Coal home heating and environmental tobacco smoke in relation to lower respiratory illness in Czech children, from birth to 3 years of age

OBJECTIVE: The objective of this study was to evaluate how indoor pollution from tobacco and home heating may adversely affect respiratory health in young children. DESIGN: A birth cohort was followed longitudinally for 3 years to determine incidence of lower respiratory illness (LRI). PARTICIPANTS: A total of 452 children born 1994-1996 in two districts in the Czech Republic participated. Evaluations: Indoor combustion exposures were home heating and cooking fuel, mother's smoking during pregnancy, and other adult smokers in the household. Diagnoses of LRI (primarily acute bronchitis) from birth to 3 years of age were abstracted from pediatric records. Questionnaires completed at delivery and at 3-year follow-up provided covariate information. LRI incidence rates were modeled with generalized linear models adjusting for repeated measures and for numerous potential confounders. RESULTS: LRI diagnoses occurred more frequently in children from homes heated by coal [vs. other energy sources or distant furnaces ; rate ratio (RR) = 1.45 ; 95% confidence interval (CI) , 1.07-1.97]. Maternal prenatal smoking and other adult smokers also increased LRI rates (respectively: RR = 1.48 ; 95% CI, 1.10-2.01 ; and RR = 1.29 ; 95% CI, 1.01-1.65) . Cooking fuels (primarily electricity, natural gas, or propane) were not associated with LRI incidence. For children never breast-fed, coal home heating and mother's smoking conferred substantially greater risks: RR = 2.77 (95% CI, 1.45-5.27) and RR = 2.52 (95% CI, 1.31-4.85) , respectively. CONCLUSIONS: Maternal smoking and coal home heating increased risk for LRI in the first 3 years of life, particularly in children not breast-fed. Relevance: Few studies have described effects of coal heating fuel on children's health in a Western country. Breast-feeding may attenuate adverse effects of prenatal and childhood exposures to combustion products.     

Formaldehyde exposure and lower respiratory infections in infants: findings from the PARIS cohort study

Background: Certain chemical pollutants can exacerbate lower respiratory tract infections (LRIs), a common childhood ailment. Although formaldehyde (FA) is one of the most common air pollutants found in indoor environments, its impact on infant health is uncertain.Objective: Our aim was to determine the impact of FA exposure on the LRI incidence during the first year of life of infants from the Pollution and Asthma Risk: an Infant Study (PARIS) birth cohort.Methods: FA was measured in a random sample of 196 infants’ dwellings, and exposure to this pollutant was estimated for 2,940 infants using predictive models based on measurements and data about potential determinants of FA levels. Health data were collected from parents by regular self-administered questionnaires. We used multivariate logistic regressions to estimate associations between FA exposure and the occurrence of LRI and wheezy LRI (wLRI), adjusting for potential confounders/risk factors.Results: During the first year of life, 45.8% of infants had at least one LRI, and LRI occurred simultaneously with wheezing in 48.7% of cases. The FA predictive models correctly classified 70% of dwellings as having high or low exposure, and we estimated that 43.3% of infants were exposed throughout the first year to levels of FA > 19.5 µg/m³. FA exposure was significantly associated with LRI and wLRI before and after adjustment for known LRI risk factors/confounders. For an interquartile increase in FA levels (12.4 μg/m³), we estimated a 32% [95% confidence interval (CI): 11, 55] and 41% (95% CI: 14, 74) increase in the incidence of LRI and wLRI, respectively.Conclusion: The findings of this study suggest that infants exposed to FA at an early age have an increased incidence of LRI.     

To use or not to use the odds ratio in epidemiologic analyses?

