Healthcare Research What Was the Thesis

Overall, 8% from the control group completed testing and counseling versus 23% from the intervention group, which is a "modest increase" according to the researchers. None of those who completed testing was positive for HIV.

With the results controlled for race/ethnicity, i.e., meaning, within the same race/ethnic group, the turnout from the incentive-driven group was significantly higher than the control group. It was also found that whether or not they came from the incentive group, African-Americans and to a lesser degree, Hispanics were significantly less likely to complete testing and counseling compared to the other racial groups.

The researchers concluded that financial incentive resulted in a moderate increase in the number of ED-referred patients completing HIV testing and counseling. They could not say if this method was going to increase the percentage of detection of undiagnosed HIV+ individuals or if it's going to be cost-effective. They recommended point-of-care testing, or testing for HIV right there and then in the ER with rapid, same-day results, to sufficiently impact the number of ethnic minorities testing for HIV infection.

4. In evaluating the research study, what were the weaknesses and strengths of the study? Was the study susceptible to bias? If so, state, define and explain how the study was susceptible to the bias (es).

First of all, the population was not properly defined in the study. It appears that anyone, regardless of age, ethnicity, presenting condition, etc., who presented himself to the ED was a potential subject for the study. Second, the sample was not randomly chosen and was in fact a "convenience sample" composed of participants pre-selected for their increased risk of HIV infection based on the attending physician's assessment. Third, it appears that a physician referral bias was present as evidenced in Table 1 by the differences in the distribution of racial group/ethnicity between the control and intervention groups. Finally, there was no control for possible clustering by physicians, i.e., the study did not control for intra-variation among physicians to account for their individual effects on the referral rate.

On the other hand, the researchers recognized the potential physician bias and addressed this using a "multivariable logistic regression model to assess the independent effect of financial incentive while controlling for race and ethnicity" (Table 2). The ability to recognize bias/limitations and consider them in the interpretation of results is one of the strengths of this study. The fact that the experiment is reproducible and has potential for improvement despite the limitations imposed is another reason to commend this study.

5. Explain the concept of randomization. Was this a randomized study?

Randomization is the method of choosing samples in a study with equal probability. This assures that the sample is homogeneous, unbiased, and truly representative of the population in question. I believe that the study was not randomized, as previously explained. Specifically, the subjects were chosen based on pre-selected criteria (hence, with unequal probability) by physicians who were neither controlled for possible biases in their selection of the subjects. Further, it was not ensured that the control and intervention groups were statistically equal.

References

Haukoos, J., Witt, M., Coil, C., & Lewis, R. (2005). The Effect of Financial Incentives on Adherence with Outpatient Human Immunodeficiency Virus Testing Referrals from the Emergency Department. Academic Emergency Medicine 12 (7); p…