Nineteen percent of bills for out-of-network visits were negotiated, and of these negotiated bills, individuals were successful in lowering their costs approximately half the time.
ABSTRACTObjectives: Out-of-network (OON) care is one area where patients might be more likely to challenge their healthcare bills due to the high out-of-pocket costs and unexpected charges related to emergency care or hospital-affiliated providers. We aimed to determine whether, and under what circumstances, patients negotiate with either insurers or providers when services are billed OON and how often patients that do engage in negotiation are successful.
Study Design: Internet-based survey.
Methods: We conducted a 2011 Internet survey on OON care on a nationally representative sample of privately insured adults (n = 721). We considered whether patients would be more likely to negotiate OON charges by demographic characteristics and under several scenarios: emergency visits, bills from hospital-affiliated OON providers at in-network hospitals, and balance bills.
Results: We found patients negotiated 19% of OON bills, were successful in lowering their costs 56% of the time, and were more likely to be successful negotiating with providers compared with insurers (63% vs 37%; P <.01). Men were more likely than women to be successful in lowering their costs (76% vs 50%; P <.05). OON bills for emergencies, providers at in-network hospitals, and with a balance bill were more likely to be negotiated, although bills from providers at in-network hospitals and with balance bills were less likely to be successfully negotiated.
Conclusions: Patients had low rates of success in negotiating OON bills for emergency care and for OON providers at in-network hospitals. Policy makers aiming to protect patients under these scenarios should consider policies that allow for an easily accessible, formal, and unbiased mediation process.
Am J Manag Care. 2016;22(10):647-652
Nineteen percent of bills for out-of-network (OON) visits were negotiated; of these negotiated bills, individuals were successful in lowering their costs approximately half the time.
Patients who believe they have inappropriately been charged for healthcare services may have difficulty challenging these costs. They may have poor understanding of health insurance terminology, face a complex billing bureaucracy, not fully understand the nature and necessity of the care they are receiving, be vulnerable if facing serious illness, or not feel comfortable discussing money with their physician and worry it will impact their relationship. Relatively few patients formally voice concerns to their health plan, even when they believe they have been inappropriately charged a large out-of-pocket cost. Circumstances, such as socioeconomic status, patient empowerment, and chronic illness, can impact whether a patient will use their voice and have a satisfactory resolution.1
Out-of-network (OON) care is one area where patients might be more likely to challenge their healthcare bills due to the high out-of-pocket costs of using OON care. In addition to higher cost sharing (eg, 20% of costs vs 10% for in-network providers), OON providers may “balance bill” patients the difference between their list price and the amount the provider is reimbursed by the insurer. Although there is little objective data on consumer out-of-pocket costs for OON care, reports suggest that balance bills may be quite high (and increasing) and a source of medical debt among the privately insured.2,3
Recently, attention to this issue has increased due to consumer complaints about the use of narrow networks in plans sold in state health insurance marketplaces; overall, 41% of 2014 marketplace plans were classified as small or extra small.4 An increasing number of plans are also reimbursing for OON care based on percentage of the Medicare rate (rather than the usual and customary rate), which often leads to higher balance bills.5,6 Even cost sharing may be increasing. Some high-deductible health plans do not allow OON services to apply to the general deductible, and there is evidence that deductibles for OON care are increasing relative to in-network deductibles. A PriceWaterhouseCoopers survey found that from 2009 to 2015, average OON deductibles increased by $1000 compared with a $500 increase for in-network deductibles.7
Although the majority of OON care—especially in the outpatient setting—is an informed choice, from a patient perspective, some charges related to OON services may be viewed as unfair, even if they meet the conditions of the insurance contract.8,9 In an emergency, patients may feel it is necessary to use the most easily accessible provider; however, they may unexpectedly receive bills from an OON provider at an in-network hospital. A hospital may be in-network, but providers who are hospital-based or hospital-affiliated who are treating the patient, such as anesthesiologists, radiologists, or pathologists, may not have contracts with the patient’s insurer. For example, although a pregnant patient may check before routine labor and delivery to ensure that her obstetrician and hospital are in her insurer’s network, during her labor, the epidural may be placed by an OON anesthesiologist or complications may arise and she may receive services from an OON neonatologist, which essentially would be almost impossible to refuse. Another concern relates to price transparency. If information on a provider’s price or insurer reimbursement for a service is not readily available, patients may make the decision to use an OON provider without understanding the cost implications.
