No abstract

Replication is an essential part of the scientific method, yet replication of systematic reviews is too often overlooked, and done unnecessarily or poorly. Excessive replication (doing the same study repeatedly) is unethical and a cause of research wastage. This article provides consensus based guidance on when to replicate and not replicate systematic reviews.

Objective The purpose of this study was to investigate the association between socioeconomic status and orofacial clefts (OFC) in Scotland. Design Study of prevalence at birth over a 10-year period using an area-based measure of material deprivation. Setting Population-based study throughout Scotland. Participants Eight hundred thirty-four live births with OFC born between January 1, 1989, and December 31, 1998, ascertained from the nationwide register of the Cleft Service in Scotland, compared with the total 603,825 live births in Scotland in this period. Results There was a strong positive relationship whereby the prevalence of OFC at birth increased with increasing deprivation. This trend was statistically significant for cleft lip and/or palate (CL[P]: p = .016) but not for cleft palate (CP; p = .078). For each type of cleft, the relative risk among those resident in the most deprived areas, compared with those resident in the least deprived areas, was 2.33. Conclusions The association between OFC and socioeconomic status is consistent with a report for an earlier period in a smaller part of Scotland. Unlike the earlier study, this pattern appears to be stronger for CL(P) than for CP. These observations do not appear to be an artifact of recording. It is possible that they reflect the association between deprivation and risk factors for OFC, most likely tobacco smoking during pregnancy. Because the relationship between OFC and socioeconomic status appears to have been virtually unstudied in other populations, it would be valuable to investigate this relationship elsewhere and determine whether known risk factors account for the relationship.

Objective-To examine the relation between the use of aspirin and non-steroidal anti-inflammatory drugs and the presence of asymptomatic colorectal adenomas.Design-Case-control study of subjects participating in a randomised controlled trial of faecal occult blood screening for colorectal cancer. Data on analgesics and other drugs were obtained from a questionnaire which was mainly concerned with diet and was administered by an interviewer.Setting-Nottingham. Subjects-147 patients with positive results in faecal occult blood tests who were found to have colorectal adenomas (cases), 153 age and sex matched control subjects with negative results in such tests (negative controls), and 176 control subjects with positive results in the tests who were found not to have colorectal adenomas (positive controls).Main outcome measures-Relative risk of developing colorectal adenomas according to frequency and duration ofuse ofanalgesics.Results-Cases reported taking less aspirin and non-steroidal anti-inflammatory drugs than the negative controls, with the estimated relative risk for any use being 0 49 (95% confidence interval 0 3 to 0.8). The inverse association was less strong when cases were compared with the positive controls (0-66 (0 4 to 1.1)). The association was specific for aspirin and non-steroidal anti-inflammatory drugs there being no association with paracetamol or other drugs. Prescribed use of non-steroidal antiinflammatory drugs for longer than five years was associated with the lowest risk (0.21 (0.1 to 0 8)), although the numbers reporting prolonged prescribed use were small.Conclusions-These findings support the hypothesis that aspirin and non-steroidal antiinflammatory drug use protects against the development ofcolorectal neoplasia.

No abstract