Development of a Brief Mindfulness-Informed Cognitive Behavioral Therapy Intervention to Pair with Pharmacist-Led Benzodiazepine Tapering for Older Adults: The CSTARS Intervention

Abstract

Benzodiazepines are potentially inappropriate medications for older adults and deprescribing interventions are needed. We describe the development of a psychologist-led, mindfulness-informed cognitive behavioral therapy (CBT) intervention to pair with pharmacist-led tapering to support benzodiazepine deprescribing in older adults. Based on previous research, we first developed an intervention conceptual model. The aim of this study was to 1) gather stakeholder feedback on previous experiences with benzodiazepine tapering and on our intervention model and proposed intervention; and 2) integrate this qualitative feedback to develop an intervention manual. We conducted a) semi-structured individual interviews with older adults (N=8) that previously attempted to taper their benzodiazepines, and b) a focus group with members (N=5) from a national deprescribing patient stakeholder group. Overlapping themes emerged, including support for the mindfulness-informed CBT intervention, the importance of control over taper pace, the need for a goal- and skills-oriented intervention, the importance of normalizing side effects of the taper and building confidence to manage side effects, and the utility of fostering acceptance during the taper. These findings informed the development of a final intervention manual, named Confidence Building Strategies for Reducing Sedative Medications (CSTARS), to be tested in a single-arm pilot feasibility trial.

Plain Language Summary

This paper describes the development of a brief psychologist-led mindfulness-informed cognitive behavioral therapy (CBT) intervention to be paired with a pharmacist-led program to help older adults reduce or discontinue their benzodiazepine medication. We developed the intervention by conducting interviews with patient stakeholders (individual interviews and a focus group) and integrating feedback to develop an intervention manual. Patient stakeholders were supportive of the intervention and several themes emerged from patient interviews that directly informed the intervention. We will assess the feasibility and acceptability of the intervention in a subsequent pilot trial.

Introduction and Background

Chronic benzodiazepine use in older adults is associated with increased risk of adverse drug events, including falls, cognitive impairment, functional decline, overdose, hospitalizations, and mortality [1-5]. Benzodiazepine risks are well-known to primary care clinicians, and benzodiazepines have long been listed in the American Geriatrics Society Beers Criteria as potentially inappropriate medications for older adults [5]. Despite this, use of chronic benzodiazepines among older adults is commonplace and has largely remained constant over the last decade [1]. The 2025 Joint Clinical Practice Guidelines on Benzodiazepine Tapering, developed by a multidisciplinary group and led by the American Society of Addiction Medicine, provides a strong recommendation for tapering benzodiazepines in older adults and highlights initiatives to bolster and guide treatment [6].

Given the potential for withdrawal symptoms and adverse withdrawal events if benzodiazepines are discontinued abruptly, deprescribing interventions involving gradual tapering are needed to safely reduce or discontinue benzodiazepines in older adults. Prior research has shown that pharmacist-led tapering programs may successfully reduce benzodiazepine use in the United States and Canada [7-10]; however, these programs typically do not address psychosocial barriers or predictors of successful discontinuation. Indeed, extant research suggests that gradual tapering programs combined with non-pharmacological support were more effective than gradual tapering alone [11]. Moreover, the 2025 Joint Clinical Practice Guidelines on Benzodiazepine Tapering recommend adjunctive psychosocial interventions during the taper process [6]. Non-pharmacological support has often included time-intensive, disorder-specific cognitive-behavioral therapy (CBT) for anxiety disorders or insomnia in an outpatient mental health setting [11], which has practical limits on broader uptake, generalizability, and implementation, particularly due to shortages in mental health providers. Despite the clear need for evidence-based deprescribing approaches, few deprescribing programs have been developed for the primary care setting in the United States or internationally [12]. Indeed, the majority of benzodiazepines are prescribed in primary care [13], making it an ideal setting for deprescribing efforts. Supportive interventions in primary care need to be brief and generalizable to a wide-range of patients for successful adoption and scalability [14].

Supportive Deprescribing Interventions: Key Takeaways

Here we describe key takeaways from the existing literature on supportive benzodiazepine deprescribing programs.

