Abstract
Objective
To assess the associations of race and socioeconomic status (SES) with pressure ulcers (PU) after accounting for health care access (HCA) among persons with spinal cord injury (SCI).
Design
Cross-sectional.
Setting
Large specialty hospital in the southeastern United States.
Participant
Persons with traumatic SCI who 1) had residual effects from their injury, 2) were 18 years or older at survey, and 3) were a year or more post-injury at survey (n=2,549).
Intervention
None.
Main Outcome Measures
Outcomes were measured by mail-in survey: having a current PU (yes vs. no), having a PU in the past year with or without reduced sitting time (no PU, no reduced sitting time, month or less, 5+ weeks), and having at least 1 PU surgery since SCI onset (yes vs. no).
Results
Of participants, 39.3% reported a PU in the past year, 19.9% had a current PU, and 21.9% reported having had surgery for a PU since their SCI onset. While race was preliminarily associated with each PU outcome, it became non-significant after controlling for SES and HCA. In each analysis, household income was significantly associated with PU outcomes after controlling for demographic and injury factors and remained significant after accounting for the HCA factors. Persons with lower income had higher odds of each PU outcome. HCA was not consistently related to PU outcomes.
Conclusions
Even after accounting for HCA, household income, a measure of SES, remained significantly associated with PU outcomes after SCI; however, race became non-significant.
Keywords: spinal cord injury, pressure ulcer, race, socioeconomic factors, health services accessibility
Pressure ulcers (PU) are one of the most common secondary conditions after spinal cord injury (SCI)1 and not only are associated with significant costs2 but can have serious consequences, as they are linked with early mortality.3 However, PUs are a preventable condition, and, therefore, the identification of risk factors and areas for prevention is essential.
Previous research has linked race4,5 and socioeconomic status (SES)6 with PUs after SCI. One previous study assessed the mediating relationship of SES, as measured by income and education, on the relationship between race and PUs.6 Results showed race was significantly associated with having a current PU and with having surgery to repair a PU since injury onset. After controlling for SES, race was still significantly associated with having surgery to repair a PU since injury onset. Additionally, SES was significantly associated with PU outcomes. However, that study only included white and black participants and did not assess the impact indicators of health care access (HCA) on the relationships of race and SES with PU outcomes.
Measures of HCA have been linked with various outcomes after SCI. Even immediately after injury, HCA, as measured by insurance type, has been linked with re-hospitalization after rehabilitation and length of stay during rehabilitation.7 Type of insurance has also been linked with psychological distress and participation, where persons with private insurance had less distress8 and were more likely to be participating in work and school.9 Saladin, Krause, and Adkins4 looked at barriers to being able to see a physician to heal a PU. They found not having enough money or insurance and not having transportation to be significant barriers. However, they did not assess how important these barriers were after controlling for important confounding variables. Kroll, Neri, and Ho10 found access to primary health care at baseline was associated with lower odds of PU at follow-up 1 year later. While there has been some research on HCA and outcomes after SCI, it has been mainly focused on insurance type, and the research specifically on PU outcomes has been limited. We will expand the measure of HCA to include not only insurance status, but also having a person thought of as a personal doctor, and not being able to afford care because of cost.
Purpose
The purpose of this study is to assess the association of race, SES, and HCA with PU outcomes after SCI, while controlling for demographic and injury factors. We hypothesize that 1) HCA will be significantly related to PU outcomes, and 2) after the addition of HCA, SES will still be significantly associated with PU outcomes.
Methods
Participants
After receiving approval from the Institutional Review Board, participants were identified through 3 different sources of records at a large specialty hospital in the southeastern United States: 1) SCI Model Systems patient database, 2) Model Systems registry, and 3) outpatient directory. Although participants were identified through one of the SCI Model Systems, our data was specifically collected for this study, and we do not utilize any of the data routinely collected by the SCI Model Systems. There were 3 inclusion criteria: 1) traumatic SCI with residual effects, 2) 18 or older at assessment, and 3) minimum of 1 year post-injury. Out of 3,669 potential participants meeting these criteria, 2,614 responded (71.2% response rate). After participation, 65 persons were determined ineligible for participation due to full recovery (16), non-traumatic injury (46), or less than 1 year post-injury at survey (3), leaving 2,549.
