A probability model for evaluating the bias and precision of influenza vaccine effectiveness estimates from case-control studies.

As influenza vaccination is now widely recommended, randomized clinical trials are no longer ethical in many populations. Therefore, observational studies on patients seeking medical care for acute respiratory illnesses (ARIs) are a popular option for estimating influenza vaccine effectiveness (VE). We developed a probability model for evaluating and comparing bias and precision of estimates of VE against symptomatic influenza from two commonly used case-control study designs: the test-negative design and the traditional case-control design. We show that when vaccination does not affect the probability of developing non-influenza ARI then VE estimates from test-negative design studies are unbiased even if vaccinees and non-vaccinees have different probabilities of seeking medical care against ARI, as long as the ratio of these probabilities is the same for illnesses resulting from influenza and non-influenza infections. Our numerical results suggest that in general, estimates from the test-negative design have smaller bias compared to estimates from the traditional case-control design as long as the probability of non-influenza ARI is similar among vaccinated and unvaccinated individuals. We did not find consistent differences between the standard errors of the estimates from the two study designs.

Estimated paediatric mortality associated with influenza virus infections, United States, 2003-2010.

Death certificate reports and laboratory-confirmed influenza deaths probably underestimate paediatric deaths attributable to influenza. Using US mortality data for persons aged <18 years who died during 28 September 2003 to 2 October 2010, we estimated influenza-attributable deaths using a generalized linear regression model based on seasonal covariates, influenza-certified deaths (deaths for which influenza was a reported cause of death), and occurrence during the 2009 pandemic period. Of 32 783 paediatric deaths in the death categories examined, 853 (3%) were influenza-certified. The estimated number of influenza-attributable deaths over the study period was 1·8 [95% confidence interval (CI) 1·3-2·8] times higher than the number of influenza-certified deaths. Influenza-attributable deaths were 2·1 (95% CI 1·5-3·4) times higher than influenza-certified deaths during the non-pandemic period and 1·1 (95% CI 1·0-1·8) times higher during the pandemic. Overall, US paediatric deaths attributable to influenza were almost twice the number reported by death certificate codes in the seasons prior to the 2009 pandemic.

A set of broadly applicable microsatellite markers for analyzing the structure of Culex pipiens (Diptera: Culicidae) populations.

Microsatellite markers were isolated and developed from Culex pipiens quinquefasciatus Say (Diptera: Culicidae) sampled in Johannesburg, South Africa, to identify those that are broadly useful for analyzing Cx. pipiens complex populations between continents. Suitable loci should be 1) inherited in a codominant Mendelian manner, 2) polymorphic, 3) selectively neutral, 4) randomly associated, 5) without null alleles, and 6) applicable across broad regions and between diverse biotypes. Loci in Cx. p. quinquefasciatus from Johannesburg ranged from two to 17 alleles per locus and expected heterozygosities (H(e)) were 0.02-0.87. Loci in Cx. p. pipiens L. from Johannesburg had five to 19 alleles per locus and H(e) values ranging from 0.57 to 0.93, whereas those from George, South Africa, had five to 17 alleles per locus and H(e) values ranging from 0.54 to 0.88. Loci in North American mosquitoes were more variable. Cx. p. quinquefasciatus from South Carolina had five to 19 alleles per locus and H(e) values ranging from 0.64 to 0.90, whereas Cx. p. pipiens from Massachusetts had six to 28 alleles per locus and with H(e) values ranging from 0.65 to 0.94. All loci were associated randomly. Overall, four of nine of these new loci satisfied all six criteria for broad utility for analyzing the genetic structure of Cx. pipiens populations.

A focus of deer tick virus transmission in the northcentral United States.

We screened salivary glands from adult deer ticks collected near Spooner and Hayward, Wisconsin, to determine whether deer tick virus, a recently described flavivirus, occurs with other tickborne agents in the upper Midwest. Intraacinar inclusions suggestive of replicating virus were detected in 4 (4.6%) of 87 ticks. The virus was isolated by suckling-mouse inoculation.

Power and sample size calculations for case-control studies of gene-environment interactions with a polytomous exposure variable.