Non-pharmacological support typically included intensive, disorder-specific CBT treatment, which has limited generalizability.

The majority of programs that combined gradual tapering with support included intensive (e.g., range 5-12 sessions, 60-120 mins), disorder-specific cognitive-behavioral therapy (CBT) [e.g., 15-19]. Time intensive interventions may be difficult to implement in primary care; we are unaware of any studies examining brief CBT models. Embedding behavioral healthcare in primary care can increase access and improve outcomes [20], and brief interventions may address problems with the shortage of behavioral health providers and scalability.

Moreover, these CBT programs were focused on treating underlying panic disorder [e.g., 15, 21], generalized anxiety disorder [e.g., 19], or insomnia [see 22 for a review]. Meta-analyses suggest that gradual tapering combined with CBT is more effective than gradual tapering alone [11, 22, 23]. While these disorder-specific treatments are necessary and important, the generalizability to wider populations looking for broader support to deprescribe their benzodiazepine in primary care is more limited. Indeed, help seeking for a broader emotional problem has been shown to be a stronger predictor of benzodiazepine use than a specific DSM psychiatric diagnosis [24], suggesting that benzodiazepine use does not always map to a specific DSM diagnosis.

Self-efficacy predicts successful discontinuation.

Several studies have identified self-efficacy as an important predictor of benzodiazepine discontinuation. O’Connor and colleagues [25] found that increases in self-efficacy for coping without the benzodiazepine and in difficult situations was associated with successful discontinuation. In the EMPOWER trial, higher baseline self-efficacy was associated with successful tapering [26]. Among older adults with insomnia, those who successfully discontinued their benzodiazepine reported higher self-efficacy at weeks 6, 8, 9, and 10 of a taper program compared to individuals that did not successfully discontinue [27], highlighting the importance of bolstering self-efficacy at these timepoints. Finally, in a systematic review of barriers and enablers of benzodiazepine deprescribing in older adults using the theoretical domains framework, beliefs about capabilities emerged as an important domain influencing deprescribing [28].

Fears and inability to tolerate withdrawal are barriers to discontinuation.

In a systematic review of patient barriers and enablers to deprescribing across a variety of medications, non-specific fears, fears of the condition returning, and fears of withdrawal effects emerged as important modifiable barriers to deprescribing benzodiazepines [29]. Evrard and colleagues [28], in a systematic review of barriers and enablers to benzodiazepine deprescribing, also identified beliefs about the consequences of deprescribing as an important barrier. In the EMPOWER trial, an inability to tolerate withdrawal symptoms was associated with a failed attempt to deprescribe [26]. Interestingly, a tapering plus CBT intervention that focused specifically on management of withdrawal symptoms was not more effective in benzodiazepine discontinuation than tapering alone. However, in this study, CBT began midway through the taper, was difficult to individualize due to the delivery of the intervention in a group, and had high dropout [17].

Few non-pharmacological support programs included mindfulness strategies, which have empirical support to positively impact medication reduction and health behavior change.

While mindfulness interventions have shown promise for reducing opioid use [e.g., 30], fewer studies have incorporated mindfulness strategies in benzodiazepine deprescribing programs. In a 2021 review examining mindfulness-based interventions for sleep disturbance among individuals hospitalized for cancer, three studies reported reductions in benzodiazepine use [31]. An 8-week Mindfulness Based Relapse Prevention program demonstrated reductions in hypnotic use, inclusive of benzodiazepines, among women with insomnia [32]. Incorporating mindfulness strategies, which emphasize nonjudgmental awareness of present moment experiences [33], in benzodiazepine deprescribing programs may enhance outcomes. Indeed, mindfulness strategies may promote purposeful versus automatic responding to difficulties during the taper, as well as increased distress tolerance [34]. Moreover, theoretical models suggest that mindfulness may promote behavior change through improvements in emotion regulation, attentional and cognitive control, and self-related processes, including self-efficacy [35].

