Reporting odds ratio

The most solid demonstration of the association between analgesic abuse and renal failure has been provided by the two prospective controlled cohort studies performed in Switzerland and Belgium with a follow-up of 10 and 6 years, respectively [39, 40]. Although both studies differed substantially with respect to study populations, analgesics consumed and length of follow-up, reported odds ratio s were... 

The PRR is just one of several measures of disproportionality that have been used. A reporting odds ratio (ROR) can be calculated from the same 2X2 table and has been mostly used in the Netherlands. The WHO uses the Information Component (IC) and the US Food and Drugs Administration the Multi-Item Gamma Poisson Shrinker (MGPS), both of which are more complex measures based on Bayesian statistics. These measures tend to produce less extreme values than PRRs when the number of cases is very small. However, when the sensitivity, specificity and predictive power of these measures were compared using Dutch data in 2002, no important differences were found provided at least 3 cases had been reported. 

In a case-control study of the relation between occupational exposures to various suspected estrogenic chemicals and the occurrence of breast cancer, the breast cancer odds ratio (OR) was not elevated above unity (OR=0.8 95% 01=0.2-3.2) for occupational exposure to endosulfan compared to unexposed controls (Aschengrau et al. 1998) however, the sample sizes were very small (three exposed seven not exposed), and co-exposure to other unreported chemicals also reportedly occurred. Both of these factors may have contributed to the high degree of uncertainty in the OR indicated by the wide confidence interval. 

One meta-analysis examined the safety and efficacy of LMWH and heparinoids in 11 randomized trials of 3048 patients with acute ischemic stroke. It reported a reduction in the incidence of deep venous thrombosis (DVT) (odds ratio (OR) 0.27,... 

In a recent retrospective study of spontaneous abortions in a large population of women working in the petrochemical industry in China, Xu et al. (1998) reported a significantly increased risk of spontaneous abortion with frequent exposure to petrochemicals (odds ratio of 2.7 95% Cl 1.8-3.9). When the risk associated with exposure to specific chemicals was examined, exposure to hydrogen sulfide was found to have an odds ratio of 2.3 (95% Cl = 1.2-4.4). 

Ioannidis et al. recently showed that after initial reports of associations between polymorphic variants and disease, subsequent studies had lower odds ratios, which were often insignificant, even when meta-analysis... 

In humans, cases of hepatomas have appeared years after acute exposure to carbon tetrachloride, however, none of the cases could establish a causal link between the exposure and development of neoplasms." Epidemiological studies have also given inconclusive results. A cancer mortality study of a population of rubber workers reported a significantly elevated odds ratio relating carbon tetrachloride with lymphatic leukemia, and lymphosarcoma and reticulum cell carcinoma." A recent retrospective cohort mortality study of aircraft maintenance workers found an increased risk of... 

A nested case-control study within a cohort of rubber workers in the United States was performed to examine the relationship between exposure to solvents and the risk of cancer (Checkoway et al., 1984 Wilcosky et al., 1984). The cohort consisted of 6678 male rubber workers who either were active or retired between 1964 and 1973. The cases comprised all persons with fatal stomach cancer (n = 30), respiratory system cancer (z7 = 101), prostate cancer ( = 33), lymphosarcoma (n = 9) or lymphocytic leukaemia (z7 = 10). These sites were chosen because they were those at which cancers had been found to be in excess in an earlier cohort analysis (McMichael et al., 1976). The controls were a 20% age-stratified random sample of the cohort (z = 1350). Exposure was classified from a detailed work history and production records. An association was observed between exposure for one year or more to carbon tetrachloride and lymphocytic leukaemia (odds ratio (OR), 15.3 / < 0.0001, based on eight exposed cases) and lymphosarcoma (OR, 4.2 p < 0.05, based on six exposed cases) after adjusting for year of birth. The relative risk associated with 24 solvents was examined and levels of exposure were not reported. [The Working Group noted that overlapping exposures limit the ability to draw conclusions regarding carbon tetrachloride.]... 

Bond et al. (1986) conducted a nested case-control study of lung cancer among a cohort of 19 608 male chemical workers in the United States (Bond et al., 1985). Further details of the study are reported in Section 2.2 of the monograph on carbon tetrachloride in this volume. Ever having been exposed to epichlorohydrin was associated with a decreased risk of lung cancer (odds ratio, 0.3 95% CI, 0.1-0.9 5 exposed cases). 

In Siemiatycki s (1991) population-based case-control study of cancer in Montreal, Canada (see monograph on dichloromethane in this volume), phenol was one of the substances evaluated 1% of the entire study population had been exposed to it at some time. Among the main occupations to which phenol exposure was attributed in this study were electric motor repairmen and foundry workers. The publication reported an association between phenol and pancreatic cancer (odds ratio, 4.8 90% CI, 1.8-12.7 n = 4) for no other site was cancer risk associated with phenol exposure. [The Working Group noted that detailed results for other sites were not provided, because they were based on small numbers, and that workers typically had multiple exposures.]... 

As described in an earlier monograph (lARC, 1986), a case-control study in Sweden found a significant association between nasal and nasopharyngeal cancer and exposure to chlorophenols, independent of exposure to wood dust (Hardell et al., 1982). The same group of researchers also reported positive associations with high-grade exposure to chlorophenols in case-control studies of colon cancer (odds ratio, 1.8 95% CI, 0.6-5.3) and primary liver cancer (odds ratio, 2.2 95% CI, 0.7-7.3) (Hardell, 1981 Hardell et al., 1984). 

In a case-control study in the north of Sweden, Hallquist et al. (1993) compared 188 men and women aged 20-70 years who had thyroid cancer with age- and sex-matched controls (two per case) selected from a register of the local population. The cases were identified retrospectively from a cancer registry and excluded a proportion of patients (19%) who had died by the time of the study. Exposure to potential risk factors, including chlorophenols, was ascertained by postal questionnaire with a supplementary telephone interview if answers were incomplete. The response rates for the cases and controls were 95% and 90%, respectively. Of the 171 cases analysed, 107 had papillary tumours. Four cases and three controls reported exposure to chlorophenols (odds ratio, 2.8 95% CI, 0.5-18). [The Working Group noted that the method of statistical analysis was not the most appropriate for individually matched data, but this is unlikely to have produced serious bias.]... 

Tea is another important dietary source for flavonoids, In fact, about half of the flavonoid intake in western populations is derived from black tea. Tea was the major source of flavonoids in the Dutch [6,13] and Welsh studies [17]. Only a small number of studies investigated the association between tea consumption and cardiovascular disease risk. No association between tea consumption and cardiovascular disease risk were reported in Scottish men and women [28] and in U.S. men in the Health Professionals follow-up study [29]. However, in a Norwegian population an inverse association was reported between tea intake, serum cholesterol, and mortality from coronary heart disease [30]. Several studies reported that tea consumption did not affect plasma antioxidant activity [31] and hemostatic factors [32]. However, a recent prospective study (the Rotterdam study) of 3,454 men and women 55 years and older followed for 2 to 3 years, showed a significant, inverse association of tea intake with severe (> 5 cm the length of the calcified area) aortic atherosclerosis. Odds ratios decreased approximately 70 % for drinking more than 500 mL/day (4 cups per day). The associations were stronger in women than in men. However, the risk reductions for moderate and mild atherosclerosis were only weak or absent [33]. 

---