall SCI incidence <=0.56536.31 29.95 0.5-122239.75 38.20 1-1.523248.10 46.07 1.5-216553.10 50.00 2-2.59160.71 61.70 >2.59267.21 65.26

27. Changing What’s Possible ® STUDY 2: Neighborhood Disadvantage and Health Outcomes after SCI longevityafterinjury.com

28. Changing What’s Possible ® 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. longevityafterinjury.com

29. Changing What’s Possible ® Self-Report Physical Health Questions 1. 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? longevityafterinjury.com

30. Changing What’s Possible ® 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? longevityafterinjury.com

31. Changing What’s Possible ® Other Measurements Age at Injury Years Post Injury Racial/Ethnic Groups Gender Injury Severity Household Income Education longevityafterinjury.com

33. Table 2. Pearson correlation between neighborhood disadvantage index and health measures Pearson Coefficientsp-value Days of physical health not good in the past month 0.06 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 year 0.08 0.06 Days of mental health not good in the past month 0.09 0.05 Days of feeling sad, blue, or depressed in the past month 0.14 <.01 Days of feeling worried, tense, or anxious in the past month 0.10 <.05

34. Table 3. Multivariate regression coefficients of neighborhood disadvantage Before controlling for individual SES 1 After controlling for individual SES 2 bp-value b Days of physical health not good in the past month0.980.120.460.47 Times of doctor visits in the past year-0.120.58-0.070.76 Times of hospitalization in the past year0.23<0.050.170.13 Days of hospitalization in the past year1.40<0.050.010.99 Times of emergency room visit in the past year0.170.240.100.50 Days of mental health not good in the past month1.84<0.011.34<0.05 Days of feeling sad, blue, or depressed in the past month2.19<0.011.61<0.05 Days of feeling worried, tense, or anxious in the past month2.37<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.

35. Changing What’s Possible ® 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. longevityafterinjury.com

36. Changing What’s Possible ® Limitations 1.Objective measure, but not completely Census data collected for purposes other than health research. Subject to survey data biases. longevityafterinjury.com

37. Changing What’s Possible ® Limitations 2.People and places are dynamic Our current research is cross-sectional and does not capture: Individual-level change ‒Exposure ‒Migration Neighborhood-level change ‒Development/deterioration ‒Economic change ‒Major events longevityafterinjury.com

38. Changing What’s Possible ® Limitations 3.Selection effects The location of individuals in different places is not random. Residential segregation ‒Socio-economic ‒Race-ethnicity Accommodation needs after rehabilitation may further drive selection. longevityafterinjury.com

39. Changing What’s Possible ® Contributions Our work extends rehabilitation research by incorporating multiple data sources:  Survey  Surveillance  Census  BRFSS  GIS  Regional longevityafterinjury.com

40. Changing What’s Possible ® Contributions Clinically Target prevention and intervention efforts in the areas with the greatest risk. Public Policy Improving communities is relevant to decreasing disability social inclusion health equality longevityafterinjury.com

41. Changing What’s Possible ® STUDY 3: Self-Perceived Environmental Barriers and Subjective Health in Persons with SCI--A Cohort Study longevityafterinjury.com

42. Changing What’s Possible ® Participants Data 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. longevityafterinjury.com

43. Changing What’s Possible ® 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. longevityafterinjury.com

44. Changing What’s Possible ® 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. longevityafterinjury.com

45. Changing What’s Possible ® Measures Other 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. longevityafterinjury.com

46. Changing What’s Possible ® 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. longevityafterinjury.com

47. Changing What’s Possible ® 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. longevityafterinjury.com

48. Changing What’s Possible ® Table 1. Characteristics of Respondents (n=1635) longevityafterinjury.com

49. Table 1. Characteristics of Respondents (n=1635) MeanStandard Deviation% Days physical health not good measured at time 1 68.4 NA Days physical health not good measured at time 2 78.6 NA Days mental health not good measured at time 1 68.2 NA Days mental health not good measured at time 2 68.0 NA Age at injury 3313.8 NA Years since injury at time 1 139.6 NA Male NA 74.0 Race Non-Hispanic White NA 73.0 Non-Hispanic Black NA 21.8 Others NA 5.3 Injury Severity Non-ambulatory C1-4 NA 9.8 Non-ambulatory C5-8 NA 25.0 Non-ambulatory non-cervical NA 34.0 Ambulatory NA 31.3

50. Changing What’s Possible ® Table 2. Scores on CHIEF-SF subscales measured at time 1 longevityafterinjury.com

51. Table 2. Scores on CHIEF-SF subscales measured at time 1 VariableNMean product scoreStandard Deviation% Policies subscale 1328 0.61.1 NA Physical/structural subscale 1409 1.5 NA Work/school subscale 1043 0.51.0 NA Attitudes and support subscale 1373 0.81.1 NA Services and assistance subscale 1205 0.81.0 NA Policies subscale >1 1328 NA 19.7 Physical/structural subscale >1 1409 NA 46.0 Work/school subscale >1 1043 NA 13.2 Attitudes and support subscale >1 1373 NA 22.4 Services and assistance subscale >1 1205 NA 26.0

52. Changing What’s Possible ® Table 3. OLS Regression Analysis: Predicting Days of Physical Health Not Good at Time 2 longevityafterinjury.com

53. Table 3. OLS regression analysis: Predicting days of physical health not good at time 2 UnstandardizedStandard p-value CoefficientError Intercept -0.321.100.77 Days of physical health not good at time1 0.370.03<.01 Male -0.080.580.90 Race (Ref: Non-Hispanic White) Non-Hispanic Black -0.020.650.98 Others -1.391.120.21 Injury Severity (Ref: Ambulatory) Non-ambulatory C1-4 1.490.890.10 Non-ambulatory C5-8 -0.370.690.59 Non-ambulatory non-cervical -0.260.630.68 Age at injury 0.070.02<.01 Years since injury at time 1 0.070.030.01 Policies subscale at time 1 -0.050.290.87 Physical/structural subscale at time 1 0.420.200.03 Attitudes and support subscale at time 1 -0.090.270.74 Services and assistance subscale at time 1 1.070.31<.01

54. Changing What’s Possible ® Table 4. OLS Regression Analysis: Predicting Days of Physical Health Not Good at Time 2 longevityafterinjury.com

55. Table 4. OLS regression analysis: Predicting days of mental health not good at time 2 UnstandardizedStandard p-value CoefficientError Intercept 2.530.97<.01 Days of mental health not good at time1 0.440.03<.01 Male -0.310.510.54 Race (Ref: Non-Hispanic White) Non-Hispanic Black -0.640.570.26 Others -0.750.980.44 Injury Severity (Ref: Ambulatory) Non-ambulatory C1-4 1.020.790.19 Non-ambulatory C5-8 0.160.600.79 Non-ambulatory non-cervical 0.020.550.97 Age at injury 0.000.020.95 Years since injury at time 1 -0.040.020.08 Policies subscale at time 1 -0.070.250.77 Physical/structural subscale at time 1 0.450.17<.01 Attitudes and support subscale at time 1 0.130.240.59 Services and assistance subscale at time 1 0.640.280.02

56. Changing What’s Possible ® 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. longevityafterinjury.com

57. Changing What’s Possible ® 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. longevityafterinjury.com

58. Changing What’s Possible ® Thank you! Dr. Yue Cao Email:caoyu@musc.educaoyu@musc.edu Toll-free:1.866.313.9963 Websites www.helpafterdisability.com www.longevityafterinjury.com longevityafterinjury.com