Adherence to standardized 8-week mindfulness-based interventions among women with breast or gynecological cancer: a scoping review

: Participant adherence to standardized 8-week mindfulness-based interventions (MBIs) may be challenging, and adaptations from intervention protocols have been reported in mindfulness research. We conducted a scoping review to determine how women with breast or gynecological cancer adhered to standardized 8-week MBIs delivered in intervention studies. Searches were conducted for articles published till February 2020 in PubMed, Embase, CINAHL EBSCO, PsycINFO Ovid SP, and Cochrane Library Wiley. The following outcomes were investigated: class and silent retreat attendance, intervention completion rate (ICR), adherence to home practice, and reasons for dropping out from an MBI study. Among the 25 included MBI studies, mindfulness-based stress reduction was the most often delivered intervention and mostly women with stage I–III breast cancer were represented. The duration of classes varied from 1.5 to 3.5 hours. Planned home practice varied from 20 to 60 min/day, and silent retreat varied from 4.5 to 8 hours. Due to heterogeneity in the reporting of class attendance, the data could not be pooled. Six studies reported an average class attendance ranging from 5 to 8.2 classes. Overall, intervention completion rate (the proportion who completed all classes) varied from 26.3% to 100%; however, discontinuations were not systematically reported. Home practice time was reported in 20% of the studies and ranged from 17 to 24 min/day. The main reasons for dropping out from an MBI study were health-related problems, organizational challenges, travel distance, and lack of motivation/commitment. About 70% of the studies reported some data on participant adherence, revealing a relatively high overall frequency of class attendance. However, the monitoring and reporting of participant adherence should be improved in future studies to increase our knowledge on the required amount of participant engagement to improve health outcomes Abstract Participant adherence to standardized 8-week mindfulness-based interventions (MBIs) may be challenging, and adaptations from intervention protocols have been reported in mindfulness research. We conducted a scoping review to determine how women with breast or gynecological cancer adhered to standardized 8-week MBIs delivered in intervention studies. Searches were conducted for articles published till February 2020 in PubMed, Embase, CINAHL EBSCO, PsycINFO Ovid SP, and Cochrane Library Wiley. The following outcomes were investigated: class and silent retreat attendance, intervention completion rate (ICR), adherence to home practice, and reasons for dropping out from an MBI study. Among the 25 included MBI studies, mindfulness-based stress reduction was the most often delivered intervention and mostly women with stage I – III breast cancer were represented. The duration of classes varied from 1.5 to 3.5hours. Planned home practice varied from 20 to 60min/day, and silent retreat varied from 4.5 to 8hours. Due to heterogeneity in the reporting of class attendance, the data could not be pooled. Six studies reported an average class attendance ranging from 5 to 8.2 classes. Overall, intervention completion rate (the proportion who completed all classes) varied from 26.3% to 100%; however, discontinuations were not systematically reported. Home practice time was reported in 20% of the studies and ranged from 17 to 24min/day. The main reasons for dropping out from an MBI study were health-related problems, organizational challenges, travel distance, and lack of motivation/commitment. About 70% of the studies reported some data on participant adherence, revealing a relatively high overall frequency of class attendance. However, the monitoring and reporting of participant adherence should be improved in future studies to increase our knowledge on the required amount of participant engagement to improve health outcomes and facilitate the implementation of effective interventions on a larger scale.