These key takeaways informed the development of a simple conceptual model for a brief, generalizable, psychologist-led CBT intervention to pair with an existing pharmacist-led tapering program in primary care [10] to enhance benzodiazepine deprescribing in older adults (Figure 1). We proposed that directly targeting one’s self-efficacy to discontinue the benzodiazepine would lead to successful benzodiazepine reduction or discontinuation. To increase self-efficacy, we proposed combining a patient-centered, pharmacist-led, gradual tapering program with psychologist-led, mindfulness-informed CBT sessions. We hypothesized that the CBT sessions would increase self-efficacy by three main pathways: 1) reducing fears and negative beliefs about the taper process, 2) increasing one’s capacity to mindfully respond to difficulties that may arise during the taper process, and 3) providing strategies to cope with difficulties that may arise during the taper process.

Figure 1.

CSTARS Conceptual Model

Study Aims

In line with guidance for improving transparency in intervention development [36], this paper describes our intervention development process. The aim of the current study was to 1) understand patient experiences with tapering and gather feedback on our conceptual model and prototype intervention and 2) integrate this qualitative data to develop an intervention manual.

Materials and Methods

We followed the Standards for Reporting Qualitative Research (SRQR) reporting guideline [37]. The study was conducted in accordance with the Basic & Clinical Pharmacology & Toxicology policy for experimental and clinical studies [38].

Study Design and Participants

Individual interviews

We conducted semi-structured individual interviews with older patients at a large US academic medical center. Patients were identified from a registry for the hospital-based general medicine clinic who previously completed a benzodiazepine taper with the clinic as part of a pharmacist-led tapering program. Eligible patients were ≥65 years and had previously participated in the tapering program, regardless of success with tapering. Prior to recruitment, an informational letter was sent to all patients that allowed them to opt-out from being contacted. Prior to the interview, participants completed a questionnaire over the phone on their demographics and medication use. Each interview was attended by at least two study investigators (TA, BW, DB, KK) including at least one clinician (KK, TA). Participants were given the option for phone or video interviews that lasted approximately 45 minutes. Participants provided verbal informed consent. All participants were provided $50 after completing the interview. Interviews took place between May 2023 and June 2023.

National stakeholder group

We conducted a virtual focus group session with a national stakeholder group over video that lasted approximately 60 minutes in April 2023. This group consisted of stakeholders representing patient, family, and organizational interests that were provided to us by the US Deprescribing Research Network (USDeN), and all participants provided verbal consent to participate.

The study was determined to be exempt human subjects research by the Beth Israel Deaconess Medical Center institutional review board (exempt category 2).

Interview Guides

We used a mixed inductive and deductive approach to understand factors that impact benzodiazepine deprescribing and perceptions of our prototype intervention, with a predominant grounded theory approach to inform our intervention conceptual model [39]. The interview guide was developed by two study investigators (KK and TA). For the individual patient interviews, the interview guide included open-ended questions on prior experiences with benzodiazepine tapering and questions to provide thoughts and feedback on the prototype intervention, which included pharmacist-led tapering and psychologist-led CBT, for benzodiazepine tapering (see Supplemental File for interview guide).

The focus group participants were asked only to provide their thoughts and feedback on a prototype intervention for benzodiazepine tapering. All participants were sent a handout before the interview that included a simplified version of the intervention conceptual model (Figure 1), a timeline of the proposed intervention (Figure 2), and a brief overview of proposed session content for three CBT sessions.

Figure 2.

Structure of the CSTARS Intervention

Qualitative Analysis and Integration

All individual sessions and the focus group were audio-recorded, de-identified, and transcribed. Transcripts were not reviewed by participants. Thematic analysis was conducted using an iterative, multistage, mixed deductive and inductive coding process. The initial codebook was created and refined by two investigators (KK, BW). Two investigators (KK, BW) reviewed transcripts for accuracy and independently coded the data using NVivo software, v12. After data was preliminarily coded, the research team met to compare data and a final codebook was created. Three investigators (KK, BW, GY) reviewed transcripts again using the final codes and created a list of themes, subthemes, and illustrative quotes. The research team met to clarify themes and resolve discrepancies to produce a final list of themes, subthemes, and illustrative quotes.

