10 Ways to Better Understand How Shifting State Policy Contexts Affect Americans’ Health

The impact of US state policies on population health is increasingly evident.¹ States’ earned income tax credits, cigarette sales taxes, right to work laws, Medicaid expansion, firearm safety, and paid family leave are just a few examples of specific policies that affect population health. Better understanding these effects has become urgent as states continually grow farther apart on both their policies and their population’s health. Although research has started to connect the dots between the divergence in states’ policies and their populations’ health,^2,3 it has become clear that better understanding the connections requires new approaches and greater attention to certain issues. I offer here 10 ways to accelerate research on the role that the seismic shifts in states’ policy contexts may have played in the troubling trends and growing disparities in Americans’ health. The list is informed by years of collaborative research on the role of state policy contexts on population health, thoughtful questions from academic and nonacademic audiences following presentations of that research, and insightful critiques from journal reviewers.        

1. Examine Policy Bundles. Many states have shifted their policies in recent decades to cohere with either a liberal or conservative orientation.¹ Some states raised the minimum wage, offered an earned income tax credit, expanded Medicaid, increased cigarette sales taxes, enacted firearm safety laws, ensured that workers have access to paid sick or family leave, and more. Many other states did not enact these policies and prohibited their localities from doing so. Residents of these two state examples experience two very different sets, or “bundles,” of policies.¹ Understanding the role of shifting state policies on population health requires more than single-policy studies—it requires examining policy bundles. Bundles reflect the reality that people do not experience one policy at a time.⁴ Bundles can also reveal synergistic or countervailing effects of policies on health and account for the fact that multiple policies often are enacted simultaneously.^1,5
2. Establish Guidelines for Developing and Analyzing Policy Bundles. Assessing the causal effects of policy bundles on population health can be challenging. For instance, a bundle that sums the number of 10 economic policies in each state in each year could be used effectively to estimate how changes in the economic policy context are associated with changes in population health, but it poses challenges for estimating causal effects. As an example, two states could have the same score for the bundle, but the specific policies that comprise the score may differ between the states. Guidelines would facilitate the development of policy bundles that are valid and interpretable, and would ensure that studies using them can provide policy-relevant, actionable results.
3. Examine Intersections between Federal, State, and Local Policies. Policies at each level of government can affect population health individually but it is unclear how they collectively impact health. The health impact of states’ policies may simply “add to” the health impacts of federal and local policies, or the impact may be accentuated or dampened by federal and local policies.
4. Investigate Lag Times. The effect of some state policies on population health may occur rapidly while the effect of others may take years to materialize. Investigating lag times is important for capturing the true effects of policies on population health and building an explanation of the effects. For instance, if the impact of a policy on population health takes five years to manifest, a study using a two-year lag may incorrectly conclude that the policy has no impact.
5. Evaluate Life Course Timing of Policy Exposure. The state policy context that people experience changes over time, regardless of whether people continuously reside in their state of birth. Someone may have spent two decades of employment earning $7.25 per hour minimum wage and not receiving an earned income tax credit, but, after several policy changes in their state of residence, they spent the next two decades earning $15 minimum wage and receiving earned income tax credits each year. A point-in-time measure of the state’s labor policies for this individual obscures decades of prior exposure to the policies. Developing cumulative measures of policy exposures could provide more rigorous estimates of their health impacts.
6. Assess Population Heterogeneity. State policies may disproportionately affect some populations. For example, minimum wage, earned income tax credits, and paid sick leave are particularly salient for less-educated individuals and their families. State policies may also be more consequential for certain types of local areas such as urban versus rural areas. If the health impacts of certain policies are stronger for some populations and areas than for others, then estimates that do not account for this heterogeneity likely will underestimate the impact of those policies on the populations and places disproportionately affected.
7. Develop Methods to Account for Interstate Migration. Some people move across state lines, which could bias estimates of the health impact of state policies if healthier people leave certain states and move to other states. Accounting for migration is hindered by a lack of data on people’s migration histories within datasets large enough to examine state policy effects on population health. A recent study using a novel counterfactual approach concluded that widening disparities in working-age mortality among states in recent decades were not due to interstate migration patterns.³ While encouraging, the importance of accounting for migration may depend on the time period studied. For example, the importance may rise if migration becomes strongly driven by income levels, political views, or other factors correlated with health. Better data on people’s migration histories and guidelines on how best to account for migration are needed.   
8. Avoid Controlling for Downstream Population Characteristics. Studies of the impact of state policies on population health generally, and wisely, control for differences in the age, sex, and race-ethnic composition of states. This is distinct from controlling for population characteristics like education, income, and health behaviors, which are influenced by state policies. Controlling for such characteristics downwardly biases estimates of the health impact of state policies.
9. Identify Pathways between Policies and Health. Understanding the reasons why certain policies impact population health requires identifying casual pathways. The pathways are likely to be varied and both intended and unintended. For example, the pathway between minimum wage policy and population health is only partly financial. Raising minimum wage has also been shown to improve birth outcomes, reduce teenage pregnancy, reduce smoking, and more.   
10. Analyze the Actors Influencing State Policies. State policies are influenced by special interests such as corporations, their lobbying groups (e.g., the American Legislative Exchange Council), and wealthy donors.⁶ Fully fleshing out the role of states in affecting population health requires incorporating the role of these actors into research studies. Doing so can identify the root causes of the seismic changes in states’ policy contexts and, consequently, in population health.

In conclusion, emerging research points to a potential role of state policy contexts in the alarming trends and growing disparities in population health in the United States.^2,3 Doing this type of research is challenging, particularly given data limitations and the vast number of state policies that exist, have changed over time, and have become correlated. Advancing this research could benefit by considering these 10 suggestions. Although no single study is expected to incorporate all 10 suggestions, the field at large could benefit by systematically incorporating them into future research.

References

1. Montez JK, Grumbach JM. U.S. State Policies and Population Health. The Milbank Quarterly. 2023:in press.
2. Montez JK, Beckfield J, Cooney JK, et al. US state policies, politics, and life expectancy. The Milbank Quarterly. 2020;28(3):668-699.
3. Couillard BK, Foote CL, Gandhi K, Meara E, Skinner J. Rising geographic disparities in US mortality. J Econ Perspect. 2021;35(4):1-27.
4. Beckfield J, Bambra C, Eikemoc TA, Huijtsd T, McNamarac C, Wendte C. An institutional theory of welfare state effects on the distribution of population health. Social Theory & Health. 2015;13(3/4):227-244.
5. Matthay ECM, Gottlieb LM, Rehkopf D, Tan ML, Vlahov D, Glymour MM. What to do when everything happens at once: analytic approaches to estimate the health effects of co-occurring social policies. Epidemiol Rev. 2021;43:33-47.
6. Hertel-Fernandez A. State Capture. New York: Oxford University Press; 2019.

About the Author