Abstract
Context
Health care costs have increasingly shifted to patients, and financial distress caused by medical care has increased. Patients may wish to discuss costs with their clinicians.
Objective
Describe patient preferences for communication about cost in the clinical setting.
Design
Cross-sectional, self-administered survey of a stratified random sample of the population insured in an integrated health care system in Washington State. Our sampling frame was the entire membership aged 21 years or older. Sampling was stratified by sex and group practice enrollment.
Main Outcome Measures
Preference for discussing health care costs with one’s physician. We conducted regression analyses to determine predictors of communication preference; potential predictors included demographic characteristics, financial burden, delay in seeking care because of cost, and socioeconomic variables. Survey responses were weighted to adjust for nonresponse and sampling.
Results
Of 7200 invitations sent, 2200 survey responses were returned. Ninety-two percent wished to know their out-of-pocket costs before beginning treatment. Most respondents preferred their physician talk with them about out-of-pocket costs (81.4%) and expressed comfort with discussing costs with their physician (75.6%). Overall, 43.7% reported any delay in seeking care in the previous 12 months. One in 5 respondents (21.6%) reported family medical debt. Delay in seeking care was positively and independently associated with preferring to discuss costs with one’s physician; current medical financial burden was not.
Conclusion
Patient preferences for communication about costs with physicians are high, and medical debt and delay in care-seeking are prevalent. Delay in care-seeking independently predicts cost communication preferences.
INTRODUCTION
As health care costs increase, people using health care services face increasing out-of-pocket costs and medical financial burden.1,2 In 2014, more than 22 million Americans delayed medical care because of costs, and in 2012, more than 1 in 4 US families experienced any medical financial burden.3,4 Medical financial burden and delay in seeking care are related to higher comorbidity and lower quality of life,5–7 and out-of-pocket costs can cumulatively have an impact on patients and families, regardless of insurance status.8–10
Patients’ understanding of expected out-of-pocket responsibility at the time care is delivered may improve patient-centered care,11 but discussions about health care costs have traditionally been kept out of the clinical arena. However, patients increasingly wish to discuss costs in the clinical setting.12–14 Patient-centered communication is associated with improved treatment decisions, patient experience, and health outcomes.15–17 Patient preferences for discussing costs with one’s clinician have not been well described.
Our aim was to describe preferences for discussing cost in the clinical setting and to identify predictors of preferences for communication about cost in an insured population. We hypothesized that people experiencing either medical financial burden or delays in seeking care because of cost concerns would be more likely to report strong preferences for communicating about cost with their physicians. Therefore, we sought to 1) describe preferences for communication about health care costs in the clinical setting, 2) determine the association between communication preference about cost and the financial burden and delay of treatment because of cost, and 3) explore other patient characteristics associated with cost communication preference.
METHODS
We conducted a self-administered cross-sectional survey of a stratified random sample of the adult population insured by Group Health Cooperative (Group Health), an integrated health care system serving nearly 600,000 people in Washington State. Our sampling frame was the entire adult membership aged 21 years or older. Sampling was stratified by sex and health plan type. Members can choose a health plan to receive care exclusively at Group Health’s primary care clinics (group practice), or at their choice of clinics within a broader contracted network (network). Study recruitment also included an oversampling of members with children aged 9 to 14 years for an unrelated study aim.
A 68-item survey (available online at: www.thepermanentejournal.org/files/16-070-survey.pdf) was mailed to 7200 people between August and November 2014. So we could investigate a second, unrelated study aim looking at predictors of survey response, the mode of recruitment approach varied; some groups received a $2 preincentive, reminder follow-up calls, or invitations to complete the survey online. The analyses reported here include all survey respondents, regardless of how they were approached.
We obtained administrative data on the entire sample to determine factors related to survey response and control for potential response bias: age, sex, race, insurance type, enrollment duration, zip code, and comorbidity measured as expected health care resource utilization.18 The study design and methods were approved by Group Health’s institutional review board.
Measures
The overall aim of the survey was to describe social determinants of health in a representative sample of the organization’s membership; the survey covered a broad range of topics, but this report summarizes an analysis of the survey’s questions pertaining to patients’ communication preferences about cost. The measures used in this analysis include demographic characteristics, financial burden, delay in care measures, and other variables hypothesized to be associated with communication preferences.
Communication Preferences
We included seven items related to communicating about cost in the clinical setting, adapted from published work.19 Our main item of interest was patient preference about communicating with physicians at the point of care (“I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment”). A second item asked about communications from team members other than a physician. Other questions assessing communication preferences included two questions about the role of out-of-pocket costs in treatment decision making: one question about preferences for knowing out-of-pocket costs before treatment and one about the ease of finding cost information when needed. There was also an item about comfort discussing costs with a physician. Response options for all seven items were presented from “strongly agree” to “strongly disagree” on a five-point Likert scale.
Financial Burden and Delay in Care-Seeking
We included 4 dichotomous items adapted from the National Health Interview Survey.3 The financial burden item asked about current medical debt for anyone in the family (“Does anyone in your family currently have any medical bills that are being paid off over time?”). Three questions asked about delay of care: “In the past 12 months, was there a time when you did not take your medication as prescribed because of cost?”, “In the past 12 months, have you delayed seeking medical care because of worry about the cost?”, “In the past 12 months, has anyone in your family delayed medical care because of worry about the cost?”3 We combined the 3 delay items into a single composite dichotomous variable for “any delay in care,” because of the correlation between the individual items, and for ease of interpretation in the multivariable model.
