Employment Among Working-Age Adults With Multiple Sclerosis: A Data-Mining Approach to Identifying Employment Interventions

Abstract

Purpose:To examine demographic, functional, and clinical multiple sclerosis (MS) variables affecting employment status in a national sample of adults with MS in the United States.Method:The sample included 4,142 working-age (20–65 years) Americans with MS (79.1% female) who participated in a national survey. The mean age of participants was 51.93 years (SD= 8.7). The dependent variable was employment status. The predictor variables included a set of demographic, functional, and MS variables.Results:The chi-squared automatic interaction detector (CHAID) analysis indicated that participants who were receiving Social Security Disability Insurance (SSDI) had significantly lower rates of employment (8.6%) than those who were not receiving SSDI (53.9%). For those not receiving SSDI, the most important factor predicting employment status was MS impact on physical functioning, as measured with the Multiple Sclerosis Impact Scale Physical Impact scale.Conclusion:The data-mining approach (i.e., CHAID analysis) provided detailed information and insight about interactions among demographic, functional, and clinical variables and employment status through the segmentation of the sample into mutually exclusive homogeneous subgroups. Implications for rehabilitation intervention, based on these subgroupings, are discussed.