Evaluation of signal detection theory on the effects of psychotropic drugs on critical flicker-fusion frequency in normal subjects.

Critical flicker-fusion frequency (CFF) has often been applied to psychotropic drug evaluation as a measure of cortical arousal. There are several variables that must be considered for proper conclusions to be obtained from CFF experimental data. Psychological variables such as subject response bias are especially difficult to control. We studied the effects of single oral doses of 10 mg diazepam and 10 mg amphetamine sulfate on CFF values obtained by a block up-down spatial forced-choice method on 13 healthy volunteers (seven males and six females). Signal detection theory was also used to obtain the non-parametric value P(A) as a measure of sensory function and B as a measure of the psychological function of response bias. As a group, amphetamine increased CFF (Hz) and P(A) and diazepam decreased CFF (Hz) and P(A), with the most significant effects observed at 2 and 3 h after drug administration. B scores showed more individual variation with a trend towards a low score, or a tendency to report more flicker responses after diazepam. Separated by sex, the males had a higher percentage of subjects that demonstrated a reduction of CFF after diazepam, while the females had a higher percentage that demonstrated an increase in CFF after amphetamine. The results suggest that CFF changes following diazepam and amphetamine are mostly changes in sensory function and not changes in response bias. It is possible to apply signal detection theory to flicker-fusion studies and the accounting of bias by controlling, measuring or eliminating it is essential in interpretation of CFF data.

Needs assessment for reducing infant mortality in Baltimore City: the Healthy Start Program.

Needs assessments are essential for policy formulation and the appropriate design of intervention programs. Recent nationwide data show that among large metropolitan areas of the United States, Baltimore has one of the highest infant mortality rates and ranks in the worst top 10 for blacks and the top 5 for whites for most indicators of poor pregnancy outcome. In this paper, we present the methods and results of a needs assessment that used multiple sources of routinely collected data and was conducted for the purpose of identifying intervention factors contributing to infant mortality in Baltimore City. This needs assessment was used by the Baltimore City Health Department to successfully secure funding for the federal Healthy Start Infant Mortality Prevention Initiative. We present the results of the analyses, along with some of the proposed interventions that resulted from the needs assessment. We also discuss the limitations of this type of needs assessment as well as suggestions for future needs assessments for the design of interventions to improve perinatal health.