Salzman C, Fisher J, Nobel K, et al. Cognitive improvement following benzodiazepine discontinuation in elderly nursing home residents. International Journal of Geriatric Psychiatry 1992;7:89-93.
Full text article: https://onlinelibrary.wiley.com/doi/abs/10.1002/gps.930070205
| Methods | Study design: Prospective cohort study
Number of groups: Two groups |
| Participants | Number of participants: 25 enrolled
Intervention deprescribing group: 13 enrolled · Residents whose physicians consented to benzodiazepine deprescribing were the active group. Control group: 12 enrolled · Residents whose physician did not consent to benzodiazepine deprescribing were the control group. Age: 83 years Sex: 20 female, 5 male Participants with dementia: all · Mild: all participants complained of forgetfulness – not elaborated to see if this is mild cognitive impairment or was a diagnosis of dementia Inclusion criteria: · None mentioned Exclusion criteria: · Moderate-to-severe dementia Concurrent medicines: 6.1 medicines per person, not including the benzodiazepine Country: United States of America Setting: Residential aged care facilities |
| Interventions | Medicine: Benzodiazepine (triazolam 0.125 mg, temazepam 30 mg, clonazepam 0.5 mg, lorazepam 0.5 mg, oxazepam 10 mg, oxazepam 30 mg)
Withdrawal schedule: There was no uniform schedule of tapering, nor were blood levels obtained. All subjects were discontinued from benzodiazepines after two weeks of gradual tapering without obvious discomfort. |
| Outcomes | Memory
Dementia Mood Assessment Scale to measure changes in sleep and affect (depression and anxiety) Successful deprescribing |
| Dates | Dates: Not stated
Follow-up duration: One year |
| Funding sources | Not described |
| Notes |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | High risk | Residents were allocated based on their physician’s recommendation.
“All residents of a nursing home who were taking benzodiazepines were identified, and their clinicians were asked whether or not discontinuation of benzodiazepines was clinically appropriate and feasible. Residents whose physicians agreed to benzodiazepine discontinuation and who were willing to give consent to discontinuation became the experimental group ( N = 13). Those whose physicians did not agree, or who did not give consent, became the control group ( N = 12).” |
| Allocation concealment (selection bias) | High risk | The patients knew whether they were being discontinued, but the test administrators were blind to whether benzodiazepines had been discontinued. |
| Blinding of participants and personnel (performance bias) | 5 out of 5 | “The patients knew whether or not they were being discontinued.” |
| Blinding of outcome assessment (detection bias) | 3 out of 5 | Not clear how blinding was achieved.
” the test administrators were blind to whether benzodiazepines had been discontinued.” |
| Incomplete outcome data (attrition bias) | Unclear risk | Unclear how many people withdrew from the study, if any, and for what reasons.
No statement was made about drop-outs. The implication was that all subjects completed the study. |
| Selective reporting (reporting bias) | 4 out of 5 | Unclear whether the researchers had a pre-specified protocol and analysis plan.
The stated objectives are reported. However, there is limited detail and no pre-published protocol. “Because of the preliminary nature of this study, ratings from the Dementia Mood Assessment Scale were available only for 11 discontinuers and five continuers.” First, did memory improve following benzodiazepine discontinuance? Both outcomes were reported,but the results were not complete. Only 16 of 25 subjects had the Dementia Mood Assessment Scale done |
| Confounding (non-randomized) | 5 out of 5 | All reasons for benzodiazepine prescription were lumped together. It seems likely that the results of this method would be affected by underlying illness. Those patients more likely to suffer adverse effects in the opinion of their prescriber were excluded from the study. |
| Other bias | Low risk | Limited detail. |
| Newcastle-Ottawa scale | ||
| Selection bias | Representativeness of the exposed cohort | Somewhat representative of older in-patient adults prescribed benzodiazepines. |
| Selection of the non-exposed cohort | The non-exposed cohort was drawn from a different population. | |
| Ascertainment of exposure | Ascertainment of exposure was by secure record.
|
|
| Demonstration that outcome of interest was not present at start of study | Demonstrated that outcome of interest was not present at start of study. | |
| Comparability bias | Comparability of cohorts on the basis of the design or analysis | The study controlled for medical practitioner’s opinion of whether the benzodiazepine can be withdrawn safely and feasibly. |
| Outcome bias | Assessment of outcome | Assessment of outcome was by independent blind assessment. |
| Was follow-up long enough for outcomes to occur | Follow-up may have been long enough for outcomes to occur.
|
|
| Adequacy of follow-up of cohorts | No statement on the Adequacy of follow-up of cohorts. | |
Leave a Reply