SGO Ethics Series: Risk for Conflicts of Interest in Physician Satisfaction Scores | Lois M. Ramondetta, MD

Lois M. Ramondetta, MD

The Situation:

I, like many of you, pride myself on good satisfaction scores—actually, on all my “scores”. However, one month, I received the following message:

“The oncologist was distant and gave me no hope. I came to this doctor seeking help. I came away devastated and hopeless.”

All survey responses are of course anonymous. Still, I tried to recall which patient might have felt this way, and after looking at the prior month’s new patients, I knew in my gut which patient had sent the review. I immediately recognized her name, recalling the entire conversation.

She was a cachectic-looking woman with a platinum-resistant cancer and, I suspected, at least a partial bowel obstruction. I had thought I had used my conversational tools, starting with gauging her understanding of the disease (she was in denial, or at least she professed to be during this consult). We discussed the response rates of 3+-line chemotherapy agents, as well as the nature of and her eligibility for clinical trials. We also discussed methods for managing her symptoms of carcinomatosis.

After my assessment and our discussion, I expressed my concern that she needed hospitalization for hydration and control of nausea and to rule out a possible large bowel obstruction requiring surgery. She was initially resistant, but with persistent nausea and distention during the visit, she agreed to go to the ER. She was released the next morning and flew home after receiving hydration. A CT scan showed extensive carcinomatosis but no large bowel obstruction. Because the consultation had focused more on her symptoms rather than the next treatment options, I had tried to call her the next day and left a message that I wanted to follow up on our visit. I left a message that I would follow up after a few weeks by phone to see how the recommended dietary and bowel management was going (even though I suspected there would be little improvement). The satisfaction survey report came back before we had followed up.

I’ll be honest—what she wrote stung, even though I suspected her anger was never just directed at me. I replayed the conversation as I remembered it in my mind. Could I have been gentler? Was all the information I gave necessary to share? Did I ask permission to share prognostic information? As I would never want to be the cause of someone feeling devastated and hopeless, although I wasn’t sure it was this patient, I decided to call her. My intent was not to get her to change her words, but rather to apologize for any pain she experienced during the visit. Mercifully, she was kind and receptive to my call. She mentioned her primary oncologist had discussed similar concerns with her before she had even come for her consultation. She said the whole visit was overwhelming and disappointing and she was simply angry when she received the survey. I felt better, and I hope she did, too, but this experience left its mark with me.

Thankfully, I can count on one hand the number of times this type of miscommunication has happened, and I believe I am much more conscious of my words (and perhaps not as forthright) during these unique one-time consultations.

Satisfaction vs. Expectation

As a physician working in a specialized cancer center, it is not uncommon that new patients have flown in for a second opinion (or third or fourth opinion) and thus my only interaction with the individual is condensed into a 60-minute meeting, during which I try to assess their entire tumor history, their understanding of their situation, and their performance status as it relates to their eligibility for clinical trials and molecularly targeted trial opportunities. Additionally, I must find the time to explain the nature of clinical trials and the accompanying socioeconomic and logistic travel challenges of trial participation.

However, most importantly, I feel it is my obligation to introduce the concept of balancing quality and quantity of life and encourage patients with incurable disease to reflect on their definitions of meaning and well-being. This may involve discussing, in a sense, eudaimonic well-being, defined as reflecting on purpose and value, versus hedonic well-being, defined as a connection to pleasure and transitory experiences.¹ For example, I remember a treasured note from the sister of a patient who thanked me for having a straightforward conversation with her sister (the patient), encouraging her to return home to spend quality time with her family. That was my one and only visit with this patient, who had died 2-3 weeks later.

The intense exchanges at the extremes of life and the existential reflections on what a good life is part of what drew me to gynecologic oncology. Establishing a strong therapeutic alliance and letting the patient know I deeply care for their well-being and will provide the information they need to make tough decisions and live their best life, despite the “cards they were dealt”, has been of primary importance to me, as I’m sure is true for every one of my SGO colleagues. In my experience, establishing a doctor-patient alliance that allows these conversations to be effective (and feel “good”) often takes multiple visits. I find having these types of discussions during one-time consultations with patients who have come from so far and with so much hope to be much more difficult to navigate.

Conflicts of Interest and Satisfaction Scores

The COVID-19 pandemic environment increased the importance of determining patients’ medical power of attorney and discussing some aspects of their goals of care in clinics and on admission, as patients were often unaccompanied and at risk for a sudden change of status. Documenting advance care planning notes (ACPs)and goals of care became an institutional priority. During this time, monthly physician satisfaction score reports also gained prominence as a metric. Notably, the survey questions themselves are relatively benign (see side bar), it is the free text comment where physicians may experience joy, pride, or shame and disappointment or even anger at the responses.

My concern is that the emphasis on satisfaction scores has the potential to introduce complex ethical dilemmas. The potential paradox between patients’ hopes and a doctor’s advice is especially highlighted by, but not limited to, one-time visits. Hypothetically, satisfaction scores could affect a physician’s willingness to talk to patients about EOL preparation vs offering a minimally effective chemotherapy agent that seems to offer hope, or even discuss other sensitive topics, such as addressing lifestyle changes for patients with unhealthy BMIs. When important conversations are tempered, the risks include giving and receiving ineffective treatments because the physician has not given them all the information they need to decide how they wish to spend their life. What is potentially lost is time with family and, in some cases, financial stability.

Weeks et al. found that patients with stage IV cancer who were under the mistaken impression that the goals of their chemotherapy treatments were curative were more likely to give a favorable rating to communication with their physician than were patients who knew that their treatments were palliative; thus, I wonder if satisfaction scores can ever be a good measure of this interaction.²

We must work to preserve the art of medicine and not fall victim to forced measurements of aspects of care that don’t impact quality care and could cause harm. For instance, patient satisfaction is not necessarily the same as patient experience or high-quality or safe care. Patient experience tends to measure what is expected in a visit (eg, reviewing of medications, determining medical histories, and spending enough time), while patient satisfaction may measure whether the patient was happy with the visit and whether they felt the physician was a good match; they may also tell the physician what they are expecting to hear.

