China 
Africa 
Explore chapters and articles related to this topic
Introduction
Published in Shein-Chung Chow, Innovative Statistics in Regulatory Science, 2019
When performing a hypotheses testing, basically two kinds of errors (i.e., type I error and type III error) occur. Table 1.3 summarizes the relationship between type I and type II errors when testing hypotheses.
Statistical Analysis
Published in Abhaya Indrayan, Research Methods for Medical Graduates, 2019
Compare this with statistical errors. There are two types: (1) errors that are known and acknowledged, such as type I and type II errors; and (2) errors that occur due to lack of expertise and carelessness. Not many realize that a genuine type I error means an ineffective regimen is proclaimed effective and many deaths can occur due to this error. Similarly, a type II error means an effective treatment is denied to patients, and this also can cause deaths as lives that could be saved are not.
Bayesian Multi-Stage Designs for Phase II Clinical Trials
Published in Harry Yang, Steven J. Novick, Bayesian Analysis with R for Drug Development, 2019
Multi-stage designs have long been used for Phase II efficacy studies to increase trial efficiency. Early two-stage designs include the work by Gehan (1961) and Simon (1989). Both designs render the sponsor an early termination of the study due to futility. Simon’s optimal design was constructed to minimize the expected sample size under the null hypothesis. An alternate design called minimax design is also considered by Simon (1989). It is aimed at minimizing the maximum sample size. In either case, Type I and Type II errors are controlled. Using different optimization criteria, many other multi-stage designs have been proposed. For example, Fleming (1982) proposed a one-sample multiple testing procedure which allows for early termination for futility or efficacy while preserving the size, power, and simplicity of the single-stage procedure; Bryant and Day (1995) suggested a two-stage design that simultaneously assesses both clinical response and toxicity; Jung et al. (2004) developed a family of two-stage designs that are admissible according to a Bayesian decision-theoretic criterion based on an ethically justifiable loss function. Several group sequential designs were suggested by Pocock (1977, 1982), and O’Brian and Fleming (1979) to allow for repeated testing. Wang and Tsiatis (1987) and DeMets and Lan (1994) extended the early work by introducing the concept of alpha spending, which rendered greater flexibility in conducting interim analyses.
Factors influencing decisions about neurogenic bladder and bowel surgeries among veterans and civilians with spinal cord injury
Published in The Journal of Spinal Cord Medicine, 2023
Denise G. Tate, Edward J. Rohn, Martin Forchheimer, Suzanne Walsh, Lisa DiPonio, Gianna M. Rodriguez, Anne P. Cameron
The narratives of the DM process for surgery following NBB dysfunction in this sample may not be representative of veterans and civilians with SCI more broadly, due to the study’s small sample size and the fact that most study participants were from one geographic area. These DM processes may be particular to the type of surgeries described here in terms of common practices at healthcare facilities, socioeconomic structures, or geography. There is also the possibility of bias in the selection of quotes or categorizations of DM enactment styles based on the narrative provided. Individual interviews can only present part of a person’s larger life narrative. Also, quantitative comparisons were based on a relatively small sample size and need to be interpreted with caution. These analyses were conducted to highlight trends in the data rather than to draw hypotheses-driven conclusions. Since these were exploratory in nature, correction for the conduct of multiple tests, was deemed unnecessary. While Type I and Type II errors may have occurred, since null hypotheses testing was not a primary study purpose, this does not affect the primary findings. Our findings are based only on participants who decided to have surgery and thus findings do not address decisions not to have surgery. Finally, the time between these surgeries and the data collection varied greatly. Those with a longer time since surgery may have had less ability to recall details, emotions, and feelings associated with DM. Longer times since surgery may have also allowed some participants to be more detached in their assessments.
Is self-regulation key in reducing running-related injuries and chronic fatigue? A randomized controlled trial among long-distance runners
Published in Journal of Applied Sport Psychology, 2022
Luuk P. van Iperen, Jan de Jonge, Josette M. P. Gevers, Steven B. Vos, Luiz Hespanhol
To test Hypothesis 3 (i.e., app intervention effects across risk profiles), we (re-)calculated latent profiles in Mplus (version 8.5; Muthén & Muthén, 1998−2017). To establish boundary conditions across profiles, we tested profiles in relation to group assignment, intervention use, and outcomes (i.e., RRIs and chronic fatigue). We employed the stepwise BCH and DCAT approaches as advocated by Asparouhov and Muthén (2020). To test whether the app effects differed across these profiles, we used model constraints and Wald tests (see Asparouhov & Muthén, 2018). The latter entails a subgroup analysis in evaluating a randomized controlled trial. If faced with potential power issues, as may be the case in the current study, this analysis can have increased rates of type I and type II errors and a limited precision of estimates (Keller, 2019). To address these limitations, we corrected for multiple testing using the conservative Bonferroni correction in deciding upon significance (Armstrong, 2014), and we explicitly note that this analysis is post-hoc and exploratory. Still, this risk profile analysis is important to answer the “what works for whom” question (Keller, 2019), which is considered relevant because it may provide directions for future studies and provide important insights into how psychological aspects can affect intervention effects.
Diurnal and seasonal differences in cardiopulmonary response to exercise in morning and evening chronotypes
Published in Chronobiology International, 2021
Kateřina Červená, Veronika Spišská, David Kolář, Katarína Evansová, Kateřina Skálová, Jiří Dostal, Stanislav Vybíral, Zdeňka Bendová
The main limitation of this study is the small sample size, leaving the analyses vulnerable to Type I and Type II errors (i.e., false positive and false negative findings). In principle, our data do not contradict previous studies; nevertheless, the risk that small effects concerning the seasons were not detected due to an underpowered sample exists, and future studies with a larger participant sample should be conducted. Moreover, although subjects were asked to adhere to their habitual sleep–wake timing, this was not controlled by actigraphy or sleep logs. Furthermore, our settings did not consider parameters like body fat percentage, baseline VO2max, or the standardized rating of perceived exertion or fatigue levels. In terms of fatigue, it was apparent the subjects were feeling tired when measured during a suboptimal time of day (i.e., M-types in the evening and E-types in the morning). Although this study provides promising data, we are aware that this is a pilot study that should be further verified on a larger sample with the limitations taken into account.