Mary Lou Thompson

Prevalence odds ratio or prevalence ratio in the analysis of cross sectional data : what is to be done?

OBJECTIVES: To review the appropriateness of the prevalence odds ratio (POR) and the prevalence ratio (PR) as effect measures in the analysis of cross sectional data and to evaluate different models for the multivariate estimation of the PR. METHODS: A system of linear differential equations corresponding to a dynamic model of a cohort with a chronic disease was developed. At any point in time, a cross sectional analysis of the people then in the cohort provided a prevalence based measure of the effect of exposure on disease. This formed the basis for exploring the relations between the POR, the PR, and the incidence rate ratio (IRR). Examples illustrate relations for various IRRs, prevalences, and differential exodus rates. Multivariate point and interval estimation of the PR by logistic regression is illustrated and compared with the results from proportional hazards regression (PH) and generalised linear modelling (GLM). RESULTS: The POR is difficult to interpret without making restrictive assumptions and the POR and PR may lead to different conclusions with regard to confounding and effect modification. The PR is always conservative relative to the IRR and, if PR > 1, the POR is always > PR. In a fixed cohort and with an adverse exposure, the POR is always > or = IRR, but in a dynamic cohort with sufficient underlying follow up the POR may overestimate or underestimate the IRR, depending on the duration of follow up. Logistic regression models provide point and interval estimates of the PR (and POR) but may be intractable in the presence of many covariates. Proportional hazards and generalised linear models provide statistical methods directed specifically at the PR, but the interval estimation in the case of PH is conservative and the GLM procedure may require constrained estimation. CONCLUSIONS: The PR is conservative, consistent, and interpretable relative to the IRR and should be used in preference to the POR. Multivariate estimation of the PR should be executed by means of generalised linear models or, conservatively, by proportional hazards regression.

Recreational Physical Activity During Pregnancy and Risk of Preeclampsia

Abstract— The potential benefits and risks of physical activity before and during pregnancy are not well studied. We studied the relation between recreational physical activity and the risk of preeclampsia in a case‐control study of 201 preeclamptic and 383 normotensive pregnant women. Participants provided information about the type, intensity, frequency, and duration of physical activity performed during the first 20 weeks of pregnancy and during the year before pregnancy. Women who engaged in any regular physical activity during early pregnancy, compared with inactive women, experienced a 35% reduced risk of preeclampsia (odds ratio, 0.65; 95% confidence interval [CI], 0.43 to 0.99). Compared with inactive women, those engaged in light or moderate activities (ie, activities with metabolic‐equivalent scores <6) experienced a 24% reduced risk of preeclampsia (95% CI, 0.48 to 1.20). The corresponding reduction for women participating in vigorous activities (metabolic equivalent scores >=6) was 54% (95% CI, 0.27 to 0.79). Brisk walking (average walking pace >=3 mi/h), when compared with no walking at all, was associated with a 30% to 33% reduction in preeclampsia risk. Stair climbing was inversely associated with the risk of preeclampsia (P for trend=0.039). Recreational physical activity performed during the year before pregnancy was associated with similar reductions in preeclampsia risk. These data suggest that regular physical activity, particularly when performed during the year before pregnancy and during early pregnancy, is associated with a reduced risk of preeclampsia.

A review of statistical methods for the meteorological adjustment of tropospheric ozone

Abstract A variety of statistical methods for meteorological adjustment of ozone have been proposed in the literature over the last decade for purposes of forecasting, estimating ozone time trends, or investigating underlying mechanisms from an empirical perspective. The methods can be broadly classified into regression, extreme value, and space–time methods. We present a critical review of these methods, beginning with a summary of what meteorological and ozone monitoring data have been considered and how they have been used for statistical analysis. We give particular attention to the question of trend estimation, and compare selected methods in an application to ozone time series from the Chicago area. We conclude that a number of approaches make useful contributions to the field, but that no one method is most appropriate for all purposes and all meteorological scenarios. Methodological issues such as the need for regional-scale analysis, the nonlinear dependence of ozone on meteorology, and extreme value analysis for trends are addressed. A comprehensive and reliable methodology for space–time extreme value analysis is attractive but lacking.