Currently, there are few federal or state protections or formal complaint or mediation processes specifically related to unexpected OON bills, leaving most patients to initiate a complaint on their own with their insurer or provider. The Affordable Care Act (ACA) partially addressed the issue of OON emergency department (ED) care on the federal level by requiring that insurers bill in-network cost-sharing rates for patients who use an OON ED. Out-of-pocket costs to these patients may still be high since balance billing by providers is still allowed.10 At the state level, patients may submit a complaint to their state for denial of services from their insurer and request an external review process; however, this process is for what is covered under the health plan and it is unclear whether patients challenging OON bills are covered under this mechanism. More recently, some state legislators have strengthened consumer protections using a variety of approaches, including requiring that the insurer, rather than the patient, pay the balance bill for unexpected OON care (sometimes referred to as "hold harmless provisions"), ensuring adequate payment for providers to deter balance billing, and requiring that providers disclose the potential for an OON bill at the point of service.11
Because of the complexity of negotiating between the insurer, provider, and patient, the use of an independent mediator has also been suggested at the state level. In Texas, in addition to disclosure rules, patients themselves may initiate the mediation process for unexpected OON bills over $500.12,13 A New York state law that went into effect April 2015 holds patients harmless from unexpected OON bills for emergencies. For other unexpected OON billing scenarios (ie, OON provider at in-network hospital), the New York law allows patients to assign their benefits to the insurer. The OON provider is then prohibited from seeking payment from the patient (except for cost-sharing amounts). After this assignment of benefits, if the OON provider and insurer disagree about reimbursement, the provider or insurer can initiate an independent mediation process.14
Little is known about how often patients negotiate OON bills and the resolution of the dispute. Consumer Reports National Research Center recently conducted a survey, which indicated that about one-third of patients have received a bill where the insurer paid less than expected, with many cases not resolved to the patients' satisfaction.15 Another study examining pre-service denials from 1998 to 2000 in 2 large HMO’s found that among pre-service denials that were appealed, approximately 1 in 5 were related to OON care, and these appeals were less likely to be won in favor of the enrollee compared with medical necessity appeals.16 Postservice appeals in the same dataset often involved failure to obtain authorization to use an OON provider.17
In this paper, we attempt to address the question of whether, and under what circumstances, patients attempt to negotiate with either insurers or providers when services are billed OON and how often patients that do engage in negotiation are successful. We conducted a national survey of users of OON care and determined what proportion self-reported that they negotiated with insurers, providers, or both, and how often they were successful. We considered whether, and which, patient or treatment characteristics predict an individual will negotiate an OON bill. We paid particular attention to those patients who report the OON service was an emergency, at an in-network hospital, or included a balance bill.
Data were obtained from a 2011 Internet survey on patient experiences with OON care on a nationally representative sample of privately insured English-speaking US adults aged 18 to 64 years. Detailed methods have been published previously.18 The survey was constructed and tested through cognitive interviews19 and pretesting, and then administered via the Internet by GfK Knowledge Networks. GfK’s online research panel consists of approximately 50,000 US households selected using high-quality address-based sampling methods similar to those used by US government surveys.20 Panelists are not excluded if they do not currently own a computer, and only invited individuals are included in the panel. The probability-based sampling used to construct the panel and its representativeness of the US population have been validated.21 Demographic information was provided by GfK for all individuals in the panel.
A series of screener questions was sent to 21,754 panelists to identify respondents enrolled in a private health insurance plan with a provider network who had seen a physician and/or mental health professional in the past 12 months. Enrollment was closed when a pre-determined number of panelists screened in and began the survey, resulting in a completion rate of 64% (13,900 panelists). Panelists who had used an OON provider (n = 721) took a 10-minute survey on their experiences.
For each OON visit, we asked respondents, “Did you try to bargain with the doctor for a lower price (either before or after seeing the doctor)?” If they answered affirmatively, we asked the follow-up question, “Were you successful in bargaining for a lower price with the doctor?” Similarly, for each visit, we also asked about trying to bargain and success in bargaining with the insurance company. For patient-level analyses (Table 1 and eAppendix [available at www.ajmc.com]), patients were defined as negotiators if they negotiated with either the insurer or the provider in any of their reported OON visits. Successful negotiators were defined by having any reported success with either the insurer or the provider for any of their reported OON visits.
