An interrupted time series analysis of parenteral antibiotic use in Colombia

Pérez, Adriana; Dennis, Rodolfo; Rodrı́guez, Benigno; Castro, Amparo; Delgado, Victoria; Lozano, Juan Manuel; Castro, M.

doi:10.1016/s0895-4356(03)00163-x

Cited by 29 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Non-linear trends can be modelled and provide better control of confounding, but require data for many time-points and may result in less-precise effect estimates. Furthermore, even after modelling trends and allowing for seasonality, there may be autocorrelation between outcome levels at adjacent time-points 38. This autocorrelation can lead to an overestimation of the precision of intervention effects.…”

Section: Causal Effects and Confounding In Non-randomised Evaluationsmentioning

confidence: 99%

Alternatives to randomisation in the evaluation of public-health interventions: statistical analysis and causal inference

Cousens

Hargreaves

Bonell

et al. 2009

Journal of Epidemiology & Community Health

View full text Add to dashboard Cite

show abstract

Section: Causal Effects and Confounding In Non-randomised Evaluationsmentioning

confidence: 99%

Alternatives to randomisation in the evaluation of public-health interventions: statistical analysis and causal inference

Cousens

Hargreaves

Bonell

et al. 2009

Journal of Epidemiology & Community Health

View full text Add to dashboard Cite

show abstract

“…This method can be problematic for several reasons: parameter estimates of intervention effects take the form of odds ratios, which are more difficult for a general audience to interpret; assumptions are violated when covariates are missing not at random, which is a frequent occurrence [8]; and computation is daunting when population size is large. Another method is the autoregressive integrated moving average model (ARIMA) [9,10] , which uses aggregate outcome estimates at each time to model intervention effects. However, ARIMA models assume a complex error correlation structure and require sample sizes of at least 50 consecutive time points [9], which is often impossible in health services studies.…”

Section: Introductionmentioning

confidence: 99%

Methods for estimating confidence intervals in interrupted time series analyses of health interventions

Zhang

Wagner

Soumerai

et al. 2009

Journal of Clinical Epidemiology

207

170

View full text Add to dashboard Cite

Objective Interrupted time series is a strong quasi-experimental research design that is increasingly applied to estimate the effects of health services and policy interventions. We describe and illustrate two methods for estimating confidence intervals around absolute and relative changes in outcomes calculated from segmented regression parameter estimates. Study Design and Setting We used multivariate delta and bootstrapping methods to construct confidence intervals 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 auto-correlated errors. Results Using previously published time series data, we estimated confidence intervals 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 and the bootstrapping method produced similar results. Conclusion The bootstrapping method is preferred for calculating confidence intervals of relative changes in outcomes of time series studies since it does not require large sample sizes when parameter estimates are obtained correctly from the model. Caution is needed when sample size is small.

show abstract

“…The example of external factor intervention can be found in Montgomery and Weatherby [3] who analyzed the effect of Arabic oil embargo on the consumption level of electricity in United States and Enders et al [4] studied the effects of metal detector technology to the number of plane pirated and Suhartono and Hariroh [5] studied on the impact of WTC's bomb in New York to stock market values around the world. Other study on internal factor intervention can be found in the study of Box and Tiao [3] , analyzing the effect of machine design law to oxidant pollution level in Los Angeles, McSweeny [6] researched on the effect of new law about constancy value in Cincinnati Bell Telephone company to the number of emergency calls, Leonard [7] observed the effect of promotion and product cost increased by a company and analyzed the effect of promotion and cost increased in customer pulse consumption in Telkom Regional Division. These models have been applied to various problem domains such as medicine [6,8] , fisheries research [9] , economic [10] , air transport and waste management [8,11] .…”

Section: Introductionmentioning

confidence: 99%

Intervention Model for Analyzing the Impact of Terrorism to Tourism Industry

Ismail¹,

Suhartono

Yahaya

et al. 2009

J. of Mathematics and Statistics

View full text Add to dashboard Cite

Problem statement: It is common in time series data with extreme change in its mean caused by an intervention which comes from external and/or internal factors. This extreme change in mean is known as regime change or structural change. Problems of external factor intervention such as the effect of the Arab oil embargo to consumption level of electricity in United States. The issue of interest here is on the impact of terrorist Bali's bomb to tourism industry in Indonesia. Approach: A theoretical and empirical studies on the intervention model, particularly pulse function of intervention is carried out focusing on the differential statistics that can be used to determine the order of intervention model. A case study of the first Bali bomb that occurred on October 12th, 2002 is an intervention of external factor that has affected the occupancy level of five star hotels in Bali. Results: The results of this theoretical study were applied to construct a model procedure of intervention model. The empirical study showed that intervention model is used to describe and explain the quantity and the length of the first Bali bomb effect towards the occupancy level of five star hotels in Bali. It shows a decreasing trend in tourist arrival in Bali, Indonesia. Conclusion: This study was focused on the derivation of some effect shapes, i.e., temporary, gradually or permanent on the arrival of tourist into Bali. A new model building procedure with three main iterative steps for determining an intervention model was used for data with extreme change in mean. The results from this theoretical study will give an opportunity for further research related to time series model that contains regime change, caused by intervention of pulse function and/or step function.

show abstract

An interrupted time series analysis of parenteral antibiotic use in Colombia

Cited by 29 publications

References 29 publications

Alternatives to randomisation in the evaluation of public-health interventions: statistical analysis and causal inference

Alternatives to randomisation in the evaluation of public-health interventions: statistical analysis and causal inference

Methods for estimating confidence intervals in interrupted time series analyses of health interventions

Intervention Model for Analyzing the Impact of Terrorism to Tourism Industry

Contact Info

Product

Resources

About