Study Design and Paired Comparisons: Individualized Education Fails to Change Practice—Or Was It Only Poor Matching?

We should commend Malone et al for submitting this AHRQ-supported* study [1] for publication when a flaw in its design or execution could be the authors’ main reason for concluding that “the current study was not able to demonstrate a significant beneficial effect of the educational outreach program on [the primary performance outcome measure].” This blog’s “Back-to-School” service campaign did not exclude studies reporting negative outcomes because these studies can potentially inform continuing education in the health professions (CEhp) as much as positive studies can.

CEhp/CME educational proposals, audience-generation strategies, and outcomes reports now specify relevant “target audiences,” recognizing that not all practitioners with a certain degree, specialty, or other professional demographic description would benefit from the same educational activity or design. With this more recent recognition of the importance of targeting specific clinicians and learning about their needs has come greater recognition that many CE participants should not be included in aggregated data. This is even truer in studies with matched pairs, where the step of greatest importance lies in setting match criteria. On September 15th, I discussed an opioids-education study where matching criteria were so stringent that the authors were not able to match certain participants (physicians in the intervention group), and these participants’ data and group assignments were handled nicely and reported clearly in the paper [2] (see post at http://fullcirclece.blogspot.com/2015/09/eight-year-canadian-study-on-opioid.html).

Conversely, the first result listed in this study’s abstract indicates a matching flaw for a study on education on drug-drug interactions (DDIs): “The 2 groups were significantly different with respect to age, profession, specialty, and geographic region.” This finding undermines other benefits to the study, namely, that large samples (19,606 prescribers) were recruited to both groups (educational intervention vs. control) and matched on prescribing volume. Individualized education (also known as academic detailing) was delivered by trained pharmacists as clinical consultants who met with prescribers to “provide one-on-one information … promote evidence-based knowledge, create trusting relationships, and induce practice change.” This study’s performance (behavioral) measure was a reduced rate of prescribing potential DDIs. The prescribing of 25 clinically important, potential DDIs increased more in the intervention group than it did in the control group.

In conclusion, when we look at this presumably negative finding, we are left to wonder whether the educational intervention was not effective—or whether a better matching process might have revealed different results on reducing potential DDIs and improving health care quality and utilization. One could argue that with nearly 20,000 prescribers in both samples, more matching criteria could have been applied without sacrificing so many data points that results would be inconclusive. The study’s design as a retrospective study could also explain recruitment and matching practices. In social sciences research (including educational outcomes research), a core expectation is generalizability of a sample to a population of interest; when reasonably achieved, generalizability lets us apply findings to practical needs and future decisions. 

Recall the study conclusion quoted above: “The current study was not able to demonstrate a significant beneficial effect …” (emphasis added). A secondary analysis with different pair-matching practices might yet inform national initiatives in improving quality while reducing costs through academic detailing, both of which help patients. Now let’s remember to thank Malone, Liberman, and Sun for sharing their data and methods with the healthcare quality and educational research communities in the Journal of Managed Care & Specialty Pharmacy.

* AHRQ = United States Agency for Healthcare Research and Quality

References cited:
1. Malone DC, Liberman JN, Sun D. Effect of an educational outreach program on prescribing potential drug-drug interactions. J Manag Care Pharm. 2013;19(7):549-557. http://www.ncbi.nlm.nih.gov/pubmed/23964616. [Featured Article]
2. Kahan M, Gomes T, Juurlink DN, et al. Effect of a course-based intervention and effect of medical regulation on physicians’ opioid prescribing. Can Fam Physician. 2013;59(5):e231-e239. http://www.cfp.ca/content/59/5/e231.full.pdf+html.
Free Full Text: http://www.amcp.org/JMCP/2013/September_2013/17103/1033.html
MeSH “Major” Terms: Drug Interactions; Drug Prescriptions; Education, Medical, Continuing; Health Education; Physician's Practice Patterns; Prescription Drugs/administration & dosage

Eight-year Canadian study on opioid prescribing among regulator- and self-referred physicians to intensive course

This educational study in a clinical journal by Kahan et al at the University of Toronto examined “the effects of an intensive 2-day course on physicians' prescribing of opioids” [1]. The most impressive feature of this study is its eight-year-plus data-gathering period of opioid-prescribing levels among participating physicians, most of whom were family physicians. Other interesting features are worth mentioning, in both instructional design and study design.

