糖心vlog

Key Points

Question听 How do households respond to medical bills they disagree with or cannot afford?

Findings听 In this cross-sectional survey of a representative sample of people in the US, 1-in-5 reported receiving a medical bill they disagree with or cannot afford; 61% of them reached out to the billing office to address their concern; less educated respondents, those with lower financial literacy, and the uninsured were less likely to reach out to a billing office. Most respondents who reached out reported financial relief, bill corrections, or better understanding of the bill.

Meaning听 Differences in self-advocacy may be exacerbating socioeconomic inequalities in medical debt burden.

Importance听 People in the US face high out-of-pocket medical expenses, yielding financial strain and debt.

Objective听 To understand how households respond to medical bills they disagree with or cannot afford.

Design, Setting, and Participants听 A retrospective cohort study was carried out using a survey fielded between August 14 and October 14, 2023. The study included a random sample of adult (aged 鈮�18 years) survey respondents from the Understanding America Study (UAS). Participant responses were weighted to be nationally representative. The analysis took place from November 3, 2023, through January 8, 2024.

Main Outcomes and Measures听 Respondents reported if their household received a medical bill that they could not afford or did not agree with in the prior 12 months, and if anyone contacted the billing office regarding their concerns. Those who did reach out were asked about their experience and those who did not were asked why.

Results听 The survey was sent to 1233 UAS panelists, of which 1135 completed the survey, a 92.1% cooperation rate. Overall, 1 in 5 of the 1135 respondents received a medical bill that they disagreed with or could not afford. Leading bill sources were physician offices (66 [34.6%]), emergency room or urgent care (22 [19.9%]), and hospitals (31 [15.3%]), and 136 respondents (61.5%) contacted the billing office to address their concern. A more extroverted and less agreeable personality increased likelihood of reaching out. Respondents without a college degree, lower financial literacy, and the uninsured were less likely to contact a billing office. Among those who did not reach out, 55 (86.1%) reported that they did not think it would make a difference. Of those who reached out, 37 (25.7%) achieved bill corrections, better understanding (16 [18.2%]), payment plans (18 [15.5%]), price drop (17 [15.2%]), financial assistance (10 [8.1%]), and/or bill cancellation (6 [7.3%]), while 32 (21.8%) said that the issue was unresolved and 23.8% reported no change. These outcomes aligned well with respondents鈥� billing concerns with financial relief for 75.8% of respondents reaching out about an unaffordable bill, bill corrections for 73.7% of those who thought there was mistake, and a price drop for 61.8% of those who negotiated.

Conclusions and Relevance听 This cross-sectional survey of a representative sample of patients in the US found that most respondents who self-advocated achieved bill corrections and payment relief. Differences in self-advocacy may be exacerbating socioeconomic inequalities in medical debt burden, as those with less education, lower financial literacy, and the uninsured were less likely to self-advocate. Policies that streamline the administrative burden or shift it from patients to the billing clinician may counter these disparities.

More than half (57%) of US adults have been burdened by medical bills within the past 5 years, and the burden is higher among those with lower levels of education, lower income, women, and Black and Hispanic adults.¹ Socioeconomic and demographic disparities in the financial burden of health care are shaped by insurance coverage, utilization, and ability to pay. There may also be differences in individuals鈥� willingness and ability to navigate avenues of financial relief and bill correction that contribute to disparate burden.

With high out-of-pocket costs, the onus is on patients to self-advocate by identifying billing errors, seeking financial assistance, and setting up payment plans. These processes can be complex, confusing, time intensive, and administratively cumbersome for patients.² These challenges are widely recognized, and books, blogs, and podcasts have emerged in recent years informing people about how to navigate their medical bills.^3-5 Yet, administrative barriers have been shown to yield disparate effects on enrollment in social programs and benefits from welfare promoting policies.^6,7

Recent surveys have documented uninsurance, underinsurance, patients鈥� financial strain, and medical debt.^8-10 These surveys describe utilization and resulting cost burden, but do not delve into the intermediate step of patients鈥� receiving and responding to bills they disagree with or cannot afford. To bridge this gap in knowledge about how patients navigate medical bills, a nationally representative panel was surveyed to probe experiences with medical billing.

Participants in the Understanding America Study (UAS), which is maintained by the Center for Economic and Social Research (CESR) at the University of Southern California, were sampled in this study. The UAS is a probability-based internet panel of approximately 13鈥�000 respondents residing in the United States (Supplement 1).¹¹ Recruitment of the initial panel was by paper survey (41.3% response rate). Respondents answered surveys on a computer, tablet, or smartphone. The University of Southern California institutional review board approved this study and written informed consent was obtained. This report follows the reporting guidelines for survey studies. The analysis took place from November 3, 2023, through January 8, 2024.