Procedures
All data were collected by mail-in survey. Participants were sent cover letters describing the study and alerting them of materials to be sent 4-6 weeks later. After the initial materials were sent, a second set of materials was sent to all non-respondents. Phone calls were initiated for non-respondents, and additional mailings were implemented for those who had misplaced or discarded materials but consented by phone to participate and requested an additional set of materials (no survey information was collected by phone). Extensive efforts were made to identify current addresses of all potential participants, including the use of multiple search engines. Participants were offered $50 remuneration for participation.
Measures
Measurements were taken by a mail-in survey tool. We used a subset of measures from a larger study of protective and risk factors associated with adverse health outcomes and secondary conditions after SCI.11 Specifically for this study, HCA, income, education, race, and PUs were examined.
Our outcome in this study was PUs, and 3 items were used to assess this outcome; 1) having 1 or more open PUs at present, 2) number of days in the past year where sitting time was adversely affected by a PU, and 3) history of at least 1 surgery to repair a PU since SCI onset. Both current PU and PU related surgery were treated dichotomously (i.e., none, 1 or more), whereas sitting time was grouped into 4 categories: 1) no PU in the past year, 2) had a PU but no reduced sitting time, 3) had a PU and a month or less of reduced sitting time, and 4) had a PU and more than a month of reduced sitting time. We only asked participants to report open PUs as it might be difficult for some participants to recognize a less severe PU, especially persons with pigmented skin.12 Therefore, all results in this study are referring to open PUs. This measurement has been used in previous studies of persons with SCI.6,13 Additionally, these PU measures have been found to measure similar constructs through confirmatory factor analysis.11
Demographic (race, sex, age, years post-injury, age at injury) and injury characteristics were assessed. Injury severity was categorized as 1) C1-C4, non-ambulatory; 2) C5-C8, non-ambulatory; 3) non-cervical, non-ambulatory; and 4) ambulatory regardless of level.6 Race was categorized as white, black, and other. There were 2 indicators of SES, education and income. Participants were asked the exact number of years of education they had completed, and the years were anchored with educational milestones (i.e., 12=High School certificate, 16=4 year college degree). Annual household income was asked consistent with the Behavioral Risk Factor Surveillance System (BRFSS),14 which was recoded into 3 categories: 1) <$25,000, 2) $25,000-74,999, and 3) >$75,000. Lastly, we included 3 measures of HCA from the BRFSS. Participants were asked 1) if they had any kind of health care coverage, 2) if they have someone they thought of as their personal doctor, and 3) if there was a time in the past 12 months they needed to see a doctor but could not because of cost.
Analysis
Preliminary descriptive analyses were used to summarize the participant sample in terms of PUs, demographic, injury, educational, SES, and HCA characteristics. The χ2 statistic was used to evaluate the statistical significance of categorical variables, whereas the t-test was used for continuous variables.
For the model building, we used multivariable logistic regression with 2 of the outcomes: current PU (yes vs. no), and ever had surgery for a PU (yes vs. no). Since the third outcome had 4 levels, we used a multinomial logistic regression model. We tested the proportional odds assumption, and this assumption was rejected, thus we used a generalized logits model. For each model, we used a 2 stage model building strategy. In the first stage of each model (Model 1), variables from Saunders et al.6 were entered (race, gender, injury severity, years post-injury, age at injury, education, household income). We wanted to see the effect of HCA on SES and race from the Saunders et al.6 article, so in the second stage (Model 2), HCA variables (insurance, personal doctor, could not see doctor b/c of cost) were added to Model 1. We used p<.05 as the cutoff for statistical significance in the final model. Hosmer-Lemeshow15 and global χ2 tests were used to assess goodness-of-fit of the binary logistic regression models, and Pearson chi-square was used for the multinomial logistic regression model. The C-statistic, measuring area under the Receiver Operating Characteristic curve, was used to assess discriminatory ability.15 Odds ratios (ORs) with 95% confidence intervals (CIs) are reported.
Results
Of 2,549 participants, 75.2% were male and 72.3% were white. Average age at survey was 45.6 (SD=14.3), and participants were, on average, 12.5 (SD=9.7) years post-injury. Of participants, 30.9% were ambulatory. Of those non-ambulatory, 10.8% had a C1-C4 level injury, 25.1% C5-C8, and 33.3% non-cervical. Having less than a high school degree was reported by 15.3% of participants, while 27.3% had at least a bachelor’s degree. Twenty-two percent of participants reported average household income of $75,000 or more and 39.2% less than $25,000.
