Jessica Enogieru_Week 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).
He VY, Condon JR, Baade PD, Zhang X, Zhao Y.Different survival analysis methods for measuring long-term outcomes of Indigenous and non-Indigenous Australian cancer patients in the presence and absence of competing risks. Popul Health Metr. 2017 Jan 17;15(1):1
2.What was the definition of the construct?
ANSWER: The authors measured cancer survival in Indigenous vs NonIndigenous patients using four cancers (head and neck, breat, colorectal, and lung). Net survival (removes the effect on non-cancer deaths) is often used to compare cancer survival rates across regions, nations and ethnic groups.
However, this becomes a problem when patients and/or clinicians want a true or “real world” prognosis of death from cancer. The authors decided to evaluate another risk of death measure called “crude probability of cancer death” that takes into account competing risk and identifies the risk of death from cancer IN THE PRESENCE OF other causes of death.
3.How did the authors provide evidence on the validity of the measure? Could you think of additional approaches to validating the measure?
ANSWER: The authors used two ways to calculate crude probability of cancer death (the Cronin-Feuer method and the Fine-Gray method). They compared two measures against net survival AND compared the two measure to each other.
I cannot think of other methods to validate the measure itself.
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?
ANSWER: The authors discussed the problem sparseness of life-table-based methods to calculating crude probability of death particularly the sparseness of data in Indigenous patients. This problem was overcome by evaluating Indigenous patients in the Northern Territories. Indigenous data from the Nortern Territories have more complete life tables and high-quality case of death data (because 30% of the population in the Northern Territories is made up of Ingedenous peoples vs 2-4% in other parts of Australia). The authors also gave 95% confidence intervals for both measures of crude probability of death (Cronin-Feuer and the Fine-Gray) to evaluate the precision of the estimate.
5.Describe the implications of a lack of measurement validity or reliability for future research applications.
ANSWER: Crude probability of cancer death estimates give clinicians and patients a more realistic view/prognosis of their survival compared with net survival estimates.
In my opinion, not only do net survival estimate ignore risk of competing death, but it also ignores the effects of non-cancer conditions on improving or worsening cancer prognosis. For example, the chronic stress of having heart failure of end stage renal disease requiring dialysis can affect inflammation levels in the body which can effect how cancer progresses or regresses in a patient AND how effective drug/chemo treatments are on treating the cancer itself. Evaluate risk of cancer death in a bubble is not good enough.
It is important for both research and clinical applications that crude probability of death estimates be accurate and precise. A lack of reliability could give cause a two fold problem. Estimates that are to low produce a favorable prognosis which could give clinicians and patients false hope. Estimates that are too high would produce an unfavorable prognosis that could cause clinicians to recommend aggressive treatments that cause undue side effects or cause clinicians to recommend hospice/palliative care when there is still a good chance of survival with treatment.
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)
Rangrass G, Ghaferi AA, Dimick JB. Explaining racial disparities in outcomes after cardiac surgery: the role of hospital quality. JAMA Surg. 2014 Mar;149(3):223-7..
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).
ANSWER: Primary outcome measure of hospital quality used was risk-adjusted mortality rate within 30 days of cardiac (CABG) surgery or in-hospital death. The mortality rate between hospitals was compared, the mortality rate between white and nonwhites within each hospital after cardiac surgery was compared. Multivariate Logistic regression was used to evaluate the relationship between race and mortality rates accounting for confounders such as socioeconomic status and other patient characteristics.
3.What is the evidence for the validity and reliability of the measures?
ANSWER: The source of the data was patients with CABG surgery from the Medicare database (2007-2008, Medicare analysis provider and review data files). They used ICD codes to identify patient who got the procedure and used hospital discharge abstracts for Medicare recipient to compile patient characteristics and vital status at 30 days. ICD codes are a commonly used and reliable method to identify patient who have a disease or have undergone a procedure. For Medicare reimbursement, hospitals are required to submit discharge abstracts for each Medicare patient, thus these abstracts are commonly used and reliable source of patient characteristics data.
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?
ANWER: Whites were used as the comparator group to nonwhite when they performed within hospital mortality rate analysis.
Hospitals that treated the lowest percentage of nonwhite patients were used as the reference group when comparing mortality rates between hospitals. Hospitals organized into 3 terciles according to the proportion of nonwhite patients undergoing CABG surgery.
These reference groups are valid because the research in the cardiac field has established that whites have better outcomes compared with non-white ethnic groups.
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.
ANSWER: The measure itself is a quantitative measure (the proportion of patients who died within 30 days of having CABG surgery) taken from the Medicare databases. The measure was analyzed and compared in different groups using logistic regression, from which odds ratio’s were calculated (thus absolute and relative calculations are provided). I definitely prefer both because absolute terms help to understand the scope of the measure itself (how many people die in total post CABG surgery) but relative terms give a sense of the depth of the disparity of the measure between different groups (relative increase in risk between racial groups for the measure within hospitals AND relative increase in the measure in hospitals that serve more nonwhite patients vs white patients).
Part 3:
- 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.
ANSWER: I agree with Amy Lockwood’s response regarding the need to re-test the MIRE survey (self reported racism in Indigenous populations) in order to determine if there is a temporal effect on the responses to the questionnaire. As she stated, there SHOUDN’T be based on the type of questions asked, but its important to validate that that does not occur.
I think it is also important to re-test not only the same people at a different time, but also test Indigenous peoples in different areas of Australia to determine if the survey is robust towards geography. For instance, self-reported racism in the Northern Territories (30% of population is Indigenous) may be different compared with self reported racism in other parts of Australia (2-4% of population is Indigenous).