China
Africa
Explore chapters and articles related to this topic
What If the Data Is Not All There?
Published in Mitchell G. Maltenfort, Camilo Restrepo, Antonia F. Chen, Statistical Reasoning for Surgeons, 2020
Mitchell G. Maltenfort, Camilo Restrepo, Antonia F. Chen
A particular missing data situation can occur in a randomized study, where patients who receive a particular treatment arm may drop out. Perhaps they were not compliant; perhaps they transferred to another treatment. In this particular case, you want to run your statistical model as if the patients were in the arm they were assigned to. This is called an intention to treat analysis and allows you to test the results of the policy of assigning a patient to treatment [53]. Although this would arguably bias results in indicating no effect between treatments, that result may be what you want if the study was intended to show potential benefits of a new treatment. If you are designing a study, you should include not only the potential use of an intention to treat analysis but also ways of ensuring that you continue to receive data from patients who discontinue or switch treatments.
Statistics, Research and Governance
Published in Manit Arya, Taimur T. Shah, Jas S. Kalsi, Herman S. Fernando, Iqbal S. Shergill, Asif Muneer, Hashim U. Ahmed, MCQs for the FRCS(Urol) and Postgraduate Urology Examinations, 2020
Hamid Abboudi, Erik Mayer, Justin Vale
A Kaplan-Meier curve is a commonly used analysis to calculate survival after a treatment using lifetime data. Censoring correctly describes the process of removing subjects from a survival analysis if they have not completed follow-up at the stipulated time-point and can be incorporated in a Kaplan-Meier analysis. Intention-to-treat analysis describes an analysis methodology that, instead of removing subjects that deviate from an agreed protocol, includes them all in the final analysis based on their initial treatment intent. Listwise deletion describes removal of a record if a single data item is missing. It is a generic method for handling missing data in statistics and is not particular to Kaplan-Meier analysis. Rounding up or rounding down a figure is commonly used in mathematics to simplify interpretation.
Epidemiology
Published in John C Watkinson, Raymond W Clarke, Louise Jayne Clark, Adam J Donne, R James A England, Hisham M Mehanna, Gerald William McGarry, Sean Carrie, Basic Sciences Endocrine Surgery Rhinology, 2018
Jan H.P. van der Meulen, David A. Lowe, Jonathan M. Fishman
The results of trials can be analyzed in two ways. First, the comparison of the intervention can be carried out according to the intervention to which the patients were randomized (‘intention-to-treat’ analysis) or according to the treatment they actually received (‘per protocol’ analysis). The advantage of an intention-to-treat analysis is that the question that is being addressed corresponds exactly with the one clinicians and patients try to answer in clinical practice. The disadvantage is that if many patients do not receive the treatment they were randomized to, this would obscure the difference between the trial arms. Per protocol analysis, on the other hand, addresses which intervention is better more directly. With this form of analysis, the treatments are being compared according to the treatments that the patients actually received. The problem with this approach is that if many patients do not receive the treatment to which they were randomized, the study no longer represents an experimental study.
The effectiveness of comprehensive physiotherapy compared with corticosteroid injection on pain, disability, treatment effectiveness, and quality of life in patients with subacromial pain syndrome: a parallel, single-blind, randomized controlled trial
Published in Physiotherapy Theory and Practice, 2023
Maryam Daghiani, Hossein Negahban, Mohammad Hosein Ebrahimzadeh, Ali Moradi, Amir Reza Kachooei, Javad Raeesi, Akram Divandari
Data were analyzed using SPSS software version 20.0 and the level of statistical significance was set at P < .05 unless stated. Descriptive statistics were calculated for all variables. The results of Kolmogorov–Smirnov (K-S) test showed that all the outcome measures had a normal distribution. Based on the between-subject factor of group (i.e. two study groups) and within-subject factor of time (i.e. pretest, posttest, 3 and 6-month follow-up), 2 × 4 (group × time) mixed model analysis of variance (ANOVA) was conducted to determine the main effects and interactions of group and time factors for each outcome measure. In the case of significant interaction between group and time, between-group differences within each time level were tested, using chi-square test for nominal variable, and an independent sample t-test for normally distributed data. Data were analyzed using both the per-protocol and intention-to-treat analysis. As a result of small missing values (3.1% in three month and 5.7% in six month follow-up), missing data were imputed at follow-ups (5 imputations using multiple imputations-with fully observed data-by SPSS), and intention-to-treat analysis was performed.
Anti-cytomegalovirus preemptive therapy to prevent cytomegalovirus disease in HIV-infected patients: a systematic review
Published in Infectious Diseases, 2023
Prenali Dwisthi Sattwika, Yanri Wijayanti Subronto, Heni Retnowulan, Karina Ambar Sattwika, Detty Siti Nurdiati
We performed statistical analysis using the Review Manager 5.4 software. The random-effects model was used to calculate effect sizes. Dichotomous outcomes were analyzed using the risk ratio (RR) measure with 95% confidence intervals (CIs) and presented using forest plots which summarized treatment effect on CMV disease, all-cause mortality and adverse effects. Intention-to-treat analysis was performed if applicable. The statistical heterogeneity among studies was assessed by a Chi2 test on N-1 degrees of freedom with an alpha of 0.05 for statistical significance and the I2 analysis to detect the magnitude of variation attributable to heterogeneity rather than to chance. I2 values of 25%, 50% and 75% correspond to low, medium and high levels of heterogeneity.
What makes a great clinical trial in physiotherapy?
Published in Physiotherapy Theory and Practice, 2022
Joshua R. Zadro, Anita B. Amorim, Giovanni Ferreira, Xiaocong Hu, Rosa E. Becerra, Broti S. Reza, Samar Khan, Anne M. Moseley
The PEDro top five trials and physiotherapy trials published in the top medical journals have characteristics that may inform the design, conduct, and reporting of future trials. These trials often use concealed allocation and intention to treat analyses and have a low loss to follow-up. Concealed allocation reduces selection bias as the researcher randomizing participants does not know what the next treatment allocation will be (Elkins, 2013). Use of sealed, opaque, and consecutively numbered envelopes or an off-site investigator who randomizes participants are simple, low-cost ways to achieve this. Intention to treat analysis involves analyzing participants in the groups they were initially allocated to, regardless of whether they stopped treatment or crossed over to another group (Elkins and Moseley, 2015). This preserves the prognostic balance afforded by randomization and makes the effect estimates more realistic (McCoy, 2017). The loss to follow-up that exceeds 15% poses serious threats to the validity of a trial (Dettori, 2011; Maher et al., 2003). Strategies to reduce loss to follow-up include obtaining funding to employ research assistants who will rigorously implement trial follow-up procedures, offering participants in the control group the intervention at the end of the trial (Zadro et al., 2019), offering incentives for recruitment, or to complete assessments (e.g., payment and prize draw), tailoring the communication strategy with participants (e.g., text or e-mail reminders), and reducing the burden of follow-up assessments (Brueton et al., 2014).