Puustinen J, Lahteenmaki R, Polo-Kantola P, et al. Effect of withdrawal from long-term use of temazepam, zopiclone or zolpidem as hypnotic agents on cognition in older adults. European Journal of Clinical Pharmacology 2014;70:319-29.
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/24337417
Full text article: https://link.springer.com/article/10.1007%2Fs00228-013-1613-6
Lähteenmäki R, Puustinen J, Vahlberg T, et al. Melatonin for sedative withdrawal in older patients with primary insomnia: a randomized double-blind placebo-controlled trial. British Journal of Clinical Pharmacology 2014;77(6):975-85.
PubMed link: https://www.ncbi.nlm.nih.gov/pubmed/24286360
Full text article: https://bpspubs.onlinelibrary.wiley.com/doi/full/10.1111/bcp.12294
| Methods | Study design: Historical cohort study
Number of groups: Two groups |
| Participants | Number of participants: 89 enrolled, 89 completed
Age: 66.7 ± 7.2 years Sex: 59 female, 30 male Participants with dementia: No Inclusion criteria: · Men and women · Aged 55 years or older · Long-term users of benzodiazepine as hypnotics, defined as 1 month or longer regular night-time use. · Benzodiazepine used as hypnotics must be temazepam, zopiclone or zolpidem · Must have been prescribed according to DSM-IV criteria for primary insomnia Exclusion criteria: · Concurrent use of antipsychotic or anti-epileptic medications · Use of a benzodiazepine other than temazepam, zopiclone or zolpidem · History of, or active alcohol or drug abuse · Severe anxiety disorder or other severe psychiatric disorder · Severe neurological disease · Smoking more than ten cigarettes a day · Autoimmune disease · Galactosaemia · Use of medication that potentially interacts with melatonin Concurrent medicines: median 4.0 (LQ 3.0, UQ 5.0) Country: Finland Setting: Community – Participants were primary health care outpatients living in the Province of Satakunta, in western Finland. The trial was performed in the City of Pori, at the Medical Teaching and Research Health Center of the Department of Family Medicine, University of Turku. |
| Interventions | Medicine: Benzodiazepines (temazepam=9, zolpidem n=14, zopiclone n=23)
Duration of regular benzodiazepine use: less than five years n=9, five to 10 years n=17, more than 10 years n=20 Withdrawal Schedule: Withdrawn over one month – replaced either with melatonin or a placebo Comparator: Results from a previous study of cognitive performance in a non-insomniac group of Finnish females who did not use benzodiazepines |
| Outcomes | Cognitive performance using the computerized test battery of attention, vigilance and controlled psychomotor processing |
| Dates | Study dates: Not described
Follow-up duration: Six months |
| Funding sources | Satakunta Hospital District Grant EVO, Härkätie Grant EVO, Southwest Finland Hospital District Grant EVO, Finnish Cultural Foundation and Pharmaceutical Company Nycomed. |
| Notes | Puustinen 2014 and Lähteenmä 2013 both report the same study |
Risk of bias table
| Bias | Authors’ judgment | Support for judgment |
| Random sequence generation (selection bias) | High risk | Not randomized |
| Allocation concealment (selection bias) | High risk | Allocation not concealed |
| Blinding of participants and personnel (performance bias) | 5 out of 5 | No.
The deprescribing group comprised participants in a randomized, double-blind trial, so the participants were blinded during that study. However, the control cohort was un-blinded. |
| Blinding of outcome assessment (detection bias) | 5 out of 5 | The Satauni benzodiazepine withdrawal study was blinded however this study was open to personnel and assessors.
The outcome assessors and personnel were aware of the members of the control cohort for this second phase of the study. |
| Incomplete outcome data (attrition bias) | 1 out of 5 | No missing data apparent.
The Puustinen study does not describe drop-outs or withdrawals. Lahteenmaki (2014) paper does describe withdrawals and drop-outs. As the Puustinen paper is the paper directly relevant to this study, it is considered high risk that the paper does not account for these drop-outs and is as-treated analysis. |
| Selective reporting (reporting bias) | 5 out of 5 | All described outcomes are reported.
The study does not have a pre-published protocol. However, the stated objectives are all reported. |
| Confounding (non-randomized) | 5 out of 5 | Lifestyle factors are not balanced between the groups. Similarly, other medications besides Z-drugs were not considered.
The two groups are not entirely comparable at baseline. For example, the comparison cohort was 100% female compared to active cohort was 66% female. The two groups are separately drawn from two unrelated clinical trials: · The first study examined the effect of hormone replacement therapy and one studying the effect of substituting melatonin for benzodiazepines. The deprescribing group had education, lifestyle and behavioral modification as well as the deprescribing. · The second study was focused on menopausal women, and as women may experience sleep disturbances during menopause, any outcomes directly relating to sleep are non-comparable. |
| Other bias | High risk | This study has pooled the data from the two groups (benzodiazepine withdrawal substituted with placebo compared to benzodiazepine withdrawal substituted with melatonin 2mg), rather than keep the data from those two groups independent.
“As melatonin 2 mg did not improve benzodiazepine withdrawal results and as no adverse cognitive effects related to melatonin have been detected in older adults (≥55 years), our previously unpublished cognition data from the melatonin and placebo groups were pooled to study the effect of benzodiazepine withdrawal on cognition. |
| Newcastle-Ottawa scale | ||
| Selection bias | Representativeness of the exposed cohort | Selected group of users. |
| Selection of the non-exposed cohort | No concurrent control group. | |
| Ascertainment of exposure | Ascertainment of exposure by written self-report. | |
| Demonstration that outcome of interest was not present at start of study | Demonstrated that outcome of interest was not present at the start of the study. | |
| Comparability bias | Comparability of cohorts on the basis of the design or analysis | The study controlled for withdrawal of hypnotics. |
| Outcome bias | Assessment of outcome | Assessment of outcomes by record linkage. |
| Was follow-up long enough for outcomes to occur
|
Follow-up was long enough for outcomes to occur.
|
|
| Adequacy of follow-up of cohorts | Follow-up unclear – and not reported numerically in the Puustinen paper. | |
Leave a Reply