Work productivity loss among adults aged 18–64 years with Multiple Sclerosis in the United States: A propensity score matched study

Abstract

Scarce data exists regarding United States (US) national-level work productivity loss among adults with Multiple Sclerosis (MS). To address this significant knowledge gap, we examined the national-level productivity loss among adults (18≤ age ≤64 years) with MS compared to propensity score matched non-MS controls. We adopted a retrospective, cross-sectional, matched cohort study design with pooled data from alternate years (2005/2007/2009/2011/2013/2015) of the Medical Expenditure Panel Survey (MEPS). We included adults who were employed and alive during the calendar year. Clinical Classification System code of "80" was used to identify individuals with MS. We matched adults with MS to non-MS adults utilizing propensity scores generated based on age, gender, and race/ethnicity using a greedy matching algorithm (8:1-digit matching). Missed workdays measured productivity loss of MEPS respondents. We selected Negative Binomial Regression (NBR) analysis as the count data model for this study. Analyses were conducted using SAS 9.4 and STATA 15.0 and accounted for the complex survey design of MEPS to generate US national-level estimates. The final propensity-score matched sample consisted of 104 and 312 (unweighted) adults with and without MS, respectively. US national-level mean [Standard Error (SE)] annual missed workdays among individuals with MS [8.94 (SE:1.59)] was significantly higher (p=0.001) compared to propensity score matched non-MS controls [3.15 (SE:0.40)]. After adjusting for several factors, NBR showed an approximately two-fold higher rate of missed work days among individuals with MS compared to propensity score matched non-MS controls (Incidence Rate Ratio: 1.98, 95% Confidence Interval: 1.18-3.33). Severity of pain, marital status, region, and hypertension also negatively impacted work productivity in this sample. Individuals with MS in the US experience significantly higher productivity loss compared to propensity score matched non-MS controls. Interventions (e.g., improved management of MS symptoms) are warranted to reduce productivity loss among individuals with MS.