Current PU
Twenty percent of participants had a PU at the time of survey. Through the preliminary descriptive analyses (not shown), with the exception of age at injury (p=0.4728), all demographic and injury variables were initially associated with having a current PU. Additionally, education and income were significantly related with having a current PU, but none of the HCA variables were significant.
In Model 1 (table 1), only injury severity and years post-injury were significant out of the demographic and injury variables. Both education and household income were significantly related to having a current PU. After the addition of HCA predictors in Model 2, education just missed significance (p=0.0600). Persons having an income of less than $25,000 had 2.03 the odds of having a current PU than those with an income of $75,000 or more (95% CI=1.42-2.91). There was no difference between the middle and high income categories. Also, persons who could not see a doctor because of cost had higher odds of a current PU than those who could (OR=1.46; 95% CI=1.02-2.01). Both Models 1 and 2 had acceptable goodness-of-fit as measured by the Hosmer-Lemeshow goodness-of fit-test and discriminatory ability above 0.70 (Table 1).
Table 1.
Logistic regression model results for having a current PU (yes vs. no)
| Characteristic | Model 1* | Model 2† | ||
|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Race (vs. white) | 0.4401 | 0.3686 | ||
| Black | 1.19 (0.91-1.56) | 1.22 (0.93-1.60) | ||
| Other | 1.07 (0.68-1.69) | 1.06 (0.67-1.69) | ||
| Gender (vs. female) | 0.0881 | 0.1398 | ||
| Male | 1.27 (0.97-1.66) | 1.23 (0.94-1.61) | ||
| Injury severity (vs. Ambulatory) | <.0001 | <.0001 | ||
| C1-C4, Non-A | 6.00 (3.85-9.33) | 5.93 (3.76-9.36) | ||
| C5-C8, Non-A | 5.78 (3.95-8.47) | 6.04 (4.08-8.95) | ||
| Non-C, Non-A | 6.10 (4.22-8.81) | 6.05 (4.16-8.81) | ||
| Years post-injury | 1.02 (1.01-1.03) | 0.0002 | 1.02 (1.01-1.03) | 0.0002 |
| Age at injury | 1.01 (1.00-1.02) | 0.0617 | 1.01 (1.00-1.02) | 0.0717 |
| Education (vs. Bachelor+) | 0.0138 | 0.0600 | ||
| <High School | 1.85 (1.27-2.69) | 1.66 (1.13-2.44) | ||
| High School/GED | 1.28 (0.91-1.80) | 1.24 (0.88-1.75) | ||
| <Bachelor | 1.38 (1.01-1.89) | 1.37 (1.00-1.88) | ||
| Household income (vs. >$75K) | <.0001 | 0.0001 | ||
| <$25K | 2.10 (1.48-3.00) | 2.03 (1.42-2.91) | ||
| $25K-$74,999 | 1.36 (0.96-1.92) | 1.34 (0.94-1.89) | ||
| Insurance (vs. yes) | 0.9018 | |||
| No | 0.97 (0.62-1.52) | |||
| Personal Doctor (vs. Yes) | 0.4665 | |||
| No | 0.81 (0.47-1.42) | |||
| Could not see a doctor b/c of cost (vs. no) | 0.0379 | |||
| Yes | 1.46 (1.02-2.01) | |||
MODEL 1: Hosmer-Lemeshow chi-squre=7.55, DF=8, p=0.4789; C-Statistic=0.722 (variables: race, gender, injury severity, years post-injury, age at injury, education, household income)
MODEL 2: Hosmer-Lemeshow chi-squre=6.93, DF=8, p=0.5446; C-Statistic=0.721(variables: race, gender, injury severity, years post-injury, age at injury, education, household income, insurance, personal doctor, could not see a doctor b/c of cost)
Reduced Sitting Time due to a PU in the Past Year
Sixty-one percent of participants reported having no PUs in the past year. There were 10.4% who reported having their sitting time reduced by more than a month due to a PU in the past year, 12.2% had sitting time reduced by a month or less, and 16.6% had a PU in the past year but no reduced sitting time. As in the analysis of current PU, reduced sitting time was not significantly related to any of the HCA variables in preliminary descriptive analyses but was related to all demographic and SES variables.