A sample of 1233 panelists were recruited to respond to this survey between August 14 and October 14, 2023. Respondents鈥� sample weights were calculated using a 2-step approach to make the survey dataset representative of the US population aged 18 years and older with respect to a predefined set of sociodemographic variables. The first step computes base weights to correct for unequal probabilities of sampling UAS members. The second step generates a final poststratification weight, allowing the sample to align with the reference population along certain sociodemographic dimensions.

Respondents were asked if their household received a medical bill that they could not afford or did not agree with in the prior 12 months. If yes, they were asked about the source of the bill and if anyone reached out to the billing office to address their concerns. Respondents who did not reach out were asked why not. Those who did reach out were asked a series of questions about their experience, including the objective for reaching out and the outcome, modality, time, and feelings associated with the interaction. The full survey instrument is shown in Supplement 1.

Information about respondents from previous UAS surveys was used in analyses, including demographics (age, gender, marital status, race and ethnicity, national origin), socioeconomic characteristics (education, household income), insurance type (private, Medicaid, Medicare, military, uninsured), number of chronic conditions, financial literacy score (score range, 0-14),¹² and Big Five personality scores: extroversion (score range, 8-40), agreeableness (score range, 9-45), conscientiousness (score range, 9-45), neuroticism (score range, 8-40), and openness (score range, 10-50).^13,14 Details about these measures are provided in Supplement 1.

Raw count values and weighted percentages with 95% CIs are reported in describing the sample characteristics and survey responses. 蠂² tests were used to compare the attributes of respondents who did and did not reach out to the billing office, conditional on receiving a problematic bill. Weighted multivariable probit regression models were used to assess the association between respondent characteristics and the outcomes of reporting receipt of a problematic bill and, conditional on receiving a problematic bill, whether or not anyone reached out to address the concern. For each of these 2 outcomes, 1 model was fit assessing association with demographic and socioeconomic characteristics and a second model was fit assessing association with financial literacy and personality. Separating these 2 types of characteristics addresses concerns about overfitting and collinearity, and also enables discrete measurement of traits that are more easily observed and traits that require in-depth skill and personality testing. Regression results are reported in Supplement 1 and presented as predicted probabilities with 95% CIs in Figure 1 and Figure 2. The predicted probabilities of continuous characteristics were calculated at the 10th, 50th, and 90th percentiles of the sample distribution. Analyses were conducted using Stata statistical software (version 17, StataCorp). A 2-sided significance level of .05 was used.

The survey was sent to 1233 UAS panelists, of which 1135 completed the survey, a 92.1% cooperation rate. The demographics, health status, socioeconomic characteristics, financial literacy skills, and personality traits of the study sample are described in Table 1. Weighting for national representation, the sample included 652 women (51.2%) (95% CI, 47.0%-55.4%), 672 were married (59.6%) (95% CI, 55.3%-63.7%), 1051 (95% CI, 87.0%-92.7%) were born in the US (90.2%), and the age distribution was 103 (95% CI, 15.8%-23.8%) aged 20 to 34 years (19.5%), 366 (95% CI, 36.4%-44.8%) aged 35 to 54 years (40.5%), 260 (95% CI, 13.7%-19.2%) aged 55 to 64 years (16.3%), and 426 (95% CI, 20.8%-26.9%) aged 65 years or older (23.7%). The racial and ethnic composition was 89 (95% CI, 9.4%-15.5%) non-Hispanic Black (12.1%), 16.8% (95% CI, 13.3%-20.9%) Hispanic, 89 (95% CI, 56.8%-65.6%) non-Hispanic White (61.3%), and 88 (95% CI, 7.4%-13.2%) other non-Hispanic racial identity (9.9%). Whereas 57 (95% CI, 6.1%-11.3%) of those sampled were uninsured (8.3%), the remainder had private health insurance (682 [57.0%]; 95% CI, 52.8%-61.2%), Medicaid (140 [14.7%]; 95% CI, 11.8%-18.2%), Medicare (187 [11.9%]; 95% CI, 9.8%-14.4%), or Military coverage (27 [3.0%]; 95% CI, 1.8%-4.9%), and health insurance status was unknown for 42 (95% CI, 3.2%-7.6%) in the sample (5.0%). More than half (545 [54.4%]; 95% CI, 50.2%-58.5%) of respondents had no chronic conditions, though 276 (21.9%; 95% CI, 18.7%-25.5%) had 1, 147 (10.9%; 95% CI, 8.7%-13.6%) had 2, and 167 (12.8%; 95% CI, 10.4%-15.7%) had 3 or more. Income and education also vary in the study sample.

