Myers MG, Weingert ME, Fisher RH, et al. Unnecessary diuretic therapy in the elderly. Age and Ageing 1982;11(4):213-21.
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/7180724
| Methods | Study design: Randomized double-blind placebo-controlled study with a parallel-design
Number of groups: Two groups |
| Participants | Number of participants: 77 randomized
· Active deprescribing group: 38 randomized · Control group: 39 randomized Age, females: 84.5 years Age, males: 79.1 years Sex: 17 female, 60 male Participants with dementia: Yes, the number of participants with dementia and the severity was not noted Inclusion criteria: · Resident at one of the two participating residential aged care facilities · Taking diuretics Exclusion criteria: · Concurrent digoxin therapy · On diuretic therapy less than 3 months · Uncooperative/no consent · ‘Active’ hypertension · ‘Active’ heart failure · Death before entry · Terminal illness unrelated to diuretics · Under age 60 Country: Canada Setting: Residential aged care facilities – two |
| Interventions | Medicine: Diuretics
Withdrawal schedule: Not described Comparator: Placebo or continued treatment |
| Outcomes | Hypertension
Congestive heart failure Biochemical abnormalities Weight Ankle edema Events |
| Dates | Dates: Not described
Follow-up duration: One year |
| Funding sources | Ontario Heart Foundation |
| Notes | Authors statement of conflict of interest: “Dr. M. G. Myers is a Senior Research Fellow of the Ontario Heart Foundation.” |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | Unclear risk | Randomization stated, but method not described.
“Subjects were randomly allocated to either continued diuretic therapy or placebo and were followed at regular intervals for one year.” |
| Allocation concealment (selection bias) | Unclear risk | Method not described. Probably done.
“The study design was double-blindand the randomization code was kept in the Department of Pharmacy.” |
| Blinding of participants and personnel (performance bias) | Low risk | Probably done. The placebo was described.
“Matching placebo tablets were available for each diuretic, Slow K and potassium chloride solution. The potassium chloride solution did not have the same taste as the active compound otherwise the placebo and active tablets were identical. None of the three subjects on placebo potassium chloride commented on a change in therapy during the study. “The study design was double-blind…” |
| Blinding of outcome assessment (detection bias) | Low risk | As above |
| Incomplete outcome data (attrition bias) | Low risk | Intention-to-treat analysis
The possibility of a ‘survivor effect’ was examined by combining the values for each subject completing the full 12 months with the final values of subjects dying or experiencing an end-point |
| Selective reporting (reporting bias) | High risk | It is unclear if there was a pre-specified protocol and data-analysis plan.
The primary end-points in the study were the development of hypertension or congestive heart failure and changes in mean blood pressure levels in the placebo and diuretic treated groups. These outcomes were not reported per se. Most measurements allowing judgments to be made were reported. The discussion makes statements based on ranging outcome measures, so it is hard to say if the specified outcome measures were actually specified before the study started or after the data were gathered |
| Other bias | Low risk |
Leave a Reply