In Model 1, sex, injury severity, years post-injury, and income were significantly related to reduced sitting time in the past year (table 2). The odds that those with less than $25,000 in household income had more than a month of reduced sitting time rather than not have a PU in the past year were about 3.45 times the odds for those with $75,000 or more in household income (95% CI=2.12-5.62). In Model 2, household income remained significant. Additionally, not being able to see a doctor because of cost was associated with reduced sitting time due to a PU in the past year. The odds of someone who could not see a doctor because of cost having reduced sitting time of a month or less rather than not have a PU in the past year was 1.88 times the odds for those who could see a doctor (95% CI=1.24-2.87). Both Models 1 and 2 had acceptable goodness-of-fit as measured by the Pearson chi-square statistic (Table 2).
Table 2.
Multinomial logistic regression model results for having reduced sitting time due to a PU in the past year P
| Model 1* | Model 2† | |||||||
|---|---|---|---|---|---|---|---|---|
| 5+ WEEKS vs. No PU | Month or less vs. No PU | Never vs. No PU | 5+ WEEKS vs. No PU | Month or less vs. No PU | Never vs. No PU | |||
| OR (95% CI) | p-value | OR (95% CI) | p-value | |||||
| Race (vs. White) | 0.4541 | 0.3695 | ||||||
| Black | 1.16 (0.81-1.65) | 1.09 (0.78-1.52) | 1.29 (0.96-1.73) | 1.17 (0.82-1.68) | 1.05 (0.74-1.48) | 1.35 (1.00-1.82) | ||
| Other | 0.70 (0.36-1.35) | 0.77 (0.43-1.39) | 1.21 (0.76-1.92) | 0.74 (0.38-1.43) | 0.77 (0.43-1.41) | 1.24 (0.77-1.99) | ||
| Gender (vs. female) | 0.0002 | 0.0002 | ||||||
| Male | 1.61 (1.11-2.33) | 1.86 (1.31-2.64) | 0.88 (0.67-1.14) | 1.59 (1.10-2.30) | 1.84 (1.30-2.63) | 0.87 (0.67-1.14) | ||
| Injury Severity (vs. Ambulatory) | <.0001 | <.0001 | ||||||
| C1-C4, Non-A | 18.36 (8.32-38.16) | 3.74 (2.23-6.29) | 3.145 (2.09-4.74) | 18.76 (8.93-39.39) | 4.08 (2.39-6.97) | 3.17 (2.08-4.83) | ||
| C5-C8, Non-A | 14.98 (7.57-29.63) | 4.52 (3.00-6.82) | 3.31 (2.41-4.57) | 15.00 (7.51-29.96) | 5.04 (3.31-7.35) | 3.37 (2.42-4.70) | ||
| Non-C, Non-A | 17.87 (9.21-34.66) | 4.68 (3.17-6.90) | 2.21 (1.61-3.02) | 17.94 (9.21-34.94) | 4.93 (3.31-7.35) | 2.19 (1.59-3.02) | ||
| Years post-injury | 1.03 (1.02-1.05) | 1.01 (1.00-1.03) | 1.01 (0.99-1.02) | 0.0007 | 1.03 (1.02-1.05) | 1.01 (1.00-1.03) | 1.01 (0.99-1.03) | 0.0005 |
| Age at injury | 1.01 (1.00-1.02) | 1.01 (0.99-1.02) | 1.00 (0.99-1.01) | 0.2053 | 1.01 (1.00-1.02) | 1.01 (1.00-1.02) | 1.00 (0.99-1.01) | 0.1603 |
| Education (vs. Bachelor+) | 0.8558 | 0.8429 | ||||||
| < HS | 1.27 (0.77-2.10) | 1.08 (0.68-1.70) | 1.13 (0.75-1.69) | 1.26 (0.75-2.10) | 1.01 (0.63-1.61) | 1.05 (0.69-1.59) | ||
| HS | 1.11 (0.71-1.73) | 0.86 (0.57-1.30) | 0.98 (0.70-1.38) | 1.11 (0.71-1.75) | 0.84 (0.56-1.27) | 1.01 (0.71-1.43) | ||
| < Bachelor | 1.31 (0.87-1.98) | 1.08 (0.75-1.56) | 0.93 (0.68-1.27) | 1.32 (0.87-2.00) | 1.11 (0.77-1.60) | 0.94 (0.68-1.29) | ||
| Income (vs. $75,000+) | <.0001 | <.0001 | ||||||
| <$25K | 3.45 (2.12-5.62) | 2.79 (1.82-4.27) | 1.38 (0.97-1.97) | 3.19 (1.94-5.24) | 2.57 (1.67-3.69) | 1.29 (0.90-1.85) | ||
| $25-75K | 1.89 (1.17-3.04) | 1.60 (1.06-2.41) | 1.26 (0.91-1.74) | 1.81 (1.12-2.93) | 1.51 (1.00-2.28) | 1.22 (0.88-1.68) | ||
| Insurance (vs. yes) | 0.7540 | |||||||
| No | 0.72 (0.38-1.37) | 0.89 (0.52-1.52) | 1.03 (0.64-1.64) | |||||
| Personal doctor (vs. Yes) | 0.9770 | |||||||
| No | 1.03 (0.50-2.12) | 1.09 (0.59-2.00) | 0.92 (0.52-1.62) | |||||
| Could not see a doctor b/c of cost (vs. no) | 0.0170 | |||||||
| Yes | 1.59 (0.98-2.58) | 1.88 (1.24-2.87) | 1.37 (0.91-2.05) | |||||
MODEL 1: Pearson chi-square=6766.04, DF-6858, p=0.7832 (variables: race, gender, injury severity, years post-injury, age at injury, education, household income)
MODEL 2: Pearson chi-square=6669.91, DF=6723, p=0.6748 (variables: race, gender, injury severity, years post-injury, age at injury, education, household income, insurance, personal doctor, could not see a doctor b/c of cost)
