Abstract
Background: A core outcome set (COS) has been developed through international consensus to reduce selective reporting and research waste, and guide outcome choice in brief alcohol intervention research. This study aimed to estimate order effects among questions in the COS.
Methods: Adults, aged over 18 years who searched online for alcohol-related help were invited to complete a survey containing the COS items. The order of four item clusters was randomised following a factorial design. Primary outcomes were order effects among the COS items and patterns of abandonment of the questionnaire.
Results: We randomised 7334 participants, of which 5256 had responded to at least one question and were available for primary and sensitivity analyses. Current non-drinkers were excluded. Median completion time for the COS was 4 min 16 s. We found evidence of order effects among COS clusters, including higher self-reported average consumption and odds of harmful and hazardous drinking among those who first answered questions on recent consumption and impact of alcohol use. Lower self-reported recent consumption was found among those first asked about average consumption. Quality of life was reported as lower among those who first responded to questions on impact of alcohol use, which in turn was lower among those who first answered question on average consumption and quality of life. Attrition was lowest when average consumption was asked first, and highest when quality of life or impact of alcohol use asked first.
Conclusions: Researchers designing studies should note that question order effects may exist. At a minimum, all study participants should be asked the same questions in the same order. There is no perfect question order; rather, researchers should be guided by the nature of the studied population, recruitment, additional questions, concerns about under-reporting, screening for inclusion, and retention concerns.
Methods: Adults, aged over 18 years who searched online for alcohol-related help were invited to complete a survey containing the COS items. The order of four item clusters was randomised following a factorial design. Primary outcomes were order effects among the COS items and patterns of abandonment of the questionnaire.
Results: We randomised 7334 participants, of which 5256 had responded to at least one question and were available for primary and sensitivity analyses. Current non-drinkers were excluded. Median completion time for the COS was 4 min 16 s. We found evidence of order effects among COS clusters, including higher self-reported average consumption and odds of harmful and hazardous drinking among those who first answered questions on recent consumption and impact of alcohol use. Lower self-reported recent consumption was found among those first asked about average consumption. Quality of life was reported as lower among those who first responded to questions on impact of alcohol use, which in turn was lower among those who first answered question on average consumption and quality of life. Attrition was lowest when average consumption was asked first, and highest when quality of life or impact of alcohol use asked first.
Conclusions: Researchers designing studies should note that question order effects may exist. At a minimum, all study participants should be asked the same questions in the same order. There is no perfect question order; rather, researchers should be guided by the nature of the studied population, recruitment, additional questions, concerns about under-reporting, screening for inclusion, and retention concerns.
| Original language | English |
|---|---|
| Article number | O5 |
| Journal | Addiction Science and Clinical Practice |
| Volume | 18 |
| Issue number | Suppl 1 |
| DOIs | |
| Publication status | Published - 13 Apr 2023 |
| Event | 18th Annual Conference of the International Network on Brief Interventions for Alcohol and Other Drugs (INEBRIA) - Edinburgh, United Kingdom Duration: 21 Sept 2022 → 23 Sept 2022 https://inebria.net/meetings-and-activities/conference/past/18th-annual-conference-of-inebria-21st-23rd-september-2022-edinburgh-scotland/ |
