RESUMEN
Post hoc power estimates are often requested by reviewers and/or performed by researchers after a study has been conducted. The purpose of this commentary is to provide a heuristic explanation of why post hoc power should not be used. To illustrate our point, we provide a detailed simulation study of two essentially identical research experiments hypothetically conducted in parallel at two separate universities. The simulation demonstrates that post hoc power calculations are misleading and simply not informative for data interpretation. As such, we encourage authors and peer-reviewers to avoid using or requesting post hoc power calculations.
Asunto(s)
Modelos Genéticos , Simulación por Computador , HumanosRESUMEN
OBJECTIVE: The aim of this study was to ascertain whether age at surgery has an impact on later neurodevelopmental outcomes for children with sagittal synostosis (SS). METHODS: The developmental outcome data from patients who had surgery for SS and who attended their routine preoperative, 6-7 months postoperative, and 5-year-old developmental assessments (yielding general quotients [GQs]) (n = 50), 10-year-old IQ assessment (n = 54), and 15-year-old IQ assessment (n = 23) were examined, comparing whether they had surgery at < 7 months, 7 to < 12 months, or ≥ 12 months). RESULTS: There was no significant effect for age at surgery for GQ at 5 years of age, but there was a significant effect (p = 0.0001) for those undergoing surgery at < 7 months in terms of preoperative gross locomotor deficit that resolved by 6-7 months postoperatively (increase of 22.1 points), and had further improved by 5 years of age (total increase of 29.4 points). This effect was lessened when surgery was performed later (total increase of 7.3 points when surgery was performed at ≥ 12 months). At 10 years of age, 1-way ANOVA showed a significant difference in Full Scale IQ (FSIQ) score (p = 0.013), with the highest mean FSIQ being obtained when surgery was performed at < 7 months of age (score 107.0), followed by surgery at 7 to < 12 months (score 94.4), and the lowest when surgery was performed at ≥ 12 months (score 93.6). One-way ANOVA for the Performance IQ (PIQ) was very similar (p = 0.012), with PIQ scores of 101.4, 91.4, and 87.3, respectively. One-way ANOVA for Verbal IQ (VIQ) was again significant (p = 0.05), with VIQ scores of 111.3, 98.9, and 100.4, respectively. At 15 years, 1-way ANOVA showed a significant difference in PIQ (p = 0.006), with the highest mean PIQ being obtained when surgery was performed at < 7 months (score 104.8), followed by surgery at 7 to < 12 months (score 90.0), and the lowest when surgery was at performed at ≥ 12 months of age (score 85.3). There were no significant results for FSIQ and VIQ, although there was a similar trend for better outcomes with early surgery. CONCLUSIONS: The findings of this study add to the literature that suggests that early surgery for SS may result in improved neurodevelopmental outcomes, with surgery optimally undertaken when patients are < 7 months of age, and with those undergoing surgery at ≥ 12 months performing the least well. These results also have potential implications for ensuring early diagnosis and referral and for the type of surgery offered. Further research is needed to control for confounding factors and to identify the mechanism by which late surgery may be associated with poorer neurodevelopmental outcomes.