Qualitative data was integrated to inform the development of an intervention manual for CSTARS, Confidence-Building Strategies for Reducing Sedative Medications. We created a joint display to demonstrate how qualitative findings informed intervention development. We described the resulting intervention manual based on this integration.

Results

Individual Interviews

Eight participants completed individual interviews (Table 1). Participants were, on average, 74 years of age (SD=7.1), 63% male, and 87% White. A majority of participants (63%) were previously prescribed lorazepam. Four participants had reduced their benzodiazepine dose during their prior taper, three attempted to taper but had no change in their dose, and one participant had successfully tapered off benzodiazepines. Five themes are described around prior experiences with tapering, and four themes on feedback of the proposed intervention.

Table 1.

Participant characteristics for individual interviews

	N=8
Demographics
Age (years), mean (SD)	74 (7.1)
Male, No. (%)	5 (63)
Hispanic ethnicity, No. (%)	1 (13)
Race, No. (%)
White	7 (87)
Black	1 (13)
Benzodiazepine Use
Benzodiazepine Medication, No. (%)
Lorazepam	5 (62.5)
Oxazepam	1 (12.5)
Clonazepam	1(12.5)
Diazepam	1(12.5)
Previous Taper Result, No. (%)
Tapered to reduced dose	4 (50)
Attempted taper and returned to prior dose	3 (37.5)
Tapered off entirely	1 (12.5)

Prior Experiences with Tapering

Theme 1: Concerns prior to tapering

Many participants expressed concerns prior to tapering their benzodiazepine. The two major concerns included: 1) worry that the symptoms for which the medication was prescribed (anxiety, sleep difficulty) would re-emerge; and 2) the possibility of experiencing withdrawal symptoms.

Theme 2: Factors that influenced patients’ decision to taper

Several participants decided to taper their benzodiazepine medication after hearing information from their provider or other sources about the risks of benzodiazepine use in older adults. Participants were often unaware of the potential risks of benzodiazepines and some expressed that they had not personally experienced any adverse effects. Other participants described internal reasons for tapering, including wanting to be on fewer medications or because they experienced unpleasant side effects from the medication (e.g., grogginess).

Box 1. Illustrative Quote for Theme 2.

“My [Primary Care Physician] (PCP) said that it would be beneficial because of my age and that there are different kinds of side effects or issues that may pop up.”

Theme 3: Challenges experienced with the taper

Sleep difficulties were the most commonly cited challenges during the taper process. Some participants experienced fears around not being able to sleep. One participant was surprised by the withdrawal symptoms they experienced, as they were not educated by their doctor on what to expect. One participant experienced a worsening of the symptoms during their taper.

Box 2. Illustrative Quote for Theme 3.

“I ended up getting on the Internet saying, I'm going crazy. Something's wrong with me. I didn't know that was really connected to the withdrawal of the med. So that would have been helpful to say these are some of the things you may experience.”

Theme 4: Desire for control of the taper pace

It was very important for patients to be in control of the pace of the taper and to collaborate with their provider on the taper plan. Together, autonomy and collaboration made participants feel more confident in their ability to deprescribe. Some participants wanted a slower taper plan to help mitigate withdrawal symptoms. One participant chose a slow taper plan due to uncertainty around withdrawal symptoms, which allowed them to gain confidence and subsequently increase the pace of the taper. Some participants reported that they received very little education from their provider on the taper process, what to expect, or on strategies to help manage the taper.

Box 3. Illustrative Quote for Theme 4.

“It was between the two of us, discussing what I was comfortable with. And she was good with that, too. So we…pretty much worked, you know, I felt like I had somebody on my team.”

Theme 5: Resources/strategies used to help with the taper and perceived gaps

Patients used several strategies to help with the taper, including distraction, engaging in enjoyable activities, mindfulness/meditation, sleep hygiene tools, psychotherapy, online resources, physical activity, and social support. Participants also identified gaps and provided suggestions to improve the taper process. Overall, patients wanted more tools for specifically managing sleep difficulties, anxiety, and withdrawal symptoms; emphasized the importance of close collaboration with a health professional; and wanted more information on what to expect during the taper.

Box 4. Illustrative Quotes for Theme 5.

