The experience of women in a phase II trial of hypnosis and progressive muscle relaxation for body image: informing future research questions

Abstract This study examines feedback from two interventions, hypnosis and progressive muscle relaxation (PMR), to improve body image in a randomized phase II trial. Eighty-seven women were randomized either to hypnosis or PMR. Sixty-three women (72%) were motivated to write comments about their study experience. These comments were explored in an unplanned qualitative analysis. Thematic analysis generated five themes, suggesting both hypnosis and PMR may improve body image through the ability to relax and manage stress, sleep better, improve mood and create a mind-body connection. Sexual health emerged as a theme for participants in only the hypnosis group which suggests hypnotic suggestions for body image may improve overall sexual health. Additional research is needed to assess this further.


Introduction
Over 5 million women have a history of breast or gynecological cancer in the United States, 1 many face long-term consequences from cancer and/or cancer treatment.][5][6][7] Breast and gynecological cancer survivors commonly report negative body image.In one study, 77% met the criteria for sexual dysfunction with body satisfaction being among the most prevalent symptoms. 8Body image is one of 15 factors impacting sexual relationships. 9Despite being among the most difficult challenges for survivors to adjust to following treatment, it has been described as modifiable, or able to be influenced by a healthcare provider. 9Research examining mindfulness, cognitive behavioral therapy, and psycho-education interventions supports the usefulness of a mind-body approach to interventions targeting body image.
Following a successful feasibility study that showed a large effect size for body image improvement, our research team conducted a randomized phase II trial evaluating hypnosis for body image in women with breast or gynecological cancer. 10Progressive muscle relaxation (PMR) was chosen as the control arm to control for the effects of simple relaxation.The primary outcome was the Impact of Treatment Scale (ITS), a measure that asks about body image change perception and has demonstrated reliability and validity in women with cancer. 8here were no statistically significant differences between the hypnosis and PMR groups on the ITS in the randomized phase II study.Both groups reported significant improvements in perception of body image (the primary outcome) from baseline to 6 wk, had similar practice rates and showed good adherence.There were non-statistically significant differences between the groups in secondary outcomes with the hypnosis group reporting greater improvements than the PMR group in perceived sexuality and sexual satisfaction.This incongruence in outcome impact is puzzling and, the benefit from PMR was surprising.
Therefore, in an exploratory, unplanned analysis, we used qualitative data provided by women in both arms of the study throughout the six weeks to attempt to understand the relationship between body image and PMR and body image and hypnosis.We aimed to describe the experiences of women in each arm to help inform future research questions.

Participants
The randomized phase II study methods and results have been reported in detail elsewhere 10 but will be described here briefly.The study was approved by the University of Michigan Institutional Review Board.All participants provided written informed consent.Eighty-seven women were randomized in a 2:1 fashion, 59 to hypnosis (68%) and 28 to PMR (32%).Participants who were motivated to leave qualitative feedback were included in this secondary analysis.

Eligibility
Women were eligible to participate in the parent study if they met the following criteria: being diagnosed with breast or gynecological cancer at any stage, undergoing or completed treatment, and reporting a negative change in body image since diagnosis with a desire to improve that perception.

Study arms
In each arm, participants attended 3 face-to-face sessions with a therapist trained in hypnosis and PMR.The sessions were spaced two weeks apart and spanned the six weeks of the study.For both groups, sessions lasted approximately 60 min, with about 20 min dedicated to hypnosis or PMR.The PMR intervention focused on assessing sources of stress and teaching the basics of PMR practice but did not address body image or sexual health.The hypnosis intervention focused on wellness, confidence and body image related to sexuality.Intervention sessions were recorded for home practice throughout the week.

Data collection
Specific open-ended questions were not a required component of the study.However, the following paper and pen questionnaires left space for comments: 1.A daily practice log. 2. A weekly side effect questionnaire to monitor negative experiences with the intervention.3. Global Impression of Change Scale, completed at the end of the study.

Analysis
Two study team members, both with research degrees and training in qualitative methods, extracted all comments from the data sources listed above.Analysis followed a hybrid, six-stage process incorporating inductive and deductive thematic analysis: (1) develop a code manual; (2) test the reliability of the codes; (3) summarize data and identify preliminary themes; (4) apply template of codes and additional coding; (5) connect the codes and identifying themes; and (6) corroborate and legitimate coded themes. 11The reviewers independently coded the data.Deductive logic, or the use of a priori codes suggested by the parent study was introduced in stage four, resulting in the addition of three possible codes.

Preliminary analysis
Of the 87 participants in the parent study, 63 women (72%) wrote comments elaborating on their study experience; 44 in the hypnosis arm (75%) and 19 in the PMR group (68%).Demographic characteristics of the sample are described in Table 1.There were no statistically significant differences between the hypnosis and PMR groups at baseline.Participants were either currently undergoing or had completed primary treatment.The majority were in a relationship and slightly over half reported previous attempt(s) to improve self-image.Women who reported a mental health diagnosis that might interfere with hypnosis, including a major depressive episode, an acute anxiety disorder, psychosis, PTSD or schizophrenia were not eligible for this study.
Comments were submitted on all three questionnaires; 61 women responded on the daily practice log, 39 on the weekly side effect questionnaire and 14 wrote in the margins of the Global Impression of Change Scale.No differences were found between groups concerning demographics.

Main analysis
Thematic analysis generated five themes.Two a priori themes were supported: relaxation/stress management and mind-body connection.
Normalizing the problem, a third a priori theme, was not supported by the comments.Three additional themes were identified: sleep, improved mood, and sexual health.Themes and supportive quotations, appear below.

