Engaging patients in matters of their health and health care is important to me, but I never considered that I might have a special advantage in this area. Then one day, perusing abstracts in a recent issue of Academic Emergency Medicine, I spotted this rather startling finding: patients were nearly 5 times more likely to want an active role in their care when the health care provider was female.1
Shared decision making – collaborative medical decision making between health care provider and patient that weighs both the best available evidence and the patient’s values and preferences – has been somewhat slow to take hold, particularly in the ED. Despite widespread endorsement from on high (the Institute of Medicine, the Centers for Medicare and Medicaid Services, the Affordable Care Act, the American Medical Association, the American College of Emergency Physicians) and increasing acceptance by both providers and patients in principle, actual participation rates are not believed to have kept pace. The reasons are only partly understood. Could the fact that women still comprise only about a third of the physician workforce – and a quarter of emergency physicians – have something to do with it?
It was an intriguing possibility.
For one thing, it made a certain intuitive sense. Mightn’t there be systematic differences in practice style between male and female providers? Might not patients in any case perceive there to be differences? It isn’t so much of a stretch to imagine that the patient of a provider perceived as having a decisive or authoritative manner would be less likely to assert his or her personal preferences, while the patient of a provider who encourages patients to speak up would be more likely to do so. But are male providers in fact more likely to fit into the first category and females into the second?
Conveniently, there is a large body of literature to help answer this question, helpfully summarized (through 2008) in Boulis and Jacobs’ book, “The Changing Face of Medicine: Women Doctors and the Evolution of Health Care in America2.” The authors ask, among other questions, whether female and male physicians differ systematically in their interactions with patients, and the answer they arrive at is an emphatic yes. Visits with female physicians last longer, on average,3,4 and involve more talking – by both physician and patient.3 Female physicians appear to talk more than their male counterparts, but patients with female physicians talk longer, too, with the result a more equal ratio between physician and patient. Conversations between female physicians and their patients appear to involve more partnership building, question asking, and information giving.3,4 Female residents, in one study, asked more open-ended questions than male residents.5 Female physicians express more empathy.5,6 Importantly, such attempts at supportive communication do in fact correlate with greater patient engagement,7 and patients of female physicians rate their visits as significantly more participatory than those of male physicians.8
Shared decision making, it would seem, should be a slam dunk for patients with female physicians and sizable challenge when the physician is male – on average, of course. But the possibility that this is so has received surprisingly little attention in the recent wave of studies looking for predictors of, or barriers to, shared decision-making. Alas, the study by Holland et al that spurred this post sheds no light on the matter whatsoever.
The study surveyed patients age 60 and over within 24 hours after ED treatment for acute musculoskeletal pain and then again at one week. The main objective was “to assess the impact of shared decision making on change in pain at one week,” but an important secondary objective was to determine, retrospectively, patients’ desired level of involvement or “control preference” in choosing an analgesic for home use. Also assessed or recorded were the degree to which shared decision making took place, patient satisfaction, health literacy, medication side effects, and patient and provider characteristics “which might confound the relationship between shared decision making and pain recovery.”
In short, the study was not actually designed to look for predictors of shared decision making.
Why should that matter, if it did this anyway? Well, for starters, patients were dropped from all analysis if they couldn’t be reached for the follow-up survey at one week, since this portion of the survey addressed the study’s primary objective, even if it didn’t happen to be ours. And as it happens, a whopping 40% of the original 157 were lost this way.
Then there’s the issue of how patients’ control preferences were measured. The authors used the Control Preferences Scale,9 a single-question measure with five possible responses:
- I make the final selection about which treatment I will receive (doctor provides medical information; no personal exchange)
- I make the final selection of my treatment after seriously considering my doctor’s opinion
- My doctor and I share responsibility for deciding which treatment is best for me
- My doctor makes the final decision about which treatment will be used but seriously considers my opinion
- I leave all decisions regarding my treatment to my doctor
And then they collapsed patients’ responses into just three categories:
- Active (responses 1 or 2 above)
- Collaborative (response 3)
- Passive (responses 4 or 5)
Now wait a second. Patients having the final say over what’s done to them, even if only after seriously considering the doctor’s opinion, disqualifies the interaction as collaborative? And doctors making the final decision, after weighing the patient’s opinion, is the same as making the decision without doing so? As an avid practitioner of what I believe to be collaborative medical decision making, I consider the process fluid by its very nature. In many cases, I probably couldn’t even tell you on leaving a patient’s room which of us was ultimately the “decider.” When patient and doctor are truly open about sharing information and, where appropriate, opinion, it is inevitable that each will be influenced by the other; the mechanics of the ultimate decision are, ironically perhaps, almost beside the point.a
Because the authors merged disparate response categories into a confusingly heterogeneous “active” category, ditto the “passive” category, the results are next to impossible to make sense of. Of the 94 patients included in the analysis, just over half (53%) preferred an “active” (16%) or “collaborative” (37%) role in choosing an analgesic; the rest (47%) preferred a “passive” role. And yes, 24% of patients with a female provider, vs. only 5% with a male provider, desired an active role in decision-making, a nearly fivefold relative difference. But consider this: patients with a male provider were not only more likely to want a so-called passive role in the decision-making they were also more likely to want a “collaborative” one!
