Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study
Abstract Background Aspirin has been considered to be beneficial in preventing cardiovascular diseases and cancer. Several pharmaco-epidemiology cohort studies have shown protective effects of aspirin on diseases using various statistical methods, with the Cox regression model being the most commonl...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2017-08-01
|
| Series: | BMC Medical Research Methodology |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12874-017-0405-6 |
| _version_ | 1817984042048421888 |
|---|---|
| author | In Sung Cho Ye Rin Chae Ji Hyeon Kim Hae Rin Yoo Suk Yong Jang Gyu Ri Kim Chung Mo Nam |
| author_facet | In Sung Cho Ye Rin Chae Ji Hyeon Kim Hae Rin Yoo Suk Yong Jang Gyu Ri Kim Chung Mo Nam |
| author_sort | In Sung Cho |
| collection | DOAJ |
| description | Abstract Background Aspirin has been considered to be beneficial in preventing cardiovascular diseases and cancer. Several pharmaco-epidemiology cohort studies have shown protective effects of aspirin on diseases using various statistical methods, with the Cox regression model being the most commonly used approach. However, there are some inherent limitations to the conventional Cox regression approach such as guarantee-time bias, resulting in an overestimation of the drug effect. To overcome such limitations, alternative approaches, such as the time-dependent Cox model and landmark methods have been proposed. This study aimed to compare the performance of three methods: Cox regression, time-dependent Cox model and landmark method with different landmark times in order to address the problem of guarantee-time bias. Methods Through statistical modeling and simulation studies, the performance of the above three methods were assessed in terms of type I error, bias, power, and mean squared error (MSE). In addition, the three statistical approaches were applied to a real data example from the Korean National Health Insurance Database. Effect of cumulative rosiglitazone dose on the risk of hepatocellular carcinoma was used as an example for illustration. Results In the simulated data, time-dependent Cox regression outperformed the landmark method in terms of bias and mean squared error but the type I error rates were similar. The results from real-data example showed the same patterns as the simulation findings. Conclusions While both time-dependent Cox regression model and landmark analysis are useful in resolving the problem of guarantee-time bias, time-dependent Cox regression is the most appropriate method for analyzing cumulative dose effects in pharmaco-epidemiological studies. |
| first_indexed | 2024-04-13T23:40:24Z |
| format | Article |
| id | doaj.art-046be2984b1c493cb8d08513b33c908c |
| institution | Directory Open Access Journal |
| issn | 1471-2288 |
| language | English |
| last_indexed | 2024-04-13T23:40:24Z |
| publishDate | 2017-08-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Research Methodology |
| spelling | doaj.art-046be2984b1c493cb8d08513b33c908c2022-12-22T02:24:31ZengBMCBMC Medical Research Methodology1471-22882017-08-011711710.1186/s12874-017-0405-6Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation studyIn Sung Cho0Ye Rin Chae1Ji Hyeon Kim2Hae Rin Yoo3Suk Yong Jang4Gyu Ri Kim5Chung Mo Nam6Yonsei University, College of MedicineYonsei University, College of MedicineDepartment of Biostatistics and Medical Informatics, Yonsei University College of MedicineDepartment of Biostatistics and Medical Informatics, Yonsei University College of MedicineDepartment of Preventive Medicine, Eulji University College of MedicineDepartment of Biostatistics, Graduate School of Public health, Yonsei UniversityDepartment of Biostatistics and Medical Informatics, Yonsei University College of MedicineAbstract Background Aspirin has been considered to be beneficial in preventing cardiovascular diseases and cancer. Several pharmaco-epidemiology cohort studies have shown protective effects of aspirin on diseases using various statistical methods, with the Cox regression model being the most commonly used approach. However, there are some inherent limitations to the conventional Cox regression approach such as guarantee-time bias, resulting in an overestimation of the drug effect. To overcome such limitations, alternative approaches, such as the time-dependent Cox model and landmark methods have been proposed. This study aimed to compare the performance of three methods: Cox regression, time-dependent Cox model and landmark method with different landmark times in order to address the problem of guarantee-time bias. Methods Through statistical modeling and simulation studies, the performance of the above three methods were assessed in terms of type I error, bias, power, and mean squared error (MSE). In addition, the three statistical approaches were applied to a real data example from the Korean National Health Insurance Database. Effect of cumulative rosiglitazone dose on the risk of hepatocellular carcinoma was used as an example for illustration. Results In the simulated data, time-dependent Cox regression outperformed the landmark method in terms of bias and mean squared error but the type I error rates were similar. The results from real-data example showed the same patterns as the simulation findings. Conclusions While both time-dependent Cox regression model and landmark analysis are useful in resolving the problem of guarantee-time bias, time-dependent Cox regression is the most appropriate method for analyzing cumulative dose effects in pharmaco-epidemiological studies.http://link.springer.com/article/10.1186/s12874-017-0405-6Cox regressionGuarantee-time biasLandmark methodTime-dependent Cox regression |
| spellingShingle | In Sung Cho Ye Rin Chae Ji Hyeon Kim Hae Rin Yoo Suk Yong Jang Gyu Ri Kim Chung Mo Nam Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study BMC Medical Research Methodology Cox regression Guarantee-time bias Landmark method Time-dependent Cox regression |
| title | Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study |
| title_full | Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study |
| title_fullStr | Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study |
| title_full_unstemmed | Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study |
| title_short | Statistical methods for elimination of guarantee-time bias in cohort studies: a simulation study |
| title_sort | statistical methods for elimination of guarantee time bias in cohort studies a simulation study |
| topic | Cox regression Guarantee-time bias Landmark method Time-dependent Cox regression |
| url | http://link.springer.com/article/10.1186/s12874-017-0405-6 |
| work_keys_str_mv | AT insungcho statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT yerinchae statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT jihyeonkim statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT haerinyoo statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT sukyongjang statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT gyurikim statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy AT chungmonam statisticalmethodsforeliminationofguaranteetimebiasincohortstudiesasimulationstudy |