HW#6
Identify a policy that is not usually intended to be a health policy but that you think may have important health implications.
Describe why an evaluation of that policy is informative (e.g., determining effects of the policy, or primarily a test of hypothesized mediators).
Specify the outcomes and populations you think most affected or least affected by the policy.
Propose a study design to evaluate the policy.
Describe biggest challenge to implementing and drawing inferences about the impact of the policy on health.
The Merchant Marine Act of 1920, better known as the Jones Act, is a protectionist regulation that prevents goods shipped between American ports from being made, owned, or crewed by non-Americans. When the law was passed in the wake of WWI, the original intent was 2-fold: to support the domestic shipping industry and to ensure the American government had access to ships and maritime personnel, in case of emergency or war. Many argue that the law is antiquated and should be repealed, but until recently, the most the law had done in modern day was make prices for imported goods higher in the non-contiguous US (Hawaii, Alaska, and Puerto Rico) and made a few Florida-based shipping companies a bit richer. [Disclaimer: This is true only on the surface. In actuality, the cost of goods from the mainland US is at least double that in neighboring islands, and a 2012 report estimated that the Jones Act caused $17 billion loss to the island’s economy between 1990 and 2010.]
In September 2017, the island of Puerto Rico was devastated by Hurricane Maria, leaving 3.4 American citizens without electricity and thousands of homes destroyed. Even 5 months after the impact, a third of the population is still without power. In this time of need, when FEMA is still needed to deliver food and water, a repeal or suspension of the law would be helpful to lower the cost of delivering that aid. Unfortunately, President Trump temporarily waived the Jones Act for only 10 days in October 2017. A bill to repeal the act is still stuck in the Senate. Waiving this act could impact access to food and water, but also to medicine for the sick, oil for truck to deliver supplies throughout the country, and raw materials to rebuild a nation’s infrastructure.
The US Virgin Islands, a US territory that is exempt from the Jones Act, was also devastated by Hurricane Maria (in addition to the damage done by Hurricane Irma). In a territory with a little more than 100,000 residents, over 33,000 individuals and families have applied for FEMA assistance, and 2 months after the hurricanes, 73% of the residents still had no power. Of course, there are differences between the nations, but it would be interesting to compare the impact of the price of imported goods/the Jones Act on the recovery of Puerto Rico vs. that of the US Virgin Islands. The largest challenges would be attributing the difference in recovery to the impact of the Jones Act, since other factors play a role (media coverage, state of island pre-hurricane, dependence on foreign goods, etc), and determining an appropriate recovery marker (power, food security, health?).
HW#7
Part 1:
1. Choose a paper describing the development or validation of a measure of relevance in health disparities research (please give the full citation and/or upload the paper if that's possible).
2. What was the definition of the construct?
3. How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?
4. How did the authors provide evidence on the reliability of the measure? Could you think of additional approaches to evaluating the reliability of the measure?
5. Describe the implications of a lack of measurement validity or reliability for future research applications.
Paper: Psaki, Stephanie R et al. “Measuring Socioeconomic Status in Multicountry Studies: Results from the Eight-Country MAL-ED Study.” Population Health Metrics 12 (2014): 8. PMC. Web. 27 Feb. 2018.
The construct defined was a measurement of socioeconomic status in a multi-country study, named the WAMI index. It includes access to improved water and sanitation; eight priority assets, selected via random forest; maternal education; and household income. They constructed this index with information provided by a questionnaire given in the 8 diverse countries to the parents of participants in the MAL-ED study, a cross-sectional study of 800 children that aimed to find relationships between early malnutrition and enteric infections and child physical and cognitive development. To validate, they compared 4 existing measures of SES: maternal education, principal component analysis, Multidimensional Poverty Index, and a novel variable selection approach using random forests. When the random forest method yielded the lowest prediction error, they combined the best components of each method to form the WAMI index, and found a significant correlation with the child height-for-age Z-score (HAZ). The study does mention that the tool is verified only for use in these 8 countries and for this disease correlation, but the method proves that an international SES index can be developed and be reliable, even between disparate cultures. For a more long-standing measure, both validity and reliability would be extremely important in future studies. Without a valid measurement tool, the predictor/modifier cannot be correctly accounted for and could result in wildly different conclusions. Without a reliable measurement tool, the study could not be repeated with the same result, which gives the audience less reason to trust the first result. In health disparities work, arguably more than in any other clinical research, the measurement tool must be beyond reproach. Otherwise, measurement error can explain away disparities that no one wanted acknowledged in the first place.
Part 2:
1. Find a paper describing a health disparity (please give the full citation or, even better, upload the paper so everyone else can look at it too)
2. Summarize the construct and measurement of the dimension of disparity (e.g., racial inequalities?, SES inequalities?) and the outcome measured (e.g., self-rated health).
3. What is the evidence for the validity and reliability of the measures?
4. What is the reference category used for the disparity measure (ie, who is the comparison group)? Why does this reference category make sense (or not) for this research question?
5. How is the disparity quantified or measured? Is this an absolute or relative measure or are both provided? Describe which type of measure you would prefer for this research area, or, if both, why.
Paper: Grobman, William A. et al. “Racial and Ethnic Disparities in Maternal Morbidity and Obstetric Care.” Obstetrics and gynecology 125.6 (2015): 1460–1467. PMC. Web. 27 Feb. 2018.
This paper analyzed data from a cohort of pregnant women who delivered at 25 US hospitals between 2008 and 2011 to determine if there is a racial/ethnic disparity in maternal morbidity and mortality. They collected information on race (Non-Hispanic White, Non-Hispanic Black, Hispanic, or Asian), on adverse events (severe postpartum hemorrhage, peripartum infection, severe perineal laceration at spontaneous vaginal delivery), and on types of obstetric care (episiotomy, general anesthesia at cesarean, labor induction, etc), as documented in the patients’ charts.
Unfortunately, race/ethnicity was simply transcribed from the patient records (not self-reported), and there is research that has argued this is not a valid measurement method (Caterino et al 2013). Non-Hispanic White is the reference group, which make sense because they are the “best-off” group for the study outcomes. And the disparity is measured via unadjusted and adjusted odds ratios for adverse events, which could increase morbidity and mortality, and for types of obstetric care, which may partially account for those differing adverse events. The ORs provided a relative measure of the disparity, which is appropriate for this cross-sectional study. The tables also provided counts for each of the adverse events by racial group, which allows the reader to calculate an absolute disparity and gives him/her a sense of the effect size. However, the ORs are more generalizable to a larger population.