Search

In 1915, Greenwood and Yule noted that for valid vaccine efficacy studies, exposure to infection in the vaccinated and the unvaccinated must be equal (Proc R Soc Med 1915;8(part 2):113-94). The direct effect of a vaccine, however, needs to be defined by the protection it confers given a specific amount of exposure to infection, not just a comparable exposure. In this paper, two classes of parameters are distinguished along lines differing from the conventional distinction between efficacy and effectiveness. Efficacy parameters attempt to control for exposure to infection and represent direct effects on individuals. Direct effectiveness parameters represent a mixture of direct effects on individuals and indirect effects in the population.

Glutathione, a tripeptide found in a variety of foods, may function as an anticarcinogen by acting as an antioxidant and by binding with cellular mutagens. The association between dietary glutathione intake and risk of oral and pharyngeal cancer was investigated using data from 1,830 white participants (855 cases and 975 controls) in a population-based case-control study conducted in New Jersey; metropolitan Atlanta, Georgia; Los Angeles County, California; and Santa Clara and San Mateo counties, south of San Francisco-Oakland, California, during 1984-1985. The estimated relative risk of cancer among people with the highest quartile of glutathione intake from all sources was 0.5 (95% confidence interval 0.3-0.7). When analyzed by dietary source, however, glutathione intakes derived from all vegetables and from meat were not related to risk of cancer. Only glutathione derived from fruit and from vegetables commonly consumed raw was associated with reduced oral cancer risk. Relative to the lowest level of combined intake of fruit and of fruit-derived glutathione, risk of cancer decreased slightly with increasing intake of fruit glutathione. This analysis was limited, however, by the small numbers of subjects with extreme combinations of intakes. Further studies are needed to distinguish the potential effect of glutathione from that of fruit and raw vegetables per se or from the influence of other constituents in these foods.

A common source of bias in evaluating vaccine efficacy following a disease outbreak is the presence of persons who had the disease prior to the outbreak. This paper examines the effects of including and excluding pre-outbreak disease cases from the calculation of vaccine efficacy based on the cumulative incidence at the end of an outbreak. Using a five-stage model, the effects of the following factors on the bias of vaccine efficacy estimates are examined: the true protective efficacy of the vaccine, the prevaccination infection rate, differences in vaccine uptake among the previously diseased and nondiseased, differences in pre-outbreak exposure to infection between vaccinees and nonvaccinees, and differences in exposure during the outbreak between vaccinees and nonvaccinees. Numerical calculations of the bias are performed for a hypothetical outbreak of measles in a developing country. Exclusion of pre-outbreak disease cases requires accurate data on disease rates prior to the outbreak, and such data are often unreliable or nonexistent. Inclusion of pre-outbreak cases contributes to the bias of the estimated vaccine efficacy, especially when there is a high prevaccination infection rate and vaccine uptake among the previously diseased is considerably lower than that among the nondiseased. In most practical cases, however, this bias is not very large.

Interpretation and estimation of vaccine efficacy is complicated when the vaccine effect is heterogeneous across vaccinated strata. If a person has a certain susceptibility, or probability of becoming infected conditional on a specified exposure to infection, then one effect of a vaccine would be to reduce that susceptibility, possibly to zero. Vaccine efficacy is a function of the relative susceptibilities in the vaccinated and unvaccinated persons. Under heterogeneity of vaccine effect, a general expression for a summary vaccine efficacy parameter is a function of the vaccine efficacy in the different vaccinated strata weighted by the fraction of the vaccinated subpopulations in each stratum. Interpretation and estimability of the summary vaccine efficacy parameter depends on whether the strata are identifiable, and whether the heterogeneity is host- or vaccine-related. Bounds are derived for the summary vaccine efficacy when the strata are not identifiable for the case of an outbreak of an acute infectious disease. The upper bound assumes that everyone is equally affected by the vaccine, and the lower bound assumes that some are completely protected while others have no protection. The biologic interpretation of the two bounds is different.

A statistical model is presented for the analysis of infectious disease data from family studies in the community. The model partitions the sources of infection into those from within the household and those from the community at large. The parameters reflecting these sources of infection are estimated as functions of the risk factors. This new model is used to overcome problems associated with the lack of independence of observations in infectious disease data and negative confounding due to the association of unmeasured exposures and immunity. An example of how this new statistical model is used to provide a clearer and less confounded description of risk factor effects is presented for data from influenza A(H3N2) epidemic seasons in the Tecumseh Respiratory Illness Study. The risk factors examined are age and pre-epidemic season antibody level as measured by the hemagglutination-inhibition test, while the outcome is the infection rate. A standard analysis of the data indicates that the efficacy of protective antibodies is 70% in children and only 47% in adults. However, such an efficacy measurement is negatively confounded by past exposure which is age dependent. By means of the model, the true, unconfounded, efficacy of protective antibodies is shown to be 90% in both adults and children.