Impact statements

  • Inadequate reporting of healthcare interventions in the scientific literature is a long-standing issue and hinders the potential for replication and implementation of evidence-based interventions in clinical practice.

  • To date, no studies have been conducted into the completeness of reporting of antidepressant tapering methods in published trials.

  • The study highlights a lack of detailed reporting of antidepressant tapering methods in published trials to date.

  • These findings highlight the need for intervention reporting tools to be used by researchers in the write up of future studies to ensure tapering methods are adequately described, enabling their replication, adaptation and translation into practice.

Introduction

Approximately 11% of the population take antidepressant medication daily, and the number of antidepressant prescriptions has almost doubled over the last decade [1, 2]. A key driving factor behind increased antidepressant prescribing rates is increased treatment duration [3]. There are growing concerns that antidepressants may be over-prescribed or prescribed for excessive durations. For example, it has been reported that 30–60% of long-term antidepressant prescriptions lack an evidence-based indication to support continued use and could potentially be deprescribed [4].

The importance of tapering is increasingly recognised when discontinuing antidepressant medication [5, 6]. This involves gradually reducing the dose over time to minimise the potential for withdrawal symptoms. However, guidance and instructions on tapering antidepressants varies [7]. In order for patients and healthcare professionals to be able to implement evidence-based tapering interventions, it is important that sufficient detail is reported in published study reports. However, a lack of detailed descriptions of healthcare interventions has been an issue in the scientific literature to date which hampers the potential for replication and can lead to research waste [8].

The Template for Intervention Description and Replication (TIDieR) checklist was developed to facilitate comprehensive reporting of interventions [9]. The checklist consists of 12 items considered essential to enable the replication of interventions (Table 1). The tool can also be used to evaluate the completeness of reporting in published studies included in systematic reviews [10, 11]. A recent Cochrane review examined the effectiveness and safety of approaches for discontinuing long-term antidepressant use [12]. However, the reporting of antidepressant tapering methods in included studies was not reviewed in detail.

Table 1 TIDieR checklist for assessment of intervention reporting
Full size table

Aim

The aim of this study was to assess the completeness of reporting of tapering methods in antidepressant discontinuation trials in a published systematic review, using the TIDieR checklist.

Ethics approval

The study involved application of the TIDieR checklist [9] to studies included in a systematic review and, therefore, ethical approval was not required.

Method

A secondary analysis was conducted of studies identified in a Cochrane systematic review of randomised controlled trials (RCTs) comparing the effectiveness and safety of approaches for discontinuing long-term antidepressant use (> 6 months) versus continuation of the medication [12]. The review included 33 studies comprising interventions involving abrupt discontinuation (n = 13 studies), discontinuation involving tapering alone (n = 18 studies), discontinuation with minimal intervention (n = 1 studies), and discontinuation with psychological support (n = 4 studies).

For the current study, full-text articles of all studies included in the systematic review were obtained and independently reviewed by two researchers (AMcG, HB) to identify studies that implemented a tapering regimen as part of the discontinuation intervention. This was supplemented by a search for publications of studies that were classified as ‘ongoing’ in the original review and had since been completed. Using the TIDieR checklist (Table 1), the researchers independently extracted data from the intervention group in each of the identified studies to assess the completeness of reporting. In studies where the tapering intervention was supplemented by a co-intervention (e.g. psychological support), the TIDieR assessments focused specifically on the tapering component of the intervention. Each TIDieR checklist item was independently rated by the researchers as either ‘reported, ‘not reported’, or ‘unclear’ using the explanations provided for each item in the checklist. No modifications were made to the original checklist items. However, where studies made no reference to an assessment of adherence to tapering (item 11), item 12 was marked as ‘not applicable’. Study authors were not contacted for missing information as the focus of this assessment was the completeness of reporting in published study reports. Any disagreements between the researchers throughout the data extraction/assessment process were resolved through discussion and, consultation with a third researcher (CC) who assisted with reviewing and finalising all checklist assessments.

Cohen’s kappa (κ) co-efficient of inter-rater reliability was used to assess the level of agreement between the two independent assessors in applying the TIDieR checklist to each study prior to consensus discussion (< 0 = poor agreement; 0–0.20 = slight agreement; 0.21–0.40 = fair agreement; 0.41–0.60 = moderate agreement; 0.61–0.80 = substantial agreement; 0.81–1.00 = almost perfect agreement) [13]. The reported kappa co-efficient was calculated based on the degree of agreement between both assessors across all TIDieR checklist item assessments across included studies.