This paper argues that the use of the odds ratio parameter in epidemiology needs to be considered with a view to the specific study design and the types of exposure and disease data at hand. Frequently, the odds ratio measure is being used instead of the risk ratio or the incidence-proportion ratio in cohort studies or as an estimate for the incidence-density ratio in case-referent studies. Therefore, the analyses of epidemiologic data have produced biased estimates and the presentation of results has been misleading. However, the odds ratio can be relinquished as an effect measure for these study designs; and, the application of the case-base sampling approach permits the incidence ratio and difference measures to be estimated without any untenable assumptions. For the Poisson regression, the odds ratio is not a parameter of interest; only the risk or rate ratio and difference are relevant. For the conditional logistic regression in matched case-referent studies, the odds ratio remains useful, but only when it is interpreted as an estimate of the incidence-density ratio. Thus the odds ratio should, in general, give way to the incidence ratio and difference as the measures of choice for exposure effect in epidemiology.     

Distributional interaction: Interpretational problems when using incidence odds ratios to assess interaction

It is well known that the incidence odds ratio approximates the risk ratio when the disease of interest is rare, but increasingly overestimates the risk ratio as the disease becomes more common. However when assessing interaction, incidence odds ratios may not approximate risk ratios even when the disease is rare. We use the term "distributional interaction" to refer to interaction that appears when using incidence odds ratios that does not appear, or appears to a lesser degree, when using risk ratios. The interpretational problems that arise from this discrepancy can have important implications in epidemiologic research. Therefore, quantification of the relationship between the interaction odds ratio and the interaction risk ratio is warranted.In this paper, we provide a formula to quantify the differences between incidence odds ratios and risk ratios when they are used to estimate effect modification on a multiplicative scale. Using this formula, we examine the conditions under which these two estimates diverge. Furthermore, we expand this discussion to the implications of using incidence odds ratios to assess effect modification on an additive scale. Finally, we illustrate how distributional interaction arises and the problems that it causes using an example from the literature.Whenever the risk of the outcome variable is non-negligible, distributional interaction is possible. This is true even when the disease is rare (e.g., disease risk is less than 5%). Therefore, when assessing interaction on either an additive or multiplicative scale, caution should be taken in interpreting interaction estimates based on incidence odds ratios.     

Correction of logistic regression relative risk estimates and confidence intervals for systematic within-person measurement error

Errors in the measurement of exposure that are independent of disease status tend to bias relative risk estimates and other measures of effect in epidemiologic studies toward the null value. Two methods are provided to correct relative risk estimates obtained from logistic regression models for measurement errors in continuous exposures within cohort studies that may be due to either random (unbiased) within-person variation or to systematic errors for individual subjects. These methods require a separate validation study to estimate the regression coefficient lambda relating the surrogate measure to true exposure. In the linear approximation method, the true logistic regression coefficient beta* is estimated by beta/lambda, where beta is the observed logistic regression coefficient based on the surrogate measure. In the likelihood approximation method, a second-order Taylor series expansion is used to approximate the logistic function, enabling closed-form likelihood estimation of beta*. Confidence intervals for the corrected relative risks are provided that include a component representing error in the estimation of lambda. Based on simulation studies, both methods perform well for true odds ratios up to 3.0; for higher odds ratios the likelihood approximation method was superior with respect to both bias and coverage probability. An example is provided based on data from a prospective study of dietary fat intake and risk of breast cancer and a validation study of the questionnaire used to assess dietary fat intake.     

Air pollution and school absenteeism among children in Mexico City.

To determine the acute effects of ozone exposure, the authors conducted a short follow-up study of respiratory illness in a population of 111 preschool children frequently exposed to ozone levels that regularly exceed 0.120 parts per million (ppm). The children attended a private kindergarten in the southwestern part of Mexico City. Parents completed a questionnaire on demographic data, medical history, and potential sources of indoor air pollution. To determine the relation of ozone and respiratory-related school absenteeism, the authors used a logistic regression model for longitudinal data. During the 3-month follow-up, 50% of the children had at least one respiratory-related absenteeism period, and 11.7% had two or more. Children exposed for 2 consecutive days to high ozone levels (> or = 0.13 ppm) had a 20% increment in the risk of respiratory illness. For children exposed for 2 consecutive days to a high ozone level and the previous day to low temperature (< or = 5.1 degrees C), the risk reached 40% (odds ratio = 1.44, 95% confidence interval 1.37-1.52). This study suggests that ozone exposure might be positively associated with the risk of respiratory illness in children and that it may have an interactive effect with low temperature exposure.     

