Performance-improvement CME
Performance-improvement CME is a format of continuing medical education that asks physicians to measure and improve their own practice. It aims beyond physician learning at the upper levels of the CME outcomes hierarchy, runs like a research project and has been slow to take up. It gives the writer customised gap analyses, grant proposals and outcomes write-ups to prepare, and it offers the physician continuing education credits and credit toward maintenance of certification.
What performance-improvement CME is
The American Medical Association defines performance improvement CME (PI-CME) activities as "a process by which evidence-based performance measures and quality improvement (QI) interventions are used to help physicians identify patient care areas for improvement and change their performance". All PI-CME activities share three steps: comparing current practice against recognised, evidence-based standards to set a baseline and identify performance gaps, planning and implementing a practice-improvement intervention, and reassessing practice to measure the results.
PI-CME is designed to produce measurable improvements not just in physician competence but also in patient outcomes and community health: against the hierarchy of CME outcomes, traditional CME sits at levels 1 to 3 — participation, satisfaction and learning — and PI-CME at levels 4 to 7 — competence, performance, patient health improvement and community health.
The format began as a pilot project of the American Medical Association's division of continuing professional development. Responding to criticism about the effectiveness of traditional CME, the association approved both PI-CME and internet search as novel learning formats worthy of continuing education credit in 2005, by one account. Another account gives the association's acceptance of PI-CME as an approved learning format as 2004. The internet search format was much simpler: a learner only had to identify a relevant practice-related question, find the answer online from a trusted source, apply the new knowledge to patient care and demonstrate what had been learned, with no requirement to demonstrate an improvement in patient care, whereas PI-CME sets a higher standard. Around the same time the association's officials began to promote PI-CME in particular as the most powerful way not just to educate clinicians but also to improve patient and community health, and proponents even claimed it would become "the core of the new CME".
In 2014 PI-CME was the newest, most complicated and most expensive form of continuing medical education. In the words of one writer whose clients deliver the format, it is "a fairly complex beast": the many clinical and research stakeholders involved, and the frequent requirement for institutional review board approval, make it "a bit like doing a research project", and it is not for the fainthearted. PI-CME activities are highly customised learning projects that often require cooperation among half a dozen organisations, two or three years to design and complete, and budgets of more than $1 million.
Uptake among physicians and providers
Until about 2010 CME providers struggled to find an audience for activities in PI-CME's unusual and highly structured format, while internet search grew explosively from the start, drawing more than 50,000 physicians in its first year against only 3,000 for PI-CME.
Barriers to uptake among physicians included a general unfamiliarity with the concept, the difficulty of shifting from knowledge-based to performance-based outcome measurement, and time constraints on teachers and learners alike; the format also had a reputation for being too expensive per learner for education companies to develop compared with other formats. In the view of the director of one medical school's office of continuing medical education, it took the CME community a while to warm up to the format conceptually and to learn how to do it right, and the format has been a growth and learning opportunity for educators. Attrition remains a common problem — physicians often start an activity with the best intentions and then do not finish — and time-consuming, tedious data entry continues to be a key barrier to participation.
By 2014, after its slow start, the format was gaining acceptance nationwide: accreditation council figures give annual physician participation in internet search as 75,844 in 2010, 92,114 in 2011 and 108,856 in 2012, and in PI-CME as 29,371, 44,275 and 141,860 in the same three years. Participation in PI-CME more than tripled from 2011 to 2012, and in 2012 it drew more physician participants than internet searching and learning, although traditional live meetings, self-paced internet modules and regularly scheduled events such as hospital grand rounds remained far more popular. Nonphysicians may technically participate but few do: accreditation council data for 2012 show physicians outnumbering them 15 to 1. The same CME office director predicted that the 2013 figures would show continued growth.
In 2014 several factors appeared aligned to drive increased physician participation.
- Medical specialties. Among the membership categories of the Alliance for Continuing Education in the Health Professions, the medical specialty category is where the format has its strongest foothold in physician participation: nearly 80% of physicians taking part nationwide in 2012 did so through nonprofit physician membership organisations.
- Hospitals. It seems a natural fit within larger hospitals and groups of hospitals, where a culture of continuous quality improvement already has deep roots; by accreditation council data, 57% of the 631 PI-CME activities offered nationwide in one recent year were offered by hospitals and health care delivery systems.
- Federal mandates. From 2015 the Centers for Medicare and Medicaid Services' Physician Quality Reporting System was to impose penalties for non-reporting of performance data, and the Affordable Care Act required health plans and insurance issuers to show how health outcomes were being improved, including through quality indices — the same kinds of quality data used to drive and evaluate PI-CME activities.
