Australian Parental Supply of Alcohol Longitudinal Study (APSALS) Analysis Code

R Analysis Code

This repository contains R code used in a number of articles using the Australian Parental Supply of Alcohol Longitudinal Study (APSALS).

Code for all analysis in the article by Clare et al 2020 published in Addiction: https://doi.org/10.1111/add.15005

DescriptionR-code
1 - Multiple imputationMultiple imputation
2 - Final data creationFinal data creation
3 - LTMLE analysis of parental supply of alcohol on harms using the package ‘ltmle’ (1).LTMLE analysis
4 - LTMLE marginal structural model analysis of earlier initiation of supply.LTMLE MSM analysis
5 - Sensitivity analysis using naive analysis (GLMs)Naive analysis
6 - E-Value sensitivity analysisE-value analysis
7 - Secondary analysis of exposure (parental supply) beginning at age 15.LTMLE - supply from age 15
8 - Sensitivity analysis with lagged predictors.LTMLE - lagged predictors
9 - Sensitivity analysis controlling for past obervations of outcomeLTMLE - control for past outcomes
10 - Sensitivity analysis with continuous outcomes.LTMLE - continuous outcomes

Changes in mental health and help-seeking among young Australian adults during the COVID-19 pandemic: a prospective cohort study

R and Stata code for all analysis of APSALS COVID-19 Alcohol paper, Upton et al 2021 published in Psychological Medicine: https://doi.org/10.1017/S0033291721001963

DescriptionR-code
1 - Missing data patternsMissing data
2 - Multiple imputationMultiple imputation
3 - Final data creationFinal data creation
4 - Import MI data into Stata for analysisStata import
5 - Primary analysis with data in long-formPrimary long-form analysis
6 - Primary analysis with data in wide-formPrimary wide-form analysis
7 - Additional analysis comparing COVID subsample to full APSALS cohortAdditional table

Alcohol use among young Australian adults during the COVID-19 pandemic: a prospective cohort study

R and Stata code for all analysis of APSALS COVID-19 Alcohol paper, Clare et al 2021 published in Addiction: https://doi.org/10.1111/add.15599

DescriptionR-code
1 - Multiple imputationMultiple imputation
2 - Final data creationFinal data creation
3 - Import MI data into Stata for analysisStata import
4 - Cross-sectional descriptives in RCross-sectional descriptives
5 - Longitudinal descriptives in StataLongitudinal descriptives
6 - Primary analyses using mixed effects models with discrete timePrimary analyis
7 - Sensitivity analysis using continuous time and ‘high risk’ consumption variableSensitivity analysis

Tobacco and vaping characteristics over 5 years in the Australian Parental Supply of Alcohol Longitudinal Study (APSALS)

R and Stata code for all analysis of APSALS COVID-19 Tobacco paper, Boland et al 2024, in progress

DescriptionR-code
1 - Multiple imputationMultiple imputation
2 - Final data creationFinal data creation
3 - Descriptive statisticsDescriptives
4 - Analysis of raw trendsTrends
5 - Multivariable regression modelsModels
6 - Correlation analysisCorrelation

R and Stata code for all analysis of APSALS Initiation trajectories paper, Clare et al 2024, being presented at KBS2024.

DescriptionR-code
1 - Create data subset from APSALS cohort dataData creation
2 - Multiple imputationMultiple imputation
3 - Test model fit of nonlinear termsModel fit
4 - Primary analysisPrimary analysis
5 - Secondary analysisSecondary analysis
6 - Pool MI results using Rubin’s rubles to get final estimatesPool results
7 - Create primary analysis figures using ggplotCreate primary figures
8 - Create secondary analysis figures using ggplotCreate secondary figures
9 - Descriptive statisticsDescriptives
10 - Generate missing data summary for appendixMissing data
  1. Lendle SD, Schwab J, Petersen ML, van der Laan MJ. ltmle: An R Package Implementing Targeted Minimum Loss-Based Estimation for Longitudinal Data. Journal of Statistical Software. 2017;81(1):1-21.
Avatar
Dr. Philip J Clare, PhD

Biostatistician at the Prevention Research Collaboration, University of Sydney.

Related