Links between the amount of antipsychotic medication prescribed per population at general practice level, local demographic factors and medication selection

Abstract

BACKGROUND Antipsychotic medications are the first-line pharmacological intervention for severe mental illnesses(SMI) such as schizophrenia and other psychoses, while also being used to relieve distress and treat neuropsychiatric symptoms in dementia. Since 2014 many antipsychotic agents have moved to generic provision. In 2017_18 supplies of certain generic agents were affected by substantial price increases. Our aim was to examine factors relating to antipsychotic prescribing in general practices across England and how cost changes in recent years have impacted. METHODS The study examined over time the prescribing volume/prices paid for antipsychotic medication by agent in primary care. The NHS in England/Wales publishes each month the prescribing in general practice by BNF code. This was aggregated for the year 2018_19 using Defined Daily doses (DDD). Cost of each agent year-on-year was determined. Monthly prescribing in primary care was consolidated over 5 years(2013-2018) and DDD amount from WHO/ATC for each agent was used to convert the amount to total DDD/practice. RESULTS Description We included 5,750 general practices with practice population >3000 and with >30 people on their SMI register. In 2018_19 there were 10,360,865 prescriptions containing 136 million DDD with costs of £110 million at an average cost of £0.81/DDD issued in primary care. Effect of price In 2017_18 there was a sharp increase in overall prices. There was a gradual increase in antipsychotic prescribing over 2013-2019 which was not perturbed by the drug price increase in 2017/18. Regression Demographic factors The strongest positive relation to increased prescribing of antipsychotics came from higher social disadvantage, higher population density(urban), and comorbidities e.g. chronic obstructive pulmonary disease(COPD). Higher %younger and %older populations, northerliness and non-white (Black and Minority Ethnic (BME)) ethnicity were all independently associated with less antipsychotic prescribing. Prescribing Factors Higher DDD/general practice population was linked with higher %injectable, higher %liquid, higher doses/prescription and higher %zuclopenthixol depot. Less DDD/population was linked with general practices using higher %risperidone and higher spending/dose of antipsychotic. CONCLUSIONS Levels of antipsychotic prescribing at general practice level are driven by social factors/comorbidities. We found a contrasting link between depot prescriptions with higher DDD and risperidone prescriptions with lower DDD.