Journal of General Internal Medicine

, Volume 29, Issue 1, pp 223–229

Exploring Public Attitudes Towards Approaches to Discussing Costs in the Clinical Encounter

  • Marion Danis
  • Roseanna Sommers
  • Jean Logan
  • Beverly Weidmer
  • Shirley Chen
  • Susan Goold
  • Steven Pearson
  • Elizabeth McGlynn
Original Research

DOI: 10.1007/s11606-013-2543-9

Cite this article as:
Danis, M., Sommers, R., Logan, J. et al. J GEN INTERN MED (2014) 29: 223. doi:10.1007/s11606-013-2543-9



Patients’ willingness to discuss costs of treatment alternatives with their physicians is uncertain.


To explore public attitudes toward doctor–patient discussions of insurer and out-of-pocket costs and to examine whether several possible communication strategies might enhance patient receptivity to discussing costs with their physicians.


Focus group discussions and pre-discussion and post-discussion questionnaires.


Two hundred and eleven insured individuals with mean age of 48 years, 51 % female, 34 % African American, 27 % Latino, and 50 % with incomes below 300 % of the federal poverty threshold, participated in 22 focus groups in Santa Monica, CA and in the Washington, DC metro area.


Attitudes toward discussing out-of-pocket and insurer costs with physicians, and towards physicians’ role in controlling costs; receptivity toward recommended communication strategies regarding costs.


Participants expressed more willingness to talk to doctors about personal costs than insurer costs. Older participants and sicker participants were more willing to talk to the doctor about all costs than younger and healthier participants (OR = 1.8, p = 0.004; OR = 1.6, p = 0.027 respectively). Participants who face cost-related barriers to accessing health care were in greater agreement than others that doctors should play a role in reducing out-of-pocket costs (OR = 2.4, p = 0.011). Participants did not endorse recommended communication strategies for discussing costs in the clinical encounter. In contrast, participants stated that trust in one’s physician would enhance their willingness to discuss costs. Perceived impediments to discussing costs included rushed, impersonal visits, and clinicians who are insufficiently informed about costs.


This study suggests that trusting relationships may be more conducive than any particular discussion strategy to facilitating doctor–patient discussions of health care costs. Better public understanding of how medical decisions affect insurer costs and how such costs ultimately affect patients personally will be necessary if discussions about insurer costs are to occur in the clinical encounter.


communicationdecision makinghealth care costspatient engagement

In efforts to promote the economic and fiscal sustainability of health care, physicians are increasingly expected to take costs into account when offering recommendations and making health care decisions for patients.14 Wise stewardship of resources is now considered an essential component of medical professionalism.57 When physicians consider costs in medical recommendations, there are several grounds for arguing that they need to discuss this openly with patients. First, patients should be aware of and participate in decisions affecting their own expenses, decisions that have profound consequences for patients’ health care access and adherence, health status and financial well being.814 Second, for the sake of procedural fairness and transparency, some authors argue that patients should be notified when costs to others influence recommendations, particularly if some benefit has been forgone.1517 Third, discussing costs with patients may contribute to enhancing their awareness about limited resources and familiarize them with cost-conscious medical decision making.

Several strategies have been proposed for broaching the topic of insurer and out-of-pocket costs with patients. Hardee and colleagues recommend that discussions of health care costs use empathic communication techniques,18 emphasizing that doctors and patients are a team working together to address the issues of cost.1921 Pearson22 emphasizes that the processes involved in medical decision-making need to be fair, with everyone doing their part to address costs. A third approach would emphasize that a less expensive option is often “good enough” to meet patients’ needs and seeks to debunk the notion that the newest, most expensive care is always the right or “best” choice.23 Finally, an educational approach seeks to alert patients that rising health care costs increase their premiums, lead to foregone wages, crowd out other public expenditures such as education, and contribute to increased taxes.

While these strategies for discussing cost in the clinical encounter seem plausible, we are aware of no studies that have tested them explicitly. We therefore convened focus groups of insured individuals to (a) explore attitudes toward discussing and considering insurer and out-of-pocket costs in the doctor–patient encounter and (b) examine whether proposed communication strategies would enhance receptivity toward discussing costs.