Genetic polymorphisms may appear to the epidemiologist most commonly as different levels of susceptibility to exposure. Epidemiologic studies of heterogeneity in exposure susceptibility aim at estimating the parameter quantifying the gene-environment interaction. In this paper, the authors use a general approach to power and sample size calculations for case-control studies, which is applicable to settings where the exposure variable is polytomous and where the assumption of independence between the distribution of the genotype and the environmental factor may not be met. It was found through exploration of different scenarios that in the cases explored, power calculations were relatively insensitive to assumptions about the odds ratio for the exposure in the referent genotype category and to assumptions about the odds ratio for the genetic factor in the lowest exposure category, yet they were relatively sensitive to assumptions about gene frequency, particularly when gene frequency was low. In general, to detect a small to moderate gene-environment interaction effect, large sample sizes are needed. Because the examples studied represent only a small subset of possible scenarios that could occur in practice, the authors encourage the use of their user-friendly Fortran program for calculating power and sample size for gene-environment interactions with exposures grouped by quantiles that are explicitly tailored to the study at hand.

A new tick-borne encephalitis-like virus infecting New England deer ticks, Ixodes dammini.

To determine if eastern North American Ixodes dammini, like related ticks in Eurasia, maintain tick-borne encephalitis group viruses, we analyzed ticks collected from sites where the agent of Lyme disease is zoonotic. Two viral isolates were obtained by inoculating mice with homogenates from tick salivary glands. The virus, which was described by reverse transcriptase polymerase chain reaction and direct sequencing of the amplification products, was similar to, but distinct from, Powassan virus and is provisionally named "deer tick virus." Enzootic tick-borne encephalitis group viruses accompany the agents of Lyme disease, babesiosis, and granulocytic ehrlichiosis in a Holarctic assemblage of emergent deer tick pathogens.

The relation of self-reported back pain to psychosocial, behavioral, and health-related factors in a working population in Switzerland.

Back pain causes a considerable loss of working days as well as health care costs and therefore represents a major public health problem in industrialized countries. Psychosocial factors have received increasing attention from researchers studying the causal factors of non-specific back pain. However, most studies focus on few dimensions, like individual or work-related factors. We studied the simultaneous association of various factors representing psychosocial, behavioral, and health-related dimensions to self-reported back pain. Data from the Berne Workplace Health Project on 850 employed men and women was analyzed. Back pain was operationalized by a dichotomized variable (having suffered moderately to severely from back pain in the preceding four weeks). The theoretical model guiding the underlying project was a general demand-resource model. Variables that--according to that model--were hypothesized to be related to back pain as well as more specific factors--like physical work load--were analyzed by stepwise logistic regression analysis. In men, there was a statistical trend (P < 0.1) for several work-related factors (low job discretion, high job demands, low job satisfaction). In women, dissatisfaction with salary was the only work-related factor associated with back pain. There was no significant association between private context factors, like poor social network or high demands/low control, and back pain. Only in men, the likelihood of back pain increased with age, while only in women, back pain was associated with emotional problems (individual factors). Among the behavioral factors, smoking was associated to back pain in men, while in women none of the behavioral factors was significant. In both men and women reporting more than two functional symptoms and a history of intestinal problems were associated to back pain. All of our findings were in the expected direction, i.e., it was invariably unfavorable categories of explanatory variables that were associated with higher prevalence of back pain. However, most associations seem to be quite unspecific. There is a need for theoretically guided research aiming at the development of a more complex process model of back pain.

[Occupational status and prevalence of cardiovascular risk indicators in employed men in German-speaking Switzerland]. / Berufliche Stellung und Prävalenz kardiovaskulärer Risikoindikatoren bei berufstätigen Männern in der deutschsprachigen Schweiz.

Based on a sample of 623 employed men from the Berne Workplace Health Project ("Härz-As-Project") we studied the relationship between the occupational status and prevalence of cardiovascular risk indicators. Besides "biological" risk indicators, like high total cholesterol, low HDL-cholesterol, hypertension, and overweight, we also studied two behavioral risk indicators (current smoking, physical inactivity in leisure time) and a summary risk score. Odds ratios for several risk indicators controlled for age, were markedly different among different occupational status groups. High cholesterol/HDL-cholesterol ratios were more common in lower occupational status groups, while the likelihood for hypertension and high total cholesterol was highest among qualified workers. There was a strong association between occupational status and the behavioral risk indicators smoking and physical inactivity in leasure time. Similarly, 1 or more risk indicators, and more than 2 risk indicators, respectively, were also found to be strongly related to occupational status. Our results confirm former findings of unequal distribution of cardiovascular disease risk indicators among groups of different occupational status. Future studies will have to focus upon the underlying causes for these inequities.