The study design grouped participants into self-referred physicians vs. physicians who were referred by medical regulators, and added a control (nonparticipant) group. Undertaking a challenging matching procedure, researchers matched nonparticipants according to specific variables, including quarterly rates of opioid-prescribing, expressed as milligrams of morphine equivalent. Subgroups of participant groups with very high opioid-prescribing patterns were also identified; unfortunately, nonparticipants to match these participants were difficult to find. Yet this targeted approach to matching is appropriate and represents a significant investment of the researchers’ time, allowing the comparative group findings shown below. Nonparticipants were added to the study concurrently with their matched participants, per an “index date” defined as “the date of course completion for participating physicians. Control physicians were assigned the same index date as their matched pair.” In one deviation from the primary outcome measure, matching was done by number of opioid prescriptions rather than milligrams of morphine equivalent. Another study design feature is the specific comparison of opioid-prescribing rates for 2 years before vs. 2 years after the educational intervention, again by group and subgroup vs. nonparticipants; participants who could not be matched were analyzed separately from participants with matched pairs.

The instructional design of the 2-day course incorporated several educational settings and modalities. Planners used didactic presentations but added problem-based case discussions and mock-interview learning interactions with standardized patients who offered feedback. Pros and cons of changing prescribing patterns were discussed in a session at the end of the course, featuring a faculty interview with a patient. The course also provided a detailed syllabus with notes and references before the course, as well as office materials. It should be noted that benzodiazepine-prescribing was also addressed in course content. Finally, each 2-day course enrolled up to 12 participants, a limit that would confer an individualized learning environment and some professional privacy in what might be a sensitive concern among participating physicians.

The authors noted in the introduction, “Medical education has been suggested as one strategy to improve opioid prescribing among physicians” [2,3] and “Educational interventions focused on opioid prescribing lead to positive improvement in physicians’ knowledge and self-reported practices” [4]. Let's look at results by reported subgroup.

Among physicians referred by medical regulators, “the rate of opioid prescribing decreased dramatically in the year before course participation compared with matched control physicians,” and “the course had no added effect on the rate of physicians' opioid prescribing in the subsequent 2 years.” It seems that these physicians might have changed their behavior by arbitrarily reducing prescribing rates because of the regulatory investigation, even without an educational intervention to inform their clinical decision-making. In fact, the authors acknowledge this, noting, “We measured only the quantity of opioids prescribed, not the quality of opioid prescribing.” The regulatory concerns may have created a false baseline for the educational study that measured only quantity of opioid prescribed rather than patient-selection or other measure of competence.

Among the self-referred physicians who were matched to nonparticipants, “there was no statistically significant effect on the rate of opioid prescribing observed” from baseline to 2-year follow-up, although there had been a temporary decrease, particularly in prescribing for patients aged 15 – 64 (here’s a nice graph with patient ages: http://www.cfp.ca/content/59/5/e231/F4.expansion.html). On the other hand, “the rate of opioid prescribing decreased by 43.9% in the year following course completion” among self-referred physicians with high prescribing rates who could not be matched, suggesting that these physicians “might have responded to what was taught in the course.”  

References cited:
1. Kahan M, Gomes T, Juurlink DN, et al. Effect of a course-based intervention and effect of medical regulation on physicians’ opioid prescribing. Can Fam Physician. 2013;59(5):e231-e239. http://www.cfp.ca/content/59/5/e231.full.pdf+html.
[Featured Article]
2. College of Physicians and Surgeons of Ontario. Avoiding Abuse, Achieving a Balance: Tackling the Opioid Public Health Crisis. Toronto, ON: College of Physicians and Surgeons of Ontario; 2010.
3. National Opioid Use Guideline Group. Canadian Guideline for Safe and Effective Use of Opioids for Chronic Non-Cancer Pain. Hamilton, ON: National Opioid Use Guideline Group; 2010.
4. Midmer D, Kahan M, Marlow B. Effects of a distance learning program on physicians’ opioid- and benzodiazepine-prescribing skills. J Contin Educ Health Prof. 2006;26(4):294-301.
Free full text PDF: http://www.cfp.ca/content/59/5/e231.full.pdf.
MeSH *Major* terms: Analgesics, Opioid/therapeutic use*; Drug Prescriptions/standards*; Education, Medical, Continuing*; Physician's Practice Patterns/standards* 