Results

Study characteristics

Twenty-two trials from the aforementioned Cochrane review implemented a tapering regimen as part of the discontinuation intervention and provided data for analysis [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35], including one study that was originally classified as ‘ongoing’ and published subsequently [18].

Table 2 provides an overview of the characteristics of studies included in the current analysis. Briefly, these studies were published between 1982 and 2021. Study sample size ranged from 18 to 478 patients. Fourteen studies focused on a single antidepressant or class of antidepressants: serotonin and noradrenaline re-uptake inhibitors [SNRIs] (6 studies), selective serotonin re-uptake inhibitors [SSRIs] (1 study) and tricyclic antidepressants [TCAs] (7 studies). Sixteen studies involved the use of placebo substitution. Most studies (19 studies) reported taper durations ranging from 1 to 8 weeks. The most commonly reported taper duration was four weeks (10 studies). Eight studies provided details on tapering rates with reductions typically ranging from 25 to 50% at weekly or monthly intervals. Recurrence/relapse was the most commonly evaluated primary outcome (20 studies).

Table 2 Overview of study characteristics
Full size table

TIDieR reporting

A summary of the finalised results from the TIDieR assessments is shown in Fig. 1, with a breakdown for each individual study provided in Table 3.

Fig. 1
figure 1

Summary overview of TIDieR checklist assessments

Full size image
Table 3 TIDieR checklist assessments for individual studies (n = 22)
Full size table

None of the study reports described all checklists items. No study clearly reported what materials had been provided (item 3) or whether tailoring had occurred (item 9). With the exception of providing a name for the intervention or study procedures (item 1), only a minority of studies clearly reported on any of the remaining checklist items.

Only three studies provided a clear rationale for tapering and, where applicable, the use of placebo as part of the tapering process (item 2). Apart from referring to the use of placebo medication, studies provided limited details on the use of any additional materials that were provided to patients (item 3). In cases where a tapering resource or information letter was referred to, no information was provided on where these materials could be accessed.

Eight studies clearly reported on the tapering regime (item 4) in terms of both the taper rate and duration. Seven studies reported details of the healthcare providers involved in delivering the intervention (item 5), which were typically general practitioners (GPs) and/or psychiatrists. Only one study clearly reported on how the intervention was delivered (item 6) whereby a hospital pharmacy posted the medication to either patients’ home or GP surgery at fixed intervals. Only two studies clearly reported on where the intervention was delivered (item 7). Five studies reported on how frequently patients were seen during the tapering process (item 8) which was typically monthly. Only one study reported details of a modification to the intervention (item 10), whereby the tapering process was conducted over six months as opposed to the recommended four-week timeframe as most patients did not find the latter to be feasible. Two studies described methods for assessing adherence to study procedures (item 11) and a further six studies reported on the number of participants that adhered to the tapering regime without detailing how adherence was assessed (item 12).

Inter-rater reliability of TIDieR scoring

Cohen’s kappa was 0.452, indicating moderate level of agreement between both independent applications of the checklist [13]. On further review, it was found that most of the differences between the independent assessors occurred in the scoring of items as ‘unclear’ versus ‘no’ due to the lack of detail in study reports. Merging these assessments resulted in substantial agreement between the assessors (κ = 0.72).

Discussion

Statement of key findings

This study assessed the completeness of antidepressant tapering interventions in studies included in a Cochrane review [12] using the TIDieR checklist and found that the descriptions of tapering methods in published studies lacked detail. This could negatively impact the potential for replication in future studies and implementation of effective tapering interventions in clinical practice.

Strengths and limitations

To our knowledge, this is the first study to assess the completeness of reporting of antidepressant tapering methods in randomised controlled trials. The TIDieR checklist was a useful tool for the assessment of intervention reporting as it highlighted specific aspects of antidepressant tapering methods which require more detailed descriptions in future studies. Each of the studies was assessed by two assessors working independently. Although the initial inter-rater reliability assessment was moderate, this was found to be largely due to the lack of detailed reporting in study reports and the associated challenge of categorising checklist assessments as ‘not reported’ or ‘unclear’. It must be noted that the scope of this analysis was limited to the reporting of tapering interventions in studies included in a single Cochrane review that focused on antidepressants and with the exception of searches for publications of studies that were classified as ‘ongoing’ in the original review, no updated searches were performed. It is possible that other intervention arms in the included studies may have been reported in greater detail. Finally, most of the included studies were published before the TIDieR checklist was made available.

