The effect of compliance with antihypertensive medications on the risk of cardiovascular outcomes in a population without a known history of cardiovascular disease has been addressed by a large population-based prospective, cohort study carried out by linking Italian administrative databases. The cohort of 242 594 patients aged 18 years or older, residents in the Italian Lombardy Region, who were newly treated for hypertension during 2000-2001, was followed from index prescription until 2007. During this period patients who experienced a hospitalization for coronary or cerebrovascular disease were identified (outcome). Exposure to antihypertensive drugs from index prescription until the date of hospitalization or censoring was assessed. Proportional hazards models were fitted to assess the association between persistence on and adherence with antihypertensive drug therapy and outcome. Data were adjusted for several covariates. During an average follow-up of 6 years, 12 016 members of the cohort experienced the outcome. Compared with patients who experienced at least one episode of treatment discontinuation, those who continued treatment had a 37% reduced risk of cardiovascular outcomes (95% confidence interval 34-40%). Compared with patients who had very low drug coverage (proportion of days covered ≤ 25%), those at intermediate (from 51 to 75%) and high coverage (>75%) had risk reductions of 20% (16-24%) and 25% (20-29%), respectively. Similar effects were observed when coronary and cerebrovascular events were considered separately. In the real life setting, fulfillment compliance with antihypertensive medications is effective in the primary prevention of cardiovascular outcomes.

Objective.: Oral antidiabetic drugs (including metformin and sulfonylurea) may play a role in the relationship between type 2 diabetes and cancer. To quantify the association between metformin and sulfonylurea and the risk of cancer, we performed a meta-analysis of available studies on the issue. Materials and Methods.: We performed a MEDLINE search for observational studies that investigated the risk of all cancers and specific cancer sites in relation to use of metformin and/or sulfonylurea among patients with type 2 diabetes mellitus. Fixed- and random-effect models were fitted to estimate the summary relative risk (RR). Between-study heterogeneity was tested using χ2 statistics and measured with the I2 statistic. Publication bias was evaluated using funnel plot and Egger's regression asymmetry test. Results.: Seventeen studies satisfying inclusion criteria and including 37,632 cancers were evaluated after reviewing 401 citations. Use of metformin was associated with significantly decreased RR of all cancers (summary RR 0.61, 95% confidence interval [CI] 0.54–0.70), colorectal cancer (0.64, 95% CI 0.54–0.76), and pancreatic cancer (0.38, 95% CI 0.14–0.91). With the exception of colorectal cancer, significant between-study heterogeneity was observed. Evidence of publication bias for metformin-cancer association was also observed. There was no evidence that metformin affects the risk of breast and prostate cancers, nor that sulfonylurea affects the risk of cancer at any site. Conclusions.: Metformin, but not sulfonylurea, appears to reduce subsequent cancer risk. This has relevant implications in light of the exploding global epidemic of diabetes.

In this paper, the authors describe fractional polynomials and cubic splines with which to represent smooth dose-response relations in summarizing meta-analytical aggregate data. Use of these two curve-fitting families can help prevent the problems arising from inappropriate linearity assumptions. These methods are illustrated in the problem of estimating the shape of the dose-response curve between alcohol consumption and all-cause mortality risk. The authors considered aggregate data from 29 cohort studies investigating this issue (1966–2000). J-shaped curves with a nadir at approximately 5–7 g/day of alcohol consumption and a last protective dose of 47–60 g/day were consistently obtained from fractional polynomials and cubic splines. The authors conclude that both of the curve-fitting families are useful tools with which to explore dose-response epidemiologic questions by means of meta-analytical approaches, especially when important nonlinearity is anticipated.