OBJECTIVE: Interrupted time series (ITS) is a strong quasi-experimental research design, which is increasingly applied to estimate the effects of health services and policy interventions. We describe and illustrate two methods for estimating confidence intervals (CIs) around absolute and relative changes in outcomes calculated from segmented regression parameter estimates. STUDY DESIGN AND SETTING: We used multivariate delta and bootstrapping methods (BMs) to construct CIs around relative changes in level and trend, and around absolute changes in outcome based on segmented linear regression analyses of time series data corrected for autocorrelated errors. RESULTS: Using previously published time series data, we estimated CIs around the effect of prescription alerts for interacting medications with warfarin on the rate of prescriptions per 10,000 warfarin users per month. Both the multivariate delta method (MDM) and the BM produced similar results. CONCLUSION: BM is preferred for calculating CIs of relative changes in outcomes of time series studies, because it does not require large sample sizes when parameter estimates are obtained correctly from the model. Caution is needed when sample size is small.
OBJECTIVE: Interrupted time series (ITS) is a strong quasi-experimental research design, which is increasingly applied to estimate the effects of health services and policy interventions. We describe and illustrate two methods for estimating confidence intervals (CIs) around absolute and relative changes in outcomes calculated from segmented regression parameter estimates. STUDY DESIGN AND SETTING: We used multivariate delta and bootstrapping methods (BMs) to construct CIs around relative changes in level and trend, and around absolute changes in outcome based on segmented linear regression analyses of time series data corrected for autocorrelated errors. RESULTS: Using previously published time series data, we estimated CIs around the effect of prescription alerts for interacting medications with warfarin on the rate of prescriptions per 10,000 warfarin users per month. Both the multivariate delta method (MDM) and the BM produced similar results. CONCLUSION: BM is preferred for calculating CIs of relative changes in outcomes of time series studies, because it does not require large sample sizes when parameter estimates are obtained correctly from the model. Caution is needed when sample size is small.
Authors: Adriana Pérez; Rodolfo J Dennis; Benigno Rodríguez; Amparo Y Castro; Victor Delgado; Juan Manuel Lozano; Maria Clara Castro Journal: J Clin Epidemiol Date: 2003-10 Impact factor: 6.437
Authors: Adrianne C Feldstein; David H Smith; Nancy Perrin; Xiuhai Yang; Steven R Simon; Michael Krall; Dean F Sittig; Diane Ditmer; Richard Platt; Stephen B Soumerai Journal: Arch Intern Med Date: 2006-05-08
Authors: Marcus A Bachhuber; Ellenie Tuazon; Michelle L Nolan; Hillary V Kunins; Denise Paone Journal: Pharmacoepidemiol Drug Saf Date: 2019-03-28 Impact factor: 2.890
Authors: Elizabeth A Brown; Clara E Dismuke-Greer; Viswanathan Ramakrishnan; Trevor D Faith; Edith M Williams Journal: Arthritis Care Res (Hoboken) Date: 2020-02 Impact factor: 4.794
Authors: Alyce S Adams; Stephen B Soumerai; Fang Zhang; Daniel Gilden; Marguerite Burns; Haiden A Huskamp; Connie Trinacty; Margarita Alegria; Robert F LeCates; Jennifer J Griggs; Dennis Ross-Degnan; Jeanne M Madden Journal: Clin Ther Date: 2015-01-22 Impact factor: 3.393
Authors: Marguerite E Burns; Alisa B Busch; Jeanne M Madden; Robert F Le Cates; Fang Zhang; Alyce S Adams; Dennis Ross-Degnan; Stephen B Soumerai; Haiden A Huskamp Journal: Psychiatr Serv Date: 2014-03-01 Impact factor: 3.084