Socioeconomic Disparities in Low Birth Weight, Article Critique Example

In a critique of quantitative research, it is important to focus on the facts as they are presented. In addition, it is vital that you fully understand the facts as presented by the work you are critiquing, as well maintain an open mind to the possibilities of factual disparity and understanding. Through this effort the research that was done will be reviewed and critiqued. In addition, other sources of information have been reviewed that provide supplemental data that either refutes or supports the findings.

Abstract: a work of research is only as good as the abstract that identifies the key goals of a study. In this instance, the abstract was brief, direct and concise. It effectively laid out the  problem, the objective, the types of data used and the method of analysis that was conducted. The abstract does include important points covered in the article, but I feel it could have went further in identification of conclusion as I thought what was written was too vague. In addition, it did not fully address one of the most significant conclusions of this study, and an area where this study delved further than others of its kind. Although it mentions the lack of a father’s name on a birth certificate, we find out later in the study that this was a significant finding.

Problem statement and purpose: the purpose as stated in the paper is to identify variables that account for an elevated risk of postneonatal mortality within the Alaskan Native population. Multiple variables were tested among a population that was divided between Alaskan Natives and non-natives, and the analysis was conducted with methodology described as  univariate, stratified and regression.

Within the literature review, this study focused on prior studies from both primary and secondary sources, noting that in previous efforts the focus has been primarily on behavioral, socioeconomic, and genetic explanations. The review further states that studies have identified high poverty rates, lack of health care access and cultural dislocation as reasons for poor health status of Alaskan Natives. This was cited in one article, stating that high rates of infant mortality among disadvantaged populations has been defined as a social issue for the past 130 years (Hogue and Hargraves, 1993). Another article cites socioeconomic status among other major factors such as nutrition, mother’s age, birth weight, and lack of prenatal care (Saenz, 2007). Furthermore, the three primary ways that different societies have attempted to address this problem is through elimination of poverty, providing free, high quality healthcare, and elimination of high risk behaviors among those who are poor (Hogue and Hargraves, 1993.)

One gap identified in the literature review is that in Alaska, healthcare is delivered by a native regional health corporation that will provide care regardless of ability to pay. The study further notes that because of this, the issue of infant mortality may assumed to be less attributable to health care access and socioeconomic status than elsewhere. One article cites that among the largest risk factors for death in Native Alaskan infants includes low birth weight, maternal prenatal smoking and maternal alcohol use during pregnancy (Walsh, 2012.)

Hypothesis and sampling: the stated belief within the abstract was that by targeting Alaska Native women who display the characteristics of a small number of potentially modifiable factors, that you can significantly reduce the postneonatal mortality gap within this population. The primary hypothesis was the increased risk of postneonatal death among Alaskan Native infants was due to higher prevalences of other known risk factors among the population, confounding the relationship between mortality and race.

The variables considered were derived from birth certification information, which included maternal age, prenatal care, education, presence of a father’s name on a birth certificate, and marital status. As there was an established method to differentiate the sample based on multiple factors, the three sampling categorizations were married, unmarried with father’s name present, and not married with father’s name missing. This was done for both Native Alaskans and non-natives. These results were compared based on a sample size of 10,300 births and 72 infant deaths between the years of 1992-2004. Given that this data spanned twelve years, it should be considered an adequately sized sample. Additionally, the findings within this study should be considered appropriate for both native and non-native people residing within Alaska.

Research design: The research design is described as logistic regression model-building to identify confounding variables that affected the association between postneonatal mortality and race. There are precedents cited within the research where similar studies used comparable methodology to examine potential risk factors. Two models were built which are referred to as selection and deletion. In selection, one confounding variable was entered into a model where postneonatal mortality was the dependent variable, and race was an independent variable. In the second deletion model, all confounding variables were entered, but then each variable was removed to note changes in the model. This methodology utilizing two different modeling approaches would appear to coincide with the research problem, theoretical framework, literature review and hypothesis.

Internal and external validity: the paper excludes some data that could potentially either conflict with the efforts to achieve effective results or could distract from the findings. For example, birth weight and pre-term birth were excluded as variables because they were not vital and could lead to bias. Additionally, a cursory review determined there is no significant difference in these factors between native and non-native populations. The paper also identifies the source of the cause of death information within Alaskan state records. The determination is made by a state committee that follows a national model for review. This would appear to be a credible committee to make the proper determination of cause of death. Although there is always potential for some error in data, especially when this data is from an outside, uncontrolled source, it does appear reasonable to assume this data is reliable.

Data collection methods: the data used for the study was obtained from birth and death certificate records for all Alaska resident live births and infant deaths between the years 1992-2004. The paper states that the Alaska Native population represents the largest minority in Alaska, accounting for 25% of all state births. Although twelve years of data is used, the relatively small number of Alaskan Natives as compared to non-natives is one potential area of concern (Day, Provost and Lanier, 2009). Regardless of quantity of population, state records should be considered a reliable source of data for purposes of this study.

Legal and ethical issues: although generalizations were necessary specific to Alaskan Natives, utilization of state records and anonymity of those who are sampled in the study should largely address legal and ethical concerns. By utilizing state record data the information used to formulate the study is a matter of public record. No individuals were singled out for interviews or questions.

Reliability and validity: all of the data utilized is a matter of public record and accessible for review. The methodology undertaken was explained fully should someone wish to verify the validity of the statements that were made as a result of this study. The methodology and modeling were properly explained so they could be clearly understood.

Analysis of data: variables were considered confounding when there was at least a 10% differential in the odds ratio between race and postneonatal mortality within the two models. Based on this definition, confounding variables were identified. Alaskan Native women were more prone to be in high risk categories of having less education and inadequate prenatal care. Another confounding variable is the lack of inclusion of a father’s name on the child’s birth certificate. These three confounding variables were identified as those that explain the majority of an increased risk of postneonatal mortality among Alaskan Natives. These statistics are included within the study and would appear both appropriate and consistent with abstract and hypothesis. Five tables of data are also included within the study showing the quantifiable results, inclusive of appropriate titles and headings.

References

Day, G.E., Provost, E. and Lanier, A.P. (2009, Jan./Feb.). Alaska Native Mortality Rates and Trends. Public Health Reports, Vol. 124, 54-64.

Hogue, C.J.R. and Hargraves, M.A. (1993, Jan.). Class, Race, and Infant Mortality in the United States. American Journal of Public Health. Vol. 83, No. 1. 9-12.

Saenz, R. (2007, Oct.). The Growing Color Divide in U.S. Infant Mortality. Population Reference Bureau. Retrieved from http://www.prb.org/Articles/2007/ColorDivideInfantMortality.aspx.

Walsh, N. (2012, Jan. 13). Mortality High in Native Alaskan Infants. MedPage Today. Retrieved from http://www.medpagetoday.com/Pediatrics/GeneralPediatrics/30646.