“But even just hearing it [education on what to expect with withdrawal symptoms] beforehand would have been helpful to have on the info letter or pamphlet or something that would have outlined that, that would have been helpful to me.”

“Well, the most useful thing would be to achieve the goal of getting back to sleep without medication and how that could be accomplished.”

“But it all comes down to, if I could feel confident, within a certain amount of time I was going to be able to handle more the side effects mentally of not being able to rely on something for sleep.”

Feedback on the Proposed Program

Theme 6: General support for the proposed program

Overall, participants were supportive or had a positive reaction to the proposed program. Participants believed the program was needed, that it provided a helpful framework for deprescribing, and that it provided frequent touch points for support.

Box 5. Illustrative Quotes for Theme 6.

“I think this program is great that what you've come up with gives the patient, you know, goals to strive for.”

“I think it's good that they're doing it in steps. And checking in every couple of weeks with somebody to see how it's going.”

Theme 7: Feedback on the pharmacist sessions

When providing feedback on the proposed taper plan and pharmacist session content, participants expressed the importance of having autonomy over the taper pace. They believed the taper pace should be individualized and that the patient should have control over when reductions to the medication are made. Some patients believed a slower taper pace was necessary to reduce withdrawal symptoms and manage fear/uncertainty over the process. They also emphasized the importance of early education on what to expect in regards to withdrawal symptoms.

Box 6. Illustrative Quotes for Theme 7.

“So the major thing would be to let the patient do it at their own pace, I would say, because it's sort of scary to see that someone's going to take you off of something…”

“I’m just going to reiterate, just to go very slow with people and very gentle and to give them lots of support when they need it and give them the amount of time they need until they feel when they reach the next lower plateau and trust their feelings when they say that, okay, I feel stable now.”

“You know, further educating someone about what those side effects are that…I don't know how many people are fully aware of what side effects they may be facing.”

Theme 8: Feedback on the CBT sessions

There was general support for the CBT sessions. A small number of participants wondered if this component would be necessary for everyone, expressing disinterest in this component, but stating that it could be helpful for others. Overall, participants reported that the CBT sessions would be helpful for providing support and helpful strategies (e.g., relaxation) for the taper process. Participants emphasized that the CBT sessions should be goal- and skills-oriented, should normalize potential side effects of the taper, and should build confidence for managing potential side effects. One participant thought that having handouts for patients would be helpful. Participants also discussed the importance of establishing rapport with the CBT provider. Some participants thought more sessions might be needed, particularly at later stages of the taper, but that the number of sessions should be based on individual preferences and progress.

Box 7. Illustrative Quotes for Theme 8.

“I'm looking at this as, here are the strategies you can develop to taper and eventually eliminate your usage. Here are some strategies. Try these out, see what works and move forward. So very narrowly focused.”

“Pharmacist might be able to tell me whatever drugs are and their interactions but won't necessarily change it. The psychologist will deal with the feelings and just it's a place to express the complicated withdrawal feelings.”

“I mean, what I would hope is that those sessions would be okay - you're waking up at 2:00 in the morning and you don't want to medicate. So here are some things you can do…”

Theme 9: Implementation and delivery

Participants believed the length of the program should be individualized based on one’s specific needs. A few participants thought the program should be longer to realistically achieve goals and to increase confidence.

In addition, to help inform future studies, we asked participants what they thought about the CBT content being delivered in online modules. Several participants reported that they would be open to online content, stating that it would be helpful to access the content whenever they wanted. A few participants explained that they would not like online content because it is “cold,” because they prefer live support, or because online support would be technologically difficult to navigate.

Box 8. Illustrative Quotes for Theme 9.

“That [online content] would be a very, very helpful thing to have it because it's some place to go. I know on my website there's many topics and stuff and I would go there first before I look outside.”

“Honestly, I think it [the length of the program] would be case dependent. You know, depending upon…how effective the first session is. It, to me, would depend.”

National Stakeholder Group

Five participants were in the stakeholder focus group, convened by the US Deprescribing Research Network, which included older adults in the roles of both patients and caregivers. Three main themes emerged in response to the proposed intervention. Many subtheme topics overlapped with those that emerged in the individual interviews.