Other Potential Predictors of Communication Preference
Group Health administrative data provided information on age, sex, health plan type (group practice vs network), duration of enrollment with Group Health, insurance type (commercial, Medicare, or Medicaid), and comorbidity measured by expected health care utilization (resource utilization band [RUB] or The Johns Hopkins Adjusted Clinical Groups System).18 These data were based on diagnostic and procedure codes in the prior year. The survey collected information on other potential predictors, including demographic and socioeconomic measures (race and Hispanic ethnicity, education, household income, marital status, employment status, and home ownership), likelihood of recommending Group Health to friends or family, and the number of days with poor physical or mental health in the previous month.
Statistical Analysis
We used administrative data on the entire Group Health-enrolled population to compute sampling probabilities in each of our sampling strata, defined by sex, health plan type (group practice vs network), and parent of an enrolled child aged 9 to 14 years. From the original sample of 7200, we received 2220 responses (30.8%). The probability of responding to the survey adjusting for characteristics related to survey response, including age, sex, group practice, and parental status, was estimated using logistic regression. We used the product of the inverse of the sampling and the estimated response probabilities to create a combined weight to use in our analyses, to control for response bias, and to create estimates generalizable to the entire sampling frame.
To summarize the demographic and socioeconomic characteristics as well as the financial burden, delay in care, and communication preference measures of the study population, we used descriptive statistics computed using the analytic weights described earlier. Spearman rank correlation coefficients were used to assess the correlation between the 7 cost communication measures. We examined the bivariate distributions between all study covariates and preference for communication with one’s physician about costs. To identify variables independently associated with our primary communication outcome, we fit modified Poisson regression models to directly estimate unadjusted relative risks associated with each variable. Next, we fit a multivariable model with all covariates significantly related to our communication item at p value ≤ 0.10. Age and sex were retained in the model regardless of their bivariate associations, as were financial burden and delay in seeking care because of their hypothesized relationship to communication preference. Analytic weights were used throughout the analysis.
Analyses were conducted using STATA version 13 statistical software (StataCorp LLC, College Station, TX).20
RESULTS
With use of weighted estimates, the study population (N = 2220) was predominantly women, married, and white, and more than 60% were aged 45 years or older (Table 1). Fifty-two percent (51.8%) reported having earned a 4-year college degree or greater. Most were private insurance holders (79.3%) rather than recipients of Medicare or Medicaid; 65.3% were enrolled in the group practice division compared with the external contracted network, and nearly half were enrolled with Group Health for 10 years or more. Most reported owning their own home and being employed. Forty-two percent reported a household income of $75,000 or more (see Table 1).
Table 1.
Sample characteristics (N = 2220)
| Characteristic | na | Weighted % (SE) |
|---|---|---|
| Womenb | 1198 | 55.6 (1.2) |
| Age, yearsb | ||
| 18–44 | 551 | 38.5 (1.3) |
| 45–64 | 914 | 39.4 (1.2) |
| ≥ 65 | 755 | 22.1 (0.8) |
| Race | ||
| White non-Hispanic | 1802 | 81.5 (1.0) |
| Asian non-Hispanic | 163 | 7.5 (0.6) |
| Other | 227 | 11.0 (0.8) |
| Marital status | ||
| Married or living with partner | 1587 | 71.2 (1.1) |
| Divorced, separated, or widowed | 420 | 15.0 (0.8) |
| Single | 198 | 13.8 (1.0) |
| Home ownership | ||
| Own | 1696 | 72.2 (1.2) |
| Rent | 374 | 22.3 (1.1) |
| Other | 91 | 5.6 (0.6) |
| Insurance typeb | ||
| Medicare | 617 | 18.2 (0.7) |
| Medicaid | 32 | 2.5 (0.5) |
| Commercial | 1571 | 79.3 (0.8) |
| Resource utilization band (RUB)b | ||
| Low (0–2) | 574 | 33.0 (1.2) |
| Medium (3) | 1068 | 46.8 (1.2) |
| High (4–5) | 521 | 20.2 (0.9) |
| Days in poor physical health | ||
| 0 | 1138 | 52.1 (1.2) |
| 1–15 | 822 | 39.0 (1.2) |
| ≥ 16 | 191 | 8.9 (0.7) |
| Days in poor mental health | ||
| 0 | 1191 | 52.3 (1.2) |
| 1–15 | 818 | 39.8 (1.2) |
| ≥ 16 | 146 | 7.9 (0.7) |
| Other characteristics | ||
| Education, 4-year college or more | 1165 | 51.8 (1.2) |
| Household income in 2013 of $75,000 or more | 893 | 42.6 (1.2) |
| Not employed | 915 | 34.9 (1.1) |
| Enrolled in group practice (yes)b | 1642 | 65.3 (1.2) |
| Group Health enrollment duration ≥ 10 yearsb | 1260 | 47.8 (1.2) |
| Would not recommend Group Health (score 0–7) | 842 | 42.4 (1.2) |
Missing values: race (n = 28), marital status (n = 15), education (n = 15), income (n = 173), employment (n = 38), home ownership (n = 59), enrollment duration (n = 5), RUB (n = 57), physical health (n = 69), and mental health (n = 65).
From administrative data.
SE = standard error.
Communication Preferences
Responses to the cost communication items appear in Table 2. Most (81.4%) either strongly agreed or agreed with preferring that their physician talk with them about their out-of-pocket costs when treatment is recommended, and most (92%) wished to know their out-of-pocket costs before beginning treatment. Three-fourths (75.6%) reported being comfortable discussing the costs of care with their physician; 20.9% preferred to discuss costs with a nonphysician team member. The majority (62.9%) indicated that their physician should consider the patient’s out-of-pocket costs when making medical decisions, and 74.8% reported considering out-of-pocket costs before making decisions for their own care. However, less than half (47.3%) reported that they could find information about health care costs when needed.
