Impact of hospital admission upon patterns of primary care prescribing

Problem

Prescribing of multiple medications is becoming increasingly common in primary care due to an ageing, multi-morbid population, and a culture of single-condition guideline-driven prescribing. Such complex prescribing may be further influenced by hospital admissions leading to changes in patient’s usual medicines. However, little is known about the impact of hospitalisation on prescribing in UK clinical practice.

Approach

A retrospective cohort study of 100,000 randomly selected adult patients admitted to hospital in 2014 was conducted using data from the Clinical Practice Research Datalink (CPRD). Weekly drug counts were calculated from one-year prior to one-year post hospital admission. Admissions were stratified by urgency (elective vs. emergency), and difference in drug counts by reason for admission and hospitalisation speciality were investigated. Age, gender, socioeconomic status, and long-term morbidities were also examined.

Findings

Of the 99,288 eligible hospital admission, 38.0% were emergency admissions, and 61.9% elective. The most and least common reason for emergency admissions were for cardiovascular conditions (12.8%) and neoplasms (2.1%), respectively, and for elective, gastrointestinal conditions (21.3%). and mental health conditions (0.3%), respectively. For treatment speciality, the most frequent was general medicine (36.2%) and orthopaedics (17.0%), and the least, obstetrics and gynaecology (4.0%) and geriatrics (0.1%), for emergency and elective admissions, respectively.Emergency attendees were prescribed more drugs one-year prior to hospitalisation compared to elective attendees (2.5±2.6 vs. 2.08±2.2). For both emergency and elective admissions, drug count increased in the 6 months preceding hospitalisation (0.28±1.68 vs. 0.19±1.49 drugs), remained stable 6 weeks after discharge (-0.01±1.74 vs. 0.01±1.15 drugs), before falling 6 months’ post admission (-0.23±2.33 vs. -0.24±1.85 drugs). Over the two years’, drug count fell by -0.09±2.53 and -0.15±2.16 drugs, for emergency and elective admissions, respectively.Similar patterns were generally observed when stratified by reason for admission. The greatest reductions in medications 6 months after hospitalisation, compared with admission, were for respiratory (-0.50±2.75 drugs) and musculoskeletal (-0.26±1.87 drugs) conditions for emergency and elective attendees, respectively. In contrast, emergency admissions for cardiovascular disease had an increase in drug count 6 months after hospitalisation, compared to admission (0.42±2.85 drugs). By treatment speciality, there was an increase in drug count between admission and 6 weeks after for emergency attendances treated in other medicine (0.23±2.27 drugs), though this difference had gone 6 months’ post admission (0.01±2.90 drugs). Comparable patterns were observed for other treatment specialities.

Consequences

Initial findings from this study indicate that prescribing increased prior to hospitalisation, with a short period of higher levels of prescribing after admission, followed by a steady decline, with fewer drugs being prescribed one-year following hospitalisation, compared to one-year prior, in both emergency and elective hospital admissions. This is in contrast to previous studies which have found an overall increase in drug count following hospital admission.