 |
 |
4 April 2012
A pragmatic randomised trial of 72 general practices in the UK found that pharmacist-led information technology intervention (PINCER) significantly reduced the frequency of some clinically important prescription and medication monitoring errors compared with simple computer-generated feedback. The study suggests that PINCER may be a cost- effective strategy for preventing medication errors if a cost of about £75 per error avoided over six months is considered acceptable.
Level of evidence:
Level 3 (other evidence) according to the SORT criteria.
Action
This study is of interest to primary care health professionals involved in prescribing and medicines optimisation. Although it has some limitations, this study adds to the evidence looking at the effectiveness of pharmacist-led interventions at reducing medication errors. It should be considered alongside other evidence when organising local services for medicines optimisation.
What is the background to this?
Building a safer NHS for patients: Improving Medication Safety (2004)’ is a report produced by the Department of Health that explores the causes and frequency of medication errors, highlights drugs and clinical settings that carry particular risks, and identifies models of good practice to reduce risk. The aim of this study was to test whether a pharmacist-led information technology intervention (composed of feedback, educational outreach, and dedicated support) could improve prescription safety and medication monitoring in general practices compared with a simple computer-generated feedback system.
What does this study claim?
The PINCER intervention was found to be an effective method for reducing a range of potentially harmful medication and monitoring errors. At six months’ follow-up, patients in the PINCER group were significantly less likely to have been prescribed a non-selective NSAID if they had a history of peptic ulcer without gastro-protection (odds ratio [OR] 0·58, 95% confidence interval [CI] 0·38 to 0·89); a beta-blocker if they had asthma (OR 0·73, 95% CI 0·58 to 0·91); or an ACE inhibitor or loop diuretic without appropriate monitoring (OR 0·51, 95% CI 0·34 to 0·78). PINCER was found to have a 95% probability of being cost effective if the decision-maker’s ceiling willingness to pay reaches £75 per error avoided at six months.
So what?
The evidence to support the use of pharmacist-led interventions to reduce medication errors in general practice is limited. This pragmatic study is a valuable addition to the evidence base. It suggests that a 12-week pharmacist-led intervention (PINCER), in addition to simple computer-generated feedback, was more effective than simple feedback alone for reducing a number of potentially hazardous medication and monitoring error. The probable direct cost of PINCER associated with avoiding one error over a six month period was estimated at about £75 per error.
The absolute reduction in risk of at least one prescription problem was reduced from 2.86% in the control group to 2.23% in the PINCER group; an absolute difference of 0.63%. The authors suggest that the reduction in errors seen is likely to be greater across GP practices generally as the control intervention used in this study is superior to routine models of care used in the UK. The absolute frequency of prescribing errors leading to patient harm is not known. As this study did not record actual adverse events (or any other patient-oriented outcomes) resulting from the errors it is not possible to say whether or not PINCER reduced patient harm by an amount that would be cost-effective according to currently acceptable cost-effectiveness thresholds. No attempt was made to quantify costs associated with adverse consequences of the medication or monitoring errors recorded in the study.
There are a number of other limitations in the study to consider when interpreting the results and its generalisation to UK practice. The study was designed to reduce allocation bias where possible. However, there may have been some selection bias as practices who declined from participating were more likely to be smaller and busier than those who agreed to participate. Not all medication/monitoring errors would have been identified in the study. Twelve were chosen (three as primary outcomes) and were restricted to those detectable by the general practice computer system, but only seven were included in the cost-analysis. An estimate of outcomes at 12 months was made to identify whether the benefit was retained over the longer term. Although the study was not powered for this time period, for one primary outcome measure (prescription of an NSAID without a PPI for patients with a peptic ulcer) the difference was no longer statistically significant.
Design
Multicentre, two-group, pragmatic, cluster randomised trial
Patients
72 UK general practices (computerised with electronic prescribing) with a combined list size of 480,942 patients.
Intervention and comparison
The practices were randomly allocated to either computer-generated simple feedback for at-risk patients (control) or a pharmacist-led information technology intervention (PINCER) composed of feedback, educational outreach, and dedicated support.
Outcomes and results
Outcomes were chosen ‘on the basis of medication management difficulties that are important in overall burden and severity of iatrogenic harm in primary care, and those detectable from general practice computer systems’. Primary outcomes were the proportions of patients at six months after the intervention who had had any of three clinically important errors shown in the table below.
Table: Prevalence of prescription and monitoring problems for the primary outcomes at six months’ follow-up
| Patient characteristics | Simple feedback | PINCER | Adjusted odds ratio (95% CI) |
| History of peptic ulcer prescribed an NSAID without a PPI (proton pump inhibitor)/history of peptic ulcer without a PPI | 86/2014 (4%) | 51/1852 (3%) | 0.58(0.38-0.89) |
| Asthma prescribed a beta-blocker/asthma | 658/22,224 (3%) | 499/20,312 (2%) | 0.73(0.58-0.91) |
| Aged ≥ 75 years receiving a long-term ACE inhibitors or loop diuretics without urea and electrolyte monitoring in the previous 15 months/aged ≥ 75 years receiving long-term ACE inhibitors or diuretics | 436/5329 (8%) | 255/4851 (5%) | 0.51(0.34-0.78) |
Secondary outcomes included the proportion of patients with at least one prescription problem/at risk of at least one prescription problem (2.30% PINCER, 2.86% control; adjusted OR 0.71, 95% CI 0.59 to 0.86).
PINCER was estimated to have a 95% probability of being ‘cost-effective’ if the ceiling of willingness to pay reached £75 per error avoided (at 6 months) or £85 per error avoided (at 12 months).
Sponsorship
Patient Safety Research Portfolio, Department of Health, England.
Further information can be found in the NPC e-learning section on improving safety and managing risk in medicines management systems which includes useful implementation resources for reducing medication errors.
Please comment on this rapid review using our feedback form.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.