Network Adequacy Standards and Enforcement
The Affordable Care Act (ACA) requires qualified health plans (QHPs) offered through the Marketplace to ensure a sufficient choice of providers and provide information to enrollees and prospective enrollees on the availability of in-network and out-of-network providers. Health plan networks are a key factor determining whether patients can actually get needed care, as claims for out-of-network services may be denied altogether or covered at a reduced rate.
Insurers can design provider networks to control utilization and reduce costs, and there are signs that “shrinkflation” (the practice of reducing the size of a product while maintaining its sticker price) may be taking place in the marketplace today. Describing recent trends in the marketplace, the Centers for Medicare and Medicaid Services (CMS) notes
“the proliferation of narrower networks … presents a number of potential consumer protection concerns including whether a narrow network has sufficient capacity to serve plan enrollees, or whether providers may be too geographically dispersed to be reasonably accessible.”
Studies find nearly 8 in 10 QHPs are health maintenance organizations (HMO) or exclusive provider organizations (EPO); both types of plans have closed networks, meaning nonemergency care from out-of-network providers generally is not covered. And many QHP provider networks are narrow. One study reviewing 2017 QHP physician networks found 21% of plans included less than one-fourth of available providers, while another 20% included fewer than 40% of available providers. Another study of 2017 QHP hospital networks found 21% of plans included fewer than one-third of available hospitals, and a further 28% of plans covered fewer than 70% of available hospitals. That study also observed that 29% of marketplace participants had only narrow network plans available. By contrast, job-based plans tend to have more robust provider networks; only 6% of employers offering health benefits say their most popular plan network is narrow.
There is no national standard for network adequacy,1 and standards that are applied vary significantly across states and types of coverage. Evaluation of health plan networks relies on plan provider directory data, which often have been found to be inaccurate or out of date. There is no national standard for the accuracy of information in health plan provider network directories; a new federal law establishing national standards for private health plans has yet to be implemented. Finally, there also is no standard measure of network size or breadth, nor any way for consumers or regulators to easily discern differences in network size.
To date federal regulation and oversight of QHP provider networks has been limited. For the 2023 plan year, CMS has proposed new network adequacy standards through regulations and guidance. The agency has also proposed a pilot network transparency indicator. This brief reviews background on federal network adequacy regulation, the availability of information about QHP networks, and options for strengthening oversight and enforcement.
Network Adequacy Regulation Overview
The federal government certifies QHPs offered in 30 federal marketplace states. Initially federal marketplace issuers were required to submit provider networks for CMS review and certain federal standards applied. Beginning with the 2018 plan year, the Trump Administration ended direct federal oversight of the adequacy of QHP networks, deferring to state oversight, accreditation by private organizations, or the issuer’s attestation. A federal court subsequently ruled this change was arbitrary and capricious, and as a result, federal oversight is scheduled to resume for the 2023 plan year. The regulation of health plan network adequacy can involve the use of various quantitative standards, as well as other oversight activities. Examples of quantitative standards include:
Time/distance standards – This type of standard is used to determine whether participating providers are geographically accessible to plan enrollees. For marketplace plans beginning in 2023, CMS has proposed time/distance standards for various types of providers and facilities. (Table 1) At least 90 percent of enrollees must live within the maximum distance to at least one provider of each type. For QHPs with tiered networks, only providers on the lowest cost-sharing tier would be counted.
Time/distance standards measure geographic proximity but not the breadth of a network. For example, in Cook County, Illinois – which measures approximately 60 miles north to south and 40 miles east to west, and where approximately 113,500 marketplace enrollees reside – a QHP network conceivably could satisfy this metric if it included just a handful of each of the provider and facility types shown in Table 1.
|Table 1: Proposed Federal Marketplace Time/Distance Standards for Selected Specialties, 2023|
|Maximum Time and Distance Standards (minutes/miles) Per County Type|
|Specialty Area||Large Metro||Metro||Micro||Rural||Counties with extreme access considerations|
|Outpatient Clinical Behavioral Health||10/5||15/10||30/20||40/30||70/60|
|Acute Inpatient Hospitals||20/10||45/30||80/60||75/60||110/100|
|Inpatient Behavioral Health Facility Services||39/15||70/45||100/75||90/75||155/140|
|The full list of specialties and facilities for which proposed time/distance standards would apply and detail on county types is found in the CMS 2023 Draft Letters to Issuers in the Federally-facilitated Exchanges.|
As late as 2017, CMS applied time/distance standards in the federal marketplace. Reviews took place during the annual certification process. During the 2016 certification process, CMS reports more than 90% of issuers passed for each of these metrics. Issuers failing to meet numeric standards could submit a “justification” explaining why its network provided reasonable access given the local availability of providers, or why offering the plan would nonetheless be in the interest of marketplace consumers. CMS did not publish data on QHPs approved pursuant to justification or on the metrics plans failed to meet.