We considered whether patients would be more likely to negotiate or dispute OON charges under several scenarios: emergency visits, bills from hospital-affiliated OON providers (eg, anesthesiologists, pathologists) at in-network hospitals, and balance bills. A visit was defined as an emergency if the respondent answered “Yes” to “When you went to the emergency room or were admitted to the hospital, was it for a medical emergency?” Hospital-affiliated provider OON bills were defined as inpatient or ED visits, where the hospital was reported to be in-network and the patient was unaware the provider was OON prior to receiving care. A visit was considered balance billed if the respondent answered “Yes” to “Did the doctor charge you more than what your insurance plan covered? Do not include any co-payments, coinsurance, and deductible you may have been responsible for.” Balance bills could be associated with any OON service, including informed use of OON providers and unexpected OON service use, such as for an emergency, hospital-affiliated providers at an in-network hospital, or when no in-network provider was available. For comparison, we also constructed a category for visits where none of these 3 issues were present.
All reported analyses were weighted to match the sample to the US population based on Current Population Survey data on sex, age, race/ethnicity, education, metropolitan area, Census region, and Internet access, and to adjust for panel recruitment, attrition, oversampling, and survey nonresponse. Frequencies and χ2 tests were used for categorical variables. Where relevant, we adjusted standard errors for multiple observations for individuals. Linear regression models predicting negotiation and success in negotiation controlled for education, race, gender, income, residence in an urban area, age, and self-reported health status.
The final sample included 721 individuals aged 18 to 64 years with private health insurance who had used an OON physician or mental health professional within the last year, representing a total of 1081 OON visits/bills (662 outpatient and 419 inpatient) as some individuals reported use of more than 1 OON provider. Overall, 19% (n = 177) of the bills for OON visits were negotiated by individuals with either the insurer or the provider, 14% were negotiated with the provider, and 10% with the insurer (Figure [A]). In 5% of visits, individuals negotiated with both the insurer and provider (data not shown). Of the negotiated bills, individuals were successful in lowering their costs about half of the time, and were significantly more likely to be successful when negotiating with providers compared with insurers (63% vs 37%; P <.01) (Figure [B]).
Approximately 22% of individuals who used OON providers negotiated an OON bill with the insurer or provider, and 58% were successful in reducing their costs for at least 1 of the bills (Table 1). An individual’s demographic characteristics were not significantly associated with whether or not they negotiated a bill. However, men were more likely than women (76% vs 50%; P <.05) and those in good health versus poor health (63% vs 25%; P <.01) to be successful in lowering their costs. These differences persisted after adjusting for education, race, income, age, and residence in an urban area (eAppendix Table).
Table 2 presents subsets of OON visits where we hypothesized individuals would be more likely to negotiate their OON bill: emergency care, use of an OON provider at an in-network hospital, and visits for which the individual was balance billed. We compared the proportion of visits negotiated, and of these, the proportion successful in lowering their costs, to the comparison group of visits in which none of the 3 issues were present. OON visits for emergencies at an in-network hospital and visits with a balance bill were significantly more likely to be negotiated (21%, 40%, and 34%, respectively) compared with the group with none of the issues studied, where only 10% of visits were negotiated. Somewhat surprisingly, success rates for OON care at in-network hospitals and those with balance bills (37% and 48%, respectively) were significantly lower than among the comparison group (74%).
In this study, we found that 19% of bills for OON visits were negotiated, and of these negotiated bills, individuals were successful in lowering their costs approximately half the time. Although individual demographic characteristics were not associated with negotiation of an OON bill, we found disparities by gender and health status in whether or not an individual was successful in lowering their costs. We also found patients were more likely to negotiate if a bill was related to OON care in an emergency, at an in-network hospital, or if there was a balance bill; although for bills at in-network hospitals and balance bills, these negotiations were less likely to be successful compared with bills where no observable issue was present.