Estimability and estimation in case-referent studies.

The concepts that case-referent studies provide for the estimation of "relative risk" only if the illness is "rare", and that the rates and risks themselves are inestimable, are overly superficial and restrictve. The ratio of incidence densities (forces of morbidity)-and thereby the instantaneous risk-ratio-is estimable without any rarity-assumption. Long-term risk-ratio can be computed through the coupling of case-referent data on exposure rates for various age-categories with estimates, possibly from the study itself, or the corresponding age-specific incidence-densities for the exposed and nonexposed combined-but again, no rarity-assumption is involved. Such data also provide for the assessment of exposure-specific absolute incidence-rates and risks. Point estimation of the various parameters can be based on simple relationships among them, and in interval estimation it is sufficient simply to couple the point estimate with the value of the chi square statistic used in significance testing.     

Detecting patterns of occupational illness clustering with alternating logistic regressions applied to longitudinal data

In longitudinal surveillance studies of occupational illnesses, sickness episodes are recorded for workers over time. Since observations on the same worker are typically more similar than observations from different workers, statistical analysis must take into account the intraworker association due to workers' repeated measures. Additionally, when workers are employed in groups or clusters, observations from workers in the same workplace are typically more similar than observations from workers in different workplaces. For such cluster-correlated longitudinal data, alternating logistic regressions may be used to model the pattern of occupational illness clustering. Data on 182 Latino farm workers from a 1999 North Carolina study on green tobacco sickness provided an estimated pairwise odds ratio for within-worker clustering of 3.15 (95% confidence interval (CI): 1.84, 5.41) and an estimated pairwise odds ratio for within-camp clustering of 1.90 (95% CI: 1.22, 2.97). After adjustment for risk factors, the estimated pairwise odds ratios were 2.13 (95% CI: 1.18, 3.86) and 1.41 (95% CI: 0.89, 2.24), respectively. In this paper, a comparative analysis of alternating logistic regressions with generalized estimating equations and random-effects logistic regression is presented, and the relative strengths of the three methods are discussed.     

Ordinal regression models for epidemiologic data

Health status is often measured in epidemiologic studies on an ordinal scale, but data of this type are generally reduced for analysis to a single dichotomy. Several statistical models have been developed to make full use of information in ordinal response data, but have not been much used in analyzing epidemiologic studies. The authors discuss two of these statistical models--the cumulative odds model and the continuation ratio model. They may be interpreted in terms of odds ratios, can account for confounding variables, have clear and testable assumptions, and have parameters that may be estimated and hypotheses that may be tested using available statistical packages. However, calculations of asymptotic relative efficiency and results of simulations showed that simple logistic regression applied to dichotomized responses can in some realistic situations have more than 75% of the efficiency of ordinal regression models, but only if the ordinal scale is collapsed into a dichotomy close to the optimal point. The application of the proposed models to data from a study of chest x-rays of workers exposed to mineral fibers confirmed that they are easy to use and interpret, but gave results quite similar to those obtained using simple logistic regression after dichotomizing outcome in the conventional way.     

Logistic regression analysis of biomarker data subject to pooling and dichotomization

There is growing interest in pooling specimens across subjects in epidemiologic studies, especially those involving biomarkers. This paper is concerned with regression analysis of epidemiologic data where a binary exposure is subject to pooling and the pooled measurement is dichotomized to indicate either that no subjects in the pool are exposed or that some are exposed, without revealing further information about the exposed subjects in the latter case. The pooling process may be stratified on the disease status (a binary outcome) and possibly other variables but is otherwise assumed random. We propose methods for estimating parameters in a prospective logistic regression model and illustrate these with data from a population-based case-control study of colorectal cancer. Simulation results show that the proposed methods perform reasonably well in realistic settings and that pooling can lead to sizable gains in cost efficiency. We make recommendations with regard to the choice of design for pooled epidemiologic studies.     