The relation of reported symptoms to social, individual, and behavioral indicators of ill-health: is the number of reported symptoms a unique general dimension of ill-health?

Counts of functional symptoms are an accepted measure of ill-health in general populations. However, their relation to other indicators of ill-health are still poorly understood. The aim of this study was to investigate such relations in order to reach a better understanding of the 'meaning' of the number of symptoms. The data analyzed was drawn from a telephone health survey conducted among a representative sample of 2111 Swiss adults as part of the Intercantonal Health Indicators Project in 1989. In a logistic regression analysis the Number of symptoms (a summery score of eleven severe function symptoms) was found to be strongly associated with eight social, behavioral, and psychological indicators of ill-health ('Unable to work', 'Frequently consulting a physician', 'Use of analgesics', 'Use of sedatives', 'Poor coping with ill-health', 'Social impairment', 'Poor health', and 'Poor psychological well-being'). To control for effects due to individual symptoms, they were included in the models by stepwise procedure. Although individual symptoms became predictive for all indicators (except for Poor coping with illness), the association between the Number of symptoms and the indicators (except for Social impairment) remained statistically significant. Despite methodological provisos our results seem to support an interpretation of the Number of symptoms as a distinct and general dimension of ill-health.

[Health, health-related behavior and gainful employment]. / Gesundheit, gesundheitsrelevantes Verhalten und Erwerbstätigkeit.

Based on data collected in a survey as part of the Swiss Intercantonal Health Indicators Project (random sample stratified by age of 20-75 year old persons, n = 2106) relationships between level of employment (full, part-time, none) and health/disease as well as health behaviour were studied. Employed men and to some extent employed women as well were found to be less frequently sick and to feel more frequently well than non-employed men and women. Overall employment appears to affect health/disease in men more strongly than in women, whereas influence on health behaviour appears to be similar in both gender groups. The results of the study are discussed both with regard to further hypotheses of health research and opportunities of health promotion.

Reported disease and psychological well-being in Swiss adults.

In Switzerland, and in many other countries as well, the distribution of morbidity and perceived health in the general population and their determining factors have not been systematically studied so far. This article reports on exploratory study of prevalence of two complex health indicators, long-standing disease (Disease) and psychological well-being (Well-being) and of their environmental, person-specific and behavioural determinants. Data from a health survey conducted in five cantons and three language regions as part of the Swiss Intercantonal Health Indicators Project were used. Whereas distributions of prevalence of Disease according to gender, age, level of education and place of living confirm results of other studies, unexpected prevalence patterns were found for Well-being, especially with regard to gender differences. Multivariate analyses by general linear models (independent sub-samples of the study population were used to develop and validate models) showed different sets of environmental, person-specific and behavioural factors to explain inter-individual differences of Disease and Well-being, both in the total validation sample and in sub-samples of women and men. The results are discussed with regard to implications for socioepidemiological health research.

Oral, pharyngeal and laryngeal cancer as a cause of death among Swiss cooks.

In an analysis of Swiss mortality data (1979-1987) excess mortality due to oral, pharyngeal and laryngeal cancer was found among cooks, and the rate was very high in the age group < 55 years. The peak number of cases was observed among cooks in the age category 45-49 years. In the standard population the highest number of such deaths was observed between 65 and 69 years of age. The numbers of alcohol-related causes of death were also elevated among the cooks, while the numbers of smoking-related deaths were not. Although the dominant role of combined alcohol and tobacco consumption for the development of oral, pharyngeal and laryngeal cancer has been confirmed by many studies, other factors (eg, volatile carcinogenic compounds formed during the cooking process) may contribute to the excess mortality from oral, pharyngeal and laryngeal cancer among cooks. The question of the relevance of such factors will have to be answered by further studies.