Unexpected OON bills due to emergency care or from hospital-affiliated providers at in-network hospitals are unique for patients because they can find themselves in a difficult triangle between their insurer and their provider, where it may be unclear who is responsible for the higher costs. Insurers argue the use of networks can reduce healthcare premiums and that insurers should not be responsible for higher reimbursement to providers, particularly specialists, unwilling to accept a “market” rate. In the case of use of OON providers practicing at in-network hospitals, insurers may argue in-network hospitals have the responsibility to ensure an in-network provider is available when necessary (eg, anesthesiology). Conversely, providers might believe insurers are not offering a “fair” rate for their services, or that insurers too frequently deny or impose administrative hassle over medically necessary services, necessitating they not participate in networks to provide the highest quality care for their patients. In most states, the patient is ultimately responsible for the bill and the responsibility to negotiate with insurers/providers over bills perceived as unfair.
This study supports the concern that some patients are not aware of their ability to dispute these unexpected OON medical bills. A 2015 Consumer Reports survey found that the majority of respondents were unsure if state resources were available to dispute insurer coverage denials, and they were unaware of the state agency tasked with handling health insurance complaints.15 In Texas, patients must be informed of their right to mediation when they are balance billed more than $1000 (this amount was lowered to $500 as of September 1, 2015); yet, in 2014, only 900 cases were filed for mediation—1 of which went to actual mediation—and stakeholders have expressed concern that patients are not aware of their rights.11
It is not surprising that individuals who report OON care in an in-network hospital were more likely to negotiate these bills. Providers and insurers did not seem particularly responsive to patients' concerns about these issues, with less than half resolved to the patients’ satisfaction; it is possible, however, that the billing was valid under the terms of the contract, even if the patient views them as “unfair.” This suggests an area that regulators may want to focus on to ensure patients are treated fairly. We also found high rates of success among the group of negotiators with no observable issue studied, which may be because those that attempt to negotiate in this situation are negotiating prior to informed OON care and, if unsuccessful, opt to remain in-network or have an idiosyncratic issue that insurers or providers perceive as valid (eg, no in-network provider available).
Although we did not find any differences by demographic characteristics in who negotiated an OON bill, we did find disparities in achieving success with negotiation among women and those in poor health. That those in poor health are less likely to be successful in negotiating is consistent with prior literature1,22—they may have competing priorities with their health, impaired physical or mental capabilities, and more medical bills overall to manage. We cannot conclude there is any gender bias on the part of insurers or providers from these data; however, the fact that women are equally likely to attempt to negotiate as men but are much less likely to be successful raises concerns, particularly in conjunction with evidence of gender disparities in economic negotiation outcomes in other contexts.23 If additional research indicates these differences are truly due to gender, and not other unobservable characteristics, this further suggests changes to the appeals process are needed.
This study has several limitations. As with all survey research, this study is subject to nonresponse and recall bias, although the use of weights (adjusting for Internet usage, panel recruitment and attrition, and nonresponse), a short recall period (12 months), and use of cognitive interviewing to eliminate questions that could not reliably be answered by self-report, partially alleviate these concerns. However, for some variables, such as whether the patient was balance billed and whether the hospital was in-network, a significant number of respondents were still unsure of their responses (30% and 21%, respectively). Another limitation is sample size, limiting power to detect differences and perform subgroup analysis on whether the patient negotiated with an insurer or a provider.
Due to time constraints with the survey, we were unable to describe important details around the negotiation, thus limiting what conclusions can be drawn from our data. Some bills may not have needed to be negotiated by the patient; for example, an insurer may pre-emptively hold a patient harmless from OON costs of emergency care, requiring no action by the patient. We were also unable to characterize whether the negotiation was around a balance bill or cost sharing, such as co-payment or deductible. We never asked specifically about why a patient negotiated the bill, and we could not determine whether the complaint was objectively legitimate. Additionally, we were unable to determine whether the negotiation was informal bargaining with the provider either before or after the visit or a formal appeal through the insurer for coverage denials. However, for emergency and inpatient care at an OON hospital, we can assume that the negotiation was after receiving an unexpected OON bill. Although additional information would have been useful to provide more context to the results presented here, we believe our data are still useful and indicate there is a need for more research in this area.
Because this survey focused on OON care, we are unable to compare against rates of negotiations/success for in-network care. Our data were collected in 2011, prior to implementation of policies established by the ACA to curtail balance billing in emergency care, as well as several state-level policies to protect and assist patients in mediation of unexpected OON charges.
