China
Africa
Explore chapters and articles related to this topic
Ten Top Tips for Successful Grant Proposal Writing
Published in Lisa Chasan-Taber, Writing Grant Proposals in Epidemiology, Preventive Medicine, and Biostatistics, 2022
If instead you are proposing to launch a new study and recruit participants, you can choose the sample size you need to achieve sufficient power. However, in this case, progressing to Step #3 and calculating your budget will be critical. A common pitfall of new investigators is to be too ambitious—proposing a larger sample size than they have the budget and experience to handle. Chapter 13, “Power and Sample Size,” provides user-friendly approaches to power and sample size calculations, available software, and annotated examples with strategies and tips.
Survival Analysis
Published in Trevor F. Cox, Medical Statistics for Cancer Studies, 2022
Figure 3.11 shows the pointwise confidence intervals for the Sunitinib KM curve. A log transformation of has been used. The idea of putting a confidence band around the KM curve is to emphasise the variation that can occur when estimating the survival function. If another trial was run, under exactly the same conditions for this lung cancer trial, the KM curve obtained would be similar to the one we have now, and within these confidence bands (or most of the curve should be within these confidence bands). Of course, this is not guaranteed to be the case. As with any confidence interval, the greater the sample size, the narrower the confidence interval. Also, note that the confidence band is made up from pointwise confidence intervals. This is not the same as an overall confidence band for the whole survival curve.
Inference on Proportions
Published in Marcello Pagano, Kimberlee Gauvreau, Heather Mattie, Principles of Biostatistics, 2022
Marcello Pagano, Kimberlee Gauvreau, Heather Mattie
This tells us that over the 48-week follow-up period, children treated with acetaminophen had 2.6% more episodes of asthma exacerbation than children treated with ibuprofen. In order to calculate a 95% confidence interval for the risk difference, an estimate of its standard error is required. In practice more than one formula for the standard error of a risk difference exists, and each statistical package may employ a different one. Some packages also generate an exact confidence interval. When sample sizes are large, the resulting confidence intervals will be very similar. In Table 14.8, Stata displays the 95% confidence interval of the risk difference as , and Table 14.9 shows a similar result from R, . Since these intervals contain the value 0, the data do not exclude the possibility that the proportions of children experiencing asthma exacerbation are identical in the two groups. As expected, this is the same conclusion reached with the hypothesis test.
A systematic review of randomized controlled trial characteristics for interventions to improve upper extremity motor recovery post stroke
Published in Topics in Stroke Rehabilitation, 2023
Marcus Saikaley, Amanda McIntyre, Griffin Pauli, Robert Teasell
Previously, we have expressed concern about the relatively small sample size of stroke rehabilitation RCTs examining interventions designed to treat motor disorders involving the upper extremity.1 One drawback of the evidence, despite its large volume, is the majority of RCTs were single-site studies with small sample sizes; 827 or 63% of the 1,307 RCTs had less than 40 subjects initially randomized; multi-centered RCTs accounted for only 19% of all RCTs. Sample size is important since larger sample sizes tend to improve statistical confidence in the results. RCTs are often underpowered due to resource limitations or difficulties recruiting patients that fit a stringent inclusion criterion, leading to a potential overestimation of effect sizes.10,11 The determination of sample size is often based on statistical power calculations that are based on the primary outcome measure being used. One positive finding is that mean sample sizes have been progressively rising since 2008. The reason for this is not known but may reflect improved funding, greater research expectations and training, or more demanding publication expectations. Many of these small RCTs use ‘usual care’ or ‘conventional care’ controls; these have been described by others as more Proof-of Principle or Phase II studies.12
Training flexible conceptual retrieval in post-stroke aphasia
Published in Neuropsychological Rehabilitation, 2022
Sara Stampacchia, Glyn P. Hallam, Hannah E. Thompson, Upasana Nathaniel, Lucilla Lanzoni, Jonathan Smallwood, Matthew A. Lambon Ralph, Elizabeth Jefferies
In line with the original use of the term “semantic aphasia” by Henry Head (1926) and the inclusion criteria proposed by Jefferies and Lambon Ralph (2006), the patients in this study were selected to show deficits affecting the appropriate use of concepts presented as words and objects when control demands were high. In addition to verbal semantic problems, they were impaired on at least one non-verbal task (see section “Tests of semantic control”). The sample size was determined by the maximum number of patients available to take part in the study. These criteria for including participants were established prior to data collection. There were no other inclusion/exclusion criteria. In common with previous SA samples (e.g., Jefferies & Lambon Ralph, 2006; Stampacchia et al., 2018), the patients showed strong effects of semantic control manipulations across tasks (details below). Individual patient data and task descriptions are provided in section “Tests of semantic control”.
BST-1 as a serum protein biomarker involved in neutrophil infiltration in schizophrenia
Published in The World Journal of Biological Psychiatry, 2022
Liang-Jen Wang, Yu-Chi Huang, Pao-Yen Lin, Yu Lee, Chi-Fa Hung, Su-Ting Hsu, Lien-Hung Huang, Sung-Chou Li
This study has some limitations that should be mentioned. First, this was a cross-sectional study. Although we found differentiated protein markers between patients with schizophrenia and healthy controls, the causal relationships among protein levels and diseases remain undetermined. For example, we were unable to distinguish whether the decreased BST1 levels were derived from the natural pathophysiology of schizophrenia or from long-term exposure to antipsychotic treatment. Second, we used serum samples for iTRAQ exploration, but the protein levels in peripheral blood may not necessarily represent the action in the brain. Third, we initially collected serum from 12 patients and 12 controls to detect iTRAQ protein markers. However, the small sample size reduces the statistical power of the study and may increase the potential for type II error. Forth, some potential confounding factors, such as life habits, physical exercise or diet, were not recorded in this study. These unmeasured factors may interfere the protein biomarker levels. In addition, the patients recruited in this study were those psychotic symptoms relatively stable. However, protein biomarker signatures may be related to psychopathology in schizophrenia. Whether our biomarker panel is able to generalised into patients with acute exacerbation warrants further investigation. Finally, the samples and data came from a single site. In the future, multicenter collaborations that combine large-scale data should be applied for multiple levels of analysis, using standardised nomenclature and integration with other databases.