For Medicare Advantage (MA) plans, CMS applies time/distance standards similar to those proposed for the federal marketplace, though standards were reduced in 2020, for example, requiring that only 85% of enrollees in non-metro areas live within time/distance standards, and further reducing the standard in all counties for MA plans that include telehealth providers.
Network adequacy standards for Medicaid managed care plans2 differ by state. In 2016, CMS required states to establish time and distance standards for specific types of providers, but the Trump Administration ended this requirement effective in December of 2020, instead allowing states to establish any type of quantitative network adequacy standard. CMS is currently developing a “comprehensive access strategy” across Medicaid fee-for-service and managed care delivery systems, with a notice of proposed rule-making planned for October 2022.
Minimum number of providers – Another standard establishes minimum provider-to-enrollee ratios. Here again, standards vary, where they exist. For example, in MA plan networks serving large metro areas, plans must contract with at least 1.67 primary care physicians per 1,000 beneficiaries. Under MA rules, contracted providers must also be within the maximum time and distance of at least one beneficiary to count toward the minimum number. CMS does not require minimum ratios for Medicaid managed care plans, though this is one of the quantitative standards that states can adopt to comply with federal Medicaid managed care rules. Where they have been adopted, these standards also vary (e.g. the minimum ratio of primary care providers to enrollees in Michigan, California, and Tennessee is 1:500, 1:2,000, and 1:2,500).
To date CMS has not applied a minimum provider ratio standard for the federal marketplace, nor has it proposed one for 2023. A review of state network adequacy regulation found that 10 states (including 5 that use the federal marketplace today) established minimum provider-to-enrollee ratios; so such a standard could apply to some federal marketplace QHPs via state regulation.
Appointment wait-times – Another type of standard sets maximum appointment wait-times for certain types of services. CMS has proposed this standard for 3 types of outpatient appointments shown in Table 2. Issuers would attest that 90% of contracted providers meet the wait-time standard; CMS would conduct compliance reviews in response to complaints and random audits. CMS has not proposed to create a complaints hotline for federal marketplace enrollees.
|Table 2: Proposed Appointment Wait Time Standard for Federal Marketplace Plans, 2023|
|Provider Type||Appointments must be available within:|
|Behavioral Health||10 calendar days|
|Primary Care (routine)||15 calendar days|
|Specialty Care (non-urgent)||30 calendar days|
|Source: 2023 Draft Letter to Issuers in the Federally-facilitated Exchanges|
Maximum appointment wait-times are another type of quantitative standard that state Medicaid programs can adopt to comply with federal requirements. Many states have done so, though these also vary based on times and services. For example, state maximum appointment wait-time standard for routine primary care services can range from 10 to 45 days, routine specialty care from 10 to 60 days, and urgent appointments from 1 to 4 days. Medicare Advantage plans are not currently required to meet appointment wait-time standards.
Essential community providers – In addition to other network standards, QHPs are required to contract with a minimum number of available essential community providers (ECP) in their service area. These include community health clinics, Ryan White providers, and other specified providers that serve predominately low-income and medically underserved individuals. In 2017 federal marketplace plans were required to contract with at least 30 percent of available ECPs; the Trump Administration reduced this threshold to 20% of available ECPs beginning in 2018. For 2023, CMS has proposed to increase the threshold to 35% of available ECPs.
Other standards – CMS proposes to require QHPs to report data in 2023 on whether network providers offer telehealth services and seeks comment on whether and how telehealth availability might be incorporated into network adequacy standards. The QHP rule proposes no new standards related to cultural and linguistic competency of network providers. Under Medicaid managed care, state standards must address the capacity of providers to communicate with patients with limited English proficiency in their preferred language and to accommodate patients with disabilities. Federal rules also require Medicaid managed care plan directories to indicate providers’ cultural and linguistic capabilities and the availability of skilled medical interpreters.
