Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method
Background: The usual-frequency case-crossover method, comparing exposure before an event with typical exposure of the same person, is widely used to estimate the risk of injury related to acute alcohol use. Prior results suggest that risk estimates might be biased upward compared with other methods...
| Main Authors: | , , , , , |
|---|---|
| Format: | Journal Article |
| Published: |
2013
|
| Online Access: | http://hdl.handle.net/20.500.11937/14911 |
| _version_ | 1848748749897072640 |
|---|---|
| author | Ye, Y. Bond, J. Cherpitel, C. Stockwell, Tim MacDonald, S. Rehm, J. |
| author_facet | Ye, Y. Bond, J. Cherpitel, C. Stockwell, Tim MacDonald, S. Rehm, J. |
| author_sort | Ye, Y. |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | Background: The usual-frequency case-crossover method, comparing exposure before an event with typical exposure of the same person, is widely used to estimate the risk of injury related to acute alcohol use. Prior results suggest that risk estimates might be biased upward compared with other methods. Methods: Using data from 15 emergency room studies in seven countries, we compared the usual-frequency case-crossover method with case-control analysis, using noninjury patients as controls. Control-crossover analysis was performed to examine potential bias and to adjust risk estimates. Results: The cross-study pooled odds ratio (OR) of injury related to drinking was 4.7 (95% confidence interval = 2.6–8.5) in case-crossover analysis and 2.1 (1.6–2.7) in case-control analysis. A control-crossover analysis found an indication of bias (OR = 2.2 [1.8–2.8]), which was larger among less-frequent drinkers. Conclusion: Findings suggest that the potential overestimation of injury risk based on the usual-frequency case-crossover method might be best explained by recall bias in usual-frequency estimates. |
| first_indexed | 2025-11-14T07:10:00Z |
| format | Journal Article |
| id | curtin-20.500.11937-14911 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T07:10:00Z |
| publishDate | 2013 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-149112018-03-29T09:06:09Z Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method Ye, Y. Bond, J. Cherpitel, C. Stockwell, Tim MacDonald, S. Rehm, J. Background: The usual-frequency case-crossover method, comparing exposure before an event with typical exposure of the same person, is widely used to estimate the risk of injury related to acute alcohol use. Prior results suggest that risk estimates might be biased upward compared with other methods. Methods: Using data from 15 emergency room studies in seven countries, we compared the usual-frequency case-crossover method with case-control analysis, using noninjury patients as controls. Control-crossover analysis was performed to examine potential bias and to adjust risk estimates. Results: The cross-study pooled odds ratio (OR) of injury related to drinking was 4.7 (95% confidence interval = 2.6–8.5) in case-crossover analysis and 2.1 (1.6–2.7) in case-control analysis. A control-crossover analysis found an indication of bias (OR = 2.2 [1.8–2.8]), which was larger among less-frequent drinkers. Conclusion: Findings suggest that the potential overestimation of injury risk based on the usual-frequency case-crossover method might be best explained by recall bias in usual-frequency estimates. 2013 Journal Article http://hdl.handle.net/20.500.11937/14911 10.1097/EDE.0b013e3182801cb4 restricted |
| spellingShingle | Ye, Y. Bond, J. Cherpitel, C. Stockwell, Tim MacDonald, S. Rehm, J. Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title | Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title_full | Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title_fullStr | Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title_full_unstemmed | Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title_short | Risk of injury due to alcohol: Evaluating potential bias using the case-crossover usual-frequency method |
| title_sort | risk of injury due to alcohol: evaluating potential bias using the case-crossover usual-frequency method |
| url | http://hdl.handle.net/20.500.11937/14911 |