We recruited participants for 22 focus groups in Santa Monica, CA and in the Washington, DC metro area during July and August of 2011. Participants were purposively sampled to include a wide range of ages, race/ethnicities, education levels, and at least half had incomes below 300 % of the federal poverty threshold. To achieve this range of participants we conducted two English-speaking groups with each of the following types of participants: middle-aged adults; low-income, middle-aged adults; young adults; low-income young adults; adults with dependents; low-income adults with dependents; retirees; and low-income retirees. We conducted four Spanish-speaking groups: one with young adults, one with middle-aged adults, one with adults with dependents and one with retirees. Since we were asking about insurers’ costs, only individuals with health insurance were included (the screening protocol is available on request).

Focus Groups

All focus groups were facilitated by experienced focus group moderators at RAND. Two pilot focus groups were conducted to test the moderator guide (available on request). Subsequently, we conducted group sessions with the participant groups specified above. Moderators introduced the topic and then led participants through a series of scenarios involving treatment decisions and sample dialogue between patients and physicians. In one scenario, participants were asked to imagine they’d had the worst headache of their life for 3 months, for which their doctor recommends an imaging study, either a magnetic resonance imaging scan (MRI) or a computed tomography (CT) scan. The doctor explains that the difference between the two imaging studies is marginal: “The MRI presents a slightly more detailed picture and might find something that the CT misses, such as an extremely uncommon blood vessel problem, but nearly all problems serious enough to need treatment would be seen on either the MRI or the CT.” In one variation of the scenario, out of pocket costs varied: the CT would cost $400, while the MRI would cost $900. In a second variation, the patient would pay $70 for either test, but insurer costs would differ (the CT would cost the insurer $330 while the MRI would cost the insurer $830).

Other scenarios involved tradeoffs between cost and efficacy, convenience, and other benefits. For example, participants were asked whether they would accept a treatment regimen involving taking pills three times per day instead of once per day if the three-times-per-day regimen were less expensive for them and/or for insurers. For all scenarios, participants discussed whether and how they would want their physicians to broach the topic of costs with them, and which treatment option they would ultimately choose. The moderator first asked open-ended questions about how they would like physicians to broach the topic of cost using such prompting questions as: Do you want to talk about these costs with your doctor? How would you like that discussion with your doctor to go? How could your doctor make these conversations more comfortable? Subsequently, the moderator solicited participants’ reactions to the four communication strategies advocated in the literature using language similar to that shown in Table 1. Focus group sessions lasted approximately 2 h and were facilitated and recorded by experienced bilingual moderators.
Table 1

Communication Strategies




“Everyone is interested in keeping health   care costs down and we’re trying to do so in   a fair way. In this situation, I’d recommend   a CT scan to any of my patients.”

Shared sacrifice

“It’s good for us to share responsibility for   keeping health care costs down. At the end   of the day, we all pick up the tab when we   pursue expensive care because premiums   rise. Keeping costs down for health plans or   Medicare or Medicaid helps you personally   in the long run.”

Threshold of “good enough” rather than optimization

“I really do think that a CT scan can   be a reasonable option for you. It might   not find quite as many things as an MRI,   but I believe it will find the important   things I am concerned about. It is   certainly good enough for what we   are trying to do.”

Shared decision-  making

“What are your thoughts about going with   the CT scan? Let’s talk about whether the   MRI is worth the extra $500 that your   insurer or Medicare or Medicaid would   be paying.”


Self-administered questionnaires were used before and after focus group discussions. Items regarding socio-demographics, health status, and barriers to access were measured using survey measures that have been tested and validated in other studies.24 Attitudinal information about discussion of costs with physicians, physicians’ consideration of costs when making recommendations, and physicians’ proper role in controlling out-of-pocket costs was collected using published items or slightly modified versions of published items, and used a 5-point Likert scale.25

Ethical Approval of Studies and Informed Consent

The study was approved by institutional review boards (IRBs) at the National Institutes of Health, RAND, and the University of Michigan. Participants signed consent forms and were compensated for their participation.


Questionnaire data were analyzed using descriptive statistics: Pearson’s Chi-square for comparison of categorical data, including comparison of attitudes before and after the focus group discussion and comparisons among participants of differing characteristics. Linear-by-linear Chi-square test was used where chi-squared distribution was only approximately valid. Tukey’s honest significance test was used for single-step multiple comparisons. Kruskal-Wallis test was used to test significant differences in median responses of racial groups, since there were more than two samples that were not assumed to be normally distributed. Only comparisons with p < 0.05 were considered significant. Responses to Likert items were transformed to dichotomous variables (agree vs. disagree/undecided) for bivariate and multivariate analyses. Binary forward stepwise logistic regressions were used to determine which variables acted as independent predictors of attitudes, with p-to-enter = 0.05 and p-to-drop = 0.10. Software used was SPSS version 19.