Theme 1: Need more time to taper

Individuals in the stakeholder group believed that a 10-week taper plan was too fast. They emphasized the need for a slow and steady taper to minimize withdrawal symptoms. They also described the need for the taper to be patient-led and flexible. One participant in the group explained that if a taper is too fast, it will set people up for failure and make future tapers even more difficult.

Box 9. Illustrative Quotes for Theme 1.

“I think you have to build into that this flexibility piece of, as symptomatology increases, as they’re having more and more difficulty, and adjust that schedule and go at their own rate.”

“And I would add to that, the risks of coming off too soon sets up failure for successive reinstatements and follow-up tapers.”

Theme 2: Needs around coping during the taper process

Individuals in the group discussed the need for frequent and supportive reassurance, resources (e.g., advocacy organizations), and support for managing withdrawal symptoms even after the taper is completed. Participants described the need for whole-person care during the taper process. They also discussed the helpfulness of peer support from someone who has already completed a taper. Participants emphasized that support persons (e.g., family, friends) need to be encouraging, knowledgeable about withdrawal, and believe their lived experience of withdrawal.

Box 10. Illustrative Quotes for Theme 2.

“… the number one thing I needed, during my taper, every week, I needed reassurance… I needed to know that I am going to be healing, that I am going to be safe, and that I am going to recover from this.”

“I'm a big believer in support, so when someone’s going through something like this, their family must be involved, because they're gonna wonder what happened…”

Theme 3: Suggested approach and targeted goals of the CBT sessions

Overall, there was support for the CBT sessions, including the mindfulness component. Participants suggested that the sessions 1) be present-focused and goal-oriented; 2) help reframe thinking around withdrawal symptoms and foster acceptance and the idea of “being comfortable with the uncomfortable”; 3) set expectations for the taper process and emphasize the temporary nature of withdrawal; 4) normalize the experience of withdrawal; and 5) offer distraction (i.e., keep busy with an activity to ride out temporary withdrawal) to help manage the taper. Participants expressed the perceived helpfulness of more than three sessions, providing further resources after 10 weeks, and the possibility of online modules for both the patient and their support system (e.g., family member, friend). They expressed the importance of the intervention provider to be understanding, knowledgeable about the taper process, and not dismissive about the patient’s experience with withdrawal.

Box 11. Illustrative Quotes for Theme 3.

“It’s very important not to get into any other traumas and really just kind of stay focused on reframing the mindset of where you’re at.”

“And I also believe that being honest with the patient and, you know, realistic… someone said get comfortable with being uncomfortable. It’s true.”

“I think the biggest piece for me was acceptance, and I like all the things you listed here, the mindfulness, the relaxation, I used all of that, but really it was just achieving some level of acceptance that I was having these symptoms, I was uncomfortable but I basically got comfortable feeling uncomfortable and I stopped fighting the symptoms and that’s when things sort of progressed for me.”

Integration of Qualitative Findings to Develop Intervention Manual

We collated our intervention conceptual model and data from the individual stakeholder and focus group interviews to develop an intervention manual for CSTARS, Confidence-Building Strategies for Reducing Sedative Medications. See Table 2, which is a joint display demonstrating how qualitative data informed the CSTARS intervention. Here, we describe the intervention manual that was developed based on this integration.

Table 2.

How Qualitative Feedback Informed Development of CSTARS Intervention Components

Participant Feedback	Intervention Components
Patient-led taper and collaboration	Patient and pharmacist collaboratively decide taper goals Both pharmacist and CBT sessions emphasize that patients are in control of taper pace to reduce fears
Taper pace too fast for 10 weeks	Patients can choose reduction or discontinuation goals Offer handoff to PCP if patients want to continue with taper after 10-week program
Resources for sleep difficulties	Provide sleep strategies handout
Setting expectations, normalizing experience	Pharmacist reviews what to expect during taper CBT sessions normalize withdrawal symptoms through “common fears and worries” handout
Emphasizing temporary nature of withdrawal and “being comfortable with uncomfortable”	CBT sessions use “benzodiazepine flu”1 model to emphasize temporary nature of withdrawal CBT sessions emphasize using coping strategies to “ride out” discomfort during taper Mindfulness is introduced to promote nonjudgmental awareness of current moment experience
Coping with difficulties	CBT sessions introduce and practice mindfulness exercise for mindfully responding to difficulties CBT sessions review and practice strategies for managing side effects during taper. Provide handouts for strategies and audio recordings of mindfulness/relaxation practices
Importance of support	CBT sessions provide education on the importance of support and encourage patients to identify support person
Providing additional resources	Provide resources for outside CBT-I or psychotherapy, if needed