Interpretation

The subject of antidepressant tapering is receiving increasing attention [5]. However, the combined findings of both the original review and the current analysis highlight considerable deficits with the existing evidence base. As noted in the original review, based on the rather limited body of available evidence, no firm conclusions could be made about the effectiveness and safety of the approaches studied to date for discontinuing long-term antidepressant use [12]. The current analysis highlights the challenges in learning from and building on the existing evidence base regarding tapering approaches due to considerable deficits in the completeness of intervention reporting. Nevertheless, the findings provide some interesting insights into previously evaluated antidepressant tapering methods.

Guidance on antidepressant withdrawal symptoms and tapering methods has changed considerably over time. For example, antidepressant withdrawal symptoms were previously described as mild and self-limiting, typically lasting only 1–2 weeks [36, 37]. However, this is increasingly refuted, and it is now recognised that many patients experience symptoms for extended periods after discontinuing antidepressants ranging from weeks to months, and even longer in some cases [38, 39]. This highlights the importance of tailoring tapering approaches according to individual needs, which none of the included studies adequately reported on. The topic of appropriate tapering rates and durations is the subject of much discussion. Whereas previously it was thought that antidepressants could be tapered in fixed linear increments over 2–4 weeks, as was the case in many of the included studies, this is no longer considered the correct approach [3, 5]. Hyperbolic tapering regimens are now increasingly being recommended for antidepressant discontinuation in order to achieve linear reductions in pharmacological effect and mitigate against potential withdrawal symptoms [40, 41]. However, none of the included studies reported adopting such an approach and therefore further evaluations of hyperbolic tapering regimens are required.

The use of placebo substitution as part of the tapering intervention was common across included studies. However, only one study clearly reported a rationale for including a placebo which related directly to the study’s aims of evaluating the pharmacological effect of maintenance antidepressant treatment in preventing relapse of depression [18]. As most of these studies involved double blind conditions, the feasibility and acceptability of a tapering approach involving placebo substitution in real-world clinical practice is unclear. The use of placebos in clinical practice raises important ethical questions, particularly in terms of the potential for patient deception and loss of autonomy in the treatment decision process [42]. Open-label placebos (i.e. placebos without deception) have been proposed as an alternative to standard placebo approaches [43, 44]. However, very few studies have examined the use of open-label placebo medication in treating depression to date [45]. The role of open-label placebo as part of antidepressant tapering regimens also requires further investigation.

In addition to the issues outlined above with the reported tapering methods, very limited information was provided across included studies in terms of any additional tapering resources and supports that were provided to study patients or healthcare professionals. This is important to address as the lack of available practical evidence-based resources means that patients are increasingly turning to the internet and online discussion forums for tapering-related advice and support [46,47,48]. Healthcare professionals have also reported a lack of support and practical guidance as barriers to assisting patients with discontinuing long-term antidepressant use [49, 50]. The inclusion of intervention resources in published reports of studies targeting other psychotropic medication, such as benzodiazepines, has enabled replication and adaptation of existing interventions in other studies [51, 52].

Finally, examining and reporting on the intervention’s implementation, such as fidelity to the intervention/study procedures and whether any modifications occurred is also critical to providing insights into any issues with intervention adherence or deviations from study protocols. For example, in the only study to report on a modification to the tapering intervention, it was found that a four-week taper was not feasible for patients and that a longer taper duration was required [15]. These types of insights provide important learnings for future researchers. Future studies of tapering interventions may wish to consider the inclusion of a process evaluation which can provide detailed insights into an intervention’s implementation, as well other relevant factors such as the intervention’s mechanism of impact and the context within which it is delivered [53].

Further research

Further research is needed into optimal antidepressant tapering methods and researchers should ensure that future interventions are adequately reported to enable replication. This should include all aspects of a tapering plan, including the practical methods of achieving increasingly smaller doses in real world settings (e.g. pill cutting, liquid formulations) particularly in cases where a hyperbolic tapering regime is being used. Future research should also consider the completeness of reporting of tapering interventions for other classes of psychotropic medication (e.g. antipsychotics, benzodiazepines).

Conclusion

The findings of this analysis highlight the need for adequate reporting of antidepressant tapering methods in future intervention studies. The lack of detailed reporting of tapering methods in research to date could ultimately contribute to research waste and delay advancements in this area as poorly reported interventions hinder replication and the potential for successful translation of effective tapering interventions into clinical practice. Future studies on antidepressant discontinuation could benefit from the use of reporting tools to ensure that details of tapering interventions are adequately reported.