China
Africa
Explore chapters and articles related to this topic
Latent Class Analysis
Published in Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle, Structural Equation Modeling for Health and Medicine, 2021
Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle
Traditionally, subgroup analysis in health and medicine is used to determine whether individuals respond differently to a treatment, or have a different strength of association, based on one or more clinical characteristics [1]. In such analyses, one typically defines the subgroups a priori based on a theory or even the categories of a single observed variable. For example, let us imagine we would like to evaluate the association between length of stay and health care costs in adults with chronic obstructive pulmonary disease (COPD). We hypothesize there may be differences in this association by sex [2]. Historically, COPD had been viewed as a disease that mostly afflicts men, though recent studies have indicated that prevalence and mortality have been on the rise for women [3]. In order to test a hypothesis about male/female differences, we would create subsets of the data for males and females. Then, we could perform traditional stratified linear regression analyses within each subgroup strata. These types of a priori subgroups (e.g. male/female) are used in conducting moderation analysis (Chapter 9) and measurement invariance testing in the multigroup modeling framework (Chapter 10).
Safety Monitoring and Analysis in Oncology Trials
Published in Satrajit Roychoudhury, Soumi Lahiri, Statistical Approaches in Oncology Clinical Development, 2018
Anastasia Ivanova, Qi Jiang, Olga Marchenko, Richard C. Zink
Subgroups are frequently considered for the analysis of safety and efficacy endpoints, with 70% of clinical trials reporting at least some results within subgroups (Pocock et al. 2002). Subgroup analyses are beneficial in that they provide clinicians with information on the potential for differential treatment response within important demographic, genetic, disease, environmental, behavioral, or regional characteristics (Quan et al. 2010; Chuang-Stein et al. 2014). From a regulatory perspective, such analyses are important to show that the estimated overall effect is broadly applicable to patients and to assess risk-benefit across the proposed indication, particularly when the study population is heterogeneous (CHMP 2014). Further, examining results within subgroups allows the study team to assess the consistency and robustness of results obtained for the entire study population, as well as to generate hypotheses for future research (Cui et al. 2002). Finally, for the study of oncology, subgroup analyses are important to identify patients at increased risk for severe toxicity of the prescribed treatments. Subgroup analyses would likely be considered for important Tier I events.
M
Published in Filomena Pereira-Maxwell, Medical Statistics, 2018
Repeated significance testing on the same body of data. Subgroup analyses are a common example, where first an overall test of statistical significance is performed, with subsequent tests carried out within subgroups of individuals sharing similar characteristics. For example, in the ISIS-2 trial (Box A.1, p. 3) comparing aspirin versus no aspirin for the management of acute myocardial infarction, one may be interested in making this comparison within different age groups, since aspirin could be effective if used for instance, in younger patients, but not in older patients. If three age groups are defined, three significance tests will be performed, the chances of a type I error increasing with the number of tests carried out. Likewise, multiple outcomes may give rise to a similar problem, and the Bonferroni and other test corrections are often applied in order to decrease the probability of declaring results to be statistically significant when in reality the relevant null hypothesis is true. See also nominal significance level. POCOCK (1983) discusses the issue of multiplicity of data. See also BLAND & ALTMAN (2011).
Postoperative serum CA19-9, YKL-40, CRP and IL-6 in combination with CEA as prognostic markers for recurrence and survival in colorectal cancer
Published in Acta Oncologica, 2020
Kethe Hermunen, Leena-Maija Soveri, Mogens Karsbøl Boisen, Harri K. Mustonen, Christian Dehlendorff, Caj H. Haglund, Julia Sidenius Johansen, Pia Osterlund
Among the strengths of our study are that all patients underwent high-quality surgery followed by long-term monitoring. All received adjuvant therapy with 6 months of 5-FU, although without oxaliplatin. A further strength is the follow-up routine with clinical examination, routine laboratory measurements, including tumour markers, and radiology for 10 years within the prospective study at a single institution. Limitations are the small sample sizes in subgroup analyses, resulting in limited statistical power. Secondly, as routine CEA, CA19-9, and CRP were not systematically analysed preoperatively, we, therefore, could not study persistently elevated markers postoperatively vs. markers that postoperatively normalised. Another limitation is that some sampling of biomarkers and initiation of adjuvant therapy was delayed beyond 8 weeks, perhaps impacting results of the efficacy of adjuvant therapy but will probably not have had an influence on the prognostic effect of the postoperative markers. These patients were treated in 1997–2001 with 5-FU-based chemotherapy and the results should be validated in patients treated with oxaliplatin-based chemotherapy.
AST to ALT ratio and risk of hemorrhagic transformation in patients with acute ischemic stroke
Published in Neurological Research, 2020
Yanan Wang, Ke Qiu, Quhong Song, Yajun Cheng, Junfeng Liu, Ming Liu
In addition, we performed subgroup analyses using stratified logistic regression models. The significance of interaction (P for interaction) was tested using the likelihood ratio test. We further used a logistic regression model with restricted cubic splines to evaluate the pattern and magnitude of the association between AAR and HT with four knots (at the 5th, 35th, 65th and 95th percentiles). Net reclassification index (NRI) and integrated discrimination improvement (IDI) were calculated to assess the ability of AAR, AST and ALT to reclassify risks of HT based on conventional model with established risk factors. All of the analyses were performed with SPSS 25.0 (IBM, Chicago, IL, USA) and Stata 15.0 (StataCorp LP, College Station, TX, USA). P values less than 0.05 (two-sided) were considered statistically significant.
Subgroup identification by recursive segmentation
Published in Journal of Applied Statistics, 2018
Alexander Hapfelmeier, Kurt Ulm, Bernhard Haller
Based on the inequations 1 and 3, a clinical outcome or the difference of outcomes can be modeled to lead to prognostic or predictive subgroups [25,28], respectively. The investigation of predictive factors and identification of respective subgroups refers to the exploration of individual treatment effects and is generally termed ‘subgroup analysis’. It usually involves the use of ‘potential outcomes’ which are defined to be the expected outcome of a subject under certain treatment and for (a) given biomarker value(s) [27]. Consequently, each subject has two potential outcomes, that is