Table 2.
Cost communication items
| Survey item wording and response option | n | Weighted % (SE) |
|---|---|---|
| I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment. (Primary outcome) | ||
| Strongly agree | 771 | 37.1 (1.2) |
| Agree | 977 | 44.3 (1.2) |
| Neutral | 301 | 12.7 (0.8) |
| Disagree | 110 | 4.6 (0.5) |
| Strongly disagree | 36 | 1.3 (0.2) |
| Missing | 25 | |
| I am comfortable talking about the cost of my care with my doctor. | ||
| Strongly agree | 667 | 31.1 (1.1) |
| Agree | 996 | 44.5 (1.2) |
| Neutral | 350 | 15.4 (0.9) |
| Disagree | 159 | 8.0 (0.7) |
| Strongly disagree | 24 | 1.1 (0.3) |
| Missing | 24 | |
| I would prefer to discuss the cost of my care with someone other than my doctor, such as a nurse, social worker, or financial counselor. | ||
| Strongly agree | 129 | 6.2 (0.6) |
| Agree | 303 | 14.7 (0.9) |
| Neutral | 1039 | 48.0 (1.2) |
| Disagree | 549 | 23.9 (1.0) |
| Strongly disagree | 168 | 7.2 (0.6) |
| Missing | 32 | |
| I prefer to know about the out-of-pocket costs for my treatment before I am treated. | ||
| Strongly agree | 1319 | 62.7 (1.2) |
| Agree | 663 | 29.3 (1.1) |
| Neutral | 160 | 6.0 (0.5) |
| Disagree | 42 | 1.4 (0.2) |
| Strongly disagree | 14 | 0.5 (0.2) |
| Missing | 22 | |
| I can find information about the cost of my health care when I need it. | ||
| Strongly agree | 289 | 12.5 (0.8) |
| Agree | 800 | 34.8 (1.2) |
| Neutral | 574 | 27.1 (1.1) |
| Disagree | 409 | 19.7 (1.0) |
| Strongly disagree | 119 | 5.9 (0.6) |
| Missing | 29 | |
| My doctor should consider my out-of-pocket costs as s/he makes a medical decision. | ||
| Strongly agree | 544 | 26.7 (1.1) |
| Agree | 804 | 36.2 (1.2) |
| Neutral | 421 | 19.3 (1.0) |
| Disagree | 302 | 12.7 (0.8) |
| Strongly disagree | 123 | 5.1 (0.5) |
| Missing | 26 | |
| I consider my out-of-pocket costs when I make a decision about my care. | ||
| Strongly agree | 672 | 34.5 (1.2) |
| Agree | 885 | 40.3 (1.2) |
| Neutral | 298 | 12.6 (0.8) |
| Disagree | 266 | 10.2 (0.7) |
| Strongly disagree | 72 | 2.4 (0.3) |
| Missing | 27 | |
SE = standard error.
The cost communication questions tended to be moderately correlated (Table 3), particularly the questions about wanting to know about, talk about, and consider costs before treatment. The questions on preferring to discuss costs with someone other than their physician and being able to find information about costs were not meaningfully correlated with other questions.
Table 3.
Correlation between cost communication items
| Communication item | Spearman rank correlation coefficient by item | ||||||
|---|---|---|---|---|---|---|---|
| 1a | 2 | 3 | 4 | 5 | 6 | 7 | |
| 1. I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment. | 1.00 | 0.49b | −0.12b | 0.45c | −0.03 | 0.53b | 0.45b |
| 2. I am comfortable talking about the cost of my care with my doctor. | 1.00 | −0.14b | 0.28b | 0.15b | 0.32b | 0.27b | |
| 3. I would prefer to discuss the cost of my care with someone other than my doctor, such as a nurse, social worker, or financial counselor. | 1.00 | 0.04 | 0.04c | 0.02 | 0.06b | ||
| 4. I prefer to know about the out-of-pocket costs for my treatment before I am treated. | 1.00 | −0.03 | 0.32b | 0.48b | |||
| 5. I can find information about the cost of my health care when I need it. | 1.00 | −0.01 | 0.02 | ||||
| 6. My doctor should consider my out-of-pocket costs as s/he makes a medical decision. | 1.00 | 0.48b | |||||
| 7. I consider my out-of-pocket costs when I make a decision about my care. | 1.00 | ||||||
These questions appear with different numbers in the survey.
p < 0.01.
p > 0.01 but < 0.05.
Financial Burden and Delay in Care-Seeking
One in 5 people (21.6%) reported current family medical debt (Table 4). One in 10 (10.9%) reported delaying using medication because of cost in the previous 12 months; nearly one-third reported having delayed medical care because of worry about cost, either care for themselves (31.5%) or for a family member (32.3%). Overall, 43.7% reported one or more care-delaying behaviors. In unadjusted analyses, people with current medical financial debt were slightly more likely to prefer talking about out-of-pocket costs with their physician at the point of care compared with people who did not; this difference was not statistically significant. In contrast, people with delay in seeking care were significantly more likely to want to talk about costs with their physician, with relative risks ranging from 1.09 (95% confidence interval [CI] = 1.02–1.16, delayed medication) to 1.14 (CI = 1.09–1.19, delayed care for family member).
Table 4.
