Disability Population: Do Differences by Disability Type Matter to Public Health?

Background and Importance

Conceptual and experiential evidence strongly suggests that important differences exist among persons with disabilities based on disability type.  Historically, the health and disability scientific literature has operationalized disability as a single population or examined one or two disability subpopulations (i.e., physical or cognitive disabilities).  To the best of our knowledge, no studies have investigated whether there is statistical justification for subdividing the larger, heterogeneous group of people with disabilities into subpopulations. 

 Methods

Medical Expenditures Panel Survey (MEPS) data from 2004-2010 were pooled for these analyses.  We classified the population into one of six mutually exclusive groups (five disability and one non-disabled) to examine ten health outcomes: Pap test, mammogram, dental exam, flu shot, usual source of care, delayed or not received necessary care, self-reported physical and mental health status, and multiple chronic conditions.  For each health outcome, a likelihood ratio test was conducted to ascertain whether disability subpopulations should be modeled together

 Results

The overall joint tests for each outcome were significant at p < .01, suggesting that the disability subpopulations should be modeled separately.  Similarly, many pairwise combination tests suggest there are no two subpopulations that are appropriate to estimate together in a single regression equation, for any of the outcomes. 

 Conclusion

The disability population is not homogenous but instead comprised of empirically relevant subgroups.  These results suggest that each of five disability subpopulations should be estimated separately, confirming what we know about differences in access to services, unique needs, and specific distinct barriers to care among disability subpopulations.