Daily air pollution effects on children's respiratory symptoms and peak expiratory flow

To identify acute respiratory health effects associated with air pollution due to coal combustion, a subgroup of elementary school-aged children was selected from a large cross-sectional study and followed daily for eight months. Children were selected to obtain three equal-sized groups: one without respiratory symptoms, one with symptoms of persistent wheeze, and one with cough or phlegm production but without persistent wheeze. Parents completed a daily diary of symptoms from which illness constellations of upper respiratory illness (URI) and lower respiratory illness (LRI) and the symptom of wheeze were derived. Peak expiratory flow rate (PEFR) was measured daily for nine consecutive weeks during the eight-month study period. Maximum hourly concentrations of sulfur dioxide, nitrogen dioxide, ozone, and coefficient of haze for each 24-hour period, as well as minimum hourly temperature, were correlated with daily URI, LRI, wheeze, and PEFR using multiple regression models adjusting for illness occurrence or level of PEFR on the immediately preceding day. Respiratory illness on the preceding day was the most important predictor of current illness. A drop in temperature was associated with increased URI and LRI but not with increased wheeze or with a decrease in level of PEFR. No air pollutant was strongly associated with respiratory illness or with level of PEFR, either in the group of children as a whole, or in either of the symptomatic subgroups; the pollutant concentrations observed, however, were uniformly lower than current ambient air quality standards. Moreover, since exposure estimation based on monitoring of ambient air likely results in misclassification of the true exposure, the negative findings of this study must be interpreted cautiously.     

Contribution of RSV to bronchiolitis and pneumonia-associated hospitalizations in English children,April 1995-March 1998

Estimates of the number of hospitalizations attributable to specific pathogens are required to predict the potential impact of vaccination. All hospital admissions for lower respiratory tract infection (LRI) in children < 5 years in England in 1995-8 were reviewed. Most admissions (76.8%) were not associated with specific organisms. Seasonality in pathogens that cause bronchiolitis and pneumonia was used to predict the proportion of cases with unspecified aetiology attributable to different organisms using multiple linear regression. Of 12,298 admissions for LRI, 17.5% were due to RSV infection. An estimated 74.8% (95% CI, 72.0-77.7%) of 'unspecified bronchiolitis' admissions and 16.3% (95% CI, 13.7-18.8%) of unspecified pneumonia' admissions were RSV related. The total mean annual incidence of hospital admissions attributable to RSV is 28.3/1000 children < 1 year of age, and 1.3/1000 children 1-4 years old. The greater burden of RSV infection than indicated through discharge data is revealed through applying simple statistical methods.     

Lower respiratory tract infections in an ethnic and social context

Family size and smoking during pregnancy were studied as mediating factors for social and ethnic variation of lower respiratory tract infection (LRI) in hospital discharge data. The study population consisted of all children aged 0-4 years in the three largest metropolitan areas of Sweden during 1990-94. Maternal smoking during pregnancy increased the risk of children being admitted to hospital for LRI during their first 3 years of life, with an adjusted odds ratio (OR) of 1.3 for the age-group 0-1 years. The risk attributed to smoking during pregnancy was the same in children of mothers in ethnic groups in which smoking during pregnancy was related to social adversity as in those in which it was not. Having at least one sibling increased the risk of being admitted to hospital for LRI in the age group 0-1 years (adjusted OR 2.2). This risk was lower in children in families in which the mother was born in southern Europe, Africa, Asia or Latin America, suggesting a contextual relation to ethnicity for this risk factor. It is concluded that family size and smoking during pregnancy are important mediators of the risk for LRI related to social adversity and ethnicity in Swedish children below 2 years of age.     

