Week 7 Hwk

Week 7 Hwk

by Nicholas Rubashkin -
Number of replies: 3

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). 

Stud Fam Plann. 2014 Mar;45(1):19-41. doi: 10.1111/j.1728-4465.2014.00374.x.

Development and validation of a reproductive autonomy scale.

Upadhyay UD1Dworkin SLWeitz TAFoster DG.

 

 

2.What was the definition of the construct?

 

Reproductive autonomy is one domain in the construct of “women’s empowerment,” defined as women’s ability to make strategic life choices where this ability was previously denied them.  Specifically, reproductive autonomy is having the power to decide about and control matters associated with contraceptive use, pregnancy, and childbearing.  This includes whether and when to become pregnant, whether and when to practice consideration, and whether and when to continue a pregnancy.   This multi-dimensional scale assesses reproductive power in the interpersonal interactions along three domains: freedom from coercion, communication, and decision making.

 

3.How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?

 

To assess the construct validity, they examined the measure’s association with a history of unprotected sex among a group of women were wished to avoid pregnancy.  Freedom from coercion and communication were significantly associated with a reduction in unprotected sex.  Decision making was not significant but showed a reduction as well. 

 

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 researchers believed that they found a valid and reliable measure for future research applications.  They measured the reliability of the scale using the Cronbach’s alpha (full scale 0.78). However, the “decision-making” sub scale had low internal consistency with a Cronbach’s alpha of 0.651.

 

5.Describe the implications of a lack of measurement validity or reliability for future research applications

 

The authors hypothesize that decision making around pregnancy varies according to the context, such that autonomous decision making may be more important for contraception and abortion and less important for desired pregnancies.  In other words, when a pregnancy is desired, women tend to involve partners and support networks in collaborative decision making.  This low-reliability of the decision making scale may mean that future comparative investigations into disparities in women’s autonomy should focus on the more reliable constructs of freedom from coercion and communication. 

 

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) 

Obstet Gynecol. 2016 Oct;128(4):869-75. doi: 10.1097/AOG.0000000000001628.

Health Care Disparity and State-Specific Pregnancy-Related Mortality in the United States, 2005-2014.

Moaddab A1Dildy GABrown HLBateni ZHBelfort MASangi-Haghpeykar HClark SL.

 

 

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).

 

Racial inequalities in maternal mortality ratio (maternal deaths per 100,000 live births).

 

3.What is the evidence for the validity and reliability of the measures?

 

They don’t discuss the validity and reliability of their data sources which derive from (1) the National Vital Statistics System (maternal and neonatal information entered into the birth certificate) and the (2) CDC database of Detailed Mortality Underlying Cause of Death (maternal deaths determined by ICD-10 codes).

 

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?

 

I think the authors use white as the reference category, but they may also use the “lowest” MMR in a given state—which is not always the white group.  The state data for disparities in MMR are all over the map, making it difficult to compare across states and over time due to changes in the proportion of the population within each racial category. 

 

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. 

 

The authors conducted bivariate correlations between the MMR and demographic, lifestyle, health, and medical service utilization factors.  They followed this with a maximum likelihood factor analysis, retaining variables that her significant at p<0.05.  Then they assess the relationship between these significant factors and the MMR using regression analysis combined with the Jouckheera-Terpska test to determine the presence/absence of a trend.

 

As I look more closely at this study, I don’t think they calculated a relative or absolute measure of MMR, even though the title includes the word “disparity”.  The “Jouckheera-Terpska” test was not listed in the NIH report in our assigned reading. They did find that “black race” was the most significant predictor of a higher state level MMR with a regression coefficient of 0.39 (larger than any of the other regression coefficients).  

 

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.

 

This was a multi-dimensional scale to measure pregnancy intent.  The desire sub-scale had a high cronbach alpha indicated a reliable measure.  It proved to have a valid relationship to pregnancy planning (termination or live birth).   The authors argue that “pregnancy intent” is multi-dimensional and a Yes/No response does not adequately capture the spectrum of variables that a woman considers around her pregnancy.  Being able to predict the level of desire and the need for timing around pregnancies can help optimize delivery of family planning services.

In reply to Nicholas Rubashkin

Re: Week 7 Hwk

by Brianna Michelle Singleton -

I agree with the authors, sometimes we don't talk about coercion and how women negotiate the choice to have sex when it comes to family planning. I haven't read many studies, but the few that I have read start with the assumption that the woman was fully empowered when she had sex. I'm not saying a rape occurred, but I am alluding to the fact that there are many reasons and pressures that led to the sexual act that speaks to her level of autonomy, values, and empowerment before she was faced with the family planning decision. In short, I really like the definition of reproductive autonomy that encompasses the entire experience.

In reply to Nicholas Rubashkin

Re: Week 7 Hwk

by Maricianah -

I agree with your  evaluation. I am however concerned that in their development – they used only quantitative means to validate their construct for a scale that is going to measure a deeply entrenched social behavioural. While they make a good case by mentioning the Cronbach alpha (full scale 0.78) which implies good internal validity and the fact that they examined the measure’s association with a history of unprotected sex among a group of women who wished to avoid pregnancy, I am concerned that we are not sure that if we repeated the measurement we will get the same result (reliability of the construct). I think (and I could be wrong) that both validity and reliability are required for any good scale. This has tremendous implications of measurement for future research particularly because  reproductive autonomy is very much a context-specific concept. 

In reply to Nicholas Rubashkin

Re: Week 7 Hwk

by Maria Glymour -

Nick,

Nice examples.  Maricianah makes a good point that it is extremely helpful to evaluate face and content validity of new measures of subtle constructs, such as repro autonomy.  The Cronbach's alpha is a measure of reliability, but it is very desirable to have test-retest.  Validity assessment is often much harder though.  I don't know this field well enough to propose additional validation assessments.  There's no gold standard presumably, so most evaluations will depend on construct validity. 

The Moaddab et al paper is very surprising.  First, it seems like they "tricked" CDC wonder into delivering small cell sizes that the database developers intended to obscure.  I'm surprised this isn't against the data use agreement for Wonder, and it certainly seems questionable.  Second, their application of a factor analysis to completely disparate constructs is puzzling as a data reduction approach.  Usually you would do factor analysis if you actually thought there was a coherent underlying construct which all of the items measured.  That is not the case here, they are just trying to select among a large set of covariates with a relatively modest sized set of observations.  Factor analysis doesn't seem appropriate.  Third, the interpretation that trends are due to racial disparities is puzzling given that the trends occur in whites and blacks.  

Maria