Finally as a condition of certification in the Exchange, QHP issuers must conduct an annual survey of enrollee experiences and report data to CMS. The survey includes questions about timely access to care, patient/provider communication capacity, and patient ratings of their doctors and of the care provided. Results inform a “star-rating” system for QHPs, and aggregated results are posted publicly.
Accuracy of Provider Network Directories
Regulators evaluate plans against quantitative network adequacy standards using network directory data, which often contain errors. Plans offered on HealthCare.gov are required to include directory links showing providers’ location, contact information, specialty, and whether they are accepting new patients. Issuers are required to update directories at least monthly. As part of its annual compliance review, CMS selects a small sample of issuers and reviews the machine-readable provider directory to verify accuracy. The most recent report found inaccuracies in all directories examined in 2020, with similar compliance problems observed in prior years.
Oversight of appointment wait-time standards also rely on directory data. In a secret shopper study of California QHPs in 2015, 73% of calls to providers listed in network directories were unable to secure appointments. Failures generally related to inaccurate phone numbers or addresses for listed providers, inaccurate specialty listings, or listed providers who were not actually in the network. Similarly, a 2014 review of Medicaid managed care plans conducted by the HHS Office of Inspector General found half of listed providers called could not offer appointments, most often because they were not practicing at the location listed in the directory or participating in the network at all. A 2018 review of the accuracy of Medicare Advantage plan directories found nearly half (48.7%) contain inaccuracies.
Starting in 2022, the No Surprises Act requires all private health plans, including QHPs, to maintain accurate provider directories and requires providers to regularly update plans about any changes in their information. Plans must verify and update directories at least every 90 days and, on an ongoing basis, post any changes within 2 business days. Plans are also required to apply in-network cost sharing for covered services provided by facilities or providers mistakenly listed as in-network. However, enforcement will be delayed because implementing regulations have not yet been published; CMS expects plans to make good faith efforts to comply with new requirements beginning in 2022.
Even while meeting minimum standards, insurers can and do design provider networks of substantially different sizes. This year the average marketplace participant in HealthCare.gov states is offered 107 different plan options. The only consumer tool for evaluating networks is to look up names of individual providers in each plan’s directory. There is no easy way to compare the size of networks overall.
In 2017, CMS piloted a “network transparency” measure in 3 states (Maine,3 Tennessee, and Texas) to indicate the relative breadth of a plan’s provider network compared to other QHPs in the same area. For three categories of providers – hospitals, primary care providers, and pediatricians – CMS compares the number of providers in a QHP network to the total number participating in any of the QHP networks. Comparisons are made on a county-basis. Plan networks for each category are then labeled as “smaller,” “larger,” or “about the same” as other QHPs in the county.
For 2023, CMS has proposed modifications to this measure for QHPs in pilot states. The provider categories measured will be general acute care hospitals, adult primary care, and pediatric primary care. The number participating in any QHP network in a county will be totaled for each category. CMS will then label QHP networks based on threshold percentages. Those containing fewer than 30% of providers participating in any of the QHP networks will be labeled as “Basic.” Those containing 30-69% of available providers will be labeled “Standard,” and those containing 70% or more will be labeled as “Broad.”
This revised indicator still provides information only about the relative breadth of a network compared to other QHPs and is therefore not an absolute measure of network size. For example, assume there are 50 general acute care hospitals in a large urban county and 3 QHP options are offered in that county. Assume further that Plan-A has 10 general acute care hospitals participating in network; Plan-B has 15 hospitals (the 10 in A’s network and 5 others); and Plan-C has 20 hospitals (the 15 in B’s network and 5 others). Under the proposed network transparency indicator, Plan A would be described as basic, Plan B standard, and Plan C broad. Yet arguably all three plan networks are narrow, excluding at least 60% of the available area hospitals.
A Unique State Approach to Network Regulation and Transparency – The New Hampshire Insurance Department (NHID) developed a novel approach to assessing private health plan provider networks that highlights the actual breadth of plan networks, not just their relative breadth, by measuring the share of all available providers that participate in a health plan’s network. Regulators determine the number of all available providers in a county by analyzing claims from the state’s all-payer claims data base (APCD), then count the share of available providers in each plan’s network. Marketplace consumers can compare QHP hospital networks on the NHID site. The state also offers an interactive tool that consumers and group purchasers can use to compare hospital networks in state-regulated health plans. Consumer-facing tools to compare other categories of provider networks have not yet been developed.
