{"id":65,"date":"2021-05-22T00:32:47","date_gmt":"2021-05-22T04:32:47","guid":{"rendered":"https:\/\/pressbooks.bccampus.ca\/rickyturgeon\/?post_type=chapter&p=65"},"modified":"2022-01-12T23:06:10","modified_gmt":"2022-01-13T04:06:10","slug":"subgroups-were-additional-comparisons-made-on-segments-of-the-study-population","status":"publish","type":"chapter","link":"https:\/\/pressbooks.bccampus.ca\/rickyturgeon\/chapter\/subgroups-were-additional-comparisons-made-on-segments-of-the-study-population\/","title":{"raw":"Subgroups analysis: Were additional comparisons made on segments of the study population?","rendered":"Subgroups analysis: Were additional comparisons made on segments of the study population?"},"content":{"raw":"Most [pb_glossary id=\"704\"]RCT[\/pb_glossary] publications report on additional analyses of subgroups from the overall trial population (e.g. examining only participants with diabetes, or only those age >70 years). In theory, such analyses could uncover more individualized treatment effects; however, in practice they are much more likely to be spurious and irreproducible, and therefore misleading. Subgroup analyses are often performed (and emphasized in publications) when studies do not find a statistically significant difference in the overall study population. Therefore, subgroup analysis is often a form of data-mining.\r\n

Checklist Questions<\/h1>\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
Are statistically significant results in a subgroup being emphasized in the context of a neutral or negative trial?<\/td>\r\n<\/tr>\r\n
Was the subgroup analysis pre-defined?<\/td>\r\n<\/tr>\r\n
Was the direction of the subgroup effect pre-defined?<\/td>\r\n<\/tr>\r\n
Was the subgroup analysis one of a small number of hypotheses tested?<\/td>\r\n<\/tr>\r\n
Is the subgroup variable a characteristic measured at baseline or after randomization?<\/td>\r\n<\/tr>\r\n
Could treatment effect differences between subgroups be attributable to baseline imbalances?<\/td>\r\n<\/tr>\r\n
Is the subgroup effect statistically significant?<\/td>\r\n<\/tr>\r\n
Is the subgroup effect consistent within and across trials?<\/td>\r\n<\/tr>\r\n
[[pb_glossary id=\"1099\"]Systematic reviews[\/pb_glossary]\/[pb_glossary id=\"1101\"]meta-analyses[\/pb_glossary] only] Is the effect suggested by comparisons within rather than between studies?<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n

Are statistically significant results in a subgroup being emphasized in the context of a neutral or negative trial?<\/h1>\r\nSubgroup analyses can be used for <\/span>data-mining<\/span> when overall results are not statistically significantly different <\/span><\/span>and may be highlighted when the [pb_glossary id=\"1517\"]primary endpoint[\/pb_glossary] fails to cross the threshold of statistical significance.<\/span>\r\n

Was the subgroup analysis pre-defined?<\/h1>\r\nSubgroup analyses that were not pre-defined in the protocol may be a form of data-mining, and are vulnerable to finding a difference by chance. Avoid making clinical decisions based on unanticipated significant subgroup differences (i.e. discovered post hoc) until they have been replicated in other studies.\r\n\r\nThis is particular concerning for continuous variables, such as age or cholesterol level, that are dichotomized via non-prespecified cutoffs (Schandelmaier S et al.<\/a>). For example, LDL cholesterol could be dichotomized via numerous arbitrary cutoffs (e.g. >3.5, >4.0...). If not pre-specified, such cutoffs could be selected by whichever value showed the most impressive or statistically significant result. Such data-mining efforts are unlikely to uncover true subgroup differences.\r\n

Was the direction of the subgroup effect correctly pre-defined?<\/h1>\r\nSubgroup effects that are significant but go in the direction opposite to what was hypothesized are less credible than correct predictions.<\/span><\/span>\u00a0<\/span>\r\n

Was the subgroup analysis one of a small number of hypotheses tested?<\/h1>\r\nMore comparisons increase the likelihood of finding a difference by chance.<\/span><\/span> See the multiplicity discussion here<\/a> for more information.<\/span>\r\n

Is the subgroup variable a characteristic measured at baseline or after randomization?<\/h1>\r\nSubgroup analyses of variables measured after randomization may be affected by the interventions, thereby introducing confounding<\/a>.\r\n
\r\n\r\nExamples of variables measured at baseline: <\/em>\r\n