Overall, 203 respondents (19.6%; 95% CI, 16.4%-23.1%) reported that their household received a bill that was unaffordable or they disagreed with (Table 1). Among those who received a problematic bill, 136 (61.5%; 95% CI, 51.6%-70.5%) reached out to the billing office and 67 (38.5%; 95% CI, 29.5%-48.4%) did not (Table 2).

Respondents were asked to report the problem(s) with their bill among 4 options (select all that apply): 81 (46.6%; 95% CI, 37.1%-56.4%) could not afford to pay, 88 (43.5%; 95% CI, 34.3%-53.2%) felt the bill was unfairly high, 68 (31.8%; 95% CI, 23.6%-41.4%) felt the bill was too high because of a mistake, and 54 (30.4%; 95% CI, 22.2%-40.1%) reported that the bill seemed confusing (Table 2). The sources of the bills were physician offices (66 [34.6%]; 95% CI, 25.8%-44.5%), emergency room or urgent care (33 [19.9%]; 95% CI, 13.3%-28.8%]), hospital (31 [15.3%]; 95% CI, 9.6%-23.4%), imaging (30 [13.4%]; 95% CI, 8.2%-21.2%), laboratory (16 [6.1%]; 95% CI, 3.1%-11.4%), outpatient procedures (10 [5.1%]; 95% CI, 2.2%-11.6%), pharmacy (4 [4%]), dental (5 [5%]), and other (8 [2.6%]; 95% CI, 1.0%-6.4%). Using 蠂² tests, there was not a statistically significant difference between those who did and did not reach out to the billing office with respect to the prevalence of each of these problems nor source of the bill (Table 2).

Those who reported that nobody reached out to the billing office about the problematic bill were asked to select reasons why from a list (select all that apply) (Table 2). Fifty-five (86.1%; 95% CI, 72.3%-93.7%) reported that they did not think it would change the bill. Eighteen (34.0%; 95% CI, 20.2%-51.1%) did not feel comfortable reaching out, and one-quarter (25.7%; 95% CI, 13.4%-43.6%) did not have time. Overall, 7 (8.7%; 95% CI, 3.0%-23.0%) respondents worried reaching out would harm future care, and 4 (3.7%; 95% CI, 1.1%-11.7%) did not know how to reach out.

Those who reported that someone did reach out to the billing office were asked to select objectives from a list (select all that apply) (Table 2). Many wanted to ask questions about price (77 [56.2%]; 95% CI, 44.0%-67.8%) and/or services (56 [47.2%]; 95% CI, 35.6%-59.2%). About 1 in 5 were seeking financial assistance (21 [19.0%]; 95% CI, 11.2%-30.4%), a payment plan (26 [18.6%]; 95% CI, 11.3%-29.2%), and/or negotiating to pay less (20 [18.1%]; 95% CI, 10.4%-29.5%). Four percent (14 [4.2%]; 95% CI, 2.3%-7.4%) reached out with an insurance issue.

Having more chronic conditions and being female was associated with increased likelihood of reporting receiving a problematic bill, after controlling for insurance type, age, education, income, marital status, and race and ethnicity (Figure 1A). Respondents with 3 or more chronic conditions had a 32.4% (95% CI, 21.0%-43.9%) predicted probability of receiving a problematic bill, compared with 12.8% (95% CI, 9.0%-16.6%) for those with no chronic conditions. Female and male participants had a 22.0% (95% CI, 17.1%-27.0%) and 13.6% (95% CI, 9.5%-17.7%) predicted probability of receiving a problematic bill, respectively. In a separate model, higher neuroticism was associated with greater likelihood of reporting receiving a problematic bill, after controlling for financial literacy and other personality traits (Figure 1B).

