Mortality from decreased food intake in older adults in nursing homes: A retrospective cohort study using electronic care record data

Abstract

Aim

To our knowledge, no studies have examined the risk of mortality in nursing homes using the routinely collected electronic care record information. We aimed to assess the risk of mortality using food intake data documented in electronic care records from Japanese nursing homes.

Methods

This academic–industry collaborative research included 814 older adults from five nursing homes in Japan. The study period was from January 1, 2016 to June 23, 2020. We used the rate of food intake data recorded by care workers in the electronic care records of the residents. We defined decreased food intake as the percentage decrease between admission and 90 days after admission into the nursing home, with cutoffs of 10%, 20%, and 30%.

Results

The number of older adults whose food intake decreased between admission and 90 days after admission was 114 (14.0%) at a 10% rate, 58 (7.1%) at a 20% rate, and 35 (4.3%) at a 30% rate. All three rates were significantly associated with mortality (10%: hazard ratio [HR] 2.02, 95% confidence interval [CI] 1.30–3.15; 20%: HR 3.03, 95% CI 1.68–5.47; and 30%: HR 4.50, 95% CI 2.15–9.44).

Conclusions

We found that the risk of mortality could conveniently be assessed from the food intake data collected routinely and comprehensively and documented in electronic care records in nursing homes. Our findings can be utilized in nursing homes, where medical staff are limited, to assess high-risk residents and to encourage a discussion regarding the management of end-of-life care.