The Severity of Need Story—at a Glance
Title XXVI of the Public Health Service Act as amended by the Ryan White HIV/AIDS Treatment Modernization Act of 2006 (Ryan White HIV/AIDS Program) delivers HIV/AIDS care to those with limited resources to pay for needed care, reaching over half a million persons in the U.S. each year. HRSA’s HIV/AIDS Bureau administers the Ryan White Program and directs funds to areas with the greatest need, which demands ever greater diligence in measuring severe need and allocating scarce funds to these areas. Efforts to define severe need have evolved since the start of the Ryan White Program in 1990, guided by legislative requirements outlined in subsequent reauthorization language from 1996, 2000, and the latest iteration in 2006.
Ongoing studies have identified challenges with past efforts to gauge severe need, such as use of measures that are not available across all areas, variable sampling techniques, and use of qualitative data versus more desirable quantitative information. Among these is a 2003 HRSA-sponsored review of unmet need methods in use by urban areas funded under Ryan White Part A and a Congressionally-mandated 2003 Institute of Medicine (IOM) Committee report, Measuring What Matters: Allocation, Planning and Quality Assessment for the Ryan White CARE Act. That body called for HRSA to adopt a new, more standardized, and quantitative approach to gauge resource needs across jurisdictions, conceptualized by the IOM as follows (see Measuring What Matters, p. 137):
Severity of Need =
(Disease Burden) x (Cost of Care) – (Available Resources)
Acting upon the IOM findings, HRSA established a HIV/AIDS SON Collaboration to develop a SON Index. The Collaboration is a multi-tiered expert body broadly representative of HHS agencies and HAB staff, national experts, Ryan White grantees, and consumers. Established in 2004, the effort to analyze and develop a draft SON Index is coordinated by HRSA/HAB and its contractors (Altarum and RTI International) under:
- A Collaborative Council comprised of cross HHS agency representatives providing overall guidance.
- A HAB SON Workgroup of representatives from across HAB that works with the contractor on development of the SON Index.
- Four specialist panels, comprised of 47 members (representing grantees, consumers, academics, HAB and other Federal agencies), that reflect the above IOM equation: Area Characteristics; Patient Coverage and Need; Patient Characteristics; and Associated Costs.
In 2005-2006, the panels completed their investigation and prepared written reports, reviewing 56+ variables and forwarding 19 for inclusion in the draft SON Index, with the rest eliminated for such reasons as lack of sufficient data to document them. The range of variables is extensive—from certain data like AIDS cases to other measures of need that are undoubtedly important but illusive to measure in terms of their impact on severe need. For example:
The Burden of Disease panel recommended inclusion of HIV/AIDS cases data in the draft Index because all States have such data.
In contrast, the Cost Panel excluded the variable stage of HIV disease progression because such data are not widely available. Besides, the true cost-driver associated with disease progression is whether a client is on antiretroviral treatment—not the person’s diagnosis or CD4 count.
Subsequent supplemental studies have been completed, and more are underway, to further explore the potential use of specific variables within the SON Index. Efforts to refine the SON Index are ongoing, and include opportunities for input from the Ryan White Community as HAB works toward finalization of the Index.
The draft SON Index formula, as of October 2007, is:
Severity of need simplified algorithm = A portion of the allocation that is adjusted by direct measures + a portion of the allocation that is adjusted by indirect measures. Direct measures are defined as those for which the relationship between a change in the measure and a change in resource needs can be estimated. Indirect measures are defined as those which are expected to influence resource needs but for which we lack sufficient data to directly estimate the real impact of a one unit change.
The direct adjustment is estimated as [Total cases – Federal insurance reduction] x Geographic costs index.
The indirect adjustment is estimated as a standardized weighted function of the death rate, the poverty rate, and the prevalence rate multiplied times total cases.