Patients with mental health illness may be prescribed multiple medicines that can occasionally be hazardous in some situations. We plan to examine how many patients are affected by these potentially hazardous prescriptions using GP records. Similar work has been conducted to examine how common potentially hazardous prescribing is across the general population in primary care, but no work has been conducted specifically for patients with mental health problems or with a focus on medications to treat mental illness.
In order to detect potentially hazardous prescribing, we will be using prescribing safety indicators. A group of mental health experts have agreed on the definition of the indicators we will use. These indicators describe patterns of prescribing and drug monitoring that should not usually happen in practice because it increases the risk of harm. From the CPRD database, we will count the numbers of all patients who were exposed to a prescription or drug monitoring practice that they usually should not have (because of a diagnosis or prescription they already had). We will also explore the change of these numbers over time and examine the variation in these prescribing patterns between general practices.
The safety of prescribing specifically for patients with mental disorders in primary care has not been investigated in a large sample of the UK population. Therefore, this study aims to assess the safety of prescribing for patients with mental disorders in primary care across the UK in order to find targets for improving prescribing safety and reducing medication-related harm. Mental health related prescribing safety indicators will be used to evaluate the safety of prescribing. These indicators were developed using the Delphi consensus method to detect potentially hazardous prescribing and drug monitoring practice that may place patients with mental illnesses at significant risk of harm.
This study will be longitudinal to examine the change in the prevalence of different types of mental health related potentially hazardous prescribing and inadequate monitoring indicators between 2009-2019. In addition, a cross-sectional analysis will be conducted for 2019 data to explore the variations between practices and the reliability of each indicator. This study will also examine patient and practice-level characteristics that are associated with potentially hazardous prescribing and inadequate drug monitoring.
Data will be extracted from the Clinical Practice Research Datalink (CPRD GOLD) for all patients with the potential to trigger an indicator because of age, disease, and/or prescription. The proportion of eligible patients receiving a potentially hazardous prescription will be calculated with 95% confidence intervals (CI). Variations between practices will be quantified for each indicator and a composite indicator using mixed effects two level logistic regression model and will be estimated using the intraclass correlation coefficient (ICC). The reliability of each indicator and the composite indicator will be estimated using the Spearman-Brown Prophecy Formula. The associations between the composite and patient or practice characteristics will also be examined using multivariable two-level logistic regression.
Variations between practices in the prevalence of each prescribing safety indicator and composite indicator in 2018; reliability of each mental health related prescribing safety indicators and composite indicator at practice level; patient and practice level characteristics associated with the composite indicator; prevalence of each prescribing safety indicator and composite indicator from 2009-2019.
Douglas Steinke - Chief Investigator - University of Manchester
Wael Khawagi - Corresponding Applicant - University of Manchester
Alison Wright - Collaborator - University of Manchester
Darren Ashcroft - Collaborator - University of Manchester
Matthew Carr - Collaborator - University of Manchester
Richard Keers - Collaborator - University of Manchester
Tony Avery - Collaborator - University of Nottingham
Practice Level Index of Multiple Deprivation