What are we actually measuring?

Satisfaction score anticipation (and reward) could incentivize physicians to prioritize patient satisfaction over appropriate and evidence-based care. These scores may then jeopardize the integrity of the physician-patient relationship and lead to compromised decision-making. One could imagine that satisfaction score expectation could result in physicians feeling pressure to comply with patients’ demands—for instance, a patient who is unhappy with the outcomes of her pain contract. Furthermore, some physicians have reported their institutions use satisfaction scores to influence compensation adding yet another aspect of COI.³ Other downstream effects of emphasized satisfaction scores might involve swifter referrals of difficult patients, disparities in scores because of race or gender concordance effects, differences related to specialty as well as the presence of depression when filling out the survey as well as simply election bias because surveys are completed preferentially by those with extreme experiences, all of which may contribute to higher physician burnout rates.^4-10 Additionally, the limitations of satisfaction scores in comparison to experience scores and quality and safety metrics should be acknowledged.¹¹ Fortunately, poor satisfaction scores are usually (but not always) avoidable through patient education (pre-visit), communication skill training, and maintaining an attitude of curiosity, humility, and empathy.

What can we all agree on?

1. We can all agree that providing timely and honest end-of-life care requires physicians to engage patients effectively and compassionately in complicated discussions about challenging situations. The process of learning the skills necessary to facilitate end-of-life care may not always be formally taught. Options for learning these skills include reading books/articles, observing a great mentor, having available training within your institution and/or contracting with a group like Vital Talk (https://www.vitaltalk.org/).
2. Physician satisfaction scores have the potential to introduce conflicts of interest in discussions with patients about sensitive issues without measuring the quality of the visit experience. Regardless, the ethical imperatives of truth telling, emotional support, and advance care planning must not be compromised by the fear of negative satisfaction scores. Upholding patient autonomy while ensuring beneficence and non-maleficence may sometimes necessitate a balance among patient satisfaction, evidence-based care, and the depth of goals-of-care discussions.
3. In difficult patient visits, physicians must “read the room” and determine what must be dealt with during the current visit and what may be best dealt with in small digestible steps—perhaps in “all in one” visits or perhaps by scheduling a phone call a week later to discuss more solemn topics.
4. Institutions can research ways to support clinicians with pre-visit expectation education on the importance of goals of care discussions and advance care planning.
5. Physicians should prioritize honest and empathetic communication, ensuring that patients are fully informed about their diagnosis, prognosis, and treatment options and outcomes. We must ensure that physicians are equipped with communication skills to provide patients with often disappointing information so that they are fully aware of their medical condition and can make important decisions regarding how they spend their precious lives.

We would like your thoughts on patient satisfaction scores. Please respond to this short survey. Survey results will be shared in a future ethics blog.

References

1. Deci, E. L., and Ryan, R. M. (2008). Hedonia, eudaimonia, and well-being: an introduction. J. Happin. Stud. 9, 1–11. doi: 10.1007/s10902-006-9018-1

2. Weeks JC et al October 25, 2012 N Engl J Med 2012; 367:1616-1625.

3. Zgierska A, Rabago D, Miller MM. Impact of patient satisfaction ratings on physicians and clinical care. Patient Prefer Adherence. 2014 Apr 3;8:437-46. doi: 10.2147/PPA.S59077. PMID: 24729691; PMCID: PMC3979780.

4. Takeshita J, Wang S, Loren AW, Mitra N, Shults J, Shin DB, Sawinski DL. Association of Racial/Ethnic and Gender Concordance Between Patients and Physicians with Patient Experience Ratings. JAMA Netw Open. 2020 Nov 2;3(11):e2024583.

5. Darnall BD, Schatman ME. Autonomy vs paternalism in the emergency department: the potential deleterious impact of patient satisfaction surveys. Pain Med. 2013

Jul;14(7):968. doi: 10.1111/pme.12170. Epub 2013 Jun 11. PMID: 23758676; PMCID: PMC4803281.

6. Fenton JJ, Jerant AF, Bertakis KD, Franks P. The cost of satisfaction: a national study of patient satisfaction, health care utilization, expenditures, and mortality. Arch Intern Med. 2012;172(5):405-411. doi:10.1001/archinternmed.2011.1662

7. Manary MP, Boulding W, Stealing R, Glickman SW. The patient experience and health outcomes. N Engl J Med. 2013;368(3):201-203.

8. Rogo-Gupta LJ, Haunschild C, Altamirano J, Maldonado YA, Fassiotto M. Physician Gender Is Associated with Press Ganey Patient Satisfaction Scores in Outpatient Gynecology. Womens Health Issues. 2018 May-Jun;28(3):281-285

9. Nuyen B, Altamirano J, Fassiotto M, Alyono J. Effects of surgeon sociodemographics on patient-reported satisfaction. Surgery. 2021 Jun;169(6):1441-1445

10. Cambria B, Basile J, Youssef E, Greenstein J, Chacko J, Hahn B, Berwald N, Ardolic B. The effect of practice settings on individual Doctor Press Ganey scores: A retrospective cohort review. Am J Emerg Med. 2019 Sep;37(9):1618-1621

11. Richman BD, Schulman KA. Are Patient Satisfaction Instruments Harming Both Patients and Physicians? JAMA. 2022;328(22):2209–2210.

Lois M. Ramondetta, MD is a gynecologic oncologist at The University of Texas, MD Anderson Cancer Center in Houston, TX.