Introduction
About 35% of patients with cancer use complementary treatments in the United States. [1] Patient-centered integrative medicine refers to combining conventional medicine and evidence-based complementary treatments. [2] Globally, complementary treatments include biological (eg, phytomedicine, diet) as well as mind-body interventions (eg, hypnosis, acupuncture, and meditation including mindfulness-based stress reduction [MBSR]). [2] Mindfulness-based interventions (MBIs) are grounded in mindfulness meditation, defined as a deliberate awareness of the present experience in a non-judgmental attitude. [3] Standardized MBIs like MBSR [4] or mindfulness-based cognitive therapy (MBCT) [5] consist of 8 weekly in-group classes (up to 3.5 hour/class) with an additional silent retreat (up to 7.5 hours) delivered during this 8-week period. The classes include body scan, sitting meditation, gentle yoga, mindful walking, group discussions, and psychoeducation. Between weekly classes, home practice is required with 45 min/day of formal practice (with prerecorded guided meditations), and 15 min/day of informal practice.
Standardized MBSR or MBCT is effective for mental health outcomes (mainly anxiety and depression) in oncology. [6,7] Although most MBIs studies include women with breast cancer, results cannot yet be generalized to other gynecological cancers, less represented in published studies to date. In addition, participant adherence (eg, attendance to classes) to all components of MBIs may be challenging, and reduced versions have been adapted (eg, 1.5 hour classes or 4-6 classes). [8] Small but significant associations were reported between a higher adherence to home practice of MBSR or MBCT and better clinical outcomes like psychological functioning (eg, anxiety). [9,10] Recently, a dose-response analysis (prediction of specific outcomes in relation to the amount of use of an intervention) including 203 randomized controlled trials (RCTs) of MBIs (≥1 Funding: This publication was funded by the Leenaards Foundation, Lausanne. Authors' contributions: JS, JB, and ME participated in the review design; JS searched and screened the references; JS and ND extracted the data independently; JS drafted the manuscript; JB and ME revised the manuscript critically; and JS, JB, ND, GB, FJ, and ME reviewed and approved the final manuscript. class) reported that mindfulness (as an outcome) was also impacted by the dose of the intervention (face-to-face contact; program intensity and use). [11] In contrast, in the same study, no evidence for dose-response relationships was found for psychological outcomes like depression or anxiety. [11] Experts present controversial opinions about the expected length of formal practice, and the minimal dose of MBIs needed to impact outcomes in a clinically meaningful way is still unclear. [8] Overall, literature shows that participant adherence in MBIs studies could be better described. [9,12] Reporting guidelines for nonpharmacological interventions recommend to describe how interventions are delivered (ie, their dosage), including the degree of participant adherence to those plans. [13] This information might promote stronger conclusions on dose-related effects, [14,15] and increase the robustness of the analysis of the MBIs being tested. [16] In addition, systematic information on participants' degree of adherence to the prescribed dosage will promote evidence-based implementation strategies. In this perspective, we conducted a scoping review with an overall aim of determining the adherence of women with breast or gynecological cancer to standardized 8-week MBIs delivered in intervention studies. The specific objectives were to: 1. Describe participants class and silent retreat attendance, intervention completion rate (ICR), and adherence to home practice within MBI studies. 2. Describe the reasons for dropping out from an MBI study.

Inclusion and exclusion criteria
This review included intervention studies involving adult women with breast or gynecological cancer independent of the disease stage. The inclusion criteria were: standardized group face-toface 8-week MBIs; studies with less intense interventions in terms of duration of the classes, silent retreat, and assigned home practice; and intervention studies, that is, RCTs, quasi-experiments, and pre-post studies. Publications were included in any language that could be understood by the study team (English, French, German, or Italian). Articles with mixed cancer populations (other than breast and gynecological), articles involving interventions with mindfulness as a minor part of the treatment, conference abstracts, reviews, opinion papers, editorials, and comments on original articles were excluded. The outcomes of interest were: the class and silent retreat attendance, ICR, adherence to home practice, and reasons for dropping out from an MBI study.

Search strategy and data selection
Articles were searched following a 2-step approach. In step 1, a librarian performed a first search until November 2018 (no limitation in the timeframe) on PubMed, Embase, CINAHL EBSCO, PsycINFO Ovid SP, and Cochrane Library Wiley. The PRISMA guideline for systematic reviews was followed. In step 2, an update of the literature was performed in February 2020 by the first author with the same methodology. No filters were applied. Keywords used were: breast/genital/fallopian/vagina * /vulvar/vulval/ ovarian/ovary/uterus/uterine/endometri * /gyn(a)ecologic * ; neoplasm * /cancer * /tumo(u)r * /carcino * /sarcom * /malignan * ;f emale; mindfulness/meditation/MBSR/MBCT/MBCR. Keywords on outcomes were not included in search strategy. Search strategy is presented in the supplemental digital content (SDC) Table 1, http://links.lww.com/OR9/A25. The first author screened titles, abstracts, and relevant full texts. Overlapping results from same samples were excluded during full text review.

