Assessment of economic burden of fatigue in adults with multiple sclerosis: An analysis of US National Health and Wellness Survey data

Abstract

BackgroundFatigue, a common disabling symptom in multiple sclerosis (MS), is reported by the majority of patients. However, evidence on the economic burden of fatigue in MS by fatigue status is limited. This study aimed to evaluate the economic burden of fatigue, including healthcare resource utilization (HCRU), labor force participation, and Work Productivity and Activity Impairment (WPAI), among adults with relapsing-remitting MS (RRMS) by low fatigue (LF) vs high fatigue (HF) and compared with adults without MS. MethodsThis cross-sectional, retrospective, observational study included pooled data from the 2017 and 2019 US National Health and Wellness Survey. The RRMS sample included respondents aged ≥18 years who reported being diagnosed with MS by a healthcare provider (HCP) and reported having RRMS. Non-MS controls included respondents aged ≥18 years who did not report being diagnosed with MS by an HCP. Fatigue was measured using the Modified Fatigue Impact Scale-5 (MFIS-5). Outcomes included HCRU (HCP visits, emergency department visits, and hospitalizations in the past 12 months), labor force participation (yes vs no), WPAI (absenteeism, presenteeism, total work productivity impairment, and activity impairment), and annualized costs (direct medical, indirect, and total). Respondents with RRMS were propensity-score matched to non-MS controls (ratio 1:3). RRMS respondents were categorized as having LF (MFIS-5<15; RRMS+LF) and HF (MFIS-5≥15; RRMS+HF). Bivariate analysis compared matched non-MS controls, RRMS+LF, and RRMS+HF. Multivariable analyses were conducted among RRMS to evaluate associations between fatigue (continuous variable) and outcomes. ResultsOverall, 498 respondents with RRMS (RRMS+LF, n=375; RRMS+HF, n=123) and 1494 matched non-MS controls were included. RRMS+HF and RRMS+LF had more HCRU in the past 12 months than non-MS controls, whereas RRMS+HF had greater HCRU than RRMS+LF (all p<0.05). WPAI was also higher among RRMS+HF and RRMS+LF, compared with non-MS controls, as well as higher in RRMS+HF vs RRMS+LF (all p<0.001). RRMS+HF had significantly higher annualized direct medical costs than RRMS+LF and matched non-MS controls ($19,978 vs $10,656, p=0.007; vs $8,048, p<0.001). Among employed respondents, RRMS+HF and RRMS+LF had higher annualized indirect costs than non-MS controls, with RRMS+HF also having higher annualized indirect costs than RRMS+LF ($23,647 vs $13,738 vs $8,001; all p<0.01); total annualized costs were higher in RRMS+HF and RRMS+LF, compared with non-MS controls, as well as RRMS+HF vs RRMS+LF (all p<0.01). In multivariable models, fatigue was significantly and positively associated with the number of HCP visits in the past 12 months (p=0.002); not participating in the labor force (p<0.001); and absenteeism, presenteeism, total work productivity impairment, and activity impairment (all p<0.001). ConclusionRRMS poses a substantial economic burden on patients and society, and this burden is disproportionately associated with HF.