Medical education with EMR-based reminders reduces antibiotic prescribing and dispensing for respiratory tract infections in Norway

It is known that British guidelines for otitis media support delayed antibiotic prescribing [1], and other countries have guidelines to reduce certain antibiotic prescribing for otitis media, for example, France [2]. Conversely, Finnish guidelines do not [3]. A 2013 Norwegian study published in the British Journal of General Practice compares the varying effectiveness of 2 interventions in delaying primary care antibiotic prescribing for respiratory tract infections, including otitis [4].

Notwithstanding a complicated design for recruiting and assigning general practitioners across multiple sites, this article offers several interesting features. First, it compares an education-only intervention with the same education enhanced by pop-up reminders of a physician’s own prescribing patterns in the electronic medical record (EMR), a nice reinforcement of the educational intervention for participating physicians. While not a focus of this post, I would like to mention a new Penn study of adherence to guidelines on otitis media using EMRs for decision support at Children’s Hospital of Philadelphia [5]. This shows interest in implementation science combined with continuing medical education (CME) for changing physicians’ practice patterns.

The Norwegian study featured here [4] data collected and linked data on prescribed and dispensed antibiotics from (a) 1 year before and (b) 1 year during the intervention, which allowed prescribing practice patterns to be displayed to physicians in the EMR at the point of prescribing antibiotics for a respiratory tract infection. It also collected pharmacy fill rates by patients, which I find interesting because it may offer insights into patients’ (or parents’) agreement with the need for the prescription, after any access barriers to medication adherence. 

Both study arms showed slightly reduced antibiotic prescribing from baseline (pre-intervention) rates: 1% reduction vs. 4% reduction in “approximated risk” (risk ratio, RR) in the education-only vs. education-plus-EMR study arms, respectively. Both results report very tight ranges around a 95% confidence interval (CI), increasing confidence in the findings. (It is further nice to see the CI reported instead of the p value, for those who often hesitate to report CI because of many readers’ greater familiarity with the p value.) While reporting of “risk ratio” may be used as simply a convenient and appropriate way of reporting epidemiological data, it seems to me that its use for reporting educational outcomes with practice data is unusual and perhaps a comment on antibiotic prescribing for these infections as a risk.

The authors find that upper respiratory tract infection, sinusitis, and otitis “gave highest odds for delayed prescribing and lowest odds for dispensing,” which led them to conclude that the greatest potential for “savings” is greatest for these infections, a comment that brings this CME study with implementation science into the context of health utilization research. The article offers freely accessible full text, so enjoy reading the study.

References cited:

1. Centre for Clinical Practice at NICE (UK). Respiratory Tract Infections - Antibiotic Prescribing: Prescribing of Antibiotics for Self-Limiting Respiratory Tract Infections in Adults and Children in Primary Care. London: National Institute for Health and Clinical Excellence (UK); 2008 Jul. http://www.ncbi.nlm.nih.gov/pubmedhealth/PMH0010014/.
2. Levy C, Pereira M, Guedj R, et al. Impact of 2011 French guidelines on antibiotic prescription for acute otitis media in infants. Médecine Mal Infect. 2014;44(3):102-106. http://www.ncbi.nlm.nih.gov/pubmed/24630597.
3. [Update on current care guidelines: acute otitis media]. Duodecim. 2010;126(5):573-4. Finnish. http://www.ncbi.nlm.nih.gov/pubmed/20597310.
4. Hoye S, Gjelstad S, Lindbaek M. Effects on antibiotic dispensing rates of interventions to promote delayed prescribing for respiratory tract infections in primary care. Br J Gen Pract. 2013;63(616):e777-e786. http://bjgp.org/content/63/616/e777.full.pdf. [Featured Article]
5. Fiks AG, Zhang P, Localio AR, et al. Adoption of electronic medical record-based decision support for otitis media in children. Health Serv Res. 2015;50(2):489-513. http://www.ncbi.nlm.nih.gov/pubmed/25287670.  