Had a PU Surgery since SCI Onset
Of participants, 21.9% reported having had a surgery to repair a PU since their SCI onset. In preliminary descriptive analyses, all demographic and injury characteristics were significantly related to PU surgery. Additionally, education, income, and having someone as a personal doctor were also related to PU surgery.
In Model 1, race and age at injury became non-significant (table 3). As with the other 2 analyses, lower income was associated with increased odds of PU surgery compared with higher income ($75,000+). In Model 2, household income remained significant. Of the HCA variables, not having a personal doctor was associated with lower odds of PU surgery (OR=0.53; 95% CI=0.29, 0.97) than with those who had a personal doctor. Both Models 1 and 2 had acceptable goodness-of-fit as measured by the Hosmer-Lemeshow goodness-of-fit test and discriminatory ability above 0.70 (Table 3).
Table 3.
Logistic regression model results for having had surgery for a PU since SCI onset (yes vs. no)
| Model 1* | Model 2† | |||
|---|---|---|---|---|
| Characteristic | OR (95% CI) | p-value | OR (95% CI) | p-value |
| Race (vs. white) | 0.0503 | 0.0720 | ||
| Black | 1.27 (0.97-1.66) | 1.28 (0.97-1.68) | ||
| Other | 0.70 (0.43-1.15) | 0.73 (0.45-1.21) | ||
| Gender (vs. female) | 0.0186 | 0.0260 | ||
| Male | 1.38 (1.06-1.80) | 1.36 (1.04-1.78) | ||
| Injury severity (vs. Ambulatory) | <.0001 | <.0001 | ||
| C1-C4, Non-A | 6.87 (4.34-10.87) | 6.31 (3.95-10.98) | ||
| C5-C8, Non-A | 6.43 (4.32-9.57) | 6.05 (4.04-9.08) | ||
| Non-C, Non-A | 9.15 (6.24-13.40) | 8.77 (5.97-12.89) | ||
| Years post-injury | 1.06 (1.04-1.07) | <.0001 | 1.06 (1.04-1.07) | <.0001 |
| Age at injury | 1.00 (0.99-1.01) | 0.5451 | 1.00 (0.99-1.01) | 0.6008 |
| Education (vs. Bachelor+) | 0.2223 | 0.1807 | ||
| <High School | 1.33 (0.90-1.95) | 1.35 (0.91-1.99) | ||
| High School/GED | 1.37 (0.98-1.90) | 1.40 (0.99-1.96) | ||
| <Bachelor | 1.34 (0.99-1.81) | 1.37 (1.00-1.86) | ||
| Household income (vs. >$75K) | <.0001 | 0.0001 | ||
| <$25K | 1.98 (1.41-2.78) | 1.95 (1.37-2.76) | ||
| $25K-$74,999 | 1.30 (0.94-1.81) | 1.25 (0.90-1.75) | ||
| Insurance (vs. yes) | 0.2095 | |||
| No | 0.73 (0.45-1.19) | |||
| Personal Doctor (vs. yes) | 0.0378 | |||
| No | 0.53 (0.29-0.96) | |||
| Could not see a doctor b/c of cost (vs. No) | 0.0550 | |||
| Yes | 1.43 (0.99-2.07) | |||
MODEL 1: Hosmer-Lemeshow chi-square=7.55, DF=8, p=0.4786; C-Statistic=0.767 (variables: race, gender, injury severity, years post-injury, age at injury, education, household income)
MODEL 2: Hosmer-Lemeshow chi-square=7.11, DF=8, p=0.5244; C-Statistic=0.767 (variables: race, gender, injury severity, years post-injury, age at injury, education, household income, insurance, personal doctor, could not see a doctor b/c of cost)
Discussion
In our cohort, 39.3% of participants reported a PU in the past year, 19.9% had a current PU, and 21.9% reported having had surgery for a PU since their SCI onset. We investigated the relationships of SES and HCA with these PU outcomes and found, while income was consistently associated with PU outcomes, HCA was not. This study is the first to examine HCA with PU outcomes among an SCI population while also controlling for SES.
We found SES, specifically income, was significantly related to PU outcomes even after controlling for HCA. Previous research has found PUs were more common among persons with lower education and among those unemployed.5 We did not find a relationship between education and PU outcomes after controlling for income. One possible explanation for the differences in PU outcomes by income level is the lack of availability of certain resources that may help prevent PUs. Hunt et al.16 found those with low SES (low income, Medicaid/Medicare, less education) were more likely to have a standard wheelchair (either manual or power) rather than a customizable wheelchair, while persons with higher SES are more likely to have access to things like customizable wheelchairs, which are shown to reduce sitting pressure.17