Effect of Circumcision on Incidence of Human Immunodeficiency Virus Type 1 and Other Sexually Transmitted Diseases: A Prospective Cohort Study of Trucking Company Employees in Kenya

To determine the effect of circumcision status on acquisition of human immunodeficiency virus (HIV) type 1 and other sexually transmitted diseases, a prospective cohort study of 746 HIV-1-seronegative trucking company employees was conducted in Mombasa, Kenya. During the course of follow-up, 43 men acquired HIV-1 antibodies, yielding an annual incidence of 3.0%. The annual incidences of genital ulcers and urethritis were 4.2% and 15.5%, respectively. In multivariate analysis, after controlling for demographic and behavioral variables, uncircumcised status was an independent risk factor for HIV-1 infection (hazard rate ratio [HRR=4.0; 95% confidence interval [CI], 1.9-8.3) and genital ulcer disease (HRR=2.5; 95% CI, 1.1-5.3). Circumcision status had no effect on the acquisition of urethral infections and genital warts. In this prospective cohort of trucking company employees, uncircumcised status was associated with increased risk of HIV-1 infection and genital ulcer disease, and these effects remained after controlling for potential confounders.

Cofactors for the acquisition of HIV-1 among heterosexual men: prospective cohort study of trucking company workers in Kenya.

OBJECTIVE To determine the prevalence, incidence, and correlates of HIV-1 infection in a cohort of east African trucking company employees. METHODS HIV-1-seronegative trucking company employees were enrolled in a prospective cohort study and evaluated at 3 monthly intervals for HIV-1 seroconversion, sexually transmitted diseases, and sexual behavior. RESULTS The baseline seroprevalence of HIV-1 among 1500 trucking company employees was 17.8%. Among 752 HIV-1-seronegative men who were followed, the HIV-1 annual seroincidence was 3.1%. In univariate analysis, HIV-1 acquisition was associated with age under 25 years, 10 years or less of sexual activity, occupation as a driver/drivers assistant, occupational travel for more than 14 days per month, religion other than Christian or Muslim, uncircumcised status, sex with a prostitute, sex with a girlfriend/casual partner, extramarital sex, and enrollment seropositivity to Treponema pallidum, Haemophilus ducreyi, and Herpes simplex virus type 2 (all P values < or = 0.05). Using multivariate analysis, HIV-1 acquisition was independently associated with 10 years or less of sexual activity (hazard rate ratio (HRR) 2.0, 95% confidence interval (CI) 1.0-4.3), occupation as a driver/drivers assistant (HRR 3.9, 95% CI 1.7-9.0), religion other than Christian or Muslim (HRR 6.1, 95% CI 1.4-25.7), uncircumcised status (HRR 2.3, 95% CI 1.0-5.0), and unprotected sex with a prostitute (HRR 2.8, 95% CI 1.1-7.0). CONCLUSIONS Trucking company employees had a high HIV-1 seroprevalence rate at enrollment and a high HIV-1 seroincidence during follow-up. Risk factors for HIV-1 seroconversion included years of sexual activity, occupation, religion, uncircumcised status, and unprotected sex with a prostitute. This population is an appropriate target for HIV-1 prevention trials and behavioral interventions.

Data reduction using a discrete wavelet transform in discriminant analysis of very high dimensionality data.

We present a method of data reduction using a wavelet transform in discriminant analysis when the number of variables is much greater than the number of observations. The method is illustrated with a prostate cancer study, where the sample size is 248, and the number of variables is 48,538 (generated using the ProteinChip technology). Using a discrete wavelet transform, the 48,538 data points are represented by 1271 wavelet coefficients. Information criteria identified 11 of the 1271 wavelet coefficients with the highest discriminatory power. The linear classifier with the 11 wavelet coefficients detected prostate cancer in a separate test set with a sensitivity of 97% and specificity of 100%.

Primary Human Immunodeficiency Virus Type 1 Infection: Clinical Manifestations among Women in Mombasa, Kenya

The occurrence of clinical manifestations associated with primary human immunodeficiency virus type 1 (HIV-1) infection was evaluated in a prospective cohort study of female sex workers in Mombasa, Kenya. Among 103 women who seroconverted to HIV-1, fever, vomiting, diarrhea, headache, arthralgia, myalgia, skin rash, swollen lymph nodes, extrainguinal lymphadenopathy, inguinal lymphadenopathy, and vaginal candidiasis were noted significantly more frequently at visits in which seroconversion first became evident. Eighty-one percent of seroconverting women had >/=1 of these 11 symptoms or signs. Among 44% of the women, the acute illness was severe enough to prevent them from working. Having >/=2 of 6 selected symptoms and signs yielded a sensitivity of 51%, specificity of 83%, positive likelihood ratio of 3.2, and negative likelihood ratio of 0.5 for acute HIV-1 infection. The recognition of primary HIV-1-infection illness in high-risk populations and subsequent risk-reduction counseling could potentially reduce secondary HIV-1 transmission during this highly infectious period.

