Ethical issues in research: study design and publication-worthiness as a case in point

Abstract

Byline: T. Sathyanarayana Rao, Chittaranjan. Andrade Several issues of this Journal have, in the recent past, carried editorials related to scientific and ethical aspects of psychiatric research and publishing in India and elsewhere. [sup][1],[2],[3],[4] One of the reasons why this subject has been accorded prominence is that the Medical Council of India and many universities now require evidence of research publications as part of the eligibility criteria for the selection and promotion of medical teaching faculty, [sup][2] heralding the dawn of a publish or perish era in the country. Perhaps, as a result, there are now a large number of national, zonal, state, and even private journals in existence, and quite possibly some of the relatively unknown journals among these accommodate research that is of dubious quality. Do we have evidence for such a serious charge? Bluntly put, the answer is yes. We describe, by way of example, some manuscripts that we have reviewed in our editorial capacities. These have been suitably anonymized to respect professional proprieties. Three Manuscripts *Manuscript 1 is a randomized controlled trial (RCT) that compares two long-established antipsychotic drugs for a well-accepted indication on the grounds that there are no published Indian data on these drugs. The manuscript does not state a hypothesis nor a primary outcome measure. The sample size is too small for a superiority trial given that both drugs are active and hence between-groups differences will be small; and it is far too small for an equivalence or noninferiority design, which is probably what the authors had in mind. The diagnoses in the RCT were based on clinical assessment and not on standardized interviews. Patients were recruited in all phases of illness, and not acute illness, alone. Neither patients nor raters were blind to treatment allocation. *Manuscript 2 is an RCT that compares two long-established antidepressant drugs for a well-accepted indication. It has all the limitations of RCT and Manuscript 1 with the addition that one drug, which the authors appear to support, is adequately dosed while the other drug, no doubt the comparator, is strikingly underdosed. *Manuscript 3 describes a cross-sectional study that compares routine hematological and biochemical test results between manic inpatients and healthy controls drawn from the hospital staff. The aim of the study was to identify biomarkers of bipolar illness. Comments on the Manuscripts Patients who consent for research are inconvenienced, or put at risk, in many different ways. For example, they may receive a potentially ineffective drug and thereby suffer illness and its attendant risks for a longer period; they will need to spend more time in the hospital because of the study procedures related to diagnosis and assessment; they will need to undergo more discomfort because of blood tests and other assessments that are mandated by the study protocol; they will need to attend follow-up more often, as required by the study protocol; and so on. Patients consent for research with the expectation that the sacrifice that they make will benefit the cause of medical science in general, and future patients in particular. [sup][5] Furthermore, when studies are conducted, review and ethics committees spend time and energy in overseeing the research. A funding organization pays for the study expenses, which can often be quite considerable. Hence, all in all, it is wrong for very many reasons to conduct research that does not answer a meaningful question or advance the cause of science in any way. Let us return to the three manuscripts that we have described. RCT 1 appeared to be purely exploratory and was probably unnecessary, given that both drugs were well-established for the indication for which they were being studied, and given the absence of a specific hypothesis that the authors intended to prove or disprove. …