The relationship between antipsychotic prescribing and the proportion of those with psychosis with a care plan: a cross-sectional study of practice-level data in England.

Problem

The NHS Mandate states that every person with a mental health condition should be offered a personalised care plan as soon as possible after diagnosis, including those living with schizophrenia and bipolar disorder. With the provision of a holistic approach to individuals living with psychotic disorders, it is possible that those with a care plan also have a more rationalised approach to the medication they are prescribed.In this study, we investigate the association between the proportion of people with a diagnosis of a psychotic disorder who have a care plan and the prescribing rate of antipsychotics and lithium in primary care practices in England.

Approach

Monthly primary care prescribing data for 2018, as well as practice age and sex profile, were downloaded from NHS Digital. Prescribing by practice was aggregated over the year and filtered for antipsychotics and lithium using British National Formulary (BNF) codes. Practice-level Index of Multiple Deprivation (IMD 2015) scores were obtained from Public Health England. Quality Outcomes Framework (QOF) data for 2017/18 were obtained for the proportions of each practice with a diagnosis of psychosis, dementia and depression. Multiple linear regression was used to examine the association between the proportions of those with psychosis with a care plan and antipsychotic/lithium prescribing, after adjusting for practice sex (% male), older age (% >65s), practice list size, practice-level deprivation (using the Index of Multiple Deprivation [IMD] score), and the proportions of each practice with psychosis, dementia and depression. Practice-level prescribing was defined as items of antipsychotics or lithium per 1000 registered patients in 2018.

Findings

On univariate analysis, overall antipsychotic/lithium prescribing rates were negatively associated with the proportion of individuals with psychosis with a care plan, with less prescribing in practices with a greater proportion of individuals with a care plan (incidence rate ratio [IRR] 0.88, 95% CI 0.83-0.93, for quintile [Q] 5 vs. quintile 1). The proportion of individual with a care plan remained an independent predictor of prescribing after adjusting for all the other variables included in the model (adjusted IRR [aIRR] 0.80, 95% CI 0.75-0.85 for Q5 vs. Q1). Incidentally, IMD score was also an independent predictor of prescribing, with more prescribing in more deprived practices (aIRR 1.67, 95% CI 1.58-1.76, Q5 vs. Q1).

Consequences

These results suggest that practices with greater proportions of individuals with a diagnosis of psychosis with a care plan tend to prescribe fewer antipsychotics/lithium. It is possible that practices that are more engaged with care plans are more likely to rationalise and optimise antipsychotic/lithium prescribing. Further work with individual-level patient datasets is required to confirm this result and to explore other possible underlying reasons for this association.