Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Risk Scores topic
No spam. Unsubscribe anytime.
San Francisco health board reviews new risk-score analysis as input to 2019 rates
Summary
Health Service System staff presented third-year risk-score results showing a modest decline in aggregate commercial risk across plans but increases for Blue Shield; commissioners asked about model limits, Medicare retiree data, and how scores will influence 2019 rate negotiations.
Marina Coleridge, Enterprise Systems and Analytics Manager for the San Francisco Health Service System, presented the system—s third annual risk-score analysis to the Health Service Board, outlining how concurrent and prospective DXCG-based scores for the October 2016September 2017 period will inform the 2019 rates-and-benefits cycle.
Coleridge said the blended commercial population (actives and early retirees) showed a continued downward trend in both concurrent and prospective risk scores year over year, with Blue Shield the only plan to show an overall increase. She cautioned that the DXCG models are commercially oriented and that applying them to Medicare-eligible populations requires caveats; the staff produced a Medicare-retiree slice as a directional indicator, not definitive measurement.
The presentation broke risk by plan (Blue Shield, City Plan, Kaiser), employer (City College among higher-risk groups), and union (trades, MEA). Coleridge highlighted enrollment shifts that affect plan-level risk: City Plan saw a large enrollment rise that influenced its mix, while Kaiser attracted younger, lower-risk hires. For Blue Shield Coleridge reported roughly a 3.18% increase in the concurrent score and a 2.22% increase in the prospective score for a specific subgroup, and elsewhere cited a 1.8% concurrent and 0.8% prospective uptick for Blue Shield overall in certain slides.
Board members asked how the system will benchmark its results. Coleridge said the team is working with Truven/IBM Watson Health but that the standard West Region benchmark is too broad; staff hopes to craft a custom benchmark tailored to the City—s market. Commissioners also pressed on clinical drivers (e.g., hepatitis C treatment uptake), model opacity (some DXCG algorithms are proprietary), and how the scores feed into actuarial rate-setting. Coleridge said the risk scores help prioritize deep dives, inform wellness and disease-management strategy, and offer an independent data source to supplement plan-provided analytics.
The board heard questions about technical detail and use cases but took no formal action on the risk-score presentation itself. Staff said more granular slices and notes are available to commissioners and will be used in negotiations with actuaries and carriers as the 2019 rates cycle proceeds.
The presentation appears intended to give the board an independent, data-driven view as staff and the actuary prepare rate proposals and negotiations with health plans.
