Fahy S, Lawlor B. Discontinuation of lithium augmentation in an elderly cohort. International Journal of Geriatric Psychiatry 2001;16:1004-09.
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/11607947
Full text article: https://onlinelibrary.wiley.com/doi/epdf/10.1002/gps.453
| Methods | Study design: Case-controlstudy
Number of groups: Two groups |
| Participants | Number of participants: 21 enrolled
· Active deprescribing successful (no relapse) group: 10 participants · Active deprescribing unsuccessful (relapse) group: 11 participants Age: 77.6 ± 6.56 years (range 66 to 90 years) Sex: 20 female, 1 male Participants with dementia: No Inclusion criteria: · Major depressive disorder · Single or recurrent episodes · With or without psychiatricfeatures · Considered clinically “well” before discontinuation (well was not defined) Exclusion criteria: · Relapse in the past following discontinuation of lithium augmentation. · Index episode (event that initially warranted addition of lithium to their antidepressant regime) was considered of such a severity to suggest the ongoing need for long-term prophylaxis with both antidepressant and lithium Concomitant medicines: All took a variety of different antidepressants (from all the various antidepressant groups) Country: Ireland Setting: Community – Outpatients at a lithium clinic at a psychiatric for elderly service attached to a capital city hospital |
| Interventions | Medicine: Lithium
Withdrawal schedule: Lithium dose was tapered gradually over a period of 2 to 12 weeks Comparator: Participants who relapsed were compared to participants who did not relapse after deprescribing |
| Outcomes | Time to relapse or follow-up time
Response to re-introduction of therapy |
| Dates | Dates: Not described
Follow-up duration: 19.5 ± 6.9 months |
| Funding sources | Not described |
| Notes | The paper compared the characteristics of people who had relapsed to those who had not relapsed after lithium had been discontinued. |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | High risk | Case-controlstudy – participants,were those from a clinic who had their medication discontinued. Allocated to one of two groups depending on whether they had relapsed or whether their medication had been successfully withdrawn. |
| Allocation concealment (selection bias) | High risk | Retrospective analysis. |
| Blinding of participants and personnel (performance bias) | 5 out of 5 | Retrospective analysis of patient files. |
| Blinding of outcome assessment (detection bias) | 5 out of 5 | As above. |
| Incomplete outcome data (attrition bias) | 1 out of 5 | There does not appear to be any missing data. |
| Selective reporting (reporting bias) | 5 out of 5 | This study is a retrospective review and published pre-specified outcomes are not available as it is a “naturalistic study” to review participants’ outcomes from clinical practice. |
| Confounding (non-randomized) | 5 out of 5 | No control group available, rather a comparison of those who did and did not relapse after medication discontinuation. |
| Other bias | Low risk | |
| Newcastle-Ottawa scale | ||
| Selection bias | Is the case definition adequate? | Yes. |
| Representativeness of the exposed cohort | The exposed cohort appears to a somewhatrepresentative of older adults currently in remission after a single depressive episode currently under regular medical care.
|
|
| Selection of the non-exposed cohort | Community controls: the control group appears to be drawn from the same community as the intervention group. | |
| Definition of Controls | No history of disease. | |
| Comparability bias | Comparability of cohorts on the basis of the design or analysis | The studycontrols for remission after deprescribing lithium. |
| Outcome bias | Assessment of outcome | No description of the method used to assess outcomes. |
| Same method of ascertainment for cases and controls | Follow-upduration was probably long enough for outcomes to occur. | |
| Non-Response rate | The non-response rate of the two groups were comparable. | |
Leave a Reply