Implementing the PINCER intervention in the East Midlands to reduce hazardous prescribing in general practices
Problem
Prescribing errors in general practice are an important and expensive preventable cause of safety incidents, morbidity, hospitalisations and deaths. A recent study in English general practices identified prescribing errors in 5% of prescription items, with one in 550 items containing a potentially life-threatening error. Given the prevalence and nature of prescribing errors (and the severity of harm associated with these), there is an urgent need to implement interventions known to correct these errors. A pharmacist-led information technology intervention for medication errors in general practice (‘PINCER’) has been shown in a randomised controlled trial to be effective in reducing rates of hazardous prescribing. The PINCER intervention involves searching GP clinical systems using computerised prescribing safety indicators to identify patients at risk from their prescriptions, and then acting to correct the problems with pharmacist support.
Approach
The aim of this study was to evaluate the large-scale rollout of PINCER across the East Midlands over an 18 month period using an interrupted time series analysis. A Quality Improvement Collaborative (QIC) approach was used to implement the intervention whereby an expert team, using structured activities, engaged clinicians and pharmacy teams to effect improvement in specific areas of practice. Support was provided in the form of education, feedback and opportunities for shared learning. Improvement was measured using anonymised routinely recorded data from general practices collected retrospectively, at three monthly time points. The data will be modelled using a multilevel logistic regression with practices and repeated measurements fitted as random effects. Multilevel logistic regression adjusting for baseline practice characteristics will be conducted and, in order to access the impact of the intervention over time, the interaction between intervention and calendar time will be fitted.
Findings
Twelve CCGs implemented the PINCER intervention, data were collected from 343 general practices between September 2015 and February 2017. A total of 2.9 million patient records were searched and 21,636 instances of potentially hazardous prescribing were identified using 11 prescribing safety indicators. Preliminary results show significant reduction in hazardous prescribing between the beginning and end of the study period for indicators associated with: gastrointestinal bleeding from 3.1%(16420/532343) to 2.0%(10501/528626) (Odds ratio 0.64: 95% CI 0.62, 0.65); heart failure from 2.5%(557/22513) to 1.7%(400/23804) (0.67 : 0.59, 0.77) and kidney injury from 2.4%(716/30354) to 1.7%(567/32312) (0.74: 0.66, 0.83).
Consequences
PINCER has been widely implemented across East Midlands’ general practices with reductions in hazardous prescribing. Given that one in 25 hospital admissions relate to hazardous prescribing at an annual cost of around £650 million, if the results of this initial analysis are supported by the findings from the time series analysis, then it is likely that the intervention will result in reductions in medication-related hospital admissions, and net cost savings to the NHS.