HCA was related to PU outcomes in 2 different ways. First, in terms of historical PUs (i.e., having had surgery for a PU since SCI onset), HCA, specifically having a personal doctor, was associated with increased odds of PU surgery. This could be because someone with PUs in the past may have had more of a need to consult a physician than someone who has not had PUs as they will have needed treatment for their PU. Frequent doctor visits might enable someone to have a doctor they think of as their personal doctor. Findings of increased health care utilization have been reported in the SCI population, and persons who use the health care system more frequently may be more likely to have a person they think of as their personal doctor. Increased severity of PU has been associated with increased number of clinic or home visits.18 On the other hand, in terms of current PU outcomes, not being able to see a doctor because of costs is associated with increased odds of PU outcomes (i.e., having a current PU or reduced sitting time due to a PU in the past year). These results demonstrate that, similar to having reduced household income, those who are not able to seek care due to lack of resources were at increased risk of PU outcomes. Contrary to our hypothesis, we did not find insurance to be predictive of PU outcomes. A previous study of nursing home residents partially supports our results as they did not find Medicaid status to be predictive of PU incidence.19
After controlling for SES and HCA variables, race was not significantly associated with any of the PU outcomes. Previous research has shown blacks have a higher occurrence of PUs than whites,5,19 possibly due to increased difficulty in early detection of PUs due to skin pigmentation.20 However, previous studies did not control for the potential confounding effects of SES and HCA. Saunders et al.6 found that after controlling for SES (income, education), race was still significantly related to having a current PU; however, race became not significantly related to having a PU in the past year and having had surgery for a PU since SCI onset. We have shown that, while initially race was significantly associated with all 3 PU outcomes studied; these differences disappeared after the inclusion of both SES and HCA. Other results showed males had increased odds of having a PU in the past year and of having had surgery to repair a PU since SCI onset. Additionally, in each of our 3 analyses, increased injury severity was associated with higher odds of PU outcomes. Both of these results have also been seen in previous studies.5,21
Limitations
While this study shows the importance of SES in relation to PU outcomes after SCI, several limitations should be noted. First, we used self-report data from a mail-in survey for all measures included in this study. While it would be ideal to have clinical data on PU occurrence, we asked questions in a general manner and used a restricted time frame as to limit the recall bias in reporting PU outcomes. While we asked participants only about open PUs, some participants could have included all of their PUs in their responses. Second, we were not able to assess the severity or grade of PUs reported by participants, we were only able to ask that they report on open PUs. Third, these data were cross-sectional, so we were not able to establish the sequence of events or look at PU outcomes over time. Fourth, our participants were identified through a clinical setting in a large rehabilitation hospital, and thus our results may not be generalizable to all persons with SCI, especially those who were not seen in a rehabilitation hospital. Lastly, we were limited in our measures of HCA factors, and the inclusion of a broader scope of factors in this area would provide a better assessment of the relation of HCA with PU outcomes. Future measures could include the need for specialized equipment and access to health care providers who specialize in the treatment of persons with SCI.
Conclusion
These results have shown the importance of SES, specifically annual household income, in relation to PU outcomes after SCI. While most of the HCA variables were not significantly associated with PU outcomes, we did find having a personal doctor was associated with increased odds of having a PU in the past year and having a current PU. Health care providers should be aware of the increased risk of PU outcomes in lower SES populations, even after accounting for basic HCA factors such as insurance. Future studies should include a broader assessment of HCA factors and should use longitudinal data to assess the recurrence of PU outcomes among individuals with SCI. While we used three basic measures of HCA as used by the BRFSS, future research would benefit from more specific HCA information such as access to rehabilitation, therapy, and proper equipment.
Acknowledgments
Sponsorship: The contents of this publication were developed under grants from the Department of Education, NIDRR grant numbers H133G050165, H133B090005, and H133A080064, and the National Institutes of Health, grant number 1R01 NS 48117. However, those contents do not necessarily represent the policy of the Department of Education or NIH, and you should not assume endorsement by the Federal Government.
We certify that no party having a direct interest in the results of the research supporting this article has or will confer a benefit on me or on any organization with which we are associated AND, if applicable, we certify that all financial and material support for this research (eg, NIH or NHS grants) and work are clearly identified in the title page of the manuscript.
List of Abbreviations
- BRFSS
Behavioral Risk Factor Surveillance System
- CI
confidence interval
- HCA
health care access
- OR
odds ratio
- PU
pressure ulcer
- SCI
spinal cord injury
- SES
socioeconomic status
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Johnson RL, Gerhart KA, McCray J, Menconi JC, Whiteneck GG. Secondary conditions following spinal cord injury in a population-based sample. Spinal Cord. 1998;36:45–50. doi: 10.1038/sj.sc.3100494. [DOI] [PubMed] [Google Scholar]
- 2.Javitz HS, Ward MM, Martens L. Major costs associated with pressure sores. J Wound Care. 1998;7(6):286–290. doi: 10.12968/jowc.1998.7.6.286. [DOI] [PubMed] [Google Scholar]
- 3.Krause JS, Zhai Y, Saunders LL, Carter RE. Risk of mortality after spinal cord injury: an 8-year prospective study. Arch Phys Med Rehabil. 2009;90:1708–1715. doi: 10.1016/j.apmr.2009.04.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Saladin LS, Krause JS, Adkins RH. Pressure ulcer prevalence and barriers to treatment after spinal cord injury: Comparisons of 4 groups based on race-ethnicity. NeuroRehabil. 2009;24:57–66. doi: 10.3233/NRE-2009-0454. [DOI] [PubMed] [Google Scholar]
- 5.Chen Y, Devivo MJ, Jackson AB. Pressure ulcer prevalence in people with spinal cord injury: age-period-duration effects. Arch Phys Med Rehabil. 2005;86:1208–1213. doi: 10.1016/j.apmr.2004.12.023. [DOI] [PubMed] [Google Scholar]
- 6.Saunders LL, Krause JS, Peters BA, Reed KS. The relationship of pressure ulcers, race, and socioeconomic conditions after spinal cord injury. J Spinal Cord Med. 2010;33:387–395. doi: 10.1080/10790268.2010.11689717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.DeVivo MJ, Stover SL, Fine PR. The relationship between sponsorship and rehabilitation outcome following spinal cord injury. Paraplegia. 1989;27:470–479. doi: 10.1038/sc.1989.74. [DOI] [PubMed] [Google Scholar]
