Relation Between Narrow Networks and Providers of Cancer Care

Key Findings

Narrower provider networks offered on the health insurance exchanges are more likely to exclude oncologists associated with high-quality National Cancer Institute (NCI)-Designated Cancer Centers. These findings suggest that narrower oncology networks, while offering lower premiums, may involve a tradeoff between cost and quality of cancer care.

The Question

To reduce the cost of insurance plans, health insurers are increasingly marketing plans that restrict access to providers (both hospitals and physicians). These narrow provider networks allow insurers to offer price competitive plans, often through lower reimbursement rates or selective contracting with providers treating lower-cost patients. Whether narrow networks limit access to high quality providers is not known.

NCI-Designated Cancer Centers, and a subset of them identified as National Comprehensive Cancer Network (NCCN) Cancer Centers, are recognized for their scientific leadership, cutting-edge medical technology, and lower mortality rates among severely ill patients. These prestigious cancer centers are more likely to attract patients requiring costly care, and insurers have a strong incentive to exclude oncologists associated with NCI or NCCN Cancer Centers from narrow networks. In this study, the authors assess the extent to which narrow networks systematically exclude NCI or NCCN-affiliated oncologists, and address the implications for whether narrow networks require a tradeoff between cost and quality.

The Findings

Of 407 markets (rating areas) in the country, 51 had an NCI-Designated Center; 27 of them had an NCCN center. The supply of oncologists was greater in markets with an NCI-Designated Cancer than other markets (13.7 vs. 8.8 per 100,000 residents).

The authors identified 248 provider networks in these 51 markets. Networks in these markets were narrower than in other markets, covering an average of 39.4% of area oncologists versus 49.9% in other markets. Despite this narrowness, the average number of covered oncologists per 100,000 residents was higher than in markets containing a NCI-Designated Center than other markets (4.5 vs. 3.8). However, 33 of 248 networks did not contain a single physician affiliated with an NCI center; these networks were narrower (covering only 14.1% of local oncologists) than networks including at least one NCI-affiliated physician (covering 42.3% of local oncologists). The figure on the front shows that oncology network breadth was associated with inclusion of NCI-affiliated oncologists, indicating that narrower networks are more likely to exclude these physicians.

The Implications

This is the first study to demonstrate a correlation between narrow networks and exclusion of NCI- and NCCN-affiliated providers. These hospitals are recognized for their high-quality cancer care, education, and research programs. Thus, this finding suggests that narrow networks may not just offer fewer providers, but that the limited number of providers included may not offer the same quality care as those who have been excluded. This highlights a critical tradeoff that consumers face when purchasing a narrow network plan: they may benefit from the lower
premiums charged by narrow network plans, but they may face reduced access to higher-quality providers in their market.

These findings are relevant to replacement proposals for the Affordable Care Act, that emphasize shoppable insurance plans for consumers. As consumers seek to learn about providers and coverage included in their plans, accurate information about these providers is essential. In 2016, the Centers for Medicare & Medicaid Services enacted rules for publishing user-friendly provider directories that include a provider’s location, contact information, specialty, medical group, and any hospital affiliations. This study’s findings indicate that inclusion of quality indicators in these provider directories – such as NCI or NCCN affiliation for cancer providers – may prove useful to consumers shopping health plans.

The question this study raises, but cannot answer, is whether or not insurers specifically exclude physicians at higher rates because of their NCI or NCCN designation, or whether exclusion results from an additional correlated factor. For example, group practice size may be associated with market power and pricing, and may be a factor in an insurer’s choice to exclude a provider. Furthermore, the data do not identify differences in actual care quality between NCI and non-NCI centers. Future research should examine the relationship between narrow networks and cancer care outcomes.

The Study

The authors used a registry of all office-based physicians to identify practicing physicians with a specialty of hematology/oncology or radiation oncology, and identified oncologists affiliated with one of the 69 NCI-Designated Cancer Centers, and the subset that were NCCN Cancer Centers. They examined provider networks offered on the 2014 individual health insurance exchanges using a previously integrated dataset, and identified 51 markets (rating areas) containing at least one NCI-Designated Cancer Centers.

The breadth of each network was calculated as the number of oncologists included in the network divided by the total number of oncologists practicing in that market. They measured a network’s likelihood of including high-quality oncologists within each market by the proportion with NCI (or NCCN) affiliation among the market’s oncologists included in the network, divided by the proportion of those with NCI (or NCCN) affiliation among the market’s oncologists excluded from the network. Values greater than one indicate relative inclusion—and values less than one relative exclusion. Then, the authors assessed the relationship between network breadth and the inclusion measure for all networks offered in any market containing an NCI-Designated Cancer Center. They also ran separate analyses for NCCN centers, but the results did not differ for this subset of cancer centers.

Lead Author

Laura Yasaitis is a postdoctoral researcher in Health Policy at the Perelman School of Medicine. She completed her doctoral training in Health Policy at The Dartmouth Institute in 2013, and was a postdoctoral research fellow for two years at the Center for Population Studies and Development at Harvard’s T.H. Chan School of Public Health. Her research interests include the study of large datasets, including surveys and administrative data, to better understand the quality and outcomes of health care in the United States. She is also interested in the application of spatial analysis methods to further our understanding of local trends in health and health care.

Co-posted from Penn LDI

Laura Yasaitis, Justin E. Bekelman, and Daniel Polsky: Journal of Clinical Oncology, July 2017