Studying the specialist workforce (and why it’s difficult)
Understanding and planning a medical workforce capable of meeting patients’ specialized health care needs is an ongoing challenge. It’s something various national specialty societies (NSS), physician organizations and governments have been grappling with for years — and to greater degrees of success.
Last Friday, we hosted our annual NSS-Human Resources for Health Dialogue. This meeting is a checkpoint to discuss advances in how we collect and model physician workforce data and how to use this data to improve planning.
The meeting drew representatives from 21 specialty societies and eight national organizations. More than ever, we’re hearing a strong message that we need to work together if we’re going to get this forecasting right.
I wanted to use this message to share with you some of the work presented by the
- Canadian Rheumatology Association,
- Canadian Association of Emergency Physicians,
- Canadian Association of Gastroenterology,
- CAPER (Canadian Post-MD Education Registry),
- Physician Resource Planning Advisory Committee, and the
- Canadian Institute for Health Information (CIHI).
Here’s a bystander view of what was discussed.
As always, your comments, feedback or questions are encouraged. Please leave a comment or send me an email at firstname.lastname@example.org.
Andrew Padmos, BA, MD, FRCPC, FACP
Chief Executive Officer
GENERAL DISCUSSION POINTS
This list details some of the common challenges that impede our common understanding of the specialist medical workforce, making planning a challenge. My appreciation goes to the Royal College Health Policy Team for their assistance in putting together this summary.
Quick link – jump to presentation slides
- Multiple data sources exist: Each source has its own strengths and weaknesses, and contributes to different understandings of physician supply.
- Data complexity: Each data source captures data elements with varying levels of accuracy and completeness. For example, full-time or part-time work, active or retired, paid by fee-for-service or alternative funding, registered or not, formally certified in a specialty or not. Databases capture different aspects depending on what they were designed to do.
- Lack of standardization: There is no standardization among data source definitions or how output is measured. This calls into question the quality of the output. For example, data on billing and fee structures is not always a reliable measure of the medical workforce and its workload; some specialists bill under different titles and there are also location-specific differences in how payment is made, etc.
- Defining study populations: NSSs are often challenged to define their membership. For example, in some cases general tasks performed by specialists in their discipline are not specific to their specialty. There are also differences in clinical vs. academic roles within the same specialty.
- How best to gather data? Surveys are often a default approach. Defining mailing lists and declining response rates are just two of the challenges with this tactic. Again, lack of standardization in defining geographic regions, specialist domains, workload, payment methods, etc., add to the difficulty.
- Access to data sources: Not all data is readily accessible. For example, data on population needs is either difficult to obtain or non-existent, same with data on outcomes.
- Reporting approaches: Data has to be presented differently to resonate with different audiences. For example, peer-reviewed publications can help circumvent skepticism by some audiences, but the media and public desire more simplicity.
- Overlooked insights: Numbers alone don’t tell the full story. Without the added context and insights gleaned from physicians-in-practice about issues facing the profession, for example, data or numbers can be misinterpreted/the meaning lost.
- Retirement: We don’t have accurate data about physician retirement, which is an important piece of the planning puzzle. To start, self-reported measures are often unreliable. Many physicians don’t retire when they said they had planned to. Another important consideration vis-à-vis retirement is to determine how we can transition physicians out of active practice but still leave the door open for work as a mentor or consultant to special cases.
- Rural/urban divides: We need more insights into the cause of distribution challenges (just because the job exists, doesn’t mean it will be readily filled) and potential solutions (e.g. training options, better advertising of positions).
- Joint vs solo studies: The way things are currently structured, specialty societies are often expected to do their own research and NSSs have varying resources. Are larger NSSs with more capacity to carry out studies better positioned to gain new training positions?
The good news in all of this was that there was general agreement, expressed at the meeting, on the need to explore a better way for NSSs and physician organizations to work together on a more rigorous approach for data collection, definition and reporting. A collaborative approach would help alleviate some of the challenges listed above. More consultation is ongoing for a proposed way forward.
|CAPER (Canadian Post-MD Education Registry)||Discover trends gleaned from CAPER and employment data. (Note: This presentation includes an update from the Physician Resource Planning Advisory Committee)||[PDF]|
|Canadian Rheumatology Association||Gain insights into the 2015 study “Stand up and be counted.” Findings include that we are currently short rheumatologists and that this shortage may worsen in the next 10 years.||[PDF]|
|Canadian Association of Emergency Physicians||Check out key findings from The Collaborative Working Group on the Future of Emergency Medicine in Canada. Included are a profile of EM physicians, their distribution and staffing shortfalls.||[PDF]|
|Canadian Association of Gastroenterology||Understand the specific challenges encountered when seeking reliable sources for data on human resources in this specialty area. Some recommended next steps are also presented.||[PDF]|
|Canadian Institute for Health Information||If you like math, the statistical approach presented in these slides will be of interest. It was generated to better profile physician scopes of practice and practice patterns.||[PDF]|