Multipronged Intervention Cuts Risky Prescribing
A 3-part quality improvement intervention targeted NSAID and antiplatelet prescribing in primary care and reduced inappropriate Rx by 37%.
Curbs excessive NSAID and antiplatelet prescribing
In the wake of a three-part intervention to curtail NSAID and antiplatelet prescribing to high-risk patients, hospital admissions for adverse events and inappropriate prescribing rates declined in eight out of nine measures, Scottish researchers reported.
A primary care practice intervention involving education, electronic health record (EHR) prompts, and short-term financial incentives resulted in a 37% reduction in inappropriate prescribing a year later, Bruce Guthrie, MB BChir, PhD, of the University of Dundee in Scotland, and colleagues reported in the New England Journal of Medicine.
The effort also led to fewer hospital admissions for gastrointestinal ulcer or bleeding (from 55.7-37.0 per 10,000 person-years, P=0.002) and for heart failure (from 707.7-513.5 per 10,000 person-years, P=0.02).
Guthrie and colleagues noted that, according to the IMS Institute for Health Care Informatics, the cost of avoidable drug-related events reached $19.6 billion in 2013.
"Previous observational studies suggest that adverse drug events have multiple underlying causes, so it is logical that through a combination of provider-facing and technological approaches, the study team reduced prescription of high-risk medications and saw a reduction in some clinically significant adverse events," Urmimala Sarkar, MD, of the University of California San Francisco, told MedPage Today in an email.
"Their methodologic approach accounted for some of the common biases seen in patient safety intervention research and makes me feel confident that their effect was related to their intervention," Sarkar added.
But, she pointed out, "their design doesn't tease apart which parts of the intervention -- the provider education, the EHR prompt, or the financial incentive -- was most important. I think that will be important as others try to adopt this approach."
The intervention, known as Data-Driven Quality Improvement in Primary Care (DQIP), was tested in a stepped wedge cluster randomized trial involving 33 physician-owned primary care practices in the Tayside region of Scotland.
As part of the trial design, the practices all received the same DQIP intervention, but were randomly assigned to one of 10 designated start dates spanning 14 months. The DQIP interventions were integrated into the practices' EHRs for 48 weeks, and data collection continued for an additional 48 weeks beyond the end of the intervention.
Prior to the intervention, 33,334 patients who had one or more risk factors that made them vulnerable to adverse drug events related to NSAIDs or antiplatelet medications. During the intervention period, 33,060 remained at risk.
The three-part intervention began with education: each practice received an 1-hour visit from a pharmacist for an educational session, written materials, and a newsletter to be delivered every 8 weeks.
Part two provided a financial incentive of an initial payment of $600 (U.S.), plus a one-time $25 bonus for each high-risk patient from the trial’s target group who underwent review during the intervention period. On average, each physician earned approximately $910 during the course of the trial.
Finally, for the third part of the intervention, an informatics tool identified patients in the EHRs of each practice who were in need of a medication review, and provided weekly updates on persistent high-risk prescribing and tracked progress with the patient reviews.
Overall high-risk prescribing activity was reduced from 3.7% to 2.2% (adjusted odds ratio 0.63, 95% CI 0.57-0.68, P<0.001). High-risk prescribing for patients who had ongoing prescriptions fell from 2.6% to 1.5% (adjOR 0.60, 95% CI 0.53-0.67, P<0.001), and new high-risk prescribing was reduced from 1.1% to 0.7% (adjOR 0.77, 95% CI, 0.68-0.87).
Although admissions for GI bleeding and ulcers and for heart failure were significantly reduced after the intervention, there was no significant effect on another adverse event tracked in the study -- acute kidney injuries -- although it did show a favorable trend (101.9-86.0 per 10,000 person-years, relative risk 0.84, 95% CI 0.68-1.09, P=0.19).
Eight out of nine high-risk prescribing measures showed reductions after implementation of the intervention. The one area without reduction was the rate of NSAID prescribing for people with heart failure.
The reductions persisted when the financial incentives ran out. At the end of the trial, 2.2% of patients who had risk factors for adverse drug events were still receiving prescriptions, and the rate stood at 2.0% 48 weeks after the end of the trial.
Guthrie and colleagues reported study limitations, including the possibility that other factors in the healthcare system would influence reductions in hospital admissions or high-risk prescribing.
Donnan reported financial relationships with Novo Nordisk, GlaxoSmithKline, Gilead Sciences, and the New Drugs Committee of the Scottish Medicines Consortium.
The trial was funded by the Scottish Government Chief Scientists Office.
Reviewed by Robert Jasmer, MD Associate Clinical Professor of Medicine, University of California, San Francisco and Dorothy Caputo, MA, BSN, RN, Nurse Planner
last updated 03.17.2016
Primary Source: New England Journal of Medicine
This article was first published on MedPage Today and reprinted with permission from UBM Medica.Free registration is required.