Patients had low rates of success in negotiating OON bills for emergency care and for OON providers at in-network hospitals. Policy makers aiming to protect patients under these scenarios should consider policies that allow for an easily accessible, formal, and unbiased mediation process. The recently implemented New York state law excludes the patient from the negotiation process between the insurer and the provider after assignment of benefits, and outcomes should be evaluated and compared with other state models, such as Texas (which requires the patient to initiate the negotiation). At a minimum, patients should be better educated on their right to appeal and state resources available to them. Also, current available estimates of balance bill amounts are based on list price and allowed amounts.3 Our finding that a significant number of OON bills were negotiated suggests that these estimates may not be accurate, and future studies are needed to define the actual out-of-pocket burden for OON care.
Use of OON care, both informed and unexpected, may only increase as insurers adopt the use of narrow networks as a way to potentially reduce healthcare costs.24-26 Policy makers will need to rapidly adopt both measures to prevent unexpected bills, such as accurate provider directories, patient education, and network adequacy standards, as well as protect patients from burdensome unexpected OON charges if they do occur.
Author Affiliations: Department of Population Health, New York University School of Medicine (KAK), New York, NY; Department of Health Policy and Management, Yale School of Public Health (SHB); New Haven, CT.
Source of Funding: This study was supported by a grant from the Women’s Health Research at Yale Pilot Project Program and funding from the Yale Robert Wood Johnson Foundation Clinical Scholars Program. The sponsors were not involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Author Disclosures: Dr Kyanko has accepted consulting fees from the Consumers Union, a nonprofit organization dedicated to consumer protection. These consulting activities are in topic areas outside of the submitted manuscript, but the entity may be perceived to have interest in the manuscript content. Dr Busch reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SHB, KAK); acquisition of data (SHB, KAK); analysis and interpretation of data (SHB, KAK); drafting of the manuscript (SHB, KAK); critical revision of the manuscript for important intellectual content (SHB, KAK); statistical analysis (SHB, KAK); provision of patients or study materials (KAK); obtaining funding (SHB, KAK); administrative, technical, or logistic support (SHB, KAK); and supervision (SHB).
Address Correspondence to: Kelly A. Kyanko, MD, MHS, Department of Population Health, New York University School of Medicine, 550 1st Ave, TRB 6th Fl, Rm 646, New York, NY 10016. E-mail: firstname.lastname@example.org.
1. Schlesinger M, Mitchell S, Elbel B. Voices unheard: barriers to expressing dissatisfaction to health plans. Milbank Q. 2002;80(4):709-755, iv-v.
2. Pollitz K, Cox C, Lucia K, Keith K. Medical debt among people with health insurance. Kaiser Family Foundation website. http://kff.org/private-insurance/report/medical-debt-among-people-with-health-insurance/. Published January 7, 2014. Accessed September 22, 2016.
3. Charges billed by out-of-network providers: implications for affordability. Americas Health Insurance Plans website. https://www.ahip.org/charges-billed-by-out-of-network-providers-implications-for-affordability/. Published September 29, 2015. Accessed September 22, 2016.
4. Polsky D, Weiner J. The skinny on narrow networks in health insurance marketplace plans. Robert Wood Johnson Foundation website. http://www.rwjf.org/en/library/research/2015/06/the-skinny-on-narrow-networks-in-health-insurance-marketplace-pl.html. Published June 2015. Accessed September 22, 2016.
5. Kahn H, Parke R. Using Medicare RBRVS for reimbursing out-of-network claims in commercial insurance. Milliman website. http://us.milliman.com/uploadedFiles/insight/health-published/using-medicare-rbrvs.pdf. Published February 2012. Accessed September 22, 2016.
6. An unwelcome surprise: how New Yorkers are getting stuck with unexpected medical bills from out-of-network providers. New York State Governor website. http://www.governor.ny.gov/sites/governor.ny.gov/files/archive/assets/documents/DFS%20Report.pdf. Published March 7, 2012. Accessed September 22, 2016.
7. PriceWaterhouseCoopers Health Research Institute. Medical cost trend: behind the numbers 2016. Commission on Care website. https://commissiononcare.sites.usa.gov/files/2016/01/20151116-05-Medical_Cost_Trend-Behind_the_Numbers_2016_PWC.pdf. Published June 2015. Accessed September 22, 2016.
8. Rosenthal E. Costs can go up fast when E.R. is in network but the doctors are not. The New York Times website. http://www.nytimes.com/2014/09/29/us/costs-can-go-up-fast-when-er-is-in-network-but-the-doctors-are-not.html. Published September 28, 2014. Accessed September 22, 2016.