¹Otto MW, Jones JC. Stopping anxiety medication: panic control therapy for benzodiazepine discontinuation: therapist guide. Oxford University Press; 2004.

Note. CBT: cognitive-behavioral therapy; PCP: primary care physician; CBT-I: cognitive-behavioral therapy for insomnia

CSTARS is a 10-week pharmacist-led tapering program combined with three individual, psychologist-led CBT sessions to build confidence for the taper process. We chose a brief intervention model that could be easily implemented and scaled in a primary care setting. See Figure 2 for an overview of CSTARS. The 10-week tapering plan was adapted from guidelines from the University of Ottawa Bruyère Research Institute [40] and from feedback from our initial quality improvement project [10]. For the pharmacist-led tapering sessions, taper goals are collaboratively developed between the pharmacist and participant and can include plans for reduction or discontinuation during a 10-week period. At week 0, the pharmacist sets expectation for the taper process and works with the participant to develop their personalized taper plan. The remaining biweekly visits are focused on reviewing progress, reviewing withdrawal symptoms, problem-solving difficulties, and developing a taper plan for the next two weeks. If patients want to continue the taper after the 10-week program, the pharmacist informs the participant’s primary care physician (PCP).

The CBT sessions are focused on building confidence to complete the taper (Table 3). Week 0 focuses on addressing and reframing common fears and worries about the taper process and setting expectations, adapted from existing materials [41-43]. We discuss the temporary nature of withdrawal using the Otto and Jones [42] “benzodiazepine flu” metaphor, which emphasizes riding out temporary withdrawal symptoms like you would uncomfortable symptoms during the flu. Psychoeducation on mindfulness is provided and participants practice and process the 3-minute breathing space exercise [44]. We emphasize using mindfulness as a strategy for purposefully responding to difficulties versus automatic and unhelpful responding. The following coping strategies are introduced for managing difficulties during the taper process: scheduling pleasant/distracting activities, positive coping statements, self-compassion, diaphragmatic breathing, progressive muscle relaxation, and mindful walking [41-43]. At Week 0, we also provide education on the importance of social support and seeking professional support, if needed. Participants are provided with a sleep strategies handout [45] and several handouts reviewing content from the CBT sessions. They are also provided instructions for how to access audio recordings for the 3-minute breathing space and progressive muscle relaxation on the study website. Weeks 2 and 6 are focused on reviewing successes and challenges, problem-solving barriers to using mindfulness or coping strategies, and developing a plan for skills practice. At Week 6, we collaboratively develop a plan for living at their reduced dose or without the medication. All participants are provided with a resource handout for finding a cognitive-behavioral therapy for insomnia (CBT-I) provider or psychotherapy, if needed. While the CBT sessions include a set curriculum to target self-efficacy, because the sessions are completed on an individual basis, they are individualized by addressing each participant’s specific questions, concerns, barriers, and challenges.

Table 3.

Overview of CSTARS Mindfulness-Focused CBT Session Content

Week 0	Week 2	Week 6
Enhancing motivation and connecting with values Expectation setting and addressing fears “Benzodiazepine flu”1 Psychoeducation on mindfulness and practice of 3-minute breathing space Introduction to strategies for coping with difficulties Psychoeducation on social support Emphasizing seeking professional support, if needed	Reviewing successes and challenges Problem-solving difficulties Review mindfulness and coping strategies Practice of progressive muscle relaxation Awareness of unhelpful hypervigilance to withdrawal symptoms	Reviewing successes and challenges Problem-solving difficulties Developing plan for living at reduced dose or without medication Address potential barriers

¹Otto MW, Jones JC. Stopping anxiety medication: panic control therapy for benzodiazepine discontinuation: therapist guide. Oxford University Press; 2004.

Discussion

We used a step-wise, iterative process to develop a brief mindfulness-informed CBT intervention to pair with a pharmacist-led tapering program to help older adults deprescribe benzodiazepines in primary care. We developed an intervention model based on key takeaways and gaps in the existing literature, gathered feedback from stakeholders, and used this feedback to develop an intervention manual for testing in a pilot feasibility trial. This pilot feasibility trial, which will be reported elsewhere, is a single-arm trial of CSTARS among older adults to assess feasibility and acceptability outcomes. Findings from this pilot trial will inform further intervention modifications and the development of a larger trial to assess intervention efficacy.

Individual interview and focus group participants provided general support for the proposed intervention. Feedback from these interviews informed several key CSTARS intervention components. Based on participant feedback, CSTARS was designed to emphasize patient control over the taper process, in close collaboration with a pharmacist, to reduce fears and promote autonomous motivation, which is important in the context of behavior change [46]. This finding is in line with existing research, which has identified patient fears around not being in control of the taper pace as a significant barrier to deprescribing [29]. Participants also emphasized the importance of setting expectations and normalizing the potential for side effects during the taper, which are reinforced in both the pharmacist-led tapering and CBT sessions. Normalizing and reframing side effects of the taper as temporary and manageable may enhance one’s distress tolerance and promote persistence with the taper [42]. Interestingly, focus group participants discussed the importance of cultivating acceptance during the taper process (“being comfortable with being uncomfortable”). This is closely aligned with mindfulness-based approaches and supported the inclusion of this component in the intervention. CSTARS focuses on cultivating nonjudgmental awareness of current moment experiences to promote purposeful versus automatic responding to difficulties during the taper. Participants also emphasized goal-oriented sessions and concrete strategies for managing sleep and other difficulties that arise during the taper process, which are directly addressed in CSTARS intervention.

Our intervention development approach yielded a novel benzodiazepine deprescribing intervention. While the majority of benzodiazepine prescribing occurs in primary care [13], very few studies have examined supportive benzodiazepine deprescribing programs in the primary care setting in the United States. CSTARS was designed to be a brief, scalable intervention for primary care, which may reduce barriers to accessing care. While there is support for brief interventions for benzodiazepine deprescribing [47], future studies, including our pilot feasibility study, will need to assess whether three CBT sessions is sufficient for improving outcomes beyond pharmacist-led tapering alone. Similarly, in line with developing an ultra-brief, supportive intervention, CSTARS was designed to target one potential mechanism of action, self-efficacy. Future research is needed to assess whether the inclusion of other mechanistic targets in our conceptual model are needed to produce change above and beyond pharmacist-led tapering alone.

Moreover, contrary to existing supportive interventions focused on disorder-specific populations, CSTARS was designed to be generalizable to a wide-range of primary care patients interested in support for deprescribing. As such, CSTARS does not focus on treating a specific underlying DSM disorder, and instead focuses on building confidence for completing the taper itself. If CSTARS is shown to be effective, future research will be needed to develop decision trees to incorporate evidence-based screening procedures to determine who needs a higher level of support (e.g., disorder-specific CBT) versus more generalized support for the taper process. Overall, while CSTARS has the potential to be a scalable and generalizable intervention to help older adults deprescribe benzodiazepines, a larger efficacy trial is needed before drawing any conclusions.

Taken together, CSTARS is a novel benzodiazepine deprescribing program that was developed through a step-wise, iterative process to incorporate stakeholder feedback. Our next phase includes a pilot study to examine the feasibility and acceptability of CSTARS, which will be reported elsewhere. If feasible and acceptable, a larger efficacy study should assess whether the addition of the mindfulness-informed CBT sessions is more effective than pharmacist-led tapering alone. While our focus was benzodiazepines, additional research is needed to develop supportive deprescribing programs for other high-risk psychotropic medications.