The authors examine the largest 20 US commercial payers’ coverage policies and identify variation in how interventions are covered and the evidence reviewed in them.
ABSTRACTObjectives: To examine coverage policies for medical interventions issued by the largest US commercial payers.
Study Design: Review of publicly accessible coverage policies for medical interventions.
Methods: We categorized the 20 largest commercial payers’ medical benefit coverage policies for medical technologies—current as of August 1, 2014—with respect to technology type (eg, medical devices, pharmaceuticals, surgeries). We identified the interventions most commonly subject to coverage policies and compared payer coverage determinations in terms of whether they covered the intervention and the evidence they reported reviewing.
Results: Eighteen payers made their coverage policies publicly available and 17 reported the evidence they reviewed in formulating policies. The types of technologies considered varied across payers, although most focused on devices and diagnostics. Of the 28 interventions most commonly subject to coverage policies, the coverage of 9 varied (ie, some payers covered the intervention and others did not). On average, payers reported reviewing clinical studies in 87% of coverage policies (range = 25%-100%). Two payers did not report reviewing systematic reviews or meta-analyses in any coverage policies, and 9 reported reviewing such evidence in at least half of their policies. Fourteen payers reported reviewing cost-effectiveness analyses at least some of the time, with frequency ranging from 8% to 43%. Commercial payers’ coverage decisions did not appear to reflect direct input from patients or patient advocates, at least as stated in published coverage policies.
Conclusions: Coverage of medical interventions varies across US private payers. Payers often report reviewing different evidence when formulating coverage policies, but do not report considering input directly from patients in evidence assessments.
Am J Manag Care. 2016;22(9):e323-e328
This paper identifies important variation in the largest commercial payers’ coverage policies for medical interventions. We found:
In US healthcare, multiple public and commercial payers issue coverage policies for medical interventions.1-3 Because payers conduct their own assessments and issue their own decisions, whether and how they cover medical interventions can vary, which, in turn, can affect patients’ access to care. Researchers have highlighted variation in payer coverage policies and examined decision making.4,5 Recent research has studied trends in Medicare National Coverage Determinations (NCDs) and the consistency of coverage with the reviewed evidence.6-8 Another recent study found substantial variation in how Medicare and private payers cover medical devices.9
This study adds to the literature by examining coverage policies for medical interventions issued by the top 20 US commercial payers. First, we examined the coverage policies issued by the largest US-based commercial payers. Second, for a representative set of interventions, we compared coverage determinations across payers (ie, whether and how the payers cover the interventions and the types of evidence the payers reported reviewing when formulating their policies).
We identified the 20 largest US-based commercial payers in terms of number of covered lives.10 We searched each payer’s website to determine the availability of their medical benefit coverage policies. In many cases, payers provide memoranda, which describe the target patient population, any conditions on patient access to an intervention, and, frequently, the clinical trials, clinical guidelines, and other evidence the payer reports reviewing when formulating the coverage policy. We focused on coverage policies pertaining to the payers’ commercial line of business by excluding coverage policies pertaining to their Medicaid managed care or Medicare Advantage lines of business. When we were unable to locate memoranda, we contacted the payer to confirm their unavailability.
We identified all coverage policies issued by each of the 20 payers (n = 7372). Included policies were current as of August 1, 2014. We categorized each policy with respect to intervention type (eg, medical device, pharmaceutical, surgery, diagnostic imaging); categories were not mutually exclusive (eg, we included implantable cardiac defibrillators and artificial lumbar disc replacement in both the medical device and surgery categories).
Comparison of Payers’ Coverage Determinations for a Representative Set of Interventions
We determined the 4 intervention types that accounted for the majority of coverage policies: medical devices, pharmaceuticals, surgeries, and diagnostic tests/imaging procedures. For each, we identified the 7 interventions for which the payers most often issued coverage policies (Table 1). We included 7 interventions of each type to ensure they were adequately represented across the payers. We used this representative set of 28 interventions to determine whether the payers covered the interventions and the types of evidence payers report reviewing in their policies.
We considered whether the payers reported reviewing 1 or more of the following 6 evidence types: clinical studies (eg, randomized controlled trials or observational studies), systematic literature reviews or meta-analyses, technology assessments, cost-effectiveness analyses, budget impact analyses, and clinical guidelines. A researcher read the coverage policies to determine whether payers cited each evidence type. The researcher then searched each policy for pertinent keywords; for example, when searching for cost-effectiveness analyses, the researcher searched for the terms, “cost-effectiveness,” “economic evaluation,” “cost-benefit,” and so on. We also determined whether coverage decisions reflected input from patients (eg, whether comments from patients or patient advocacy organizations were accounted for).
