Comorbidity among patients receiving opioid substitution therapy: examination of primary and secondary healthcare utilisation, cost of care, drug-drug interactions and mortality risk.

Date of Approval: 
2021-01-15 00:00:00
Lay Summary: 
People who are opioid-dependent are addicted to heroin and other similar ‘street drugs’. Opioid substitution therapy (OST) involves the prescribing of methadone or buprenorphine. Studies conducted worldwide have shown that OST recipients are more frequently hospitalised compared to people of similar age and gender who are not opioid-dependent. They also have an elevated risk of dying due to overdose when they start and after they cease OST. These patients differ in terms of their age and health that may result in a greater need for unplanned hospitalisations and increased overdose risk. However, the potential impact of comorbidities and interactions between OST and other drugs on overdose is poorly understood. Studies have shown that many deaths occur at an early age and are potentially preventable. Using patient records that are held in the Clinical Practice Research Datalink (CPRD), we aim to determine the rate of fatal/non-fatal overdose, to see whether chronic diseases and medications put some people at higher risk of overdose or death, whether this risk differs between people who receive methadone or buprenorphine, whether people who receive methadone make more or less use of healthcare services than people who are prescribed buprenorphine, and whether the costs of healthcare use are different between these groups. This study will investigate healthcare usage, medication safety and risk of dying among OST recipients, with a specific focus on cardiovascular, respiratory, digestive system, and kidney diseases.
Technical Summary: 
Although elevated risk of drug-related death occurring during opioid substitution therapy (OST) initiation and after cessation has been widely reported, limited research has been conducted on the potential impact that chronic diseases might have on hospitalisation due to non-fatal overdose and drug-related death. A substantial number of deaths in this population are premature and are potentially preventable. There are theoretical drug-drug interactions (DDIs) between OST and commonly prescribed medication (e.g. antidepressants, anxiolytics, antibiotics, antiarrhythmics, antiepileptics), that have not been examined and that may result in serious cardiovascular or respiratory-related events. The impact of specific comorbidities on OST choice, drug safety, and healthcare costs is poorly understood. We plan to utilise interlinked primary and secondary care records, area-level deprivation, and mortality data. We will delineate a cohort of patients who have received OST, based on prescription codes in the CPRD GOLD and Aurum databases. We will calculate and apply propensity scores to balance baseline covariates between methadone and buprenorphine recipients. Hazard ratios (and their 95% Confidence Interval) will be generated by fitting Cox proportional hazards models to examine whether patients who receive methadone versus those who receive buprenorphine have an elevated risk of non-fatal/fatal overdose. Moreover, we will include an interaction term to examine the potential effect modification of specific diseases (e.g. cardiovascular, respiratory, digestive system, or kidney diseases) on non-fatal/fatal overdose compared to opioid-dependent patients who were unaffected by these diseases. We will also assess the risk of harm that DDIs may have by applying the self-controlled case-series methodology. We will also apply regression models with interaction terms to investigate if primary and secondary care resource utilisation and associated costs in England differ between methadone and buprenorphine recipients and between patients with and without specific chronic conditions.
Health Outcomes to be Measured: 
 Non-fatal/fatal overdose  Adverse events due to drug-drug interactions: - Ventricular arrhythmia - Cardiac failure - Overdose  Primary and secondary care resource utilisation and associated costs of care
Application Number: 
20_000077
Collaborators: 

Darren Ashcroft - Chief Investigator - University of Manchester
Eleni Domzaridou - Corresponding Applicant - University of Manchester
Matthew Carr - Collaborator - University of Manchester
Roger Webb - Collaborator - University of Manchester
Thomas Allen - Collaborator - University of Manchester
Tim Millar - Collaborator - University of Manchester

Linkages: 
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation