Dataset Information

SAS® Macros for Computing the Mediated Effect in the Pretest-Posttest Control Group Design.

ABSTRACT: Mediation analysis is a statistical technique for investigating the extent to which a mediating variable transmits the relation of an independent variable to a dependent variable. Because it is useful in many fields, there have been rapid developments in statistical mediation methods. The most cutting-edge statistical mediation analysis focuses on the causal interpretation of mediated effect estimates. Cause-and-effect inferences are particularly challenging in mediation analysis because of the difficulty of randomizing subjects to levels of the mediator (MacKinnon, 2008). The focus of this paper is how incorporating longitudinal measures of the mediating and outcome variables aides in the causal interpretation of mediated effects. This paper provides useful SAS® tools for designing adequately powered studies to detect the mediated effect. Three SAS macros were developed using the powerful but easy-to-use REG, CALIS, and SURVEYSELECT procedures to do the following: (1) implement popular statistical models for estimating the mediated effect in the pretest-posttest control group design; (2) conduct a prospective power analysis for determining the required sample size for detecting the mediated effect; and (3) conduct a retrospective power analysis for studies that have already been conducted and a required sample to detect an observed effect is desired. We demonstrate the use of these three macros with an example.

SUBMITTER: Valente MJ

PROVIDER: S-EPMC6133302 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

SAS® Macros for Computing the Mediated Effect in the Pretest-Posttest Control Group Design.

Valente Matthew J MJ MacKinnon David P DP

SAS global forum 20170101

Mediation analysis is a statistical technique for investigating the extent to which a mediating variable transmits the relation of an independent variable to a dependent variable. Because it is useful in many fields, there have been rapid developments in statistical mediation methods. The most cutting-edge statistical mediation analysis focuses on the causal interpretation of mediated effect estimates. Cause-and-effect inferences are particularly challenging in mediation analysis because of the ...[more]

PMID: 30215060

Similar Datasets

Project description:BackgroundSocial distancing measures to address the US coronavirus disease 2019 (COVID-19) epidemic may have notable health and social impacts.Methods and findingsWe conducted a longitudinal pretest-posttest comparison group study to estimate the change in COVID-19 case growth before versus after implementation of statewide social distancing measures in the US. The primary exposure was time before (14 days prior to, and through 3 days after) versus after (beginning 4 days after, to up to 21 days after) implementation of the first statewide social distancing measures. Statewide restrictions on internal movement were examined as a secondary exposure. The primary outcome was the COVID-19 case growth rate. The secondary outcome was the COVID-19-attributed mortality growth rate. All states initiated social distancing measures between March 10 and March 25, 2020. The mean daily COVID-19 case growth rate decreased beginning 4 days after implementation of the first statewide social distancing measures, by 0.9% per day (95% CI -1.4% to -0.4%; P < 0.001). We did not observe a statistically significant difference in the mean daily case growth rate before versus after implementation of statewide restrictions on internal movement (0.1% per day; 95% CI -0.04% to 0.3%; P = 0.14), but there is substantial difficulty in disentangling the unique associations with statewide restrictions on internal movement from the unique associations with the first social distancing measures. Beginning 7 days after social distancing, the COVID-19-attributed mortality growth rate decreased by 2.0% per day (95% CI -3.0% to -0.9%; P < 0.001). Our analysis is susceptible to potential bias resulting from the aggregate nature of the ecological data, potential confounding by contemporaneous changes (e.g., increases in testing), and potential underestimation of social distancing due to spillover effects from neighboring states.ConclusionsStatewide social distancing measures were associated with a decrease in the COVID-19 case growth rate that was statistically significant. Statewide social distancing measures were also associated with a decrease in the COVID-19-attributed mortality growth rate beginning 7 days after implementation, although this decrease was no longer statistically significant by 10 days.