9. Kyanko KA, Pong DD, Bahan K, Curry LA. Patient experiences with involuntary out-of-network charges. Health Serv Res. 2013;48(5):1704-1718. doi: 10.1111/1475-6773.12071.
10. Patient Protection and Affordable Care Act; Requirements for Group Health Plans and Health Insurance Issuers Under the Patient Protection and Affordable Care Act Relating to Preexisting Condition Exclusions, Lifetime and Annual Limits, Rescissions, and Patient Protections [final rule and proposed rule]. Federal Register website. http://webapps.dol.gov/federalregister/HtmlDisplay.aspx?DocId=23983&AgencyId=8&DocumentType=2. Published June 28, 2010. Accessed September 22, 2016.
11. Hoadley J, Ahn S, Lucia K. Balance billing: how are states protecting consumers from unexpected charges? Robert Wood Johnson Foundation website. http://www.rwjf.org/en/library/research/2015/06/balance-billing--how-are-states-protecting-consumers-from-unexpe.html. Published June 2015. Accessed September 22, 2016.
12. Mediation for out-of-network hospital-based health care provider claims. Texas Department of Insurance website. http://www.tdi.texas.gov/consumer/cpmmediation.html. Accessed August 10, 2015.
13. S.B. 481. Texas Legislature Regular Session website. http://www.legis.state.tx.us/tlodocs/84R/billtext/html/SB00481I.htm. Accessed September 22, 2016.
14. Independent dispute resolution for emergency services and surprise bills. State of New York Department of Financial Services website. http://www.dfs.ny.gov/legal/regulations/emergency/np400t.pdf. Accessed September 22, 2016.
15. Surprise medical bills survey. Consumers Union website. http://consumersunion.org/research/surprise-bills-survey/. Published May 5, 2015. Accessed September 22, 2016.
16. Studdert DM, Gresenz CR. Enrollee appeals of preservice coverage denials at 2 health maintenance organizations. JAMA. 2003;289(7):864-870.
17. Gresenz CR, Studdert DM, Campbell N, Hensler DR. Patients in conflict with managed care: a profile of appeals in two HMOs. Health Aff (Millwood). 2002;21(4):189-196.
18. Kyanko KA, Curry LA, Busch SH. Out-of-network physicians: how prevalent are involuntary use and cost transparency? Health Serv Res. 2013;48(3):1154-1172. doi: 10.1111/1475-6773.12007.
19. Willis GB. Cognitive Interviewing: A Tool for Improving Questionnaire Design. Thousand Oaks, CA: Sage Publications, Inc; 2005.
20. KnowledgePanel design summary. GfK website. http://www.knowledgenetworks.com/ganp/docs/KnowledgePanel(R)-Design-Summary.pdf. Published 2012. Accessed September 22, 2016.
21. Validity of the Survey of Health and Internet and Knowledge Network’s panel and sampling. CDC website. http://www.cdc.gov/pcd/issues/2004/oct/pdf/04_0004_01.pdf. Published May 2003. Accessed September 22, 2016.
22. Rosenthal M, Schlesinger M. Not afraid to blame: the neglected role of blame attribution in medical consumerism and some implications for health policy. Milbank Q. 2002;80(1):41-95.
23. Mazei J, Huffmeier J, Freund PA, Stuhlmacher AF, Bilke L, Hertel G. A meta-analysis on gender differences in negotiation outcomes and their moderators. Psychol Bull. 2015;141(1):85-104. doi: 10.1037/a0038184.
24. Employer Health Benefits Survey [among firms offering health benefits, percentage of firms whose largest plan includes a high-performance provider network, 2007, 2010, 2011, and 2013]. Kaiser Family Foundation website. https://kaiserfamilyfoundation.files.wordpress.com/2013/08/8465-employer-health-benefits-2013-chartpack.pdf. Accessed September 22, 2016.
25. Bauman N, Coe E, Ogden J, Parikh A. Hospital networks: updated national view of configurations on the exchanges. McKinsey&Company website. http://healthcare.mckinsey.com/hospital-networks-updated-national-view-configurations-exchanges. Accessed September 22, 2016.
26. Baicker K, Levy H. How narrow a network is too narrow? JAMA Intern Med. 2015;175(3):337-338. doi: 10.1001/jamainternmed.2014.7763.