What is the effect of co-morbid depression on the use of unscheduled hospital care by people with a long term condition?

Problem

The primary healthcare system is reported to be less able to support people living with one or more long term conditions (LTC) where they are also experiencing a mental health problem. This may be evidenced by additional use of unscheduled hospital care by people with co-morbid mental and physical health problems. We wanted to quantify the size of this effect as part of a programme of research evaluating integrated care approaches that address both mental and physical health problems.

Approach

The study used a combined routine clinical database covering the whole of the adult population (age 18 and over) of Sheffield, a medium sized city in England (n=469,368). The bespoke dataset combined extracts from routine NHS records from primary and secondary care and linked at the person level. Cross-sectional data on all GP-recorded LTCs including depression as defined by QOF coding rule sets were included, together with longitudinal data on the use of scheduled and unscheduled hospital care. Analyses identified the extent of GP identified comorbidity between each LTC and depression, and the extent to which patients with LTCs and co-morbid depression used unscheduled care over a twelve month period in comparison to those with a LTC only. Logistic regression was used to determine the effects of depression after controlling for other potential confounding variables.

Findings

75,107 (16%) had a recorded diagnosis of depression. Of all patients with an identified LTC 20.9% also had a diagnosis of depression. Of all patients diagnosed with depression 50.9% had at least one LTC. The proportion of people who accessed unscheduled care increased as the number of LTCs increased. The proportion of individuals who had one or more 'unscheduled attendances' at hospital was 24% for those with a physical health LTC only, compared to 31.5% for those with a LTC and comorbid depression. This was a statistically significant association (p<.001). A logistic regression model was developed that controlled for a number of confounding factors. Those patients with depression were 1.59 times more likely to make use of unscheduled care over the twelve month period, after controlling for other variables.

Consequences

In this current study, for people living with a LTC, the addition of depression increases use of unscheduled care such as emergency department attendance. Primary care approaches that integrate self-management of LTCs and depression may be particularly useful given the extent and impact of this comorbidity.