Byron Wm. Brown

The crossover experiment for clinical trials.

The two-period crossover or changeover design for clinical trials is compared with other simple designs in terms of statistical precision and cost. The sensitivity of the crossover to bias due to carryover effects is examined. The feasibility of using the crossover data to test for the existence of carryover effects is investigated and found to be uneconomical. A numerical example is presented.

Susceptibility rhythm to E. coli endotoxin and bioassay.

Summary Light-synchronized periodicity analysis reveals a susceptibility rhythm in C mice to E. coli endotoxin. Susceptibility varies predictably and significantly along the 24-hour time scale. A dose of endotoxin which is compatible with survival of most animals when given during middle of daily dark period is highly lethal when it is given 8-12 hours earlier or later. LD50 also was determined for 2 samples from identical batch of endotoxin tested 12 h apart on separate groups of comparable C mice kept under conditions standardized for periodicity analysis: Differences in potency significant at 1% were seen at 2 dose levels common to both assays, the computed “potency ratio” being 3.22.

Hypnotic efficacy of diphenhydramine, methapyrilene, and pentobarbital

The antihistamines diphenhydramine and methapyrilene were compared with pentobarbital for hypnotic effect in two Veterans Administration Hospital populations using subjective‐response methods. In the first part of the study, 60 mg and 180 mg of pentobarbital were compared with 50 mg and 150 mg of diphenhydramine. A positive dose‐response relationship was obtained only for pentobarbital; neither dose of diphenhydramine was significantly different from 60 mg of pentobarbital for any response variable. In the second part of the study, 100 mg of pentobarbital, 50 mg of diphenhydramine, and 50 mg of methapyrilene were compared with placebo. One hundred mg of pentobarbital and 50 mg of diphenhydramine were significantly different from placebo, but 50 mg of methapyrilene was not.

Planning a quantal assay of potency.

SUMMARY A step-by-step procedure is given for using available information and a statement of desired precision to design a quantal assay of the potency of a Test Preparation relative to a Standard Preparation. The simpler assay for estimating an L.D.50 is also considered. In each case the result is a specification of the dose levels to be used, and the number of observations to be taken at each dose level, to assure a desired precision.

Combined routes of administration to assay oral analgesia in postoperative pain.

To increase the sensitivity of the method for evaluating oral analgesics in postoperative patients, we designed a combined oral/parenteral bioassay. Drugs studied were parenteral morphine, parenteral propiram, and oral codeine at two dose levels each and oral propiram at four dose levels. Results from data on 308 patients suggest that future studies designed to establish the relative potencies of oral analgesics should use parenteral morphine as the standard in a combined oral/parenteral study because this approach provides a very sensitive measure of analgesia. Further, with one drug as the reference compound, results from many sources would be more readily compared.

A statistical analysis of data from film badge performance tests.

Abstract Data on tests of film badge services have been analyzed statistically to determine the probability that services may be expected to meet performance criteria developed under the auspices of the National Sanitation Foundation (NSF). These criteria were derived empirically on the basis of a single sample of film badge services in the United States. Ten film badge services participating in the NSF seal of approval program released their data for this analysis. The study compares the empirical criteria and statistical expectations based on newly available data. The implications of the analysis for future film badge performance criteria are also discussed.

Effects of amine oxidase inhibition on E. coli endotoxin mortality in mice.

Summary M.A.O. inhibitors P.I.H., Nialamide and S.K.F. 556 were shown to have a small but significant effect in reducing mortality to intraperitoneal injection of E. coli endotoxin. Inhibition of M.A.O. alone appeared to have an effect distinct from inhibition of both M.A.O. and D.A.O. though the 3 drugs were not of significantly different potency.