Risk factors for hospital admission in the 28 days following a community-acquired pneumonia diagnosis in older adults, and their contribution to increasing hospitalisation rates over time: a cohort study

Abstract

Objectives: To determine factors associated with hospitalisation after community-acquired pneumonia (CAP) among older adults in England, and to investigate how these factors have contributed to increasing hospitalisations over time.

Design: Cohort study.

Setting: Primary and secondary care in England.

Population: 39,211 individuals from the Clinical Practice Research Datalink, who were eligible for linkage to Hospital Episode Statistics and mortality data, were aged ≥ 65 and had at least 1 CAP episode between April 1998 and March 2011.

Main outcome measures: The association between hospitalisation within 28 days of CAP diagnosis (a 'post-CAP' hospitalisation) and a wide range of comorbidities, frailty factors, medications and vaccinations. We examined the role of these factors in post-CAP hospitalisation trends. We also looked at trends in post-CAP mortality and length of hospitalisation over the study period.

Results: 14 comorbidities, 5 frailty factors and 4 medications/vaccinations were associated with hospitalisation (of 18, 12 and 7 considered, respectively). Factors such as chronic lung disease, severe renal disease and diabetes were associated with increased likelihood of hospitalisation, whereas factors such as recent influenza vaccination and a recent antibiotic prescription decreased the odds of hospitalisation. Despite adjusting for these and other factors, the average predicted probability of hospitalisation after CAP rose markedly from 57% (1998-2000) to 86% (2009-2010). Duration of hospitalisation and 28-day mortality decreased over the study period.

Conclusions: The risk factors we describe enable identification of patients at increased likelihood of post-CAP hospitalisation and thus in need of proactive case management. Our analyses also provide evidence that while comorbidities and frailty factors contributed to increasing post-CAP hospitalisations in recent years, the trend appears to be largely driven by changes in service provision and patient behaviour.

Keywords: EPIDEMIOLOGY; PUBLIC HEALTH.

Figure 1

Mutually adjusted ORs (circles) with 95% CIs (lines) of hospitalisation in the 28 days after CAP for factors included in the final model*. The model also contained year of CAP diagnosis, but the results for year are not presented. *Baseline categories were age 65–69 years; condition or medication not present (for comorbidities, frailty factors, recent medications); unvaccinated/no record of vaccination (for influenza and pneumococcal vaccination) (CAP, community-acquired pneumonia; MI, myocardial infarction).