This tutorial by Aurelio Tobias, Yoonhee Kim, and Lina Madaniyazi provides a comprehensive guide to implementing the time-stratified case-crossover design in environmental epidemiology studies. The design is an effective alternative to conventional time-series regression for estimating short-term associations between environmental exposures and acute events. Key aspects covered include: 1. **Design Overview**: The case-crossover design compares exposure levels on the day of a health event (case day) with nearby days (control days) to identify differences in exposure that might explain case counts. 2. **Time-Stratified Approach**: This method controls for long-term trends and seasonality by selecting control days based on the day-of-week within-month and year, ensuring unbiased estimates. 3. **Conditional Regression Models**: These models adjust for time-varying confounders, such as temperature and relative humidity, and can handle overdispersion and autocorrelation in aggregated data. 4. **Adjusting Covariates**: Time-invariant subpopulation covariates require reshaping the data into a long format and conditioning them in the expanded stratum set. 5. **Effect Modification**: Interaction models can be used to investigate whether health effects vary by subpopulation characteristics, providing more precise estimates and allowing for formal hypothesis testing. 6. **Multi-Location Studies**: The space-time-stratified case-crossover design combines time and spatial dimensions to analyze multilevel data, suitable for small geographical units. 7. **Discussion**: The tutorial highlights the advantages of the time-stratified case-crossover design, including its flexibility and ability to handle aggregated data, and discusses extensions to multilevel data and effect modification analysis. The tutorial includes practical examples using real data sets from Valencia and London, demonstrating the application of these methods in environmental epidemiology research.This tutorial by Aurelio Tobias, Yoonhee Kim, and Lina Madaniyazi provides a comprehensive guide to implementing the time-stratified case-crossover design in environmental epidemiology studies. The design is an effective alternative to conventional time-series regression for estimating short-term associations between environmental exposures and acute events. Key aspects covered include: 1. **Design Overview**: The case-crossover design compares exposure levels on the day of a health event (case day) with nearby days (control days) to identify differences in exposure that might explain case counts. 2. **Time-Stratified Approach**: This method controls for long-term trends and seasonality by selecting control days based on the day-of-week within-month and year, ensuring unbiased estimates. 3. **Conditional Regression Models**: These models adjust for time-varying confounders, such as temperature and relative humidity, and can handle overdispersion and autocorrelation in aggregated data. 4. **Adjusting Covariates**: Time-invariant subpopulation covariates require reshaping the data into a long format and conditioning them in the expanded stratum set. 5. **Effect Modification**: Interaction models can be used to investigate whether health effects vary by subpopulation characteristics, providing more precise estimates and allowing for formal hypothesis testing. 6. **Multi-Location Studies**: The space-time-stratified case-crossover design combines time and spatial dimensions to analyze multilevel data, suitable for small geographical units. 7. **Discussion**: The tutorial highlights the advantages of the time-stratified case-crossover design, including its flexibility and ability to handle aggregated data, and discusses extensions to multilevel data and effect modification analysis. The tutorial includes practical examples using real data sets from Valencia and London, demonstrating the application of these methods in environmental epidemiology research.