All All-Pay ayer er Claims Claims Da Data tabas bases es: : - PowerPoint PPT Presentation

All All-Pay ayer er Claims Claims Da Data tabas bases es: : What’s Next After the Sup Supreme eme Cou Court t Dec Decision? ision? AcademyHealth Annual Research Meeting New Orleans, Louisiana June 26, 2017 Craig Schneider, senior health researcher Mathematica Policy Research

Outline • Background on APCDs • Gobeille v. Liberty Mutual Insurance Company case • Methodology • Findings • Conclusions Disclaimer : The contents of this presentation are solely the responsibility of the author and do not necessarily represent the official views of the Massachusetts Center for Health Information and Analysis or the U.S. Department of Health and Human Services or any of its agencies. 2

Background on APCDs • Relatively new – before 2008, only in 2 states (Maine and Maryland) • Growing rapidly • Mix of state agencies and multi-stakeholder nonprofits • APCD Council – a collaboration of the National Association of Health Data Organizations (NAHDO) and Institute for Health Policy and Practice at UNH – is a learning network that supports state APCDs • “All - payer” term – Commercial plans, Medicare, Medicaid, TPAs, PBMs, dental benefits 3

Data in (and not in) APCDs Information typically collected Data elements excluded • Encrypted member ID • Uninsured • Type of product (HMO, etc.) • Denied claims • Type of contract (family, single, etc.) • Workers’ comp • Patient demographics • Premium information • Diagnosis and procedure codes • Capitation and administration fees • Service provider information • Referrals • Prescribing physician, facility type • Test results • Plan and member payments • Provider affiliation with group practice • Revenue codes and dates of service • Provider networks Source: APCD Fact Sheet, NAHDO and APCD Council, UNH Institute for Health Policy and Practice, funded by The Commonwealth Fund, 2010. 4

APCD Implementation Map Sl Slide ide co courtesy sy of APCD APCD Co Council il, , www.a .apcdcouncil il.or .org 5

Value of APCDs • Contains almost complete sample of a state’s insured population – can support broad range of information needs and studies • Data used to support websites on public cost information • Calculate hospital aggregate costs • Analyze cost-shifting between payers • Evaluate access issues • Study dual-eligible population • Evaluate payment reform initiatives • Assess community health See www.apcdshowcase.org for case studies of how states are using APCD data 6

Gobeille v. Liberty Mutual Insurance Company • Supreme Court decision March 1, 2016 • Question: Does ERISA preempt state statutes that require insurers to submit data to APCDs? • ERISA covers self-insured plans • Argument against: Burden on plans to comply with varying data submission requirements of multiple states • Argument for: The “burdens” are trivial (supported by U.S. Solicitor General) • Decision: – Because different regulations from various jurisdictions “could” create wasteful administrative costs, Vermont’s statute is preempted. – SCOTUS acknowledged that Vermont doesn’t actually create such burdens. – Justice Breyer said DoL could create regulations for standardized data submission format across states/jurisdictions. 7

Methodology • Consulting engagement with Massachusetts Center for Health Information and Analysis • Semistructured interviews with 10 states • Surveys of 4 other states • Interviews and survey asked about: – organizational structure, data collection process, data release rules, budget, views on a “finished” APCD, data matching approach, provider attribution process, patient identification and re-identification procedures, website content, resources to produce analytics, contracting approach, fees for sharing data, marketing materials, presence of state laws that are more restrictive than federal law, impact of and responses to Gobeille decision, transparency strategy in the state, and anything else the state APCD wanted to mention • This presentation focuses on findings regarding Gobeille and the “finished” APCD questions 8

Findings • Vision for a completed APCD • Impact of and responses to Gobeille • Issues regarding substance use data 9

Features/Functionality for “Completed” APCD • Merge claims with clinical data, registry data, EMR data, social determinants of health, or vital records • Meticulous processes for cleaning data • Ability to change when users’ needs change • All stakeholders have confidence in data quality • Acquire self-funded plan data lacking after Gobeille • Better race/ethnicity data • Non-claims payment (i.e., value-based purchasing) • A rich set of public use files (PUFs) updated regularly that people may query • Dashboard of relevant information build on database with business intelligence tools 10

Features/Functionality for “Completed” APCD (2) • Contain all payers and uninsured (via dummy claim), sitting on server with one seamless longitudinal data set that produces standard extracts • Effectively use cost data • Enable analysis of competitiveness of insurance market • Break down data silos into integrated consolidated source, with user-friendly dashboard to visualize results • Fully financed by the stakeholders • Data users issue publications to drive transparency • Address capitation and back-end payments 11

Impact of Gobeille • States reported losing a number of claims in their database, ranging from 15% to 40% • Conversely, one state is receiving more data – the SCOTUS decision raised awareness among employers • Generally, states are experiencing a greater decline in data from national employers than from state-based employers • States commented that, although the loss of this data is disappointing, the APCD remains a valuable resource 12

State responses to Gobeille Approaches reported by states: • Reaching out to employers, brokers, coalitions, data users • Issuing regulations • Enacting a law that requires opt-in; working with employers on info needs/promotion; establishing Employer Advisory Commission • Creating opt-in for employers; reaching out to employers • Planning to move to opt-in; engaging employers, hospitals, and Chambers of Commerce 13

Next steps following Gobeille • State APCD statutes are still enforceable – fully insured plans, non-ERISA plans, and TPAs if employer supports data submission • Most TPAs are concluding that the plan sponsor (the employer) has right to determine whether TPA submits data • DoL issued proposed regulation 7/21/16 – final release not known • In late May, NAHDO and APCD Council completed Common Data Layout effort – convened states and payers to develop consensus content and format for payers to report APCD administrative data Source: NAHDO, APCD Council, and NASHP, Key Regulatory Issues Facing APCD States Post Gobeille, April 2016. 14

A Bigger Challenge for APCDs than Gobeille ? • 42 CFR Part 2, Confidentiality of Substance Use Disorder Patient Records • Final rule issued by SAMHSA on 3/21/17 • Rule limits data shared by providers • Challenges: – N o universal ID system to isolate “Part 2” providers in claims data feeds – More confusing for APCDs than hospital discharge databases because source is payers who have both Part 2 and non-Part 2 providers and might be reluctant to submit data 15

Conclusions • APCDs are a critical resource for health services researchers, value-based purchasing organizations, and policymakers • Gobeille decision problematic but “not a fatal blow” to APCDs • 42 CFR Part 2 is an additional challenge to APCDs • States, NAHDO, and APCD Council working with payers and federal agencies to resolve these issues – “Burden” and privacy concerns can be addressed, enabling APCDs to continue to be valuable assets to states 16

Contact Information Craig Schneider Senior health researcher Mathematica Policy Research cschneider@mathematica-mpr.com (617) 715-6955 Resources: www.nahdo.org www.apcdcouncil.org 17