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).
Malecki, K. C., Engelman, C. D., Peppard, P. E., Nieto, F. J., Grabow, M. L., Bernardinello, M., ... & Martinez-Donate, A. (2014). The Wisconsin Assessment of the Social and Built Environment (WASABE): a multi-dimensional objective audit instrument for examining neighborhood effects on health. BMC public health, 14(1), 1165.
*(This was too large to upload)*
This paper describes the development and implementation of the WASABE and examines the instrument’s ability to capture a range of social and built environment features in urban and rural communities. A systematic literature review and formative research were used to create the tool. Inter-rater reliability parameters across items were calculated. Prevalence and density of features were estimated for strata formed according to several sociodemographic and urbanicity factors.
2.What was the definition of the construct?
In order to define “neighborhood” environments, ArcGIS Network Analyst (ESRI, Redwood, CA) was used to define a 400-meter (about a quarter of a mile) non-Euclidian street network buffer around each selected household. This distance (equivalent of a 5–10 minute walk) was chosen because previous studies on “walkability” have found it to be the upper limit of the distance individuals are generally willing to walk to procure a service
Urban and rural communities were classified at the census block group level according to U.S. Census definitions for urbanicity (http://www.census.gov/geo/reference/ua/urban-rural-2010.html).
Dichotomous measures of presence or absence of a feature within the 400 m buffer surrounding an individuals’ household (or within a polygon) were used to estimate prevalence of non-residential destinations, walking/biking trails, parks, fitness centers, grocery stores, litter and trash, and fast food restaurants. Social environment features including presence or absence of neighborhood social or cultural signs, security warnings or signs and active engagement defined by observation of people walking or biking were also derived.
Health promoting behaviors were classified as dichotomous outcomes according to whether or not individuals self-reported having met U.S. physical activity or dietary requirements (yes vs. no).
3.How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?
In order to assess construct validity of WASABE, we examined the ability of the instrument to capture variation in exposure to built and social environment features within the SHOW sample. The prevalence of features was examined by sociodemographics, health behaviors, neighborhood perception and census block group urbanicity (urban vs. rural) for 939 participants.
I can’t think of a better approach.
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?
To better understand the reliability of the tool across different field surveyors, we explored inter-rater reliability using percent agreement across all segments in the dataset that were double rated by different raters within at most one week of one another. We used percent agreement to assess inter-rater reliability rather than kappa statistics because our goals were to assess comparability and reproducibility across a number of different pairs of raters]. Categories of inter-rater reliability were predefined as excellent (>90%), very good (80-89%), good (70-79%), moderate (60-70%) or poor (<59%). After initially testing the reliability and validity of the results 2009 data, measures with moderate to poor agreement were dropped from the final WASABE tool used for the 2010 sample collection. In addition, for more subjective measures where we saw poor reliability we improved the training and modified the manual.
5.Describe the implications of a lack of measurement validity or reliability for future research applications.
I chose this article because I believe where we live is intricately tied to wealth. Most people chose to live in the best neighborhood that their money can afford; and the qualities of the built environment of the neighborhood we can afford to live in impacts our health. Neighborhood health is more than the beauty of the area. It is also the proximity of resources and the ability to engage in positive health behaviors. I realize that I am extrapolating the connection of the qualities of a neighborhood to a person’s socioeconomic status.
The lack of validity for a tool like this can lead to gross exaggerations of how people interpret using the tool. For instance, a cornerstore with bananas and oranges is not the same as a grocery store.
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)
Perry, J., Linehan, C., Kerr, M., Salvador‐Carulla, L., Zeilinger, E., Weber, G., ... & Buono, S. (2010). The P15–a multinational assessment battery for collecting data on health indicators relevant to adults with intellectual disabilities. Journal of Intellectual Disability Research, 54(11), 981-991.
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).
Construct: Health inequities faced by adults with intellectual disabilities
Outcomes measured: Quality of life in four domains:
Demographics ( living arrangements; daily occupation; income/socio-economic status; life expectancy)
Health status Epilepsy; oral health; body mass index; mental health; sensory; mobility
Determinants: Physical activity; challenging behaviour; psychotropic medication use
Health systems: Hospitalisation & contact with health care professionals; health
check; health promotion; specific training for physicians
3.What is the evidence for the validity and reliability of the measures?
When data collection was complete in all 14 countries, each interviewer was asked to complete a questionnaire to assess the feasibility of the P15 in terms of: (1) face reliability, i.e. apparent accuracy of responses according to type of respondents (person with ID, family carer, non-family carer); (2) social and psychological acceptability, i.e. whether items were acceptable to respondents (judged mainly by whether respondents refused to answer particular items); (3) user-friendliness, i.e. whether respondents had difficulty understanding and/or responding to particular items, the extent of additional explanation required, training requirements and time taken to administer; and (4) practicality, i.e. how useful the item might be in routine practice and monitoring, whether it would be useable in general health surveys/databases, and whether interviewers required additional training/experience to administer particular items. Respondents were asked to address these issues on a section-by-section basis, and to identify items which proved problematic on one or more of these areas for 25% or more of the P15 interviews administered. For example, if an interviewer found that one in four interviewees had difficulty responding to the item which requested the frequency of visits to and from their general practitioner, that item would be flagged and rated on each of the feasibility criteria. Responses from interviewers who had conducted at least four P15s were analysed. This amounted to 23 interviewers who had conducted 674 P15s between them. Considering the P15 as a whole, the feasibility questionnaire also asked individual country project leaders or principal interviewers for ratings on (1) to (4) above plus applicability (whether the content seemed appropriate for the assessment of the health of adults with ID), efficiency (the value of the information derived from the P15 in relation to the effort required to solicit it), coding complexity and cultural transferability. Each of these was rated from 1 to 4, where lower ratings indicated greater feasibility in each respect.
High internal consistency was found for the established scales incorporated within the P15 (the ABC, the PAS-ADD and the SNS) and for the epilepsy and mobility indicators. The internal consistency of the physical activity and oral health domains was low; and alpha coefficients on the remaining domains were reasonable.
The mean Kappa value of the PAS-ADD Checklist and the ABC in all countries was 0.73 and 0.48, respectively.
Results of the bivariate tests using data from the current sample were consistent with findings reported in the wider literature in the case of six out of nine comparisons
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?
It doesn’t state specifically, but it is assumed that the comparison group is adults without intellectual disabilities. This reference group makes sense because people fall on a range of different intellectual capabilities just as the general population has a range of health and social needs.
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.
Selection of indicators was guided by the quality and quantity of supportive literature, the capacity to operationalise it and the value of the indicator for advancing knowledge about the health of adults with ID.
The researchers chose 18 factors that gave a broad picture of quality of life measures. They stated that this tool could be used in conjunction with other specific benchmarks, like measuring vaccination rates among adults with disabilities. Information gathered from this survey could be used to monitor trends in health and lifestyle to inform service planning.