It seems almost mean to mention that all the confidence intervals were wide.
There is one last reason to abort the mission – and it’s a doozy. Patients were also much more likely to prefer an active role – about 3 ½ times more – when the provider was a nurse practitioner as opposed to an attending physician or resident. Could it be that provider gender was merely a confounder for provider type? Scouring the paper for a breakdown of female and male providers by job type and finding none, I contacted one of the authors and my suspicions were confirmed: all of the nurse practitioners were female. (For patients seen only by attendings or residents, there was a trend toward a preference for collaborative decision making with females vs. males but it didn’t reach significance.)
We return to the study by Holland et al1 to see what the authors make of the apparent association between provider gender and patient control preference, and where that leaves us.
The authors, seemingly unfazed by the problems of category confusion, confounding, and capacious confidence intervals uncovered in the 2nd post in the series, take the association at face value and offer up two possible interpretations in the Discussion (hint: only one of them is flattering):
“A greater preference for an active role by patients seen by female providers and nurse practitioners may reflect that women and nurse practitioners do a better job of listening to and empowering older adults to make their own decisions…(or) older adults may feel less comfortable leaving the decision to women and nurse practitioners.”
Methodologic problems notwithstanding, the authors have in fact hit on something. If a study were to be designed to identify provider characteristics associated with a patient preference for collaborative decision making – and let me here urge the reader to consider taking up the challenge! – and it turned out that patients of us female providers do want more of a say in their care than those of our male colleagues, how should we respond? Do we take pride or offense? Design yet another study to sort out which?
I say just run with it. When patients are driven to know and understand more – more about their condition, their treatment options, and the trade-offs that come with these options – and, as a result, their treatment can be tailored to their personal preferences, values and circumstances, what’s to be gained by asking why? The research literature and my own experiences have together persuaded me that patient engagement, where feasible, is better for patients, better for providers. If there’s anything at all about me that encourages this, well I should be so lucky.
But if, on the other hand, the question of a definite female advantage in this area should go unanswered, or future research place it in doubt, no matter. Better communication with patients is not a zero sum game. There’s room for all of us providers – female and male, doctor, nurse practitioner, and physician assistant – to be good at what we do; to encourage patients to understand and speak up about their care; to promote greater trust by patients toward all providers; in short, to improve health care for all.
a Note that collaboration doesn’t alter the bottom line, which is patient autonomy; patients (those with decisional capacity) ethically and legally get to have the last word, so long as they don’t demand care the provider can’t provide or believes to be of no potential benefit (or greater potential harm than benefit).
- Holland, W.C., et al., A Prospective Evaluation of Shared Decision-making Regarding Analgesics Selection for Older Emergency Department Patients With Acute Musculoskeletal Pain. Acad Emerg Med, 2016. 23(3): p. 306-14.
- Boulis, A.K., and Jacobs, Jerry A., The Changing Face of Medicine: Women Doctors and the Evolution of Health Care in America. 2008: Cornell University Press.
- Roter, D., M. Lipkin, Jr., and A. Korsgaard, Sex differences in patients’ and physicians’ communication during primary care medical visits. Med Care, 1991. 29(11): p. 1083-93.
- Roter, D.L. and J.A. Hall, Physician gender and patient-centered communication: a critical review of empirical research. Annu Rev Public Health, 2004. 25: p. 497-519.
- Shapiro, J. and D.D. Schiermer, Resident psychosocial performance: a brief report. Fam Pract, 1991. 8(1): p. 10-3.
- Bylund, C.L. and G. Makoul, Empathic communication and gender in the physician-patient encounter. Patient Educ Couns, 2002. 48(3): p. 207-16.
- Street, R.L., Jr., et al., Patient participation in medical consultations: why some patients are more involved than others. Med Care, 2005. 43(10): p. 960-9.
- Cooper-Patrick, L., et al., Race, gender, and partnership in the patient-physician relationship. JAMA, 1999.282(6): p. 583-9.