Today’s post takes a look at the OG article that gave physician coaching its strongest leg to stand on. We’ll explore the study in a pseudo-journal club fashion with a fair share of GIFs. Journal club would be way better if memes were always included!
This article discussed in this post can be found here.
The Study in a Nutshell
In August 2019, the physician coaching world welcomed the article “Effect of a Professional Coaching Intervention on the Well-being and Distress of Physicians: A Pilot Randomized Clinical Trial.” While this predated me and my coaching journey, I can imagine that about seven people at the time went:
This study sought to examine the effect that individualized coaching had on physician well-being. The coolest part of this study is the fact that they tested an intervention (coaching) that addressed burnout! And did so using a randomized controlled trial (RCT)!
Their hypothesis was “professional coaching would result in measurable improvements in well-being, job satisfaction, resilience, and fulfillment in physicians and measurable reductions in burnout.”
The authors randomized 88 physicians from Mayo Clinic sites in Arizona, Florida, Minnesota, and Wisconsin into two arms: 1) the intervention arm, which received six coaching sessions via professionally trained coaches, and 2) the control arm, which received thoughts and prayers.
I’m kidding. The control arm was told to do more yoga as well as modules on burnout in their personal time.
I’m still kidding. The control arm received nothing during the study.
Fret not: they were made whole after the study concluded and given access to the same amount of the individualized professional coaching.
Gimme the Study Design Deets
Inclusion criteria included:
- Being a physician in the departments of medicine, family medicine, and pediatrics
- Could be a PCP or subspecialist
- In practice for 5 to 30 years
Physicians were recruited by email and departmental announcements and volunteered to participate. I find this very perplexing because physicians are notorious for not checking their work email—and are usually stuck in email jail since they never clear out their inboxes!
I can only imagine that free food was offered and they were tricked into signing up!
They closed enrollment when they hit their target study goal of 80 physicians, which they calculated would give them enough study power to detect a moderate effect size in their outcomes of interest (see below).
More like “just enough power.”
Randomization was stratified by:
- Years in practice
- Work site location
- Primary care vs subspecialty
Data collection occurred via a simple survey at the start of the study and then five months later at the conclusion.
Oh, and in order to ensure that any benefit didn’t simply come from physicians scheduling coaching calls during clinical hours (and thus reducing their burnout simply by decreasing exposure to a noxious stimulus), they enforced making up any missed patient time at other hours!
Intervention and Outcomes
The intervention arm (a.k.a. the coaching group) received a total of 3.5 hours of confidential coaching over six sessions. The coaches were trained by a coaching institution that was known to work with physicians.
Note that the coaches were not physicians.
Coaching occurred via telephone—I’m imagining rotary phones at this point since we’re talking 2017-2018, and I can’t remember anything pre-pandemic.
Where the heck was Zoom!? Who are these rubes using telephones?
The outcomes measured using standardized instruments (likert scales abound!) were burnout, quality of life, resilience, job satisfaction, engagement, and meaning in work.
Ready Table One... And Table Two
Table 1’s demographic characteristics shows that the randomization was successful…except the control group had 18% more women. However, this had a p value of 0.09 and did not achieve statistical significance.
Of note, women constituted ~55% of the study population and only ~34% of the eligible population. Thus, they were overrepresented in the study.
Table 2 documents the baseline measurements of the above-mentioned outcomes, and finds no statistically significant differences between the coaching and control groups.
So...What Happened?
I’ll cut to the chase with the results that achieved significance in comparison to the control group.
In the coaching group (interventional arm):
- Emotional exhaustion was reduced by 16% (p value <0.001)
- High emotional exhaustion was reduced by 20% (p value <0.001)
- Overall burnout was reduced by 24% (p value <0.001)
- Quality of life was increased by 20% (p value 0.005)
- Resilience was increased by 4% (p value 0.04)
Notably, there were no statistically significant reductions in depersonalization or improvements in job satisfaction, engagement, and meaning in work.
Look at that—this coaching stuff I ramble about works when held up to scientific rigor! Moreover, it works for physicians!
Strengths and Things to Point Out
First and foremost, I’m excited that folks are investigating interventions to help physicians.