Relaxation and stress management
The a priori code relaxation and stress management was evident as a theme in both study arms throughout the six-week intervention (see Table 2).

Sleep
Sleep was a consistent theme in both study arms and supports the ability of both hypnosis and PMR to induce relaxation that is complete enough to, in some cases, induce sleep.There were many comments related to falling asleep during the intervention.Additionally, some participants made comments about getting better sleep in general since starting the intervention.Examples of both can be found in Table 3.

Mind-body connection
A second a priori code was supported by data related to a mind-body connection.Unlike relaxation quotes that either used some form of the   word relax or focused on the participant's mental or emotional experience, mind-body connection quotations included some mention of the mental/ emotional experience and the physical experience (see Table 4).

Improved mood
Improved mood was reported by women in both arms.In the hypnosis group, statements reflected feelings of empowerment, happiness and being outgoing, consistent with the language used in the hypnotic suggestions.
In the PMR group, patience, energy, and emotional balance were mentioned.Participants in both groups mentioned a new willingness to share with their spouse (see Table 5).

Sexual health
Sexual health emerged among participants in the hypnosis arm, but not the PMR arm (see Table 6).Comments were consistent with the hypnotic suggestions delivered over the six-week intervention beginning with positive suggestions about one's body and sexual needs, progressing to feelings  about sexuality and the sexual experience and ending with a suggestion to select and implement a behavior consistent with meaningful self-care.

Discussion
In our randomized phase II study, PMR was planned to be an active control but was found to have more than a placebo effect on body image, our primary outcome.Our qualitative findings support that we successfully blinded our hypothesis as comments from both groups revealed engagement and positive effects on stress, relaxation, connecting the mind and body and mood.Evidence of stress reduction, relaxation, and improved mood among the PMR group is not surprising given effects demonstrated by other research.A study of first-year veterinary students found improvements in basic relaxation and stress levels after they learned to use PMR for selfcare. 12In a narrative review by Raad et al, 13 PMR was examined along with cognitive behavioral therapy (CBT) and deep breathing (DB) among adults undergoing treatment for infertility and investigators concluded that each was associated with improvements in mood and reduced stress levels.What is surprising is that PRM could positively impact body image.
The results of this exploratory qualitative analysis led us to form future research questions about body image.The Theory of Unpleasant Symptoms 14 states that in a person experiencing unpleasant symptoms, there is a continual and reciprocal relationship between influencing factors (physiologic, psychologic and situational), symptoms and performance.This qualitative data suggests that mechanisms through which body image may be improved include improved sleep, which decreases cortisol and can help the stress response; coping well with stress, which may improve feelings of control; increasing positive mood, which can lead to more self-care activities; and the ability to connect one's positive feelings to their physical body (mindbody connection) (see Figure 1).
Several of the responses from the hypnosis group directly reflected the hypnotic suggestions used in the intervention.For example, Participant 1041 stated "more relaxation, noticeable sexual feelings, openness, interest."Relaxation, sexual feelings, and interest were part of the hypnotic suggestions.Additionally, the hypnosis group included findings of improved sexual health.This may be the result of the specific hypnotic suggestions around sexual health and feeling empowered.During the final weeks of the intervention period, there was evidence in the hypnosis group of increased action to make changes.For example, participant 1043 stated "(I) felt good about myself.Starting to make changes and take control over my body."This shift into action was not seen in the PMR group's responses.Similar to Jensen et al 15 where women in treatment and post-treatment for breast cancer participated in self-hypnosis and an improvement was found for pain, fatigue and sleep problems, all symptoms targeted by hypnotic suggestions, we pose that the specific hypnotic suggestions helped women to implement change in their behaviors and improve their sexual health.These findings may explain the slightly higher but not statistically significant sexual satisfaction and sexual interest scores seen in the hypnosis group in the parent study. 10ne of the key limitations of this study is that participants self-selected by choosing to provide written comments on the questionnaires from the parent study.As such our findings represent women who were motivated to write comments and may not be a true representation of the entire study population.An additional limitation is that it is not known whether participants were coping with other chronic conditions that might affect body image.It should also be emphasized again that comment analysis was not originally planned when this study was designed.It would have been helpful to have included a qualitative component designed expressly to understand mechanisms of benefit.

Conclusions
Our study findings suggest that PMR and hypnosis both offer a strategy for self-care that nurtures the whole person and ultimately may improve body image.However, the comments made by participants also support the quantitative data 10 that suggestions embedded in the hypnosis intervention did impact the women in a positive way related to sexual health.Future research should test the relationships we propose and better explore their directional and/or causal nature.Additionally, more research is needed to continue to determine which intervention (PMR or Hypnosis) is better suited to improve cancer survivors' body image and subsequently their overall sexual health.

Figure 1 .
Figure 1.Potential research questions for improved body image.adapted from the middle-range theory of unpleasant symptoms: an update (lenz et al, 1997).a continuing and reciprocal relationship occurs between influencing factors (sleep, mind-body connection, stress and mood), symptoms (body image) and performance (a complex status of sexual health).The influences and interactions illustrate how both PMr and self-hypnosis might have contributed to improved body image.

Table 2 .
select quotations illustrating the theme relaxation and stress management.

Table 3 .
select quotations illustrating the theme sleep.

Table 4 .
select quotations illustrating the theme mind-body connection.

Table 5 .
select quotations illustrating the theme improved mood.
i have felt more calm and balanced emotionally.(Week6, Participant 1083).

Table 6 .
select quotations illustrating the theme sexual health.