QuestAnalyitcsDataDifference.pdf
WHAT MAKES QUEST ANALYTICS DATA DIFFERENT? ®
Accurate healthcare network data is essential to support effective network management, improve quality, lower costs, and deliver the best patient care. But not all data sources are equal. Quest Analytics conducted an extensive analysis of public use and payer-sourced data files, shedding light on the critical differences between these sources.
The findings: Payer-sourced data, which is what
Quest Analytics® uses, delivers dramatically different and more reliable insights. That has important implications for network accuracy, compliance, and strategic decision-making.
The Healthcare
Landscape is Complex
Just consider:
Understanding Data Differences and Risks
• The United States has more than 2,000 networks among Medicare, Medicaid, and Marketplace.
• There are more than 1 million practitioners with Medicare Advantage network adequacy requirement specialties.
• Within Medicare, Medicaid, and Qualified Health Plans (QHPs), the average number of networks per provider is approximately 60, with higher network participation in high-demand specialties.
With so many variables, comprehensive and accurate data is a competitive advantage.
Many vendors offering network intelligence tools rely on claims data and public use files instead of payer-sourced data. That can lead to problems:
Claims data, which is focused on utilization, could misrepresent the network. For example, when an in-network practitioner has no utilization for that network, it could lead to the conclusion that the practitioner is out of network.
Public use files often contain discrepancies.
For example, some providers listed in application programming interface (API) data do not appear in public-facing web directories, and the Centers for Medicare & Medicaid Services (CMS) guidance does not enforce exact alignment between the two.
Public data captures a much lower percentage of the
network universe. Quest Analytics’ analysis highlights a concerning disparity in data capture rates between public use files and payer-sourced data across Medicare, Marketplace, and Medicaid. This raises concerns about whether trends seen in public data reflect true market behavior or are the result of incomplete data capture.
Percent of Networks Captured by LOB — Quest Analytics Data vs Public Directory Files
- FFM data includes AK, AL, AZ, DE, FL, HI, IA, IN, KS, LA, MI, MO, MS, MT, NC, ND, NE, NH, OH, OK, SC, SD, TN, TX, UT, WI, WV, and WY.
Public data underreports providers in a network. Quest Analytics analyzed millions of provider instances across 97 Medicare Part C contracts. Surprisingly, only 28% of National Provider Identifiers (NPIs) were common between public use files and payer-sourced data.
Public data overreports ghost providers. One of the most concerning issues of provider directories is the prevalence of ghost providers—providers listed but not actively practicing. Public use files were nearly twice as likely to report networks with ghost provider rates exceeding 9%, which is the average for ghost providers in a network.
® The Quest Analytics Data Advantage
For more than 20 years, Quest Analytics has been a leader in the provider network management marketplace, working with 90% of payers and managing more than 7,000 health plan network iterations in its platform. This gives Quest Analytics access to an unparalleled breadth and depth of data.
What sets Quest Analytics’ proprietary data apart is not just what it contains—it’s where it originates. Health plans submit network data for analysis to Quest Enterprise Services® (QES®), a cloud-based network measurement and management platform of Quest Analytics. This means that Quest Analytics’ foundational data comes directly from network files submitted by payers for compliance with regulatory requirements, which is then validated by regulators. As a result, this data best represents a true reflection of the networks.
That data is then enriched by the Quest Analytics BetterDoctor® attestation and verification processes, affiliation, and claims information. We use this same data advantage to train our machine learning models, ensuring our artificial intelligence is grounded in authentic, real-world data that delivers uniquely valuable insights.
As just one example, data from Quest Enterprise Services captures a significant portion of the provider universe in the United States and offers a deep understanding of provider access points. There are 1.3 million providers that practice in at least one of the Medicare Advantage required specialties. The table below demonstrates the high percentage of U.S. practitioners included in Quest Analytics’ proprietary dataset. The first table represents the number of distinct providers by line of business compared to the total universe. The second table shows the number of distinct locations and providers at these locations, also referred to as access points, and the average number of locations per provider.
| In thousands | US-total providers | Quest Analytics-total providers | Quest Analytics% |
|---|---|---|---|
| Total universe | 1,320 | 897 | 68%^{1,2}$ |
| Medicare | $944^{3}$ | 744 | 79% |
| CCIIO-FFM | 693 | 693 | 100% |
| QHP-SBE^{4}$ | $741^{3}$ | 460 | 62% |
| Medicaid | 875^{3}$ | 558 | 64% |
| In thousands | Unique provider location | Provider at location5(access points) | Locations per provider |
|---|---|---|---|
| Total universe | 964 | 3,588 | 4.0 |
| Medicare | 635 | 2,543 | 3.4 |
| CCIIO-FFM | 567 | 2,267 | 3.3 |
| QHP-SBE4 | 265 | 1,040 | 2.3 |
| Medicaid | 374 | 1,670 | 3.0 |
63% of the providers not shown in Quest Analytics data are mental health providers (clinical social work, clinical psychology, and psychiatry) and chiropractors that are more likely to not contract with health plans.
Excluding mental health providers that are not part of Quest Analytics data from the total universe increases Quest Analytics % to 81%. This segment of providers typically does not participate in health plans.
Estimated based on claims activity.
State-based exchange (SBE) states, though fewer in number than states using the federal marketplace for QHP, tend to have larger populations. Examples include CA, PA, IL, GA, MA, NJ, and NY.
Practitioners reporting locations >10 are regressed to the mean derived from the provider universe that reported 10 or fewer locations.
Our analysis shows that provider network composition changes 20-30% in a year, and the composition of providers at locations in network can change 35-60% in a year in any given market. This is why Quest Analytics refreshes data on weekly and monthly cadences to keep up with the fast pace of change in the healthcare industry. More up-to-date data leads to better strategic decision-making.
The evidence is clear: The complex nature of healthcare networks requires data that is accurate, reliable, and reflective of real-time changes in provider practice locations, network affiliations, and member access. That’s what Quest Analytics delivers.