Digital recordings of focus group discussions were transcribed and Spanish transcripts were translated. Two independent raters coded and analyzed the transcripts to identify key themes, using NVivo version 10. The inter-coder agreement was over 90 % on all 22 transcripts (coding scheme is available on request).

Qualitative data underwent two separate sets of analysis. One focused on exploring study participants’ comments pertaining to attitudinal barriers they might have towards addressing health care costs in the clinical setting, and is published elsewhere.26 The second qualitative analysis focused on participants’ comments pertaining to doctor–patient communication about health care costs and their reactions to the proposed communication strategies, and is reported here. Representative comments that exemplify frequent codes were selected for inclusion here (additional comments are available in the Online Appendix Section II, and the structure of the coding scheme is available from the authors upon request).


Participant Characteristics

Two hundred and eleven individuals participated in focus groups. The mean age was 48 years, approximately half the participants were female, one third were African American, and one quarter were Latino (see Table 2).
Table 2

Participant Characteristics (N = 211)












































 High school or less



 Some college or more



Health status





 Very good












Chronic disease status





 One or more||



Marital status


 Never married






 Living with partner



 Separated or Divorced






Annual household income


 Less than $10,000




































 Not sure



Barriers to accessing health care


 No cost-related barriers



 One or more cost-related barriers



*Absolute numbers do not sum to 211 if participants were missing variables;

Percents refer to responding participants and do not include missing data

Participants listed in the ‘Other’ racial category include: one American Indian; two Asians, two Hawiian/Pacific Islanders, and five individuals who reports several racial categories

§Hispanics included Puerto Rican, Mexican/Mexican-American/Chicano, Cuban, Other Hispanic/Latino, and Non-Hispanic Latino

||Respondents who responded positively to having been told by their doctor that they had any one or more of the following chronic conditions: high blood pressure, diabetes, cancer, heart disease, lung disease, other

Attitudes Toward Discussing Out-of-Pocket Costs

In general, participants wanted to discuss out-of-pocket costs as indicated to their responses to survey items (see Table 3).
Table 3

Attitudes Before and After Focus Group Sessions








I would like my doctor to talk with me about my out-of-pocket medical costs when he/she recommends a test or treatment.







I would like my doctor to talk with me about all costs when he/she recommends a test or treatment, not just my out-of-pocket costs.







My doctors should consider my out-of-pocket costs when they make medical decisions.







My doctors should consider all costs when they make medical decisions, not just my out-of-pocket costs.







Doctors should play a role in reducing the out-of-pocket costs of healthcare for their patients.







Doctors should play a role in reducing all costs of healthcare, not just the ones that their patients pay out-of-pocket.







All ratings on a Likert scale of 1–5, with 1 indicating strongest agreement with the statement

*p < 0.05

**p < 0.01

***p < 0.001

M Mean, SD standard deviation, t t test, df degrees of freedom

In the focus group conversations, participants also expressed a clear desire to know out-of-pocket costs up front, as illustrated by the following comment:

I’m tired of getting hit by all these costs. Doctors have no clue about the costs, they just say here’s what I recommend, you do it and then 20 bills pop up in the mail. And then it’s followed by the letters from the insurance company saying we didn’t cover this and we didn’t cover that. I want to know what it’s costing, I want to know the options.

Other comments on this theme reflected a wish to manage one’s health care costs, and a fear that not talking about costs might allow doctors to make arbitrary decisions regarding costs without proper patient notification.

I don’t really see why, once again, I think anything else that you pay for, you know how much you’re paying for something, you don’t just get a mystery bill that shows up 2 months later and you’re like, what is this?

Perceived Reasons for Lack of Conversations About Costs

Participants felt that doctor visits are too impersonal for conversations about costs, that physicians are too busy to talk about financial matters, or that physicians don’t know the costs of health care. Participants spoke of a broader problem of feeling that doctors’ visits are too rushed and the quality of interactions is too low for meaningful conversation and trust-building. As one participant said:

The doctor tends to be so rushed, just because they have ten more patients in the waiting room. So I’m lucky if I have more than the initial check up and then, “Okay, the nurse’ll be in for a test,” that’s it. So I don’t even know what the test is, let alone how much of it is going to be out of my pocket versus the insurance company paying.

Attitudes Toward Discussing Insurer Costs

While participants were in favor of discussing out-of-pocket costs with physicians, comments during focus groups revealed that they largely were not interested in discussing costs borne by insurers, including Medicare or Medicaid.