Tuberculosis and silica exposure in South African gold miners

Aims: To examine the effect of silica exposure, in the absence of silicosis, on the prevalence of pulmonary tuberculosis (PTB), which is epidemic among South African gold miners. Methods: Cross-sectional study of 520 gold miners over 37 years of age. Length of service, and cumulative and average dust and quartz exposure indices were derived for each miner. Chest radiographs were read for PTB by two NIOSH “B” readers. PTB was defined as a self-reported history of PTB or PTB on chest radiograph. Logistic regression was used to adjust for age, smoking, and silicosis. PTB effects of different exposure metrics for silica, scaled on their interquartile range (IQR), were compared. Results: Means (ranges) were: age 46.7 (37.1–59.9) years; length of service 21.8 (6.3–34.5) years; average intensity of respirable quartz 0.053 (0–0.095) mg/m3. PTB prevalence was 19.4% (95% CI 16.0 to 22.8) on history alone, and 35.2% (95% CI 31.1 to 39.3) on history or on chest radiograph. Length of service was poorly predictive of PTB, while all exposure indices which included dust or quartz yielded prevalence odds ratios (PORs) of approximately 1.4 (95% CI ∼1.1 to 1.8) for changes of one interquartile range in exposure. Controlling for silicosis—by adjustment or restriction—did not modify these results. Drillers and winch operators had the highest PTB prevalences and the highest dust and silica exposures. Conclusion: Older in-service gold miners in South Africa have a high prevalence of PTB, which is significantly associated with dust and silica exposure, even in the absence of silicosis. Limitations include a survivor workforce and the use of cumulative exposures based on current exposures. Dust control is an important component in control of the PTB epidemic in South African gold mines.

Vitamin C and the risk of preeclampsia-results from dietary questionnaire and plasma assay

Background. Oxidative stress plays an important role in the pathophysiology of preeclampsia. Methods. In a case-control study of 109 women with preeclampsia and 259 controls, maternal dietary and plasma vitamin C in relation to preeclampsia risk were assessed. Dietary intake during the periconceptional period and pregnancy was ascertained using a semiquantitative food frequency questionnaire. Logistic regression procedures were used to derive odds ratios (OR) and 95% confidence intervals (CI). Plasma ascorbic acid was determined using automated enzymatic procedures. Results. After adjusting for maternal age, parity, prepregnancy body mass index, and energy intake, women who consumed <85 mg of vitamin C daily (below the recommended dietary allowance), as compared with others, experienced a doubling in preeclampsia risk (OR = 2.1; 95% CI = 1.1–3.9). The OR for extreme quartiles of plasma ascorbic acid (<42.5 vs ≥ 63.3 &mgr;mol/liter) was 2.3 (95% CI = 1.1–4.6). Compared with women in the highest quartile, those with plasma ascorbic acid <34.6 &mgr;mol/liter (lowest decile) experienced a 3.8-fold increased risk of preeclampsia (95% CI = 1.7–8.8). Conclusions. Our results, if confirmed, would suggest that current public health efforts to increase intake of fruits and vegetables rich in vitamin C and other antioxidants may reduce the risk of preeclampsia.

Nervous system effects of occupational manganese exposure on South African manganese mineworkers.

Occupational exposure to airborne manganese dust has been shown to produce adverse effects on the central nervous system. Four hundred and eighty-nine blue and white collar manganese mineworkers from South Africa were studied cross-sectionally to investigate the nervous system effects of medium to low occupational manganese exposures. The different facilities included underground mines, surface processing plants, and office locations. A job exposure matrix was constructed using routine occupational hygiene data. Exposure variables included years of service, a cumulative exposure index (CEI) and average intensity of exposure (AINT) across all jobs, and blood manganese. Endpoints included items from the Q16, WHO-NCTB, SPES, and Luria-Nebraska test batteries, and a brief clinical examination. Potential confounders and effect modifiers included age, level of education, past medical history including previous head injury, previous neurotoxic job exposures, tobacco use, alcohol use and home language. Associations were evaluated by multiple linear and logistic regression modeling. Average exposure intensity across all jobs was 0.21mg/m(3) manganese dust. Multivariate analyses showed that none of the symptom nor test results were associated with any measure of exposure including blood manganese, after adjustment for confounders. This relatively large null study indicates that manganese miners exposed on average across all jobs to MnO(2) at levels near the American Conference of Governmental Industrial Hygienists Threshold Limit Value (ACGIH TLV) are unlikely to have a subclinical neurotoxicity problem.