Strengths and Issues Associated with Randomized Controlled Trials

RANDOMIZED CONTROLLED TRIALS 1

Randomized Trials in Epidemiology

Wei et al. (2017) conducted a randomized controlled trial to measure the effectiveness of an antimicrobial stewardship program. The program, which targeted caregivers and providers in China, sought to reduce inappropriate antibiotic prescribing for respiratory tract infections in children. The study was informed by the high rate of inappropriate antibiotic prescribing among pediatric patients in China. The study population was primary care hospitals in rural China. The researchers selected all 25 hospitals within Liujiang and Rong Counties in Guangxi Province, and randomly allocated 13 to the control group and 12 to the intervention group over a six-month intervention period. Hospitals in the intervention group received caregiver education on antibiotics prescribing, monthly peer-review meetings for physicians, as well as clinician training and guidelines on appropriate antibiotics prescribing. Conversely, those in the control group offered usual care, where clinicians prescribe antibiotics at their discretion.

The study ran for six months. The outcome measure was the antibiotic prescription rate for pediatric outpatients aged between 2 and 14 receiving treatment for upper respiratory tract infections. The researchers collected baseline data on pediatric antibiotic prescription rates in all hospitals in the three months before the start of the study. To assess the effectiveness of the intervention, the researchers reviewed outpatient antibiotic prescription rates over the final three months of the six-month intervention period to determine the proportion of outpatient pediatric patients who received antibiotics for upper respiratory tract infections. They adjusted for doctor covariates, confounding patient factors, and baseline antibiotic prescription rates. The findings showed an intervention effect, with antibiotic prescription rates in the intervention group declining by 42 percent as compared to 5 percent in the control group. The study concluded that the antimicrobial stewardship program reduced unnecessary antibiotic prescribing for children by 29 percent.

The study provides crucial insights for policymakers in China. However, it faces two fundamental ethical issues. First is the question of whether it is ethical to deny children antibiotics on the basis of fulfilling scientific goals (Goldstein et al., 2018). Some randomized controlled trials attempt to minimize bias by offering placebo to control group participants as those in the intervention group receive the treatment intervention. In this study, however, children in hospitals under the intervention group are denied antibiotics for research purposes. This could raise ethical questions and may be interpreted as a violation of the non-maleficence and beneficence principles that prohibit medical professionals from deliberately causing harm to patients (Curley, 2020).

Secondly, the researchers in this study collect data by reviewing children’s medical prescription records, which raises both information privacy and informed consent concerns. Under normal circumstances, the researchers would notify participants that they intend to use their personal medical records for scientific purposes and the participants would then give their consent (Goldstein et al., 2018). This study, however, uses children, who may not be in a position to legally give their consent.

Despite these ethical issues, the adopted study design brings about crucial benefits that would not be realized using other research designs. The primary benefit is that the random assignment of items to either the control or intervention group helps to increase objectivity by minimizing researcher bias in selecting whom to place in what group (Goldstein et al., 2018). The high objectivity, in turn, enhances the generalizability of research findings (Goldstein et al., 2018). Generalizability means that the research findings generated from the sample can effectively be used to make inferences about the entire population. In this case, for instance, randomization makes the findings generated from the sample of 25 generalizable to all Chinese hospitals.