Self-rated health as a predictor of mortality and healthcare use in older adults at high risk of hospitalisation: a prospective cohort study in Sweden

Abstract

Objective: This study aimed to evaluate the predictive value of self-rated health (SRH) on mortality and healthcare use in older adults (aged ≥75 years) at high risk of hospitalisation in comparison to an objective measure of comorbidities, the Charlson Comorbidity Index (CCI).

Design: Prospective cohort study conducted within the research project 'Proactive Primary Care for Frail Elderly Persons'.

Setting: 19 primary care practices in south-east Sweden, between January 2018 and December 2019.

Participants: In total, 355 adults aged ≥75 years were included in the study. They were among the 11% older adults with the highest predicted risk of hospitalisation, as identified by a statistical prediction model for unplanned hospital admission.

Outcome measures: Outcomes were all-cause mortality and healthcare use measured as hospital care days and the number of physician visits in primary and secondary care. These were analysed for different groups of SRH and comorbidities measured using the CCI.

Results: SRH was grouped into Excellent/Very good, Good, Fair and Poor. The overall mortality rate was 26.5%. Compared with the Poor group, the adjusted HRs were significantly lower for Excellent/Very good (HR=0.2; 95% CI: 0.1 to 0.8, p=0.02) and Fair (HR=0.5; 95% CI: 0.3 to 1.0, p=0.04). Compared with the comorbidity group CCI 0-1, CCI 2-3 had an adjusted HR of 2.2 (95% CI: 1.1 to 4.6, p=0.03), CCI 4-5 had an adjusted HR of 2.6 (95% CI: 1.2 to 5.4, p=0.01) and CCI>5 had an HR of 4.9 (95% CI: 2.4 to 10.2, p<0.001). The number of hospital care days was 70% lower (adjusted relative difference=0.3; 95% CI: 0.1 to 0.8) for Excellent/Very good (3.9 days) compared with Poor (10.7 days). All groups of CCI diagnoses (2-3, 4-5 and >5) had significantly more hospital care days than CCI 0-1.For physician visits in secondary care, both the SRH Excellent/Very good (p=0.004) and Good (p=0.02) groups had significantly fewer visits compared with Poor. In the comorbidity groups, no statistical differences were found between CCI categories.

Conclusions: In a cohort of older adults at high risk of hospitalisation, the predictive value of SRH for risk stratification was limited. Objective health measures appeared to offer greater utility than SRH for guiding healthcare planning and tailoring interventions for vulnerable older adults in this cohort.

Trial registration number: Clinical Trials NCT03180606.

Keywords: Frailty; GERIATRIC MEDICINE; Hospitalization; Mortality; PUBLIC HEALTH.

Figure 1. Participant flow chart.

Figure 2. Mortality during 2 years of follow-up, comparing different groups of self-rated health.

Figure 3. Mortality during 2 years of follow-up, comparing different groups of Charlson Comorbidity Index.