Project description:Strengthening pre-adolescents knowledge and skills through an age- and culturally-appropriate intervention could prevent health issues later in life. Early interventions could influence the trajectory of future risky behaviour, and may influence health behaviour amongst their parents. The CIrCLE of Life Initiative was developed to address HIV and obesity. We evaluated whether the combined intervention increased knowledge, enhanced skills, and/or promoted healthy behaviour among students (9-12 years old) and their parents. The study was conducted from May to December 2018. Trained educators delivered 30-min lessons over ten consecutive weeks with 537 Grade 6 students at five government-run schools, in a district, in South Africa. Schools were purposively selected based on socioeconomic status and urban-rural classification. Students communicated with parents through shared homework activities. A pretest-posttest study design was used, with a 3-month follow up. Both groups completed self-administered paper-based questionnaires. A score of subscales was used in analysis. The pretest and posttest scores were compared for students and parents using a dependent t-test. Differences in outcomes by school quintile were compared using one-way ANOVA. Response rates were high for both students (80.6%) and their parents (83.4%). Statistically significant differences were observed in HIV knowledge in students pretest (mean 8.04, SD 3.10) and posttest scores (mean 10.1, SD 2.70; p < 0.01), and their parents (mean 10.32, SD 2.80 vs 11.0, SD 2.50; p < 0.01). For both students and parents, pre- and post-test obesity awareness mean scores were similar, 1.93, SD 0.92 and 2.78, SD 0.57; p < 0.01, for students; and 2.47, SD 0.82 and 2.81, SD 0.54; p < 0.01, for parents. In the posttest, statistically significant changes were also observed in both groups, enhancing skills in measuring body mass index and pulse rate, and interpreting food labels. Students had a high intention to share gained knowledge with parents who had a high intention to receive it (89.4 and 89.5%, respectively). The intervention increased knowledge about HIV and obesity-related awareness, and it enhanced skills in selected outcomes among pre-adolescents and parents. Accurate messages and enhanced communication skills could support inter-generational knowledge transfer. ClinicalTrials.gov Identifier: NCT04307966 retrospectively registered on 12 March 2020.

Project description:BackgroundNeurofeedback training (NFT) has been shown to be effective in treating several disorders (eg, attention-deficit/hyperactivity disorder [ADHD], anxiety, and depression); however, little is currently known regarding the effectiveness of remote NFT systems.ObjectiveThis retrospective study provides real-world data (N=593) to assess the efficacy of app-based remote NFT in improving brain health and cognitive performance.MethodsImprovement was measured from pre- to postintervention of in-app assessments that included validated symptom questionnaires (the 12-item General Health Questionnaire, the ADHD Rating Scale IV, the Adult ADHD Self-Report Scale, the 7-item Generalized Anxiety Disorder scale, and the 9-item Patient Health Questionnaire), a cognitive test of attention and executive functioning (ie, continuous performance task), and resting electroencephalography (EEG) markers. Clinically significant improvement was evaluated using standard approaches.ResultsThe greatest improvement was reported for the anxiety questionnaire, for which 69% (68/99) of participants moved from abnormal to healthy score ranges. Overall, adult and child participants who engaged in neurofeedback to improve attention and executive functions demonstrated improved ADHD scores and enhanced performance on a cognitive (ie, response inhibition) task. Adults with ADHD additionally demonstrated elevated delta/alpha and theta/alpha ratios at baseline and a reduction in the delta/alpha ratio indicator following neurofeedback.ConclusionsPreliminary findings suggest the efficacy of app-based remote neurofeedback in improving mental health, given the reduced symptom severity from pre- to postassessment for general psychological health, ADHD, anxiety, and depression, as well as adjusted resting EEG neural markers for individuals with symptoms of ADHD. Collectively, this supports the utility of the in-app assessment in monitoring behavioral and neural indices of mental health.

Dataset Information

SAS® Macros for Computing the Mediated Effect in the Pretest-Posttest Control Group Design.

Publications

SAS® Macros for Computing the Mediated Effect in the Pretest-Posttest Control Group Design.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets