Association of multimorbidity and changes in health-related quality of life following myocardial infarction: a UK multicentre longitudinal patient-reported outcomes study

Abstract

Background: Multimorbidity is prevalent for people with myocardial infarction (MI), yet previous studies investigated single-health conditions in isolation. We identified patterns of multimorbidity in MI survivors and their associations with changes in HRQoL.

Methods: In this national longitudinal cohort study, we analysed data from 9566 admissions with MI from 77 National Health Service hospitals in England between 2011 and 2015. HRQoL was measured using EuroQol 5 dimension (EQ5D) instrument and visual analogue scale (EQVAS) at hospitalisation, 6, and 12 months following MI. Latent class analysis (LCA) of pre-existing long-term health conditions at baseline was used to identify clusters of multimorbidity and associations with changes in HRQoL quantified using mixed effects regression analysis.

Results: Of 9566 admissions with MI (mean age of 64.1 years [SD 11.9], 7154 [75%] men), over half (5119 [53.5%] had multimorbidities. LCA identified 3 multimorbidity clusters which were severe multimorbidity (591; 6.5%) with low HRQoL at baseline (EQVAS 59.39 and EQ5D 0.62) which did not improve significantly at 6 months (EQVAS 59.92, EQ5D 0.60); moderate multimorbidity (4301; 47.6%) with medium HRQoL at baseline (EQVAS 63.08, EQ5D 0.71) and who improved at 6 months (EQVAS 71.38, EQ5D 0.76); and mild multimorbidity (4147, 45.9%) at baseline (EQVAS 64.57, EQ5D 0.75) and improved at 6 months (EQVAS 76.39, EQ5D 0.82). Patients in the severe and moderate groups were more likely to be older, women, and presented with NSTEMI. Compared with the mild group, increased multimorbidity was associated with lower EQ-VAS scores (adjusted coefficient: -5.12 [95% CI -7.04 to -3.19] and -0.98 [-1.93 to -0.04] for severe and moderate multimorbidity, respectively. The severe class was more likely than the mild class to report problems in mobility, OR 9.62 (95% confidence interval: 6.44 to 14.36), self-care 7.87 (4.78 to 12.97), activities 2.41 (1.79 to 3.26), pain 2.04 (1.50 to 2.77), and anxiety/depression 1.97 (1.42 to 2.74).

Conclusions: Among MI survivors, multimorbidity clustered into three distinct patterns and was inversely associated with HRQoL. The identified multimorbidity patterns and HRQoL domains that are mostly affected may help to identify patients at risk of poor HRQoL for which clinical interventions could be beneficial to improve the HRQoL of MI survivors.

Trial registration: ClinicalTrials.gov NCT01808027 and NCT01819103.

Keywords: EQ5D; Health-related quality of life; Multimorbidity; Myocardial infarction.

Conflict of interest statement

Prof Gale reports grants from Abbot Diabetes, personal fees from Amgen, personal fees from AstraZeneca, personal fees from Bayer, grants from BMS, personal fees from Daiichy Sankyo, and personal fees from Vifor Phamra outside the submitted work. Dr Wilkinson reports a research grant from BMS. The other authors declare that they have no competing interests.

© 2021. The Author(s).

Figures

Fig. 1

Percent of patients with conditions in each multimorbidity class. Severe multimorbidity patients tended to have high levels of all co-morbidities, moderate multimorbidity patients tended to have hypertension, and diabetes, and mild multimorbidity were patients with few co-morbidities. Note COPD indicates chronic obstructive pulmonary disease; CVSD, cerebrovascular disease; PVD, peripheral vascular disease; CRF, chronic renal failure

Fig. 2

Temporal changes of HRQoL by multimorbidity class. The EQ-VAS scores range from 0 (worst) to 100 (best) health status with a difference of 7 points considered clinically meaningful. Patients with severe multimorbidity have worse health-related quality of life, shown by EQ-VAS scores, at baseline and through 12 months of follow-up

Fig. 3

Association of multimorbidity classes and EQ-5D dimensions, (Mobility, Self-care, Activities, Pain and Anxiety/depression), odds ratios and 95% confidence intervals (reference group, mild multimorbidity). Adjusting for age, sex, ethnicity (white versus other) smoking status (never vs ex or current), past medical history of MI, angina, diagnosis (STEMI or NSTEMI), revascularisation (percutaneous coronary intervention [PCI] vs. no PCI; coronary artery bypass graft [CABG] surgery vs no CABG surgery), cardiac rehabilitation (yes/no) and interactions of time and multimorbidity, and medications