MeSH *Major* terms: Anti-Bacterial Agents/therapeutic use*; Education, Medical, Continuing*; General Practice/statistics & numerical data*; Physician's Practice Patterns/statistics & numerical data*; Respiratory Tract Infections/drug therapy* 

Pharmacy Education for Hospital Clinicians on VTE Prophylaxis Changed Performance, Bringing Guideline-Adherent Care To Most Patients

Earlier today, I wrote of interprofessional clinical education regarding team communication during cardiac surgery. Now I continue the theme of nonphysician education by highlighting contributions of pharmacy education to patient care, and one that particularly relates to (post)surgical care. While this month’s Back-to-School campaign (illustrating published educational outcomes) mainly features recent articles, this 2005 study by Dobesh and Stacy in the Journalof Managed Care Pharmacy (free full text available) is a worthy read for its contributions to quality care research from the pharmacy perspective and scope of practice.

Venous thromboembolism (VTE and/or DVT, PE) is a great concern among surgeons and other physicians. In fact, the VTE evidence-basedguideline by the Institute for Clinical Systems Improvement (ICSI; Jobin et al 2012) names 10 stakeholder groups—including physicians and pharmacists—as “intended users.” The current article used the 2004 American College of Chest Physicians (ACCP) recommendations. Effectively preventing VTE can dictate the chances of successful outcomes and reduce patient readmission rates for many conditions. Because of the challenges of selecting the optimal anticoagulant agent and dosage for individual patients, pharmacists can clearly collaborate with physicians in making decisions about VTE prophylaxis. The 2012 guideline considered pharmacological thromboprophylaxis with unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), fondaparinux, warfarin, aspirin, apixaban, dabigatran, and rivaroxaban—enough therapeutic options to suggest the need for consultation between physicians and pharmacists.  

The pharmacy intervention for nurses, pharmacists, and physicians in the community hospital was traditional in instructional format, involving reinforcing in-service and quality-assurance presentations, as well as newsletters. The educational outcomes assessment method was more notable, using retrospective chart reviews with statistically similar patients before and after the educational intervention (15 months of patient charts before, and 6 months after). Patient chart reviews showed statistically significant and clinically meaningful change in VTE prophylaxis performance in practice. Specifically, both “suitable” and “optimal” prophylaxis increased (P = .006 and P < .0001 respectively), with a fourfold increase in the optimally treated percentage of patients associated with pharmacy education of physicians, nurses, and pharmacists.  

These data show that traditional educational initiatives developed by one health care profession for others can be effective in changing performance, especially when guidelines for practice and risk categories are presented in reinforcing text-based and live formats. This intervention brought guideline-adherent care to 93% of patients with risk, up from 49% before the intervention.  

References cited:
Dobesh PP, Stacy ZA. Effect of a clinical pharmacy education program on improvement in the quantity and quality of venous thromboembolism prophylaxis for medically ill patients. J Manag Care Pharm. 2005;11(9):755-62. PMID: 16300419.
Geerts WH, Pineo GF, Heit JA, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126:338S-400S. PMID: 15383478.
Jobin S, Kalliainen L, Adebayo L, et al. Venous thromboembolism prophylaxis. Bloomington (MN): Institute for Clinical Systems Improvement (ICSI); 2012. Available at: http://www.guideline.gov/content.aspx?id=39350. Accessed September 7, 2015. 


PubMed:  http://www.ncbi.nlm.nih.gov/pubmed/16300419
Journal Free Full Text: http://amcp.org/data/jmcp/contemporary_755-762.pdf
MeSH *Major* terms: Health Personnel/education; Heparin, Low-Molecular-Weight/therapeutic use; Inservice Training; Thromboembolism/prevention & control; Venous Thrombosis/prevention & control