I think the doctor’s supposed to be my advocate completely.

What’s funny is I totally agree with that statement [about how keeping costs down for health plans helps you personally in the long run because premiums rise when we all pursue expensive care] but I wouldn’t want him to say it to me. I might say it to him but I wouldn’t want him to say it to me.

For participants in the group composed of retired older adults, the idea of physicians discussing costs borne by society raised particular alarm. Many participants expressed a worry that cost-consciousness in health care decisions hurt the elderly because older adults are perceived as no longer useful to society.

If money means that much to you where you say this person over here, his health means nothing because he’s not worth an extra $10,000 for us to spend on him…because they’re now 70 and we got some youngsters coming up here at 20. So they’re through. So what they do, you have to understand, we being senior like we are, we know they take us and throw us over here on the sideline. They got the best out of us, so there’s really nothing else we have to contribute. So why should they spend a bunch of money [on us]?

Strategies to Enhance Receptivity Toward Discussion of Costs

Participants found none of the proposed communication strategies useful for increasing their comfort with discussing costs with physicians. Overall, attitudes toward discussing costs with physicians were more negative after participating in the focus groups than beforehand (Table 3). In the focus groups, many participants said that there was no approach that doctors could take to talking about costs borne by society that would make patients more receptive to the message.

The idea is too offensive. Well, if he’s going to tell me that “I’m recommending the CT scan basically because it’s cheaper,” there’s nothing he can say to make it sound right.

Moderator: What are your thoughts on the different ways the doctor can talk to you about how the costs your health plan would have to pay factor into that recommendation?

Participant: I said it before and I’ll say it again: I’d walk [out].

While the language used by the doctor did not seem to affect participants’ willingness to discuss and consider costs, half of the Spanish-speaking focus groups and three quarters of the English-speaking groups included one or more participant comments indicating that long term doctor–patient relationships and trust in their physician would make physician recommendations of a marginally less effective treatment option more acceptable.

Moderator: Would you be receptive to him recommending the less expensive option, meaning the less expensive to your health insurance company or to Medicare or Medicaid?

Participant: I would have to really trust him…I’ve been with the same primary doctor for the last 25 years and I would trust him. If he said you don’t need to get this test, you’re fine with this test, then I would go with him because I trust him.

I would seriously consider it, if I had been going to this doctor, you know, I had built up a relationship and I trusted him, and he’s telling me that, then I would seriously consider doing the lesser cost thing.

Views of the Proper Role of Physicians

Response to the questionnaire indicated that participants agreed or strongly agreed both before and after group discussions that physicians should play a role in reducing patients’ out-of-pocket costs. Yet participants’ expressed reservations about having such conversations.

I disagree with the part about the doctor knowing what my financial situation is. It’s none of his business. His business is my healthcare only, not what’s in my pocket.

Some participants said that while they were uncomfortable discussing costs with their doctor, they’d be willing to talk to someone else about out-of-pocket costs.

Let my doctor be my doctor, and then let me step into your financial office and discuss what I need to do. Call it social worker or whomever, but let’s keep the two separate.

I don’t want to be swayed by the price. I want to know what treatment will give me the best results…If I have any questions about the cost after I make my decision, I’ll go to the insurance company.

While participant responses to survey items showed general agreement that doctors should play a role in reducing all costs (Table 3), most dialogue reflected the sense that discussing insurer costs was incompatible with their notion of the doctor’s proper role.

I really, really don’t feel that [costs to society are] a part of a doctor–patient conversation…. The physician is the patient’s advocate, and he should lay out the options, present the costs, the benefits, and likely prognosis to me and then let me make an informed decision as a consumer. As far as the cost to my insurance company, let’s debate this in the halls of Congress. I don’t want to have this debate in my physician’s office where my health is at stake. It’s not appropriate.

Association Between Participant Characteristics and pre-Focus Group Attitudes

Bivariate associations between participant characteristics and attitudes are shown in Online Appendix Tables 16. Only significant associations from bivariate analyses and from logistic regressions are summarized here.

Older participants were more likely than younger participants to agree that they would like to talk to their doctors about out-of-pocket costs both in bivariate analysis (Online Appendix Table 1) and in the logistic regression (OR = 2.85, p =  0.003).

Older participants and sicker participants were both in agreement with talking to the doctor about all costs in bivariate analyses (Online Appendix Table 2) and logistic regression (OR = 1.8, p = 0.004; OR = 1.6, p = 0.027 respectively) .

