OBJECTIVES: Multimorbidity, co-existence of two or more chronic conditions, is associated with high indirect costs. However, the extent to which differences in person-level characteristics among those with and without multimorbidity contribute to this high indirect cost burden is unknown. We examined factors that contributed to the excess indirect costs of multimorbidity using post-linear regression decomposition method.

METHODS: A cross-sectional study of employed adults (age 18-64 years; N = 11,663) using data from the 2015 Medical Expenditure Panel Survey was undertaken. Multimorbidity was identified from a list of 18 chronic conditions suggested by the Department of Health and Human Services. Annual indirect costs (i.e. productivity losses) were computed by multiplying annual missed workdays due to illness by daily wages. Blinder-Oaxaca decomposition quantified the relative contribution of differences in characteristics among those with and without multimorbidity to excess indirect costs of multimorbidity.

RESULTS: In our study, 27.7% employed adults had multimorbidity. Compared to those without multimorbidity, higher percentages of individuals with multimorbidity were older, obese, and reported poor physical and mental health. For example, a higher percentage of adults with multimorbidity (14.1% vs 3.5%, 5.9% vs 2.6%) reported poor physical and mental health respectively. In adjusted OLS regressions on both untransformed and log-transformed indirect costs, multimorbidity was associated with high indirect costs (β_adj: $906.37 vs 429.95, p<0.001 (untransformed); $2.56 vs $1.44, p<0.0001 (log-transformed)). Decomposition analysis showed that 45.2% of this difference was explained by differences in distributions of perceived physical health (29.2%), mental health (3.7%), and older age (7.4%) between those with and without multimorbidity.

CONCLUSIONS: Differences in poor perceived health and older age explained most of the excess indirect costs associated with multimorbidity. Our findings suggest that patient perception of an illness contribute to economic outcomes and interventions that incorporate patient-reported outcomes may improve both economic and humanistic outcomes among those with multimorbidity.