The fifth independent variable, handicap, is described as a disadvantage that limits or prevents fulfillment of a role that is typical, depending on age, sex, and social and cultural factors (World Health Organization, 1980). Whiteneck, Charlifue, Gerhart, Overholster, & Richardson (1992) developed the Craig Handicap Assessment & Reporting Technique (CHART) as a way of quantifying the degree to which persons with a spinal cord injury in a community setting are handicapped. A higher score implied less handicap. A total handicap score on the CHART cannot be computed without the presence of all five subscale scores. The CHART was given as part of the questionnaire to the respondents of this study. Seventy-seven respondents (48.7%) had scores for all the five subscales, 65 persons (41.1%) had scores for four subscales, and 16 (10.1%) had scores for three or fewer subscales. Due to the loss of data in excluding those respondents answers on only four subscales, in these cases, missing subscale scores were replaced with the subscale mean for the sample for any person reporting a score on four CHART subscales. This conservative approach to estimating responses from incomplete data has been recommended as one approach to this dilemma in social sciences (SPSS, 1997). All respondents with three or fewer subscale scores were removed from the analysis (n = 16).
The sixth independent variable, years since injury, is included in the demographic information provided by participants. As described earlier, participants on the average were 12 years post-injury.
The seventh variable, health, is addressed through two questions. First, participants were asked to rate their overall physical health during the past year on a scale of 1 to 5, with 1 being poor and 5 being very good. All participants responded to this question. The mean overall physical health during the past year was 4.01 with 76% indicating their health was either good or very good. Second, participants were asked to indicate the number of times they had been hospitalized during the past year. All participants responded to this question. The number of hospitalizations reported included 0 (n = 118, 75%), 1 (n = 30, 19.0%), or at least 2 (n = 10, 6%).
Results
Measurement scores, frequencies, means, and standard deviations were initially computed. The data were analyzed using multiple regression analysis. An alpha level for all statistical testing of significance was preset at the .05 level.
Of the seven independent variables, two (marital status and personal control) contributed to the variance explained by the analysis (see Table 1). Perceived control was the single largest predictor of life satisfaction in the regression analysis and was significant at the .01 level with [Beta] = .26, t = 4.50, p [is less than] .01 (see Table 1), indicating that life satisfaction may be predicted by the amount of control a person perceives.
Table 1
Multiple Regression Analysis Results for Persons with Spinal Cord Injuries on the Dependent Variable Life Satisfaction (n = 107)
R [R.sup.2] Adj. [R.sup.2] F p
Full Model .64 .41 .36 8.72 < .001
Variables in Equation
Name [Beta] t p
Constant -.29 -.09 .93
Control .26 4.50 .001(**)
Communication -.94 -.89 .37
PASI .05 1.48 .14
Health .45 1.10 .28
Marital Status -1.98 -2.44 .02(*)
Handicap -.004 -.72 .47
Years Since .05 1.14 .26
Note. (*) p < .05 (**) p < .01.
In addition, marital status also contributed to life satisfaction and was significant at the .01 level with [Beta]= -1.98, t = -2.44, p [is less than] .02 (see Table 1). Because of the dummy coding of marital status, the meaning of this contribution was that married persons with spinal cord injury reported higher levels of life satisfaction than those participants who were not married at the time of the survey. Because of the correlation between the various measures, the strongest relationships were first accounted for in the multiple regression. The remainder of the independent variables failed to make a significant independent contribution to the variance explained.
In this study, the independent variables were moderately correlated with one another, and separately correlated with the life satisfaction (see Table 2.).
Table 2
Correlation Coefficient Matrix for Persons with Spinal Cord Injuries (n=158)
COMMUNIC CONTROL HANDICAP HEALTH LIFE SATIS
AGE -.081 .032 .094 .057 .095
COMMUNIC -.045 .022 -.015 -.200(*)
CONTROL .588(**) .183(*) .597(**)
HANDICAP .121 .414(**)
HEALTH .184(*)
LIFE SATIS
PAS
SATIS HHS
SATIS OTH
PASI SATIS HHS SATIS OTH YRS SINCE
AGE .022 -.106 -.071 .441(**)
COMMUNIC -.045 -.112 -.013 -.053
CONTROL .428(**) .154 -.035 .175(*)
HANDICAP .346(**) -.035 -.007 .185(*)
HEALTH .284(**) .082 .071 -.016
LIFE SATIS .366(**) .109 .011 .201(*)
PAS .301(**) .289(**) .072
SATIS HHS .231(*) -.215(*)
SATIS OTH -.136
Note. (*) p < .05 (**) p < .01.
Years since injury (r = .201, p [is less than] .05), control (r = .597, p [is less than] .01), health (r = .184, p [is less than] .01), personal assistance satisfaction (r = .366, p [is less than] .01), and handicap (r = .414, p [is less than] .01) were all positively correlated with life satisfaction. Conversely, the variable communication skills (r = -.200, p [is less than] .05) was negatively correlated with life satisfaction. The negative correlation of communication skills with life satisfaction is a function of the coding schema. For the communications measure, a lower score indicates greater clarity and directness in the exchange of information. Therefore, the greater the clarity and directness of the participant, the greater the life satisfaction.
Other Findings
Using a five-point scale, persons with a spinal cord injury indicated that they were more satisfied with the quality of personal assistance services from other sources ([Bar] x = 3.80) than they were with services from home health care agencies ([Bar] x = 2.92).
Persons with Paraplegia were found to be more effective communicators, perceived themselves as having more control, and encountered fewer limitations than did persons with Quadriplegia. That persons with paraplegia encountered fewer obstacles than persons with quadriplegia supports the findings reported by Whiteneck et al., (1992). However it is important to note that there were no significant group differences in life satisfaction based upon injury level. This result supports the findings by Woodrich and Patterson (1983). Persons with quadriplegia can be empowered by allowing them to exercise greater choice in self directing their assistance services and home life by removing barriers that obstruct the opportunity to participate in community life.
Sixty-five percent of participants (n = 103) identified themselves as being responsible for directing their own personal assistance services. There was a wide range in the hourly amount of personal assistance services received in a typical 24-hour period depending on who provided the service: spouse (n = 47, [Bar] x = 3.27, married respondents only), recruited (n = 75, [Bar] x = 2.88), agency (n = 74, [Bar] x = 2.81), parent (n = 66, [Bar] x = 1.94), relative (n = 59, [Bar] x = .67), and friend (n = 60, [Bar] x = .89). As severity of the Disability decreased so too did the use of personal assistance services.
The questionnaire allowed for more than one source of personal assistance services. The frequency with which each source was selected indicated that persons with a spinal cord injury received assistance from family members (spouse, parent, relative) at a higher rate than any other single category. The per hour rate of pay for personal assistance services decreased when family members provided assistance (spouse = $.40, parent = $.76, other relative = $.57) as compared to agency-provided assistance ($7.14). Medicare, Medicaid, private insurance, or a combination of these three were identified by 27% of the respondents as contributing to the payment for personal assistance services. One hundred and fifteen respondents (73%) reported annual out of pocket medical expenses of $5,000 or less.
Discussion
