Variations in quality of life in people with multimorbidity: A cross-sectional survey comparing MMQ1 and EQ-5D-5L

TL;DR

This study compares two questionnaires for measuring quality of life in people with multiple chronic conditions, finding one to be more effective.

Contribution

The study validates the MMQ1 questionnaire and demonstrates its superior ability to detect variations in quality of life compared to EQ-5D-5L.

Findings

MMQ1 detected age-related variations in quality of life not captured by EQ-5D-5L.

MMQ1 showed better discriminative ability, especially for distinguishing between poor and very poor self-rated quality of life.

Both measures found higher condition count and deprivation linked to poorer quality of life.

Abstract

The Multimorbidity Questionnaire (MMQ1) is a patient-reported outcome measure assessing quality of life (QoL) in people with multimorbidity. An English-language version was recently validated for use in the United Kingdom. This study examines: (1) Whether MMQ1 detects expected variations in QoL according to individual characteristics; (2) How MMQ1 compares with EQ-5D-5L in detecting such variations, and in discriminating between different levels of QoL.

A postal survey was distributed to 2753 patients with multimorbidity. Relationships between MMQ1 and EQ-5D-5L with six independent variables (long-term condition count, mental-physical multimorbidity, deprivation, self-rated QoL, age and sex) were examined using linear regression analyses. Discriminative ability was assessed using Receiver Operating Characteristic curves and sample size calculations with respect to consecutive classes of self-rated QoL.

597 responses were received (22%). Respondents had a mean age of 69.5 years and 48% were men. Higher long-term condition count, the presence of mental-physical multimorbidity and increasing deprivation were associated with poorer QoL on both measures. In addition, three MMQ1 domains demonstrated age-related variations in QoL that were not detected using EQ-5D-5L. MMQ1 exhibited superior discriminative ability to EQ-5D-5L, especially in distinguishing between individuals with ‘Poor’ vs ‘Very Poor’ self-rated QoL, where EQ-5D-5L was particularly weak.

MMQ1 detected expected variations in QoL according to individual characteristics, supporting known-groups validity. It was superior to EQ-5D-5L in its ability to detect age-related variations in QoL and to discriminate between different levels of self-rated QoL. MMQ1 has the potential to improve the measurement of QoL in people with multimorbidity.

Figures