Wide Variation in the Number of Coverage Policies That Payers Issue
Eighteen of 20 of the largest commercial payers made their medical benefit coverage policies publicly available (Kaiser Permanente and Centene Corporation were the exceptions). The number of policies payers issued varied, from 146 by UnitedHealthcare to 698 by Aetna. Eighty-nine percent of payers issued the largest proportion of their coverage policies for medical devices; 11% issued the largest proportion for drugs (Table 2).
How Payers Cover Medical Technology Often Varies
Payers covered the representative set of interventions with different frequencies (eAppendix, available at www.ajmc.com). Highmark most often covered the interventions we examined (20 of the 23 interventions for which Highmark issued a coverage policy; 87%). Blue Shield of California least often covered the interventions (11 of 21; 52%) (Table 3).
Sixteen interventions in the representative set were covered by all the payers that issued a coverage policy for it (eg, cryosurgical ablation for prostate cancer). Three interventions in the set were not covered by any payer (eg, biochemical markers for Alzheimer’s disease). The remaining 9 interventions in the set were covered by some payers but not others; for example, of the 11 payers that issued a coverage policy for microvolt t-wave alternans testing to identify patients at risk of sudden cardiac arrest, 5 covered the technology and 6 did not.
The Types of Evidence Payers Report Reviewing in Their Coverage Policies Varies
The types of evidence that payers report reviewing in their coverage policies for interventions in the representative set varied. On average, the 17 payers that reported the evidence they reviewed cited clinical studies in 87% of coverage policies (range = 25%-100%).
Payers reported evaluating systematic reviews and meta-analyses with different frequencies. Two payers did not report reviewing systematic reviews or meta-analyses in any of their coverage policies; the remaining 15 payers did so in an average of 50% of their coverage policies (range = 22%-89%). Payers also report reviewing cost-effectiveness analyses with different frequencies. Three payers did not report reviewing cost-effectiveness analyses in their coverage policies; the remaining 14 payers did so in an average of 14% of their coverage policies (range = 8%-43%). Four payers reported reviewing a budget impact analysis in one of their coverage policies for interventions in the representative set (Table 3).
Each payer reported reviewing clinical guidelines and did so in an average of 80% of their coverage policies (range = 62%-100%). Each payer reported reviewing technology assessments in a proportion of their coverage policies and did so in an average of 60% of their coverage policies (range = 22%-89%).
Coverage Policies Did Not Directly Reflect Patient Input
We found no indication of payers having a formal process for including patients and/or consumer input into decision making. Furthermore, we found that no payers reported accounting for patient input (ie, from patients or patient advocacy organizations) in the evidence base they reported reviewing (Table 3).
Although researchers have previously examined variation in commercial payer coverage policies, this study is unique in describing the coverage policies issued by the largest US-based commercial payers.9,11-13 Our study highlights several points.
Coverage Policy Practices Among the Largest US Commercial Payers Varies
We found wide variation in the number of coverage policies promulgated by the largest commercial payers. Why some payers are more selective than others in this regard is unclear. The administrative burden of maintaining a large number of up-to-date coverage policies may lead payers to restrict those they make publicly available to big-ticket technologies or interventions for which patients’ access is most ambiguous. Some payers may believe that limiting the number of publicly available policies affords them greater flexibility and allows them to more rapidly respond to new evidence or changing circumstances. Patients, however, may benefit from having more access to information about coverage of medical interventions.
Payers Often Cover Interventions Differently
We found that for the representative set of interventions, payers’ coverage determinations often varied. Discrepancies could possibly be attributable to payers differentially weighing factors (eg, the influence of local physicians) in their decision making. Payers may interpret the supporting evidence-base differently or differentially weigh product price, value, and budget impact in decision making. Differences in how payers cover interventions may also be influenced by contractual language and/or state mandates. Interestingly, we found variation in the coverage policies issued by plans in our sample that are part of the Blue Cross Blue Shield Association. This variation suggests that despite having access to Blue Cross Blue Shield national model policies, the plans independently modify the policies to meet their own standards and priorities on occasion.
The differences among coverage policies suggest that patients may have differential access to medical interventions that, in turn, may lead to variation in health outcomes.5 Differences between payer-covered benefits may also impact provider behavior, as physicians may tailor patient care to insurance coverage. However, variation among payer coverage determinations may have some salutatory effects if payers’ coverage policies reflect their enrollees’ unique characteristics.14
Payers Report Using Different Evidence in Coverage Determinations
The evidence that payers reported reviewing in their coverage policies varied. As expected, we found that clinical evidence and clinical guidelines featured prominently in the payers’ evidence review. Technology assessments also commonly featured in the reviewed coverage policies. In contrast, review of systematic reviews or meta-analyses and economic studies varied considerably. Notably, the majority of payers reported reviewing cost-effectiveness analyses in at least some of the representative set of coverage policies, presumably reflecting the continued introduction of high-cost treatments.15 Few payers reported reviewing budget impact analyses.