In bivariate analysis, White, Hispanic, and sicker participants were all more likely than others to agree with having their doctor consider out-of-pocket costs when making decisions (Table 3). In logistic regression modeling, only sicker participants agreed that physicians should consider out-of-pocket costs (OR = 1.4, p = 0.032).

Ethnicity, health status, income, and experiencing barriers to accessing health care were all found in bivariate analysis to be significantly related to willingness to having their doctors consider all costs when making medical decisions (Online Appendix Table 4). However, in logistic regression of these four variables, only barriers to accessing health care remained independently associated with agreement that physicians should consider all costs in making medical decisions (OR = 2.4, p = 0.011).


Participants in this study wanted to know what medical care will cost them, expressed dissatisfaction with the lack of transparency around these costs, and felt that discussions of out-of-pocket costs, with some caveats, could be appropriate in the clinical encounter. However, they were generally opposed to the idea that physicians should broach the topic of costs borne by insurers. Strategies recommended in the literature for discussing costs, when presented to participants, did not enhance their willingness to discuss costs with a doctor. In contrast, a trusted physician with whom they have a strong relationship seemed to make some participants more positively disposed to discussing costs and taking costs into account when making choices among competing interventions in the clinical encounter.

As in all focus group studies, a limitation of our study was that we had a small number of participants who were not randomly selected, so results should not be interpreted as generalizable to the entire insured population. Instead, our sampling aimed to capture a wide range of insured persons, and oversample those who might be most concerned about health care costs.27 An additional limitation inherent in our study design is that attitudes expressed in focus groups in response to hypothetical cases are not necessarily highly predictive of how patients will behave in actual practice. Finally, the communication strategies we examined were designed primarily to help patients understand and come to terms with insurer costs. Other available approaches to discussing costs put more emphasis on helping patients address their own health care costs.13,14

The existing literature regarding patient attitudes toward discussing costs in the clinical encounter is limited and inconsistent in its findings.2830 One study shows that many patients want to know about any rationing of their own care, including being given a “good explanation” to enable them to judge whether the decision made had been correct so that they can protest the decision.31 While various approaches to discussing costs in the clinical encounter have been suggested,13,14,18,22,23 we are not aware of any studies that have explicitly examined such approaches to doing so.

Our study has several implications. The results highlight how strong relationships with physicians, particularly relationships that are long standing, might increase patients’ openness to including health care costs in their decision making. It implies that rushed visits, with insufficient time to talk about important issues, may undermine efforts to bring a sensitive topic like costs into the doctor–patient conversation, and may be counterproductive for overall efficiency as well as quality. The findings suggest that trust is a valuable ingredient for honest conversations about how to make decisions that are optimal from a patient and societal perspective. Given that patients are more receptive to discussing out-of-pocket costs than insurer costs, a possible strategy for enabling discussions about insurer costs may be stronger efforts to educate the public about the importance, for their own sake, of controlling insurer costs, and engaging them in deliberations outside of the clinical encounter about how to make value-laden tradeoffs.



Robert Wesley, PhD provided statistical support.


This research was supported by the Intramural Research Program of the NIH, The Office of the Director of the NIH and the Department of Bioethics at the NIH Clinical Center.

Prior presentations

The paper was presented at the 2012 annual meeting of the American Society for Bioethics and Humanities.

Conflict of Interest

The authors declare that they do not have a conflict of interest.


The views expressed here are those of the authors and do not necessarily reflect the policies of the National Institutes of Health, the US Department of Health and Human Services, RAND, or Kaiser Permanente. This research was conducted while Dr McGlynn was at RAND.

Supplementary material

11606_2013_2543_MOESM1_ESM.docx (76 kb)
ESM 1(DOCX 75 kb)

Copyright information

© Society of General Internal Medicine 2013

Authors and Affiliations

  • Marion Danis
    • 1
  • Roseanna Sommers
    • 1
  • Jean Logan
    • 1
  • Beverly Weidmer
    • 2
  • Shirley Chen
    • 3
  • Susan Goold
    • 3
  • Steven Pearson
    • 1
  • Elizabeth McGlynn
    • 4
  1. 1.Department of BioethicsNational Institutes of Health Clinical CenterBethesdaUSA
  2. 2.RANDSanta MonicaUSA
  3. 3.University of MichiganAnn ArborUSA
  4. 4.Kaiser PermanenteOaklandUSA