Bivariate associations: Communication preferences and medical financial burden and delay in care-seeking
| Question | Reported burden or delay because of costs | I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment (“Agree” or “Strongly agree”) | |||
|---|---|---|---|---|---|
| na | Weighted % (SE) | Weighted % (SE) | Relative risk (95% CI) | p value | |
| Burden | |||||
| Does anyone in your family currently have any medical bills that are being paid off over time? | |||||
| No | 1781 | 78.4 (1.1) | 80.6 (1.0) | Referent | 0.15 |
| Yes | 417 | 21.6 (1.1) | 83.9 (2.0) | 1.04 (0.99–1.10) | |
| Delay | |||||
| In the past 12 months, was there a time when you did not take your medication as prescribed because of cost? Do not include over-the-counter medication. | |||||
| No | 1988 | 89.1 (0.8) | 80.6 (1.0) | Referent | 0.01 |
| Yes | 212 | 10.9 (0.8) | 87.8 (2.5) | 1.09 (1.02–1.16) | |
| In the past 12 months, have you delayed seeking medical care because of worry about the cost? | |||||
| No | 1625 | 68.5 (1.2) | 78.5 (1.1) | Referent | < 0.01 |
| Yes | 572 | 31.5 (1.2) | 87.7 (1.5) | 1.12 (1.07–1.17) | |
| In the past 12 months, has anyone in your family delayed medical care because of worry about the cost? | |||||
| No | 1582 | 67.7 (1.2) | 78.0 (1.2) | Referent | < 0.01 |
| Yes | 608 | 32.3 (1.2) | 88.7 (1.4) | 1.14 (1.09–1.19) | |
| Any delay due to costb | |||||
| No | 1392 | 56.3 (1.2) | 77.8 (1.2) | Referent | < 0.01 |
| Yes | 818 | 43.7 (1.2) | 85.9 (1.3) | 1.10 (1.06–1.15) | |
Missing values: burden (n = 22), delay medications (n = 20), delay care-self (n = 23), and delay care-family (n = 30).
Ten respondents did not respond to any of the delay items and were coded as missing.
CI = confidence interval; SE = standard error.
Characteristics Associated with Preference for Talking about Out-of-Pocket Costs
Characteristics associated with wanting to talk about costs with one’s physician included younger age, nonwhite race, annual household income below $75,000, currently employed, Medicare/Medicaid insurance, shorter enrollment duration, low likelihood of recommending Group Health, and low resource utilization (Table 5).
Table 5.
Bivariate estimates: Characteristics associated with communication preferencesa
| Characteristic | Weighted % (SE) | Relative risk (95% CI) | p value |
|---|---|---|---|
| Age, years | |||
| 18–44 | 86.1 (1.6) | Referent | < 0.01 |
| 45–64 | 80.2 (1.4) | 0.93 (0.89–0.98) | |
| ≥ 65 | 75.1 (1.6) | 0.87 (0.83–0.92) | |
| Sex | |||
| Men | 79.9 (1.4) | Referent | 0.16 |
| Women | 82.5 (1.2) | 1.03 (0.99–1.08) | |
| Race | |||
| White non-Hispanic | 80.7 (1.0) | Referent | < 0.01 |
| Asian non-Hispanic | 89.9 (2.6) | 1.11 (1.05–1.18) | |
| Other | 83.4 (2.8) | 1.03 (0.96–1.11) | |
| Marital status | |||
| Married or living with partner | 81.0 (1.1) | Referent | 0.30 |
| Divorced, separated, or widowed | 80.9 (2.1) | 1.05 (0.99–1.12) | |
| Single | 85.1 (2.5) | 1.05 (0.99–1.12) | |
| Education | |||
| Less than 4-year college degree | 82.6 (1.3) | Referent | 0.21 |
| 4 years college or more | 80.3 (1.3) | 0.97 (0.93–1.02) | |
| Income | |||
| < $75,000 | 84.0 (1.2) | Referent | < 0.01 |
| ≥ $75,000 | 78.3 (1.5) | 0.92 (0.89–0.98) | |
| Employed | |||
| Yes | 84.1 (1.1) | Referent | < 0.01 |
| No | 76.3 (1.5) | 0.91 (0.86–0.95) | |
| Own home | |||
| No | 84.8 (1.8) | Referent | 0.02 |
| Yes | 79.9 (1.1) | 0.94 (0.90–0.99) | |
| Insurance type | |||
| Commercial | 74.2 (1.8) | Referent | < 0.01 |
| Medicare/Medicaid | 83.0 (1.0) | 1.12 (1.06–1.18) | |
| Enrolled in group practice | |||
| No | 82.8 (1.7) | Referent | 0.27 |
| Yes | 80.6 (1.1) | 0.97 (0.93–1.02) | |
| Enrollment duration, years | |||
| < 10 | 84.7 (1.3) | Referent | < 0.01 |
| ≥ 10 | 77.6 (1.3) | 0.92 (0.88–0.96) | |
| Would recommend Group Health | |||
| Yes (score 8–10) | 78.7 (1.3) | Referent | < 0.01 |
| No (score 0–7) | 85.1 (1.3) | 1.08 (1.04–1.13) | |
| Resource utilization band | |||
| Low (0–2) | 84.5 (1.7) | Referent | < 0.01 |
| Medium (3) | 82.1 (1.3) | 0.97 (0.93–1.02) | |
| High (4–5) | 74.3 (2.1) | 0.88 (0.82–0.94) | |
| Days in poor physical health | |||
| 0 | 83.2 (1.2) | Referent | 0.13 |
| 1–15 | 79.7 (1.6) | 0.96 (0.91–1.00) | |
| ≥ 16 | 78.8 (3.0) | 0.95 (0.87–1.03) | |
| Days in poor mental health | |||
| 0 | 81.4 (1.3) | Referent | 0.88 |
| 1–15 | 81.0 (1.5) | 1.00 (0.95–1.04) | |
| ≥ 16 | 82.7 (3.1) | 1.02 (0.94–1.10) | |
Communication preference defined as responding “strongly agree” or “agree” to the survey statement, “I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment.”
CI = confidence interval; SE = standard error.
In multivariable analysis, delay in seeking care remained an independent positive predictor of communication preference (relative risk = 1.06, 95% CI = 1.01–1.12; Table 6); lower income and low likelihood of recommending Group Health were the only other independent predictors. Current financial burden, age, sex, employment, home ownership, enrollment duration, insurance type, and predicted resource utilization were not independently associated with communication preferences.
Table 6.
Multivariate model: Characteristics associated with communication preferencea
| Characteristic | Relative risk (95% CI) | p value |
|---|---|---|
| Any financial burden | 0.99 (0.93–1.05) | 0.65 |
| Any delay in care because of cost | 1.06 (1.01–1.12) | 0.03 |
| Women | 1.02 (0.97–1.07) | 0.40 |
| Age, years | ||
| 21–44 | Referent | 0.65 |
| 45–64 | 0.98 (0.92–1.03) | |
| ≥ 65 | 0.96 (0.86–1.08) | |
| Race | ||
| White | Referent | 0.24 |
| Asian non-Hispanic | 1.06 (0.98–1.15) | |
| Other | 0.98 (0.90–1.06) | |
| Resource utilization band (RUB) | ||
| Low (0–2) | Referent | 0.24 |
| Medium (3) | 1.00 (0.95–1.06) | |
| High (4–5) | 0.94 (0.87–1.02) | |
| Other characteristics | ||
| Household income ≥ $75,000 | 0.93 (0.88–0.99) | 0.01 |
| Not employed | 0.95 (0.90–1.01) | 0.12 |
| Own home | 0.99 (0.93–1.04) | 0.62 |
| Medicare/Medicaid | 1.00 (0.90–1.12) | 0.93 |
| Enrolled ≥ 10 years | 0.96 (0.91–1.01) | 0.11 |
| Would not recommend Group Health | 1.07 (1.02–1.13) | 0.01 |
Communication preference defined as responding “strongly agree” or “agree” to the survey statement, “I would like my doctor to talk with me about my out-of-pocket costs when s/he recommends a test or treatment.”
CI = confidence interval.
DISCUSSION
We conducted a cross-sectional, self-administered survey of a population-based sample of an integrated health care system membership in the Pacific Northwest. We found that most participants prefer communication about cost with their physicians. Delays in seeking care because of cost and current medical financial burden were notably prevalent in our sample, and preferences for physician communication of out-of-pocket costs were higher among patients reporting delay in seeking care. In multivariable analyses, delay in care emerged as an independent predictor of communication preferences, whereas current medical financial burden did not. Other independent predictors of preferring communication with one’s physician about out-of-pocket costs were lower income and lower likelihood to recommend Group Health.
We found higher levels both for wanting to know costs before treatment (92%) and preferences for discussing costs (81%) than in previous reports.19.21 A survey of people with cancer found high preferences for knowing costs before treatment initiation (68%) and for discussing costs with one’s physician (59%)19; other surveys of cancer populations have found similar results.13,22 Prior work on clinic-based cost communication has focused on people with cancer, whereas our sample is a broad sample of an insured population. Because more than half of Americans have tried to locate cost information before receiving care,12p11 it is not clear whether the higher levels of preference for cost communication in our study reflect temporal changes in patient preferences over time, or differences between the preferences of an insured population overall compared with cancer-specific populations. The reasons for these differences should be explored in future studies.
One in 5 enrollees reported medical debt, comparable with national estimates.21 We found higher rates (31.4%) of cost-related delay in seeking care compared with estimates from the Centers for Disease Control and Prevention of 7.1% in an insured population,22 although this may reflect Washington State’s higher estimates compared with national estimates (12.5% vs 9.9% in 2011–2012).23 Delay in seeking care may be as high as 57% in people with low income and health insurance deductibles greater than $500, although financial burden measures may be only moderately reliable in this population.24 One in 10 people (10.9%) in our sample experienced a delay in using a prescribed medication compared with 6.1% of adults aged 18 to 64 years with private coverage reported in the 2013 National Health Interview Survey.3
We hypothesized that financial burden and delay in seeking care would both be independently related to communication preferences, but only delay in seeking care emerged as an independent predictor. This finding may suggest that it may be “too late” to intervene with cost communications in those who have already incurred medical debt, but it may also suggest an opportunity for physicians to intervene during clinical encounters, which is supported by other studies.25,26 It may also suggest opportunities for outreach to patients to reassure that cost concerns will be addressed proactively, and to encourage timely care. Further prospective studies can examine the best ways to meet patients’ needs for financial information at the point of care without adversely affecting clinic flow or patient experience.
We acknowledge several limitations. The response rate to the survey was low, but it was similar to large national surveys27 and offset by our use of sampling frame data to conduct weighted analyses accounting for nonresponse and for unbalanced sampling probabilities.28 Although we adjusted for age, the main predictor of response, and comorbidity in the multivariable model, unmeasured sources of bias could remain. For example, people who had experienced medical financial burden or delay in care may have been more likely to respond to the survey. However, financial burden and delay are more prevalent in younger age groups, whereas older people were most likely to respond to the survey. Furthermore, these questions were embedded in a larger survey, so we do not expect differential response rate based on financial burden or delays in seeking care alone.
Our data reflect the experiences of people in a single integrated care system and may have limited generalizability to other populations. However, we were intentionally as minimally restrictive in our sampling strategy as possible to increase generalizability, the organization’s membership mirrors the socioeconomic demographics of the state’s insured population, and we found no differences in preferences on the basis of whether care is provided in Group Health group practice clinics. Nonetheless, our findings should be explored in other populations.
Finally, our use of items from previously published studies was intended to ease benchmarking to other work, but biases in the interpretation or design of specific items designed for other surveys may remain.
CONCLUSION
Our study findings add to the growing evidence base that people are comfortable with and want direct communication about their costs in the clinical setting. Communication about costs is associated with medication adherence and out-of-pocket spending11,29 and may affect health outcomes over time.30,31 Physicians are increasingly in favor of cost data being available in the clinical setting29 and in the expansion of the traditional medical role to include screening for and assistance with financial distress.25,32–37 Early evidence suggests that it is feasible to present costs in the clinical setting,38–40 and multiple organizations have implemented price transparency initiatives that provide pricing for planned or bundled services. However, strategies for assessing patients’ financial need, optimal clinical workflow and physician health care team access to prices, and presentation of price information are still not well understood and likely vary across health care systems. Our study results affirm that health insurance does not protect from financial hardship or delay in seeking care because of medical expenses,41,42 even in an era after the Affordable Care Act,43 and these patient-reported situations remain prevalent. Our study findings also affirm that a significant portion of patients have difficulty locating information about health care costs; most prefer communication about out-of-pocket costs at the point of care, and most prefer it with a physician and not another member of the team. The degree to which assurance of communication about costs with one’s physician can reassure patients and avoid either delay in seeking care or medical debt deserves further study. However, given these clear patient preferences and the mounting evidence that physicians themselves are willing to help address patients’ financial considerations, health care systems and organizations have a clear mandate for action, and clinic-based tools to aid in communication about costs should be available.
A Social System
A physician and a patient taken together make up a social system. They do so because they are two and because they have relations of mutual dependence. Also they are heterogeneous, they manifest sentiments, they have economic interests, they talk, reason, pretend to reason, and rationalize.
— Lawrence Joseph Henderson, 1878–1942, American physiologist, chemist, biologist, philosopher, and sociologist
Acknowledgments
This study was funded by the Group Health Research Institute Director’s Development Fund.
Kathleen Louden, ELS, of Louden Health Communications provided editorial assistance.
Footnotes
Disclosure Statement
The author(s) have no conflicts of interest to disclose.
References
- 1.Shankaran V, Jolly S, Blough D, Ramsey SD. Risk factors for financial hardship in patients receiving adjuvant chemotherapy for colon cancer: A population-based exploratory analysis. J Clin Oncol. 2012 May 10;30(14):1608–14. doi: 10.1200/JCO.2011.37.9511. DOI: https://doi.org/10.1200/jco.2011.37.9511. [DOI] [PubMed] [Google Scholar]
- 2.Azzani M, Roslani AC, Su TT. The perceived cancer-related financial hardship among patients and their families: A systematic review. Support Care Cancer. 2015 Mar;23(3):889–98. doi: 10.1007/s00520-014-2474-y. DOI: https://doi.org/10.1007/s00520-014-2474-y. [DOI] [PubMed] [Google Scholar]
- 3.Cohen RA, Kirzinger WK. Financial burden of medical care: A family perspective [Internet] Hyattsville, MD: National Center for Health Statistics, Centers for Disease Control and Prevention; 2014. Jan, [cited 2016 Jul 20] Available from: www.cdc.gov/nchs/products/databriefs/db142.htm. [Google Scholar]
- 4.QuickStats: Percentage of persons who delayed or did not receive medical care during the preceding year because of cost, by health status—National Health Interview Survey, United States, 2014. MMWR Morb Mortal Wkly Rep. 2015 Dec 11;64(48):1351. [Google Scholar]
- 5.Kent EE, Forsythe LP, Yabroff KR, et al. Are survivors who report cancer-related financial problems more likely to forgo or delay medical care? Cancer. 2013 Oct 15;119(20):3710–7. doi: 10.1002/cncr.28262. DOI: https://doi.org/10.1002/cncr.28262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fenn KM, Evans SB, McCorkle R, et al. Impact of financial burden of cancer on survivors’ quality of life. J Oncol Pract. 2014 Sep;10(5):332–8. doi: 10.1200/JOP.2013.001322. DOI: https://doi.org/10.1200/jop.2013.001322. [DOI] [PubMed] [Google Scholar]