A greater appreciation of payers’ use of evidence could help providers understand how patients’ benefit packages are designed. It is essential for manufacturers to understand the role of evidence in coverage policies as they design clinical development programs to demonstrate the efficacy, safety, and value of their products.
The Patient Voice in Coverage Determinations Is Lacking
We found that payers did not report considering input from patients in their coverage policies. This finding is at odds with recent initiatives that seek to emphasize the patient perspective in the drug approval process and evidence generation.16-18 This finding is also inconsistent with other countries’ health technology assessment agencies that engage with the public at various levels of decision making.19-23
We were unable to account for policies that payers did not make publicly available and for the fact that an appeals process is typically available when coverage is denied. With respect to the reviewed evidence, payers may not disclose all sources of information that contribute to their decision. Furthermore, payers may independently perform their own analyses (eg, budget impact and cost-effectiveness analysis), but do not report those results in coverage policies.
Not all payers in our sample issued a policy for each intervention. In addition, by choosing to include interventions for which payers most commonly have written policies, we may have introduced bias to the sample. For instance, it may be that these interventions are those for which the evidence is most ambiguous, and thus, the identified variation may not be generalizable to other interventions.
We do not account for issues specific to the included intervention types that may influence payer coverage. As an example, drugs and devices are often subject to different patient cost-sharing rules, and although payers must negotiate drug prices with manufacturers, price negotiations for devices are left to hospitals. We also do not account for differences among covered patient populations. For instance, one payer may limit coverage of an intervention to a patient subgroup (eg, patients suffering from severe disease), whereas another payer may cover this subgroup and also cover patients suffering from less severe disease.
Lastly, we do not account for the fact that the payers’ coverage policies may not be entirely independent. For instance, several health plans that are part of the Blue Cross Blue Shield Association may use Blue Cross Blue Shield national model policies as the basis of their policies.
Research is needed to better understand variability between payers’ coverage policies and how payers formulate their coverage decisions. Future research should examine a more representative set of interventions (eg, the highest spend interventions) and whether the evidence that payers report reviewing in their coverage policies varies for different intervention types. Examining whether payer characteristics, such as size, geographic breadth, region, etc, correlates with different coverage policy characteristics (eg, number of issued coverage policies, determination decisions, use of evidence) would also be useful.
Coverage of medical interventions differs across payers, as does the evidence payers report reviewing in their policies. The variation has potentially important implications for patients’ access to medical interventions, providers delivering care, and manufacturers bringing their products to market. The largest payers do not directly reflect input from patients in their coverage policies.
Author Affiliations: Tufts University School of Medicine (JDC, PJN), Center for the Evaluation of Value and Risk in Health, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center (JDC, PJN, MDC), Boston, MA.
Source of Funding: This study was funded by the Medical Device Manufacturers Association.
Author Disclosures: At the time of this study, Mr Chenoweth was an employee of Tufts Medical Center; he is currently an employee of Evidera. Dr Neumann has previously served on the advisory boards of Merck, Bayer, Pacira, Novo Nordisk, Shire, and Amgen for health economics topics; he has served on the advisory board for the Congressional Budget Office; he has consulted for Boston Health Economics, Purdue, Vertex, and Precision Health Economics. The CEA Registry has been funded by NSF, NLM, AHRQ, CDC, and a variety of pharmaceutical and device companies who subscribe to the data. Unrelated to this research, Dr Chambers previously participated in an advisory board for Sanofi.
Authorship Information: Concept and design (JDC, MDC); acquisition of data (JDC, MDC); analysis and interpretation of data (JDC, MDC); drafting of the manuscript (JDC, MDC, PJN); critical revision of the manuscript for important intellectual content (JDC, PJN); statistical analysis (JDC); obtaining funding (JDC); and supervision (PJN).
Address Correspondence to: James D. Chambers, PhD, Assistant Professor, Tufts University School of Medicine, Center for the Evaluation of Value and Risk in Health, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, 800 Washington St, #63, Boston, MA 02111. E-mail: email@example.com.
1. Cebul RD, Rebitzer JB, Taylor LJ, Votruba ME. Organizational fragmentation and care quality in the U.S healthcare system. J Econ Perspect. 2008;22(4):93-113.
2. Elhauge E. The Fragmentation of U.S. Health Care: Causes and Solutions. Oxford, UK: Oxford University Press; 2010.
3. Austin DA, Hungerford TL. The market structure of the health insurance industry. Federation of American Scientists website. https://www.fas.org/sgp/crs/misc/R40834.pdf. Published November 17, 2009. Accessed August 31, 2015.
