Uncertainty in Clinical Research

"It is a fact of life that human beings find it difficult to maintain a consistent, self-conscious appreciation of the extent to which uncertainty accompanies them on their daily rounds and to integrate that uncertainty with whatever certainties inform their conduct. Physicians are not exempt from this human proclivity." Jay Katz¹     

A positive domestic violence screen predicts future domestic violence

The objective of this study was to determine if a brief screen for domestic violence (DV) predicts future violence. We conducted a cohort study of adult women who presented to an inner-city emergency department during an 8-week study period. Participants were screened for DV using the Partner Violence Screen (PVS). At 4 months, follow-up telephone interviews were conducted: rates of verbal and physical violence were measured using the modified Conflict Tactics Scale. Relative risks of violent events (physical and verbal) were calculated. Of the 215 women who enrolled, 36 (16%) had an initial screen positive for DV. Ninety-six women participated in the follow-up phase; of these women, 9% had screened positive for DV. At 4 months, women with DV were 11.3 times more likely to experience physical violence and 7.3 times more likely to experience verbal aggression. The study's screen identified women at high risk for subsequent physical violence and verbal aggression.     

Use of prostate specific antigen in the identification of semen in postmortem cases

Over the past 20 years, the use of prostate specific antigen (PSA) as evidence of the presence of semen in forensic cases has been well established. In this study, we compared a commercially available immunoradiometric assay (IRMA) for the identification of PSA with the identification of spermatozoa in swabs obtained from the vagina of deceased females. There was agreement between the 2 methods in 67 of the 80 cases (84%), including 18 cases where both the PSA was positive and sperm were observed microscopically. The remaining cases had 1 positive result and 1 negative result. We concluded that although there was substantial agreement between the presence of sperm and a positive PSA, there were enough discrepancies between the 2 methods (16%) to justify the use of both methodologies, if possible, to make the determination of sexual activity prior to death.