Critical Review of Research Article: Statistics and Research

The purpose of this paper is to review the article concerning the utility and overall accuracy and effectiveness of secondary trial data, as well as assess the practicality of secondary trial data within the field of health care. By examining numerous aspects of the article and comparing them to the standards established by Sox & Goodman, it is possible to assess the overall quality and utility of both the article itself as well as the concepts contained within it. Examining the utility of secondary trial data will help to put into perspective the numerous ways that data can be obtained, and how not all methods of obtaining data are created equal. This should allow for a greater understanding of the underlying concept of statistics and even allow for appropriate action to be taken if secondary trial data is found to be not as practical a method of obtaining data as once believed.

The basic premise behind this article is a general examination of both the concepts behind secondary trial data as well as some actual examples of the data, such as quality of life, patient utilities, cost of care, and potential sources of cost data (Simpson & Tilley, 2011). The main focal point of the research to be examined here is a comparison between the economic aspects of different treatment approaches, especially within the field of the cost-treatment analysis (Simpson & Tilley, 2011). These are usually referred to as cost-effectiveness analysis or cost-utility analysis, as these two analyses are perfect examples of the sort of secondary trial data that the research aims to examine (Simpson & Tilley, 2011). The reason these two aspects, cost-effectiveness analysis and cost-utility analysis, are such good examples of secondary trial data is because they rely on other, primary forms of data to obtain an accurate result. That is to say, the cost-effectiveness analysis, for example, relies on obtaining primary data from the source, the cost, in this case, and extrapolating that cost with the benefits of the costs in question, and, from there, forming secondary trial data by using the cost-benefit analysis. This allows for a much more thorough analysis of primary data, since it allows the most important aspects of research, such as cost and overall effectiveness, to be examined at least once more.

In order to better understand the primary points being presented by this article, it is necessary to compare the standards and methods of the research with methods put forth in the Sox & Goodman article that list some of the most effective methods of comparative effectiveness research. First, one of the most important parts of completing a study is to carefully analyze the content of the study and tailor it toward the specific audience the study is aimed at (Sox & Goodman, 2012). In this respect, the research article performs its role admirably. The research wastes no time in quickly delving into the hard facts of the matter, such as the numerous variables requires for an effective cost-utility analysis (Simpson & Tilley, 2011). The reason this is so important in terms of the audience is that this focus on cold, hard data could very well polarize a great deal of the research's potential readership, so focusing on what could be considered niche factors represents a willingness by the research authors to disregard the common reader and tailor its research and vocabulary directly to health care professionals and statisticians.

However, the very concept of utilizing the cost-effectiveness analysis in the first places exposes the research article to many biases, which the Sox & Goodman article makes a point to mention avoiding at all costs. Most prominently, the self-selection bias can easily make the data from a report or research slanted toward one direction or the other because the very means that the data was obtained in the first place is unbalanced or unfair to one party or another (Boardman, 2011). The research article in question, fortunately, does appear to take efforts to avoid these biases by spreading out the variables to be examined, and recommending that any further research for secondary trial data be similarly spread out (Simpson & Tilley, 2012). This brings up one of the faults of the research. The article by Sox & Goodman emphasizes an approach that lends itself well to the subject matter that is being examined or researched on a case-by-case basis (Sox & Goodman, 2011). Quite simply, this means the methods should complement the subject matter well. However, one study found that, for the most part, cost-benefit analyses are local almost exclusively to strictly economic research (Layard, 1994). Since the research article being examined here is most certainly not economic in nature, it makes the cost-benefit analysis, as well as the other cost-analyzing being performed, seems out of place, and might not be the best way to go about researching this secondary trial data. A more fitting type of research to be performed might be a simple comparison between many of the variables examined in the study, such as the cost of hospital admission and drug costs post-discharge (Simpson & Tilley, 2012).

Another important facet of research as dictated by the Sox & Goodman article is the utilization of observational studies, oftentimes by using organic, real-life data (Sox & Goodman, 2011). However, oftentimes the issues being researched simply do not lend themselves to the gathering of real, in-person data very well, leading to a situation seen here, where the research variables, most of which are simple statistics, might seem cold and distant (Simpson & Tilley, 2012). There is no real way around this, and the research does attempt to rectify this problem somewhat by gathering statistics that are nevertheless "real" and applicable to the real world, such as drug costs and hospital length of stay; variables that most anyone who is reading the research with an interest in health care will find useful. In fact, much of the material found throughout this research paper seems to be adapting to many of the standards of the Sox & Goodman paper, even though the actual material being covered by the research does not lend itself well to this.

The last major aspect of the paper to be compared to the template in the Sox & Goodman paper is known as the propensity score analysis, which essentially just amounts to a comparison between multiple comparable variables or factors within a particular study (Sox & Goodman, 2011). In this respect, the research passes with flying colors, as it utilizes not just several different variables, which it compares extensively with one another, but also utilizes different models to do so, such as the Markov model (Simpson & Tilley, 2012). Essentially, these Markov models examine the present state of a variable, and then extrapolate this state with previous states of the variable, making it ideal for examining variables such as cost or profit from a particular variable (Rabiner & Juang, 1986). Using this Markov model allows for the research to feel much modern and relevant to current medical interests, as it represents a gradual change within these variables, as well as a willingness to accept that change.

On the whole, the research article adheres well to the standards and methods presented in the Sox & Goodman model and does an admirable job of performing research that is both informative but also specific to the field that it is focusing on. This helps to accomplish what many, including one article, define to be the ultimate goal of research such as this, which is an overall economic and social value to as many people as possible, even those who are not necessarily interested in the research (Nelson, 1971). The research's focus on the methods itself, that is, the gathering of this secondary trial data, combined with the novel methods of reporting, gathering, and researching the data itself, allow for a great deal of leeway when performing the research, which makes the results feel much more natural, especially since these results were, themselves, based on primary results from other studies. As far as secondary trial data goes, this article demonstrates the most effective and efficient methods to gather and report it to the audience.

References

Boardman, A. E. (2010). Cost-benefit analysis. 3-4.

Layard, P. R. G. (1994). Cost-benefit analysis. R. Layard, & S. Glaister (Eds.). Cambridge University Press. 3-6

Nelson, R. R. (1971). Simple Economics of Basic Scientific Research, The. J. Reprints Antitrust L. & Econ., 3, 725.

Rabiner, L., & Juang, B. H. (1986). An introduction to hidden Markov models. ASSP Magazine, IEEE, 3(1), 4-16.

Simpson, K. N., & Tilley, B. C. (2012). Economic Analysis of Secondary Trial Data. Progress in Cardiovascular Diseases, 54(4), 351-356.

Sox, H. C., & Goodman, S. N. (2012). The Methods of Comparative Effectiveness Research. Annual Review of Public Health, 33, 425-445.