- Electronic medical records. Used properly, they can be a powerful resource for a PI-CME intervention: every such system requires the creation of a database, which can be searched to tell physicians how their patients are doing on many outcome variables, and those data can then set the baseline, plan the intervention and measure the results.
- Publication in the medical literature. Outcomes data from performance improvement initiatives were previously published mostly in nonmedical journals — a PI-CME study of improved diabetes care at the Joslin Clinic appeared in an educational journal in 2011 — but medical specialty journals in diabetes, leukaemia, lung cancer and osteoporosis had begun publishing PI-CME outcomes reports, and acceptance of the format may grow as physicians read them in their own specialty's journals.
- Word of mouth. The format requires significant time and effort upfront, but once that investment is made physicians often begin to feel a sense of ownership and, according to the same CME office director, find the outcomes deeply rewarding because they can see their own actions improving patient care.
A 2015 survey of needs-assessment writers set PI-CME beside quality-improvement CME. QI-CME is "a systematic, formal approach to the analysis of practice performance and efforts to improve performance", designed to reveal clinicians' practice gaps as determined by quality measures and to meet their individual educational needs through an educational intervention. Although the two terms may seem similar to people outside the industry, CME practitioners widely regard PI-CME as the more demanding and difficult to execute, and 44% of respondents agreed that QI-CME methodologies were changing the way they wrote needs assessments against only 21% for PI-CME. The survey's authors suggest that PI-CME's relevance to how needs assessments are written may be waning in favour of QI-CME, with its closer links to manufacturing, and that this, with the rise of managed care in the 1970s and the federal mandate in favour of electronic health records, could be interpreted as one more sign of physicians' lost professional autonomy.
The writer's work on a performance-improvement activity
"Good writing is critical for these programs", in the words of a former medical writer who co-founded a company specialising in PI-CME: the writer needs to be able to communicate at the front end about what is being measured, and at the back end for publication. In one diabetes activity, enrolled physicians reviewed the records of their own patients with diabetes and measured their performance against a nationally approved standard for quality care, haemoglobin A1C goal attainment rates. The writer's job was to write what was essentially a customised gap analysis for each physician, prepared after all participants had entered their patients' records into a database, saying where that physician's performance met the standard and where it fell short. For each section of the report dealing with a diabetes performance metric the writer had to go into the guidelines, provide a very concise summary highly relevant to that particular measure, and add language to the visual display of data showing the provider that the current status was not within the guidelines and explaining why.
The same company founder frequently writes the grant proposals that make new PI-CME projects possible, and in 2014 he expected his next hire to be a medical writer. He advises writers who would like to work on PI-CME projects to gain a working knowledge of the many national organisations that develop and publish quality standards for medical practice. A writer experienced in the format advises freelance writers and editors starting out in the field to ask many questions at the outset, to gather detailed knowledge of the various stakeholders involved and a clear vision of how they are meant to work together.
A writer whose clients deliver the format has made a subspecialty of helping them prepare PI-CME outcomes data for publication in peer-reviewed journals. According to that writer, such an initiative commonly produces enormous amounts of raw outcomes data, and the write-up then begins to resemble a clinical study report on a much smaller scale, with the writer's job becoming one of sifting through piles of documentation in search of a compelling story angle; in her words, "someone with a keen eye, some time, and some tenacity can help turn those data into a piece of gold".
Credit for the physician
Three performance-improvement initiatives show the credit a physician can earn from the format.
- Risk Evaluation and Mitigation Strategies: An Employer-Driven CME Initiative for Efficacy and Safety (REMEDIES). A mixed-methods programme of live, web-based, interactive and performance improvement activities, offering up to 24.5 continuing education credits, 20 of them from the performance improvement component, and a certificate of compliance with US Food and Drug Administration educational requirements for prescribing extended-release opioid analgesics.
- Performance Improvement Strategies in Multiple Sclerosis. An initiative in its third year in 2014, designed to help health care professionals assess their clinical practice methods according to current evidence and expert consensus, offering credit toward the maintenance of certification requirements of the American Board of Psychiatry and Neurology and 20 continuing education credits from the American Medical Association.
- Advanced Clinical Management of Diabetes and Its Complications. A programme completed in 2012 in which physicians submitted data on their patients with diabetes and received confidential reports comparing their performance with quality standards established by the American Diabetes Association and the American Medical Association, for up to 20 continuing education credits from the American Medical Association and toward the maintenance of certification requirements of the American Board of Internal Medicine.