- 7.Rehkopf DH, Berkman LF, Coull B, Krieger N. The non-linear risk of mortality by income level in a healthy population: US National Health and Nutrition Examination Survey mortality follow-up cohort, 1988–2001. BMC Public Health. 2008 Nov 10;8:383. doi: 10.1186/1471-2458-8-383. DOI: https://doi.org/10.1186/1471-2458-8-383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stump TK, Eghan N, Egleston BL, et al. Cost concerns of patients with cancer. J Oncol Pract. 2013 Sep;9(5):251–7. doi: 10.1200/JOP.2013.000929. DOI: https://doi.org/10.1200/jop.2013.000929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Valtorta NK, Hanratty B. Socioeconomic variation in the financial consequences of ill health for older people with chronic diseases: A systematic review. Maturitas. 2013 Apr;74(4):313–33. doi: 10.1016/j.maturitas.2013.01.015. DOI: https://doi.org/10.1016/j.maturitas.2013.01.015. [DOI] [PubMed] [Google Scholar]
- 10.Aiga H. Hidden out-of-pocket payments can burden patients’ families but are under-recognised. BMJ. 2015 Sep 23;351:h4990. doi: 10.1136/bmj.h4990. DOI: https://doi.org/10.1136/bmj.h4990. [DOI] [PubMed] [Google Scholar]
- 11.Zafar SY, Chino F, Ubel PA, et al. The utility of cost discussions between patients with cancer and oncologists. Am J Manag Care. 2015 Sep;21(9):607–15. [PubMed] [Google Scholar]
- 12.Schleifer D, Hagelskamp C, Rinehart C. How much will it cost? How Americans use prices in health care [Internet] San Francisco, CA: Public Agenda; 2015. Mar 9, [cited 2016 Jul 20]. Available from: www.publicagenda.org/files/HowMuchWillItCost_PublicAgenda_2015.pdf. [Google Scholar]
- 13.Meisenberg BR, Varner A, Ellis E, et al. Patient attitudes regarding the cost of illness in cancer care. Oncologist. 2015 Oct;20(10):1199–204. doi: 10.1634/theoncologist.2015-0168. DOI: https://doi.org/10.1634/theoncologist.2015-0168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Henrikson NB, Tuzzio L, Loggers ET, Miyoshi J, Buist DS. Patient and oncologist discussions about cancer care costs. Support Care Cancer. 2014 Apr;22(4):961–7. doi: 10.1007/s00520-013-2050-x. DOI: https://doi.org/10.1007/s00520-013-2050-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Greene SM, Tuzzio L, Cherkin D. A framework for making patient-centered care front and center. Perm J. 2012 Summer;16(3):49–53. doi: 10.7812/tpp/12-025. DOI: https://doi.org/10.7812/TPP/12-025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Street RL, Jr, Makoul G, Arora NK, Epstein RM. How does communication heal? Pathways linking clinician-patient communication to health outcomes. Patient Educ Couns. 2009 Mar;74(3):295–301. doi: 10.1016/j.pec.2008.11.015. DOI: https://doi.org/10.1016/j.pec.2008.11.015. [DOI] [PubMed] [Google Scholar]
- 17.Epstein RM, Street RL., Jr . Patient-centered communication in cancer care: Promoting healing and reducing suffering. Bethesda, MD: National Cancer Institute, US Department of Health and Human Services, National Institutes of Health; 2007. NIH Publication No. 07-6225 [Internet] [cited 2016 Jul 20]. Available from: http://appliedresearch.cancer.gov/areas/pcc/communication/pcc_monograph.pdf. [Google Scholar]
- 18.The Johns Hopkins ACG System [Internet] Baltimore, MD: Johns Hopkins University; 2015. [cited 2015 Dec 18]. Available from: https://acg.jhsph.org/ [Google Scholar]
- 19.Bullock AJ, Hofstatter EW, Yushak ML, Buss MK. Understanding patients’ attitudes toward communication about the cost of cancer care. J Oncol Pract. 2012 Jul;8(4):e50–8. doi: 10.1200/JOP.2011.000418. DOI: https://doi.org/10.1200/jop.2011.000418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.StataCorp. Stata statistical software: Release 13 [Internet] College Station, TX: StataCorp LP; 2013. [cited 2016 Jul 20] Available from: www.stata.com. [Google Scholar]
- 21.Irwin B, Kimmick G, Altomare I, et al. Patient experience and attitudes toward addressing the cost of breast cancer care. Oncologist. 2014 Nov;19(11):1135–40. doi: 10.1634/theoncologist.2014-0117. DOI: https://doi.org/10.1634/theoncologist.2014-0117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Table 74. Delay or nonreceipt of needed medical care, nonreceipt of needed prescription drugs, or nonreceipt of needed dental care during the past 12 months due to cost, by selected characteristics: United States, selected years 1997–2012 [Internet] Atlanta, GA: National Center for Health Statistics, Centers for Disease Control and Prevention; 2013. [cited 2015 Dec 18] Available from: www.cdc.gov/nchs/hus/contents2013.htm#074. [Google Scholar]
- 23.Table 76. Delay or nonreceipt of needed medical care during the past 12 months due to cost, by state: 25 most populous states and United States, average annual, selected years 1997–1998 through 2011–2012. Atlanta, GA: National Center for Health Statistics, Centers for Disease Control and Prevention; 2013. [cited 2016 Jan 15] Available from: www.cdc.gov/nchs/hus/contents2013.htm#076. [Google Scholar]
- 24.Penfold RB, Kullgren JT, Miroshnik I, Galbraith AA, Hinrichsen VL, Lieu TA. Reliability of a patient survey assessing cost-related changes in health care use among high deductible health plan enrollees. BMC Health Serv Res. 2011 May 27;11:133. doi: 10.1186/1472-6963-11-133. DOI: https://doi.org/10.1186/1472-6963-11-133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Moriates C, Shah NT, Arora VM. First, do no (financial) harm. JAMA. 2013 Aug 14;310(6):577–8. doi: 10.1001/jama.2013.7516. DOI: https://doi.org/10.1001/jama.2013.7516. [DOI] [PubMed] [Google Scholar]
- 26.Khera N. Reporting and grading financial toxicity. J Clin Oncol. 2014 Oct 10;32(29):3337–8. doi: 10.1200/JCO.2014.57.8740. DOI: https://doi.org/10.1200/jco.2014.57.8740. [DOI] [PubMed] [Google Scholar]
- 27.Behavioral risk factor surveillance system 2013 summary data quality report [Internet] Atlanta, GA: Centers for Disease Control and Prevention; 2014. Aug 15, [cited 2016 Jul 20]. Available from: www.cdc.gov/brfss/annual_data/2013/pdf/2013_dqr.pdf. [Google Scholar]
- 28.Johnson TP, Wislar JS. Response rates and nonresponse errors in surveys. JAMA. 2012 May 2;307(17):1805–6. doi: 10.1001/jama.2012.3532. DOI: https://doi.org/10.1001/jama.2012.3532. [DOI] [PubMed] [Google Scholar]
- 29.Tilburt JC, Wynia MK, Montori VM, et al. Shared decision-making as a cost-containment strategy: US physician reactions from a cross-sectional survey. BMJ Open. 2014 Jan 14;4(1):e004027. doi: 10.1136/bmjopen-2013-004027. DOI: https://doi.org/10.1136/bmjopen-2013-004027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Shippee TP, Wilkinson LR, Ferraro KF. Accumulated financial strain and women’s health over three decades. J Gerontol B Psychol Sci Soc Sci. 2012 Sep;67(5):585–94. doi: 10.1093/geronb/gbs056. DOI: https://doi.org/10.1093/geronb/gbs056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tucker-Seeley RD, Harley AE, Stoddard AM, Sorensen GG. Financial hardship and self-rated health among low-income housing residents. Health Educ Behav. 2013 Aug;40(4):442–8. doi: 10.1177/1090198112463021. DOI: https://doi.org/10.1177/1090198112463021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Riggs KR, Ubel PA. Overcoming barriers to discussing out-of-pocket costs with patients. JAMA Intern Med. 2014 Jun;174(6):849–50. doi: 10.1001/jamainternmed.2014.853. DOI: https://doi.org/10.1001/jamainternmed.2014.853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Guy GP, Jr, Ekwueme DU, Yabroff KR, et al. Economic burden of cancer survivorship among adults in the United States. J Clin Oncol. 2013 Oct 20;31(30):3749–57. doi: 10.1200/JCO.2013.49.1241. DOI: https://doi.org/10.1016/j.jval.2013.03.666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Meropol NJ, Schrag D, Smith TJ, et al. American Society of Clinical Oncology. American Society of Clinical Oncology guidance statement: The cost of cancer care. J Clin Oncol. 2009 Aug 10;27(23):3868–74. doi: 10.1200/JCO.2009.23.1183. DOI: https://doi.org/10.1200/jco.2009.23.1183. [DOI] [PubMed] [Google Scholar]
- 35.Schnipper LE, Davidson NE, Wollins DS, et al. American Society of Clinical Oncology. American Society of Clinical Oncology statement: A conceptual framework to assess the value of cancer treatment options. J Clin Oncol. 2015 Aug 10;33(23):2563–77. doi: 10.1200/JCO.2015.61.6706. DOI: https://doi.org/10.1200/jco.2015.61.6706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ubel PA, Abernethy AP, Zafar SY. Full disclosure—out-of-pocket costs as side effects. N Engl J Med. 2013 Oct 17;369(16):1484–6. doi: 10.1056/NEJMp1306826. DOI: https://doi.org/10.1056/nejmp1306826. [DOI] [PubMed] [Google Scholar]
- 37.Ramsey SD, Ganz PA, Shankaran V, Peppercorn J, Emanuel E. Addressing the American health-care cost crisis: Role of the oncology community. J Natl Cancer Inst. 2013 Dec 4;105(23):1777–81. doi: 10.1093/jnci/djt293. DOI: https://doi.org/10.1093/jnci/djt293. [DOI] [PubMed] [Google Scholar]
- 38.Kelly RJ, Forde PM, Elnahal SM, Forastiere AA, Rosner GL, Smith TJ. Patients and physicians can discuss costs of cancer treatment in the clinic. J Oncol Pract. 2015 Jul;11(4):308–12. doi: 10.1200/JOP.2015.003780. DOI: https://doi.org/10.1200/jop.2015.003780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Henrikson N, Tuzzio L, Ralston J, Buist DS. Designing a cost comparison tool for cancer patients: “The more people know ahead of time, the better”. Poster presentation at the AcademyHealth 2013 Annual Research Meeting; 2013 Jun 23–25; Baltimore, MD. Washington, DC: AcademyHealth; 2013. [Google Scholar]
- 40.Blumenthal-Barby JS, Robinson E, Cantor SB, Naik AD, Russell HV, Volk RJ. The neglected topic: Presentation of cost information in patient decision aids. Med Decis Making. 2015 May;35(4):412–8. doi: 10.1177/0272989X14564433. DOI: https://doi.org/10.1177/0272989x14564433. [DOI] [PubMed] [Google Scholar]
- 41.Herman PM, Rissi JJ, Walsh ME. Health insurance status, medical debt, and their impact on access to care in Arizona. Am J Public Health. 2011 Aug;101(8):1437–43. doi: 10.2105/AJPH.2010.300080. DOI: https://doi.org/10.2105/ajph.2010.300080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer treatment: A pilot study assessing out-of-pocket expenses and the insured cancer patient’s experience. Oncologist. 2013;18(4):381–90. doi: 10.1634/theoncologist.2012-0279. DOI: https://doi.org/10.1634/theoncologist.2012-0279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.The Patient Protection and Affordable Care Act of 2010. Public Law 111-148, 111th Congress, 124 Stat 119, HR 3590, enacted 2010 Mar 23.