Data extraction
The first and third authors performed the data extraction independently. Discrepancies were discussed, and the last author was solicited in case of disagreements.
Attendance referred to the number of women who participated in classes or in silent retreat. Where possible we calculated the ICR (as the proportion of women who completed all classes) [17] applying the following formula (higher rates indicate a higher proportion of participants who completed all the 8 classes): ICR ð%Þ¼ number who completed intervention X 100 number assigned to intervention group Adherence to home practice was defined as the amount of home practice completed daily (in minutes per day). Participants who dropped out from an MBI study referred to women who were assigned to the mindfulness group but did not complete the study.
The reasons for study dropouts for control groups are accessible in SDC, Table 2, http://links.lww.com/OR9/A25. Additional data on study participation (number assessed for eligibility, number eligible, number ineligible and reasons for ineligibility, reasons for refusing to take part in an MBI study, study participation rates, and study completion rates) are available in SDC, Table 3, http://links.lww.com/OR9/A25.

Results
A total of 1164 articles were screened based on their titles and abstracts. After full-text examination, 26 articles were included in the scoping review, referring to 25 intervention studies since 1 study was reported in 2 articles. All included publications were in English. The main reasons for exclusion were related to designs or interventions that did not fit our inclusion criteria (Fig. 1).
Completion of a silent retreat was reported in 5 studies. [19,23,24,26,30] The other studies were unclear or did not report any data on silent retreat attendance. Adherence to home practice was reported in 5 studies [25,26,28,31,43] and ranged from 17 to 24 min/day ( Table 2).

Reasons for dropping out from an MBI study
The main reasons provided in the studies as reported by participant who dropped out from an MBI study were: healthrelated problems, organizational challenges, travel distance and lack of motivation/commitment (Table 3). Eight studies did not report or were unclear about the reasons for dropping out from an MBI study. [18][19][20][21]36,[39][40][41]
The average number of classes attended was 6 (range, 0-9), silent retreat included, and the mean home practice duration was 25 min/day, 6 days/week. [44] These results are comparable to those of the face-to-face interventions reported in our review. Thus, adaptations of the MBIs to different contexts and patients' needs should be further investigated. However, careful attention should be paid to the description of how they are delivered, the monitoring, and reporting of participant adherence and related barriers. Our review shows that reports on the reasons for dropping out from a study were lacking. These data are important to learn which changes should be made to increase participant adherence. Similarly, nearly half of the studies on MBIs tested in various populations did not report data on dropouts, [45] thus limiting the interpretation of implementability of MBIs in specific settings.
To our knowledge, this review is the first to include qualitative and quantitative data to determine participant adherence to MBIs. However, although most of the studies provided some information on participant adherence, inconsistencies in the monitoring and reporting made comparison between the studies difficult. These inconsistencies hampered data synthesis. Thus, the exact amount of the participants' engagement to improve health outcomes still remains unclear. [43] The development of guidelines structuring the reporting of MBI studies like those existing for acupuncture interventions for example is recommended. [46,47] If the above-mentioned limitations are addressed in future studies, the preparation, interpretation, and comparison of MBI studies might improve greatly. [12] A limitation of our review is the screening of the articles that was done by the first author only. Other relevant articles might have been missed or excluded. Furthermore, only the data for 8-week MBIs were included.
To conclude, about 70% of the studies reported some data on participant adherence, revealing a relatively high overall frequency of class attendance. However, the monitoring and reporting of adherence should be improved in future studies. This could increase our knowledge on the amount of participant engagement needed to improve health outcomes and facilitate the implementation of effective interventions on a larger scale.

Conflicts of interest statement
The authors declare that they have no financial conflict of interest with regard to the content of this report.