HRs and 95% confidence intervals (CIs) were estimated using Cox proportional hazards models stratified by the factors used for the stratified log-rank tests and by treatment.
To determine if region of origin acted as an effect modifier, we conducted a stratified multivariate analysis by creating eight identical models, stratified by the eight world regions.
Further models stratified by depression status post diagnosis and sensitivity analyses excluded women who were depressed when the sleep measures were queried.
Crude linear regression models stratified by the quintiles of body weight of the subjects were generated with BMI (kg m−2) as main outcome and pasta consumption (grams per day) as independent factor, in both Moli-sani and INHES populations.
Crude and multi-adjusted models stratified by gender were generated with main outcome the BMI (Kg m-2) and independent factor, the predicted pasta-body weight residuals in both Moli-sani and INHES datasets.
We used Cox proportional hazards models stratified by age (categorised in 5-year increments) and follow-up intervals to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for incident skin cancers associated with cumulative UV flux.
Crude linear regression models stratified by gender were generated with main outcome the BMI (kg m−2) and independent factor, the pasta consumption (grams per day) in both Moli-sani and INHES populations.
We also performed exploratory analyses in which we ran multivariable-adjusted models stratified by age (<55 years of age compared to ≥55 years of age).
The analysis procedures were conducted separately in time series models stratified by status of temperature sensitivity (i.e. temperature sensitizer vs. temperature non-sensitizer groups).
We additionally fit our final models stratified by survey year.
Village was included in all models a priori, whilst in models stratified by age, sex was included a priori and age (in years) was included a priori in models stratified by sex.