Conditional on receiving a bill, respondents who were uninsured or Medicare beneficiaries, less educated, and in the 35 to 44 year age range were less likely to reach out to the billing office, whereas those with military insurance were more likely to reach out, after controlling for other demographic, socioeconomic, and health attributes (Figure 1C). The predicted probability of reaching out was 19.9% (95% CI, 鈭�6.7% to 46.6%) among those without a high school diploma, and above 60% for all higher education groups. Respondents with private insurance had a 68.1% (95% CI, 54.3%-82.0%) predicted probability of reaching out compared with 32.5% (95% CI, 4.2%-60.7%) among the uninsured and 30.1% (95% CI, 1.0%-59.3%) among Medicare beneficiaries. Higher financial literacy, higher extroversion, and lower agreeableness were also associated with greater likelihood of reaching out, after controlling for other personality traits (Figure 1D).

For 32 (95% CI, 13.7%-32.8%) respondents (21.8%) who reported that their household reached out about a household bill, their problem was not solved yet (eTable 1 in Supplement 1). Among the remaining respondents, nothing changed for 30.6% (95% CI, 19.7%-44.2%) of them. One-third (32.9%; 95% CI, 21.8%-46.4%) had a bill corrected and 23.2% (95% CI, 13.2%-37.6%) reported a better understanding of their bill. Many received financial relief in the forms of bill cancellation (9.3%; 95% CI, 2.7%-27.9%), financial assistance (10.4%; 95% CI, 4.4%-22.5%), price reduction (19.4%; 95% CI, 10.7%-32.7%), and/or setting up a payment plan (19.8%; 95% CI, 11.1%-32.9%). These outcomes are not mutually exclusive; many respondents reported multiple outcomes.

We evaluated resolutions stratified by the bill problem and reasons for reaching out (eTable 2 in Supplement 1). Among 34 participants who could not afford their bill and reached out, 14 (49.0%; 95% CI, 26.3%-72.0%) received financial aid, bill cancellation, and/or price reductions. An additional 11 (26.8%; 95% CI, 11.5%-50.8%) received a payment plan. Overall, 75.8% of the respondents who could not afford their bill and reached out received some form of financial relief. Among 37 participants who thought their bill was too high because of a mistake and reached out, 25 (73.7%; 95% CI, 50.2%-88.7%) had their bill corrected. Among 14 participants who reached out wanting to negotiate the price, 10 (61.8%; 95% CI, 10.6%-76.3%) reported that the price was reduced.

Overall, 104 survey respondents communicated with the billing office personally (69.2%; 95% CI, 57.1%-79.1%), whereas 35 reported another household member (30.7%; 95% CI, 20.9%-42.5%) or someone from an advocacy organization (4 [3.5%]; 95% CI, 1.0%-11.6%) reaching out (Table 3). Among those who reached out personally, 96 did so by telephone (87.4%; 95% CI, 74.1%-94.4%) and spent less than 1 hour reaching out (66 [60.2%]; 95% CI, 45.6%-73.2%). The majority strongly agreed (49 [45.7%]; 95% CI, 32.1%-60.0%) or agreed (39 [47.3%]; 95% CI, 33.3%-61.8%) that they were comfortable communicating with the billing office. Whereas 28 (23.4%; 95% CI, 14.0%-36.5%) strongly agreed and 33 (31.2%; 95% CI, 19.0%-46.8%) agreed that they felt treated with respect by the people in the billing office, 10 (11.0%; 95% CI, 4.6%-23.9%) disagreed, and 9 (7.8%; 95% CI, 3.3%-17.6%) strongly disagreed.

In this cross-sectional survey of a representative sample of people in the US, 1 in 5 reported that their household received a medical bill that they could not afford or disagree with and more than half of those (61.5%) reported that someone reached out to the billing office to address the concern. Whereas most respondents (86.1%) who did not reach out reported that they did not think it would make a difference, the experiences of those who did reach out provide evidence to the contrary. Most respondents who reached out reported at least 1 form of financial relief, bill corrections, or better understanding of the bill.

These findings suggest that patients benefit from reaching out to the billing office on receiving a problematic bill, and those who do not reach out may be missing financially advantageous opportunities. The likelihood of receiving problematic bills did not significantly differ by insurance type, education, or income. However, conditional on receiving a problematic bill, those who were uninsured, Medicare insured, and less educated were less likely to reach out to the billing office. These differences in self-advocacy may be widening the gap in financial burden of health care between those with higher and lower socioeconomic status. This is consistent with literature that found administrative burden yields disparate impacts across populations.¹⁵

This study also demonstrates the role of personality traits in self-advocacy behaviors, whereby a more extroverted and less agreeable personality was associated with increased likelihood of reaching out. Only 3.5% of respondents had help from an advocacy organization to reach out to the billing office, and the remaining cases where handled directly by households. Among those who reached out to the billing office themselves, the time burden stretched from an hour or less (60.2% of people) to more than 5 hours (5.0% of people). Nearly 1 in 5 (18.8%) respondents who reached out personally did not feel respected by the billing office staff. These process burden findings, paired with the finding that 38.5% of respondents did not reach out at all, suggests that there may be unmet need for formal assistance navigating bills.