- 8.Tate DG, Forchheimer M, Daugherty J, Maynard F. Determining differences in post discharge outcomes among catastrophically and noncatastrophically sponsored outpatients with spinal cord injury. Am J Phys Med Rehabil. 1994;73:89–97. doi: 10.1097/00002060-199404000-00004. [DOI] [PubMed] [Google Scholar]
- 9.Tate DG, Stiers W, Daugherty J, Forchheimer M, Cohen E, Hansen N. The effects of insurance benefits coverage on functional and psychosocial outcomes after spinal cord injury. Arch Phys Med Rehabil. 1994;75:407–414. doi: 10.1016/0003-9993(94)90164-3. [DOI] [PubMed] [Google Scholar]
- 10.Kroll T, Neri MT, Ho PS. Secondary conditions in spinal cord injury: results from a prospective survey. Disabil Rehabil. 2007;29:1229–1237. doi: 10.1080/09638280600950603. [DOI] [PubMed] [Google Scholar]
- 11.Krause JS, Reed KS, McArdle JJ. A structural anlaysis of health outcomes after spinal cord injury. J Spinal Cord Med. 2010;33:22–32. doi: 10.1080/10790268.2010.11689671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Bennett MA. Report of the task force on the implications for darkly pigmented intact skin in the prediction and prevention of pressure ulcers. Adv Wound Care. 1995;8(6):34–35. [PubMed] [Google Scholar]
- 13.Krause JS, Carter RE, Pickelsimer E, Wilson D. A prospective study of health and risk of mortality after spinal cord injury. Arch Phys Med Rehabil. 2008;89:1482–1491. doi: 10.1016/j.apmr.2007.11.062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.CDC. Behavioral Risk Factor Surveillance System Survey Questionnaire. 1996 http://www.cdc.gov/brfss/questionnaires/pdf-ques/96brfss.pdf.
- 15.Hosmer DW, Lemeshow S. Applied logistic regression. 2. New York, New York: John Wiley and Sons; 2000. [Google Scholar]
- 16.Hunt PC, Boninger ML, Cooper RA, Zafonte RD, Fitzgerald SG, Schmeler MR. Demographic and socioeconomic factors associated with disparity in wheelchair customizability among people with traumatic spinal cord injury. Arch Phys Med Rehabil. 2004;85:1859–1864. doi: 10.1016/j.apmr.2004.07.347. [DOI] [PubMed] [Google Scholar]
- 17.Sprigle S, Chung KC, Brubaker CE. Reduction of sitting pressures with custom contoured cushions. J Rehabil Res Dev. 1990;27:135–140. doi: 10.1682/jrrd.1990.04.0135. [DOI] [PubMed] [Google Scholar]
- 18.Garber SL, Rintala DH. Pressure Ulcers in veterans with spinal cord injury: a retrospective study. J Rehabil Res Dev. 2003;40:433–442. doi: 10.1682/jrrd.2003.09.0433. [DOI] [PubMed] [Google Scholar]
- 19.Baumgarten M, Margolis D, van Doorn C, et al. Black/White differences in pressure ulcer incidence in nursing home residents. J Amer Geri Soc. 2004;52:1293–1298. doi: 10.1111/j.1532-5415.2004.52358.x. [DOI] [PubMed] [Google Scholar]
- 20.Lyder CH, Yu C, Emerling J, et al. The Braden Scale for pressure ulcer risk: evaluating the predictive validity in Black and Latino/Hispanic elders. Appl Nurs Res. 1999;12(2):60–68. doi: 10.1016/s0897-1897(99)80332-2. [DOI] [PubMed] [Google Scholar]
- 21.Garber SL, Rintala DH, Hart KA, Fuhrer MJ. Pressure Ulcer Risk in Spinal Cord Injury: Predictors of Ulcer Status Over 3 Years. Arch Phys Med Rehabil. 2000;81:465–471. doi: 10.1053/mr.2000.3889. [DOI] [PubMed] [Google Scholar]