Are Community-Level Financial Data Adequate to Assess Population Health Investments?

Navigate This Article

Tim Casper, MA; David A. Kindig, MD, PhD

Suggested citation for this article: Casper T, Kindig DA. Are Community-Level Financial Data Adequate to Assess Population Health Investments? Prev Chronic Dis 2012;9:120066. DOI: http://dx.doi.org/10.5888/pcd9.120066

PEER REVIEWED

Abstract

The variation in health outcomes among communities results largely from different levels of financial and nonfinancial policy investments over time; these natural experiments should offer investment and policy guidance for a business model on population health. However, little such guidance exists. We examined the availability of data in a sample of Wisconsin counties for expenditures in selected categories of health care, public health, human services, income support, job development, and education. We found, as predicted by the National Committee on Vital and Health Statistics in 2002, that availability is often limited by the challenges of difficulty in locating useable data, a lack of resources among public agencies to upgrade information technology systems for making data more usable and accessible to the public, and a lack of enterprise-wide coordination and geographic detail in data collection efforts. These challenges must be overcome to provide policy-relevant information for optimal population health resource allocation.

Top of Page

Background

Because of the increasing attention on improving population health, policy makers in the public and private sectors are asking for better information to guide their investment decisions. The Institute of Medicine (1) has called on the federal government to issue an annual report on trends and disparities in social and environmental factors that affect health and to advance the use of system-based simulation models to explain the health consequences of the underlying determinants of health. These requests are consistent with calls for comparative effectiveness research to inform cost-effective allocation of resources among the multiple determinants of health (2,3).
Much of the variation in health outcomes among communities may result from different levels of financial and nonfinancial policy investments over time; these natural experiments should offer investment guidance for a business model of population health (4). Why have researchers not estimated the per capita level of investment at the community level for each of the determinants of health (medical care, public health, behaviors, and the social and physical environment and their related programs and policies) beyond which health does not improve at all or very much? Why haven’t researchers suggested benchmarks for each community according to its level of health and health-producing factors?
Almost nothing in the public and population health literature addresses these questions. Some empiric research has attempted to determine the balance of health determinants associated with various health outcomes; such work examines the factor itself — such as being uninsured or high school graduation rates (5) — but not the financial or policy resources producing them. One exception is the work of the Trust for America’s Health: it estimated that an investment of $10 per person per year in proven community-based programs to increase physical activity, improve nutrition, and prevent smoking and other tobacco use could save more than $16 billion annually within 5 years (6).
Other research explores health investment relationships in detail, estimating the effect of public services on state mortality while controlling for median income and income inequality (7). Total per capita expenditures in public services were significantly associated with all mortality measures, as were expenditures in education, environment, and housing. Similarly, a study of the effects of state expenditures on state-level age-adjusted mortality reported on public expenditures, tax structures, and welfare program rules and found that more generous education spending, more progressive tax systems, and more lenient welfare program rules helped to improve population health (8). However, the magnitudes of the effects were small, most likely because using the state as the unit of analysis masks much of the variation in outcomes and investments at local levels.
In Shaping a Health Statistics Vision for the 21st Century, the National Committee on Vital and Health Statistics (9) called for a better health statistics information system to help policy makers decide how to use health resources and emphasized the importance of considering factors beyond the health of the individual, to include community and socioeconomic factors. We do not believe, however, that there yet exists a national or state data set that includes comparable measures of financial investment in the multiple determinants of population health at the substate level. The absence of such a data set is probably due to the many challenges of collecting data from multiple levels of government and making them comparable.
As part of its annual reports on state expenditures, the National Association of State Budget Officers noted that comparing data, even among states, is difficult because data are collected and reported differently (10). For example, they noted that spending on juvenile corrections may be characterized as spending on corrections in 1 state and spending on human services in another.