Nearly one-third (31.8%) of respondents with a problematic bill suspected that their bill was too high because of a billing error. Among those who reached out about a billing error, most (73.7%) reported that a mistake on the bill was corrected. This indicates that billing errors are not uncommon and self-advocacy to correct errors is often successful. Consumer advocates report that common billing errors include double billing and upcoding, inaccuracies when medical billers create claims without complete clinical information, and erroneous reductions or denials on the part of insurers adjudicating claims.¹⁶

Efforts to streamline the administrative burden of addressing billing issues may improve access to financial relief for more patients. For example, as of 2021 there were 13 states that mandated hospitals screen patients for insurance eligibility and eligibility for other programs, including the entities鈥� charity or discounting policy.¹⁷ Policies that shift administrative responsibility from patients to the billing health care professional may counter the self-advocacy disparities observed in this study.

A study limitation is the relatively small survey sample, though the cooperation rate was high at 92.1% and the weighting schema was designed to yield national representation. This was a novel survey exploring patient self-advocacy experience and outcomes, and future studies could build on this one with an expanded scope of questions and larger sample size. Respondent sociodemographic characteristics, financial literacy, and personality traits were captured on a prior survey rather than concurrently with this survey. The study also captured reported behavior from the prior 12 months, which may be subject to recall bias and variation in follow-up time since bill receipt.

This cross-sectional survey study found that 40% of households that received a bill they could not afford or disagreed with did not contact the billing office. Most of those who did self-advocate achieved bill corrections and payment relief. Respondents with lower education levels, those with less financial literacy, and the uninsured were less willing or able to reach out to a billing office regarding their problematic bill, and therefore likely missed opportunities to alleviate their bills. These differences in self-advocacy may be exacerbating socioeconomic inequalities in medical debt burden.

Accepted for Publication: July 10, 2024.

Published: August 30, 2024. doi:10.1001/jamahealthforum.2024.2744

Open Access: This is an open access article distributed under the terms of the CC-BY License. 漏 2024 Duffy EL et al. JAMA Health Forum.

Corresponding Author: Erin L. Duffy, PhD, MPH, University of Southern California, 635 Downey Way, VPD 414F, Los Angeles, CA 90089-3333 (eld_805@usc.edu).

Author Contributions: Dr Duffy had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Duffy, Trish.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Duffy.

Critical review of the manuscript for important intellectual content: Frasco, Trish.

Statistical analysis: Duffy.

Obtained funding: Trish.

Administrative, technical, or material support: Frasco.

Supervision: Trish.

Conflict of Interest Disclosures: Dr Duffy reported grants from the University of Southern California (USC) Schaeffer Center for Health Policy & Economics and a gift from Richard N. Merkin, MD, during the conduct of the study; personal fees from Cornerstone Research outside the submitted work. Dr Trish reported support for the research from USC Schaeffer Center for Health Policy & Economics and from Richard N. Merkin, MD, during the conduct of the study; grants from Alexion, Amgen, Arnold Ventures, Biogen, BioMarin, Blue Cross Blue Shield of Arizona, Blue Cross Blue Shield of Massachusetts, Bristol Myers Squibb, California Hospital Association, Cedars-Sinai Health System, Charles Koch Foundation, CommonSpirit, Commonwealth Fund, Edwards Lifesciences, Eli Lilly, Gates Ventures, Genentech, Gilead Sciences, GRAIL, IVI Foundation, Johnson & Johnson, Kaiser Family Foundation, National Institutes of Health, Novartis, Pfizer, RA Capital, Roche, personal fees from Cedars-Sinai Health for system Consulting, personal fees from Centene for expert testimony, Cornerstone Research for expert testimony, Guardian Pharmacy for expert testimony, personal fees from Mallinckrodt for expert testimony, and personal fees from Varian Medical Systems for expert testimony outside the submitted work; and Serving on the editorial boards for the American Journal of Managed Care and Medical Care Research and Review. No other disclosures were reported.

Funding/Support: This study was funded by the USC Schaeffer Center for Health Policy & Economics and a gift from Richard N. Merkin, MD.

Role of the Funder/Sponsor: The USC Schaeffer Center for Health Policy & Economics and Richard N. Merkin, MD, had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.