I knew about this study when I was exploring coaching and I’d be remiss to not credit this study as helping push me towards both seeking a coach and later becoming a coach.
This study corroborated my experience and those of the other physicians around me whose transformations I witnessed.
The biggest strength of the study was the design: a randomized controlled trial. As you know, RCTs are the gold standard for “seeking the truth” and sit atop the hierarchy of study types.
That’s because they deliver the highest level of evidence in their ability to limit as much bias as possible (due to randomization; assuming they’re well designed), and thus are best at establishing causal relationships.
I next want to point out how the tremendous outcomes the coaching group experienced came from just 3.5 hours of coaching. That’s not a lot of time, right!?
That’s like the same amount of time as it would take you to watch one Lord of the Rings extended edition movie.
Side note: I once marathoned all three movies—watched the first two extended editions before heading to the theater to watch the final one—that was a long day!
This is what I looked like at the end of that day.
I appreciated that they allowed the coaching to be unscripted so that the physicians could bring anything to the coaches, and have their individual needs met. There were no preset or forced agendas. That goes a long way in building trust and rapport.
Weaknesses and Limitations
No study is without its weaknesses and limitations, and this study is no exception.
The sample size was on the smaller side. They labeled it “modest.” It also relied on volunteers, and so self selection bias may have played a role. For instance, these may have been physicians who were more at their wits end looking for help, or more amenable to coaching, or otherwise biased in some way.
This second selection bias example could have been at play since the baseline levels of burnout for the study participants was higher than other studies have noted for physicians.
Women were overrepresented in the study based on the demographics of the eligible pool of physicians. Interestingly, I feel like within the Mayo Clinic system women must make up a smaller percentage of physicians than overall in medicine—I have no evidence for this, just going off of a hunch.
I would have loved to see a crossover design after the 5-month survey mark and after they offered the control group the same coaching access outside of the formal study.
I would have appreciated another survey further along, maybe at the 12-month mark, to evaluate for the persistence of the benefit seen. I know, I know, I’m just asking for a more longitudinal study, which is more expensive.
One Big "What If..."
I’m separating out the final weakness of this study (in my books): that the coaching was provided by non-physician coaches.
I wholeheartedly believe that physician coaches would have had a much greater impact on the outcomes. Yes, I’m biased. I’ll own that, but hear me out!
When the person coaching you understands firsthand the sacrifices made to become a physician, what residency entailed, the specific challenges of the physician day to day experience, and the drivers of physician burnout, it’s not hard to imagine that such a coach would connect on a deeper level and offer better insights, no?
Heck such a coach has probably specifically dealt with a lot of the same issues being brought before them!
Moreover, in the article, they reference two other modalities in addition to coaching that have evidence in improving well-being: mentorship and peer support.
A physician coach can intrinsically offer both of those during coaching sessions as well.
The good news is that some of my fellow physician coaches are collecting outcome data on burnout and well-being in their coaching programs.
I am doing so within my charting program. Even if it’s a program targeting a specific problem (*cough*that drives the top two causes of burnout*cough*), the mindset work that helps physicians is universal.
Parting Thoughts and Takeaways
Formal literature on physician coaching is in its infancy, and this study is at its core. The literature will only expand from here and I’m excited to see the evidence base further established and bolstered.
We certainly need as many “evidenced-based approaches to promote physician well-being,” as they say in the article.
To date, the literature shows that investing in specific interventions can be beneficial and turn the tide against burnout. And those interventions are not yoga classes or modules on burnout. (I’m not bagging on yoga, just on administrators telling us that’s all we have to do to “feel better” 🙄)
It’s incumbent upon employers and institutions to invest in these interventions including coaching to help save the physician workforce.
If you’d give a construction worker a helmet as a basic measure for protection, then healthcare organizations should offer coaching to physicians to help reduce and mitigate burnout and support well-being in a similar fashion as a basic measure for protection!
Addressing moral injury requires an approach grounded in humanism, in empathy. This is what physician coaching is all about!
What do you think of this study? Let me know in the comments section below.
If you haven’t subscribed to my email list, then do so below so you don’t miss my new posts or my weekly updates (only for subscribers).
I’d also be most appreciative if you shared this post with anyone whom you think would benefit from the content or message of the blog. They may similarly be most appreciative ☺.