New Hampshire also categorizes providers based on their claims for key covered services that appear in the APCD. Regulators have identified core, common, and specialized services for each specialty. For example, to count a doctor as an adult primary care provider, the NHID counts those who have claims in the APCD for preventive and routine care provided to an adult. Regulators then compare this provider list to health plan directory data to determine the number of participating primary care providers in each plan. This approach, relying on APCD claims data, helps to correct for misclassification of specialties in provider directories and for other types of directory mistakes (e.g., continuing to list as participating a doctor who has actually retired or moved away, and so who no longer shows claims in the APCD).
The federal government does not currently have an all-payer claims database like New Hampshire to implement an approach like this. It does have a claims database of providers participating in Medicare, which includes nearly all available hospitals and physicians, although adjustments would be needed for pediatricians and other providers that participate less frequently in Medicare.
To explore the variation in network size and accessibility of information, we conducted a manual search of QHPs provider directories in the Houston, Texas area. Directories for the benchmark silver plan and for the lowest cost silver plan offered by other issuers were examined. Where it was clear that issuers offer multiple network options, these different networks were also included in the search. Participating general acute care hospitals within 25 miles of Houston (zip code 77002) were counted, with results shown in Table 3. The number participating in plans ranged from 9 to 42, with a total of 52 general acute care hospitals participating in at least one of the Houston-area QHP networks. Table 3 also shows the current network transparency indicator displayed on HealthCare.gov today and shows what the proposed revised network transparency indicator would be, based on this search.
Finally, as a proxy measure for the total number of available general acute care hospitals in the Houston area, Table 3 also shows the number participating in the most popular health plan option offered to federal employees (the FEHBP Blue Cross standard plan), which includes 56 general acute care hospitals according to that plan’s online directory.
Marketplace consumers in the Houston area appear to have a choice of at least 11 different hospital provider networks. In six of these networks (applicable to 125 of the 202 QHPs offered), at least 75% of available hospitals are not included. Broader hospital networks are for sale in the Houston area marketplace, though these plans cost more. Table 3 also shows that the added premium for a 45-year-old for a plan with a broader hospital network would cost $85-$157/month, an increment not covered by marketplace premium subsidies.
|Table 3: Participating Hospitals in Marketplace Plans vs. FEHBP, Houston, TX, 2022|
|Issuer/ Network (total number of plans offered)||Number of PAR Hospitals Shown in Directory *||Percent of all QHP PAR hospitals||Percent of all available hospitals||Current Marketplace Network Transparency Indicator||Proposed Marketplace Network Transparency Indicator||Additional Monthly Premium for This Issuer’s Lowest Cost Silver Plan** (Age 45)|
|Bright Health (28) (benchmark)||11||21%||20%||Smaller||Basic||0|
|Aetna-CVS Health (5)||13||25%||23%||About the same||Basic||$76.50|
|United Health Care (13)||14||27%||25%||About the same||Basic||$63.64|
|BCBS-TX / MyBlue Health (3)||17||33%||30%||Smaller||Standard||$28.01|
|Molina (9)||30||58%||54%||About the same||Standard||$85.89|
|Ambetter /Balance (46)||34||65%||61%||About the same||Standard||$89.15|
|BCBS-TX / Blue Advantage (10)||36||69%||64%||About the same||Standard||$156.59|
|Community Health Choice (9)||42***||81%||75%||About the same||Broad||$57.47***|
|Total hospitals in all QHPs||52|
|BCBS for Federal Employees||56|
|* Participating hospitals were counted using provider directory links for HealthCare.gov plans displayed for zip code 77002 (within 25 miles), viewed on January 12-14, 2022. Counts include only general acute cares hospitals, not rehab or psychiatric hospitals, freestanding emergency rooms, imaging centers, clinics, or other facilities that commonly display as “hospitals” in QHP provider directories.
** This column compares the unsubsidized premium for each issuer’s lowest-cost silver plan for a 45-year-old to the unsubsidized premium for a 45-year-old for the benchmark plan (Bright HealthCare Super Silver 1), which is $423.63. Marketplace subsidies are tied to the cost of the benchmark plan. Consumers can also buy plans that cost more, but must pay 100% of the additional cost.