Effect measures in prevalence studies

There is still considerable confusion and debate about the appropriate methods for analyzing prevalence studies, and a number of recent papers have argued that prevalence ratios are the preferred method and that prevalence odds ratios should not be used. These arguments assert that the prevalence ratio is obviously the better measure and the odds ratio is "unintelligible." They have often been accompanied by demonstrations that when a disease is common the prevalence ratio and the prevalence odds ratio may differ substantially. However, this does not tell us which measure is the more valid to use. In fact, the prevalence odds ratio a) estimates the incidence rate ratio with fewer assumptions than are required for the prevalence ratio; b) can be estimated using the same methods as for the odds ratio in case-control studies, namely, the Mantel-Haenszel method and logistic regression; and c) provides practical, analytical, and theoretical consistency between analyses of a prevalence study and prevalence case-control analyses based on the same study population. For these reasons, the prevalence odds ratio will continue to be one of the standard methods for analyzing prevalence studies and prevalence case-control studies.     

An approach for developing a national estimate of waterborne disease due to drinking water and a national estimate model application

In this paper, the US Environmental Protection Agency (EPA) presents an approach and a national estimate of drinking water related endemic acute gastrointestinal illness (AGI) that uses information from epidemiologic studies. There have been a limited number of epidemiologic studies that have measured waterborne disease occurrence in the United States. For this analysis, we assume that certain unknown incidence of AGI in each public drinking water system is due to drinking water and that a statistical distribution of the different incidence rates for the population served by each system can be estimated to inform a mean national estimate of AGI illness due to drinking water. Data from public water systems suggest that the incidence rate of AGI due to drinking water may vary by several orders of magnitude. In addition, data from epidemiologic studies show AGI incidence due to drinking water ranging from essentially none (or less than the study detection level) to a rate of 0.26 cases per person-year. Considering these two perspectives collectively, and associated uncertainties, EPA has developed an analytical approach and model for generating a national estimate of annual AGI illness due to drinking water. EPA developed a national estimate of waterborne disease to address, in part, the 1996 Safe Drinking Water Act Amendments. The national estimate uses best available science, but also recognizes gaps in the data to support some of the model assumptions and uncertainties in the estimate. Based on the model presented, EPA estimates a mean incidence of AGI attributable to drinking water of 0.06 cases per year (with a 95% credible interval of 0.02-0.12). The mean estimate represents approximately 8.5% of cases of AGI illness due to all causes among the population served by community water systems. The estimated incidence translates to 16.4 million cases/year among the same population. The estimate illustrates the potential usefulness and challenges of the approach, and provides a focus for discussions of data needs and future study designs. Areas of major uncertainty that currently limit the usefulness of the approach are discussed in the context of the estimate analysis.     

Respiratory syncytial virus infection: denominator-based studies in Indonesia, Mozambique, Nigeria and South Africa

OBJECTIVE: To assess the burden of respiratory syncytial virus (RSV)-associated lower respiratory infections (LRI) in children in four developing countries. METHODS: A WHO protocol for prospective population-based surveillance of acute respiratory infections in children aged less than 5 years was used at sites in Indonesia, Mozambique, Nigeria and South Africa. RSV antigen was identified by enzyme-linked immunosorbent assay performed on nasopharyngeal specimens from children meeting clinical case definitions. FINDINGS: Among children aged < 5 years, the incidence of RSV-associated LRI per 1000 child-years was 34 in Indonesia and 94 in Nigeria. The incidence of RSV-associated severe LRI per 1000 child-years was 5 in Mozambique, 10 in Indonesia, and 9 in South Africa. At all study sites, the majority of RSV cases occurred in infants. CONCLUSION: These studies demonstrate that RSV contributes to a substantial but quite variable burden of LRI in children aged < 5 years in four developing countries. The possible explanations for this variation include social factors, such as family size and patterns of seeking health care; the proportion of children infected by human immunodeficiency syndrome (HIV); and differences in clinical definitions used for obtaining samples. The age distribution of cases indicates the need for an RSV vaccine that can protect children early in life.     

