For the secondary outcomes, we used Poisson regression to analyze exacerbation rates requiring hospitalization and negative binomial to estimate the other variables covering healthcare utilization data.
Most of the research used linear regression to analyze vehicle mileage [12, 13, 14, 15, 16, 17, 18], logistic regression to mode choice [19, 20, 21, 22], negative binomial regression to self-selection [23, 24, 25], probit regression on non-work trip generation and vehicle mileage [26, 27], and propensity score matching on neighborhood design and walking trips [28, 29].
The number of falls per person-year will be analyzed using negative binomial regression to estimate the difference in fall rates between the two groups.[ 34] The proportion of fallers in the two groups will also be compared using the relative risk statistic.
We used binomial regression to calculate RRs adjusted for possible confounders.
After testing different models, we used negative binomial regression to adjust rural and urban cases.
Of 3244 subjects (return rate 52.4%), 936 men aged 40 64 years were analyzed using log-binomial regression to explore factors related to PSA-screening participation within 1 year.
Negative binomial regression was used to analyze incidence and socioeconomic deprivation.
Negative binomial regression was used to analyze trends for hospitalization rates of all encephalitis hospitalizations and for those caused by known and unknown pathogens.
Negative binomial regression was used to analyze the number of new or enlarging T2-hyperintense lesions (covariates: region, study and baseline T2-hyperintense volume) and the number of new T1-hypointense lesions (covariates: region, study, and baseline T1-hypointense volume).
Univariate and multivariate log-binomial regression were used to analyze factors affecting willingness to undergo HIV testing.
