Evidence-Based Medicine, Article Review Example

Abstract

In a study by Bangalore, et.al. (2012), the researchers investigated the use of ?-blockers in patients with, and at-risk for, coronary artery disease. The researchers concluded that ?-blockers do not reduce risks associated with coronary artery disease.  While the article appears to be valid upon first review, because it is a large qualitative study, when evaluated using an evidence-based medicine approach, the study lacks validity on several key points.

The study is qualitative research that employs a very large number of subjects.  However, this fact alone does not make the study valid.  The subjects were selected from an international database of patients who had experienced myocardial infarctions, had coronary artery disease, or had risk factors for coronary artery disease.  This database is not random, and the criteria for selecting the subjects were not random.  The subjects were men and women who were already concerned with the possibility of heart disease as the title of the database suggests.  A more accurate measure would look at a random selection of subjects from an age group in which coronary artery disease is common.

Results & Data

In their analysis, the researchers divided the subjects into three groups: those who had suffered myocardial infarctions, those who had coronary artery disease, and those who were at risk of coronary artery disease.  The validity of the study is very suspect because there is not a control group.  The researchers do not use a group of subjects without any known occurrence or risk of coronary artery disease that match the ages and genders of the subjects.  Without this grouping, it is impossible to determine whether the ?-blockers had no effect because of their inefficacy or because of some other factor.  The groups the researchers studied did not show reduced occurrence of cardiac events when given ?-blockers, but what is the occurrence of cardiac events in a similar group not identified as having coronary artery disease?  Validity would be lent to the study if the researchers had found that a control group of healthy subjects similar to the groups in their study had cardiac events at a similar rate to the control group.

Participants

Additionally, the database consists of observational data that was collected on a group of subjects that registered for the database between 2003 and 2004.  Researchers collected data for approximately five years after selection of the participants, ending in 2009.  For a study published in 2012, the period of years elapsing between the last data collection and the publication of the report is too long.  The data analysis should not have taken four years, which is almost as long as the researchers conducted the study.  Why was there such a long time period between data collection and publication?  This large gap could indicate bias or even a difficulty in finding a publisher, which could indicate concerns over the validity of the study.

Lastly, the subjects were only studied for a total of five to six years.  The researchers did not identify if that was an appropriate amount of time to wait for an occurrence of a cardiac event.  In the three groups identified by the researchers, what if the mean time of a secondary cardiac event was ten years and the use of ?-blockers extended this time frame to twenty years?  This study would be unable to identify that effect and could falsely lead physicians to not prescribe ?-blockers that could have extended their patients’ lives.  This outcome alone is reason to discard this study.

Conclusion

In the end, the study has too many error and inconsistencies to provide information for a doctor.  A physician attempting to decide whether to prescribe ?-blockers to heart patients would need to do much reading beyond this study.