Epidemiology confidence interval

A large number of epidemiology and case-control studies have examined the potential association between oral aluminum exposure and Alzheimer s disease. A number of these studies have been criticized for flawed patient selection, poor comparability of exposed and control groups, poor exposure assessment, poor assessment of health outcomes, and weak statistical correlations (Nieboer et al. 1995 Schupf et al. 1989). Studies conducted by Martyn et al. (1989), McLachlan et al. (1996), and Michel et al. (1990) have found an association between oral exposure to aluminum and an increased risk of Alzheimer s disease. In a survey study conducted by Martyn et al. (1989), the incidence of Alzheimer s disease in individuals under the age of 70 was estimated from computerized tomographic (CT) records. The 1,203 subjects lived in 88 county districts within England and Wales. Data on aluminum concentrations in the municipal water over a 10-year period were obtained from water authorities and water companies. The subjects were classified as having probable Alzheimer s disease, possible Alzheimer s disease, other causes of dementia, or epilepsy. The relative risks of Alzheimer s disease were elevated in the subjects living in districts with aluminum water concentrations of >0.01 mg/L. However, the relative risk exceeded unity only in the subjects with aluminum water concentrations of >0.11 mg/L (relative risk of 1.5, 95% confidence interval of 1.1-2.2). 

Confidence intervals (CIs) are usually the preferred means for evaluating chance in epidemiological studies because they are more informative than P values. CIs are the interval (usually 95%) around the risk estimate (which is a point estimate) that represent the upper and lower values of the true risk estimate. CIs provide information on both the precision of the point estimate and statistical significance. Wide confidence intervals indicate that there is a high degree of uncertainty on the accuracy of the point estimate and is usually a result of small study populations. CIs that exclude 1 means that there is a 95% probability that the null hypothesis is not operating. 

Figure 1. Overview of epidemiological studies investigating the renal risk of analgesic consumption. A. Description of methodological details used in the included studies. B. Presentation of the overall risk (odds ratio with 95% confidence interval) associated to the consumption of any analgesic exceeding the mentioned dose. C. Presentation of the odds ratios with 95% confidence interval published in the included epidemiological studies focussing separately on the ingredients aspirin and paracetamol.

Various claims have been made about possible associations with neurological and/or autoimmune diseases. In a recent epidemiological study of 20 000 people (84% males) with detailed exposure data, a consistent level of amalgam treatment across the cohort, and investigation of a wide range of possible health outcomes, an association was evidenced only with multiple sclerosis, and this association was relatively weak (adjusted hazard ratio of 1.24 95% confidence interval 0.99-1.53) (Bates et al., 2004). Additional epidemiological studies are needed to fully address the question of autoimmune-related health effects of dental amalgams (Weiner et al., 1990 DHHS, 1993 Cuttres, 1997). 

The association between folate intake and colorectal cancer in 27 observational epidemiological studies was analysed in a recent meta-analysis (Keimedy et al. 2011). This meta-analysis, which included 18 case-control studies and nine cohort studies, showed that high folate intake was associated with reduced risk of colorectal adenomas. The association was stronger in case-control studies [relative risk (RR) 0.85 95% confidence interval (Cl) 0.74-0.99] than in cohort studies (RR 0.92 95% Cl 0.81-1.05). 

---