*** The lowest cost silver plan offered by Community Health Choice indicates it has a “Limited Provider Network” which costs $535.58/month, or $57.47 more than the benchmark silver plan. However, the provider directory for this plan is the same as that for all other plans offered by this insurer, none of which include “limited network” in the name. The next lowest cost silver plan offered by this insurer costs $111.95 more than the benchmark plan for a 45-year-old. Also of note, the provider directory for this insurer indicates certain participating hospitals in its Silver Plan networks are not included in Bronze plan networks.
The current HealthCare.gov network transparency indicator offers a clue as to differences in network size, though the indicator provides limited information. Hospital networks are either described as “about the same” or “smaller” when compared to each other.4 The proposed revised network transparency indicator, as applied to the results shown in Table 3, would label 6 of the 11 networks as “basic,” 4 as “standard,” and one as “broad.”
While this illustration looked only at general acute care hospitals, network transparency indicators could also report on inclusion of providers of specialized services for seriously and chronically ill patients. For example, a 2016 KFF study of hospital networks in Medicare Advantage plans specifically examined the participation of National Cancer Institute (NCI)-designated cancer centers and found most MA plans in the Houston area did not include the University of Texas MD Anderson Cancer Center. None of the Houston-area QHP directories included this cancer center in network, either, though it does participate in the FEP plan network.
Insurers offer a range of network designs in marketplace plans, including some that may be exceedingly narrow. What this means for patient access to care – and for continuity of care for consumers transitioning from other coverage to the marketplace – is not clear, nor is it measured. Proposed new standards for 2023 set minimum standards for QHP provider networks, but these alone will not assure networks meet any minimum standard of breadth. Time-distance standards as proposed require at least 1 participating provider be in proximity to most enrollees, but do not assure a sufficient number will be available. Appointment wait-time standards begin to measure actual access to care but have been proposed only for 3 types of routine care, not specialized or urgent care services.
Other aspects of network adequacy have not been incorporated into proposed standards, though public comment was requested on some issues. It remains to be seen whether QHP network standards will measure provider language and cultural competencies, or accessibility for people with disabilities, or access to specialized care for specified groups such as children, patients with chronic health conditions, or other underserved communities.
Differences in the breadth of health plan networks could be more clearly measured and described. Because the marketplace primarily offers transitional coverage, it could help consumers to show which QHPs, if any, offer comparable networks to the plans they’re leaving, whether job-based plans left by workers during the “Great Resignation” or Medicaid managed care plans left by beneficiaries disenrolled as the public health emergency unwinds.
Better transparency measures can also inform oversight, helping regulators identify differences in provider networks and consider how to deploy other tools to understand how differences affect access to care. For example, data from QHP consumer experience surveys could highlight access concerns meriting further investigation. ACA transparency data could also be enhanced and used in oversight to monitor differences in out-of-network claims for specific services or by specified groups of patients. A complaints hotline for QHP shoppers and enrollees could alert regulators to potential problems, as could mid-year reviews.
Network adequacy is a key factor affecting patient access to care, yet it is challenging to operationalize. There is also an inherent trade-off with costs and affordability, as broader networks could increase premiums through higher prices and greater use of services. Close monitoring and improved transparency may provide mechanisms to help evaluate standards and refine them over time.
The National Association of Insurance Commissioners developed a Health Benefit Plan Network Access and Adequacy Model Act.. However, it includes no quantitative standards, and adoption by states has been limited.
Federal law requires Medicaid managed care plans to assure that they have capacity to serve expected enrollment in their service area and maintain a sufficient number, mix, and geographic distribution of providers. A Medicaid managed care plan must make covered services accessible to its enrollees to the same extent that such services are accessible to other state residents with Medicaid who are not enrolled with that plan.
Maine has since become a state-based marketplace.
The results in Table 3 show counts only of participating general acute care hospitals. It is not clear whether CMS also counts other types of facilities in calculating the pilot network transparency indicator. For example, using online directories to search general acute care hospitals, we found results often included other types of facilities including psychiatric hospitals, rehab facilities, addiction treatment facilities, freestanding emergency rooms, imaging centers, and clinics. While we excluded all of these other types of facilities from our count, the CMS pilot might have classified facilities differently.