Serious errors and omissions occurring in papers submitted to OEM
| Design: |
| Authors unclear of type of epidemiological study |
| Adequacy of sample size not considered |
| Bias in selection of subjects |
| Execution: |
| Data collection problems and missing data not adequately reported |
| Non-respondents not investigated |
| Sample selection and exclusions inadequately justified |
| Analysis: |
| Parametric tests carried out on obviously skewed data |
| Use of multiple paired tests |
| Inappropriate analysis of repeated measures or longitudinal data |
| Incorrect analysis of matched case-control studies |
| Modelling incorrect—for example, inadequate adjustment for confounders, interaction terms not included, only significant variables from preliminary analyses included |
| Presentation: |
| Inadequate description of the methodology and statistical procedures |
| Inappropriate summary statistics for non-normal data |
| No presentation of risk estimates—for example, odds ratios—and confidence intervals |
| Interpretation: |
| Potential bias due to sample selection, no or poor response, missing values, exclusions |
| Lack of statistical power not considered |
| No allowance made for multiple testing |
| Misunderstanding and misinterpretation of results from models |