Lymphatic and hematopoietic cancer in teachers

A recent study found high rates of leukemia and related disorders among teachers. This finding may be related to exposure to childhood infections. Therefore, epidemiologic studies on the risk of lymphatic and hematopoietic cancer among teachers were systematically reviewed. Altogether 26 relevant investigations were identified, most from ad hoc publications rather than from scientific journals. Elevated risks of leukemia, lymphoma, and multiple myeloma were found in studies using proportional mortality or mortality odds ratios as outcome measures. However, these observations may reflect low overall mortality and do not necessarily indicate high death rates from the cancers of interest. In studies deriving standardized mortality or incidence ratios, the risk estimates were generally lower. The most striking finding was for non-Hodgkin's lymphoma (approximate summary relative risk 1.36, 95% confidence interval 1.13-1.62), but it was likely to have been exaggerated by publication bias. In conclusion, no compelling epidemiologic evidence exists for a hazard of leukemia or related diseases among teachers.     

The Association Between Short Interpregnancy Interval and Preterm Birth in Louisiana: A Comparison of Methods

There is growing interest in the application of propensity scores (PS) in epidemiologic studies, especially within the field of reproductive epidemiology. This retrospective cohort study assesses the impact of a short interpregnancy interval (IPI) on preterm birth and compares the results of the conventional logistic regression analysis with analyses utilizing a PS. The study included 96,378 singleton infants from Louisiana birth certificate data (1995–2007). Five regression models designed for methods comparison are presented. Ten percent (10.17 %) of all births were preterm; 26.83 % of births were from a short IPI. The PS-adjusted model produced a more conservative estimate of the exposure variable compared to the conventional logistic regression method (β-coefficient: 0.21 vs. 0.43), as well as a smaller standard error (0.024 vs. 0.028), odds ratio and 95 % confidence intervals [1.15 (1.09, 1.20) vs. 1.23 (1.17, 1.30)]. The inclusion of more covariate and interaction terms in the PS did not change the estimates of the exposure variable. This analysis indicates that PS-adjusted regression may be appropriate for validation of conventional methods in a large dataset with a fairly common outcome. PS’s may be beneficial in producing more precise estimates, especially for models with many confounders and effect modifiers and where conventional adjustment with logistic regression is unsatisfactory. Short intervals between pregnancies are associated with preterm birth in this population, according to either technique. Birth spacing is an issue that women have some control over. Educational interventions, including birth control, should be applied during prenatal visits and following delivery.     

Number needed to immunize to prevent RSV with extended half-life monoclonal antibody

BackgroundRespiratory syncytial virus (RSV) is one of the most important respiratory pathogens in young children. Infants <6 months of age and infants and young children with extreme pre-term birth, and cardiac and pulmonary co-morbidities experience the highest incidence of severe RSV disease. There are no licensed vaccines; immunoprophylaxis is recommended for the highest risk children. Extended half-life RSV monoclonal antibodies (EHL-mAbs) are under development intended for immunization of all infants and high-risk children <2 years of age. We modeled the anticipated public health benefits of RSV EHL-mAb immunization using the number needed to immunize (NNI).MethodsWe combined RSV hospitalization, outpatient and outpatient lower respiratory tract infection (LRI) incidence estimates and a range of immunization efficacies to estimate the annual NNI. We calculated the absolute incidence rate reduction (ARR) by multiplying the incidence rates by immunization efficacy. NNI was calculated as the reciprocal of the ARR.ResultsFor an RSV EHL-mAb with 70% efficacy, 6–18 infants would need to be immunized to prevent one RSV-associated outpatient visit, and 13–33 infants would need to be immunized to prevent one RSV-associated LRI outpatient visit. To prevent one RSV-associated hospitalization, 37–85 infants 0–5 months of age, and 107–280 infants 6–11 months of age would need to be immunized.ConclusionsPublic health benefits, such as disease cases averted due to immunization, are essential elements in consideration of candidate vaccines for a national immunization program. An RSV EHL-mAb of moderate efficacy could have high impact. These data provide an additional perspective for public health decision making.