, , , , and ( 2017) Mortality in patients who discontinue low‐dose acetylsalicylic acid therapy after upper gastrointestinal bleeding. Pharmacoepidemiol Drug Saf, 26: 215– 222
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/27921360
Full text article: https://onlinelibrary.wiley.com/doi/abs/10.1002/pds.4140
| Methods | Study design: Retrospective cohort study
Number of groups: 1 |
| Participants | Number of participants: 547
Age: Not reported Sex: 207 female (38%) Participants with dementia: Unclear Inclusion criteria: · Experienced an upper gastro-intestinal bleed, and were taking low dose aspirin at the time (75-300mg/day) Exclusion criteria: Death within 30 days of the bleeding event Country: England Setting: Community |
| Interventions | Medicine: Aspirin
Withdrawal schedule: Not described |
| Outcomes | Mortality
Cessation of aspirin |
| Dates | Dates: Not stated – June 2012
Follow-up duration: 60 months |
| Funding sources | Study was funded in part by AstraZeneca and supported by Oxford PharmaGenesis (consultancy, writing support)
Investigators receive funding from AstraZeneca and Bayer Pharma |
| Notes | Age not reported |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | Low | Single arm observational study |
| Allocation concealment (selection bias) | Low | Single arm observational study |
| Blinding of participants and personnel (performance bias) | Unclear | Retrospective so unlikely patients and prescribers aware of the study |
| Blinding of outcome assessment (detection bias) | High | Outcome assessors aware of case details including aspirin cessation |
| Incomplete outcome data (attrition bias) | Unclear | Some data not collected during routine practice missing |
| Selective reporting (reporting bias) | Unclear | No protocol or pre-specified outcomes |
| Confounding (non-randomized) | High | Confounders discussed and accounted for in analysis |
| Other bias | High | Used pharmaceutical prescribing data, no ascertainment of over-the-counter aspirin use. Those who died within 30 days excluded, introducing survivorship bias. Industry funding. |
| Selection bias | Representativeness of the exposed cohort | Truly representative of patients taking aspirin at the time of an upper gastrointestinal bleed |
| Selection of the non-exposed cohort | No non-exposed cohort | |
| Ascertainment of exposure | Records | |
| Demonstration that outcome of interest was not present at start of study | Yes | |
| Comparability bias | Comparability of cohorts on the basis of the design or analysis | No comparison group |
| Outcome bias | Assessment of outcome | Record |
| Was follow-up long enough for outcomes to occur | Yes | |
| Adequacy of follow-up of cohorts | No statement | |
Leave a Reply