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).
Paper: Association of socioeconomic position and medical insurance with fetal diagnosis of critical congenital heart disease (paper uploaded)
2.What was the definition of the construct?
In this paper they used the year 2000 U.S. census data and linked each home address to their census block group. A block group is the smallest geographic census unit and is though to be relatively homogeneous with respect to economic status and living conditions of its residents. They then calculated a composite SE score for each patient based on a prior validated measure. The score is based on 6 variables for each subject’s block group. Three variables represent dimensions of wealth and income (log of the median household income, log of the median value of housing units, and the percentage of households receiving interest, dividend or net rental income). Two variables represent education (the percentage of adults 25 years or older who completed high school, and the percentage of adults 25 years or older that completed college). The last variable represents occupation (the percentage of employed persons 16 years of age or older in executive, managerial or professional specialty occupations. This was developed using a ‘factor analysis’, which I don’t know much about. The composite SE score for each subject was calculated by summing the 6 scores (they converted to z-scores).
3.How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?
In this particular article, the authors point back to a study done in 2001 (Diez Roux et al, NEJM), that used these methods to assess the incidence of coronary heart disease by neighborhood of residence. They said that the measure was validated in that article, but it wasn’t clear when I read the original NEJM article. One way to validate this type of measure is to match it to questionnaire responses from individuals living in a particular neighborhood to see if the composite SE score is an adequate representation of that persons socioeconomic position (i.e, personal or individual data).
This study was also done in the New England area and may not be representative of the entire U.S. population, thus another approach would be to test these methods in a different region in the U.S.
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?
The authors did not provide evidence on reliability of this measurement. Since this was based on census data (self-reported), I am not sure how we would go about testing the reliability of the measure. One concern about this composite score is that it is based on a factor analysis. Thus, several other measures of SEP were not included because they weren’t felt to be as powerful as the ones included. I would say that reliability of this measure is probably good since it is based on the census data, but I question the validity because of the comments above.
5.Describe the implications of a lack of measurement validity or reliability for future research applications.
Given my concerns for the validity of this composite score, it is not a score that can be utilized outside of the research arena at this point. It should also be interpreted with caution, as it does not include several other important components of SEP (i.e. race/ethnicity). I think they tried to address this in a multivariable model by looking at the SE composite score (categorized into quartiles), type of medical insurance and race/ethnicity. In this model, they found that only private insurance was associated with an increase in the odds of having a prenatal diagnosis of CHD. I find this interesting and confusing, since all of these factors are related to one another.
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)
Paper:
Hints SR, Gould J, Bennett MV, Gray EE, Kagawa KJ, Schulman J, Murphy B, Villarin-Duenas G, Lee HC. Referral of Very Low Birth Weight Infants to High-Risk Follow-Up at the Neonatal Intensive Care Unit Discharge Varies Widely across California. J Pediatr 2015;166:289-95.(paper too big to attach)
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).
In this study they looked at various predictors of NICU graduates following-up in High-Risk Infant Follow-up Programs. These programs have been declared necessary by California Children Services (CCS/Medi-cal) and are thus supported financially by public insurance. There is a mandate that these NICU’s across the state have these programs for their patients. One of the things this article assessed was the role of maternal race as a predictor of follow-up in these programs. In table 2 of the paper (multivariable logistic regression model), they found that a maternal African American or Hispanic race was associated with lower adjusted odds of follow-up in a High-Risk infant Program.
3.What is the evidence for the validity and reliability of the measures?
In this case, maternal race was self reported and entered into a central database (CPQCC-CCS) that encompasses all Very Low Birth Weight Infants that survive until discharge home. The article does not specifically talk about reliability and validity; however, since race is self reported, I believe this is the best way to assess its role in health disparities. They do mention that this disparity is likely related to “challenged support systems” in hospitals with serving the highest number of A.A. and Latina mothers.
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?
In the multivariable model, the reference category used was White race. It seems they did this to compare A.A. and Latinos to the next highest majority population (which would be whites). The Harper article talks about comparing to the “Best-off group”. In this case that would be the Native American group; however, this group is much smaller in size as compared to the other groups so this approach also has limitations.
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.
They measured this disparity by performing a multi-variable logistic regression analysis and reporting adjusted odds ratios, so it is a relative measure. I think this approach makes sense; however reported a risk difference would also be helpful.
Part 3:
1. Read someone else's response to part 1 above (identifying a construct) and comment, specifically noting whether you can see any additional implications of measurement quality for future research or whether you agree with those noted by your classmate.
I found the article posted by Tene very interesting and a good way to incorporate race into risk models for various diseases. I think it is a strategy that should be used more often in the research setting to develop risk models. This would help us move more towards personalized medicine rather than assuming that risk factors are the same across all racial groups.