4. Steiner CA, Powe NR, Anderson GF, Das A. The review process used by US health care plans to evaluate new medical technology for coverage. J Gen Intern Med. 1996;11(5):294-302.
5. Steiner CA, Powe NR, Anderson GF, Das A. Technology coverage decisions by health care plans and considerations by medical directors. Med Care. 1997;35(5):472-489.
6. Chambers JD, Morris S, Neumann PJ, Buxton MJ. Factors predicting Medicare national coverage: an empirical analysis. Med Care. 2012;50(3):249-256. doi: 10.1097/MLR.0b013e318241eb40.
7. Chambers JD, Chenoweth M, Cangelosi MJ, Pyo J, Cohen JT, Neumann PJ. Medicare is scrutinizing evidence more tightly for national coverage determinations. Health Aff (Millwood). 2015;34(2):253-260. doi: 10.1377/hlthaff.2014.1123.
8. Chambers JD, Chenoweth MD, Pyo J, Cangelosi MJ, Neumann PJ. Changing face of Medicare’s national coverage determinations for technology. Int J Technol Assess Health Care. 2015;31(5):347-354. doi: 10.1017/S0266462315000525.
9. Chambers JD, Chenoweth M, Thorat T, Neumann PJ. Private payers disagree with Medicare over medical device coverage about half the time. Health Aff (Millwood). 2015;34(8):1376-1382. doi: 10.1377/hlthaff.2015.0133.
10. Atlantic Information Services’ Directory of Health Plans. Atlantic Information Services website. http://aishealth.com/marketplace/aiss-directory-health-plans. Updated January 1, 2015. Accessed August 31, 2015.
11. Buttorff C, Andersen MS, Riggs KR, Alexander GC. Comparing employer-sponsored and federal exchange plans: wide variations in cost sharing for prescription drugs. Health Aff (Millwood). 2015;34(3):467-476. doi: 10.1377/hlthaff.2014.0615.
12. Meckley LM, Neumann PJ. Personalized medicine: factors influencing reimbursement. Health Policy. 2010;94(2):91-100. doi: 10.1016/j.healthpol.2009.09.006.
13. Hresko A, Haga SB. Insurance coverage policies for personalized medicine. J Pers Med. 2012;2(4):201-216. doi: 10.3390/jpm2040201.
14. Bailey PG. Medicare and national coverage. Health Aff (Millwood). 2005;24(1):295-296.
15. Chambers JD. Do changes in drug coverage policy point to an increased role for cost-effectiveness analysis in the USA? Pharmacoeconomics. 2014;32(8):729-733. doi: 10.1007/s40273-014-0166-6.
16. About us. Patientâ€‘Centered Outcomes Research Institute website. http://www.pcori.org/about-us/how-were-funded. Published October 6, 2014. Accessed August 31, 2015.
17. Enhancing benefit-risk assessment in regulatory decision-making. FDA website. http://www.fda.gov/ForIndustry/UserFees/PrescriptionDrugUserFee/ucm326192.htm. Updated July 7, 2015. Accessed August 31, 2015.
18. Patient preference information—submission, review in PMAs, HDE applications, and de novo requests, and inclusion in device labeling: guidance for Industry, Food and Drug Administration staff, and other stakeholders. FDA website. http://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM446680.pdf. Published August 24, 2016. Accessed August 25, 2016.
19. Gagnon MP, Desmartis M, Lepage-Savary D, et al. Introducing patients’ and the public’s perspectives to health technology assessment: a systematic review of international experiences. Int J Technol Assess Health Care. 2011;27(1):31-42. doi: 10.1017/S0266462310001315.
20. Kreis J, Schmidt H. Public engagement in health technology assessment and coverage decisions: a study of experiences in France, Germany, and the United Kingdom. J Health Polit Policy Law. 2013;38(1):89-122. doi: 10.1215/03616878-1898812.
21. Abelson J, Giacomini M, Lehoux P, Gauvin FP. Bringing ‘the public’ into health technology assessment and coverage policy decisions: from principles to practice. Health Policy. 2007;82(1):37-50.
22. Abelson J, Bombard Y, Gauvin FP, Simeonov D, Boesveld S. Assessing the impacts of citizen deliberations on the health technology process. Int J Technol Assess Health Care. 2013;29(3):282-289. doi: 10.1017/S0266462313000299.
23. Whitty JA. An international survey of the public engagement practices of health technology assessment organizations. Value Health. 2013;16(1):155-163. doi: 10.1016/j.jval.2012.09.011.