The impact of muscle relaxation techniques on the quality of life of cancer patients, as measured by the FACT-G questionnaire

Introduction Patients with cancer frequently suffer from emotional distress, characterized by psychological symptoms such as anxiety or depression. The presence of psychological symptoms combined with the complex nature of oncology processes can negatively impact patients’ quality of life. We aimed to determine the impact of a relaxation protocol on improving quality of life in a sample of oncological patients treated in the Spanish National Public Health System. Materials and methods We conducted a multicenter interventional study without a control group. In total, 272 patients with different oncologic pathologies and showing symptoms of anxiety were recruited from 10 Spanish public hospitals. The intervention comprised abbreviated progressive muscle relaxation training, according to Bernstein and Borkovec. This was followed by weekly telephone calls to each patient over a 1-month period. We collected sociodemographic variables related to the disease process, including information about mental health and the intervention. Patients’ quality of life was assessed using the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire. Bivariate and univariate analyses were performed, along with an analysis of multiple correspondences to identify subgroups of patients with similar variations on the FACT-G. Results Patients showed statistically significant improvements on the FACT-G overall score (W = 16806; p<0.001), with an initial mean score of 55.33±10.42 and a final mean score of 64.49±7.70. We also found significant improvements for all subscales: emotional wellbeing (W = 13118; p<0.001), functional wellbeing (W = 16155.5; p<0.001), physical wellbeing (W = 8885.5; p<0.001), and social and family context (W = −1840; p = 0.037). Conclusions Patients with cancer who learned and practiced abbreviated progressive muscle relaxation experienced improvement in their perceived quality of life as measured by the FACT-G. Our findings support a previous assumption that complementary techniques (including relaxation techniques) are effective in improving the quality of life of patients with cancer.


Introduction
Patients with cancer frequently suffer from emotional distress, characterized by psychological symptoms such as anxiety or depression. The presence of psychological symptoms combined with the complex nature of oncology processes can negatively impact patients' quality of life. We aimed to determine the impact of a relaxation protocol on improving quality of life in a sample of oncological patients treated in the Spanish National Public Health System.

Materials and methods
We conducted a multicenter interventional study without a control group. In total, 272 patients with different oncologic pathologies and showing symptoms of anxiety were recruited from 10 Spanish public hospitals. The intervention comprised abbreviated progressive muscle relaxation training, according to Bernstein and Borkovec. This was followed by weekly telephone calls to each patient over a 1-month period. We collected sociodemographic variables related to the disease process, including information about mental health and the intervention. Patients' quality of life was assessed using the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire. Bivariate and univariate analyses were performed, along with an analysis of multiple correspondences to identify subgroups of patients with similar variations on the FACT-G.

Results
Patients showed statistically significant improvements on the FACT-G overall score (W = 16806; p<0.001), with an initial mean score of 55.33±10.42 and a final mean score of 64.49±7.70. We also found significant improvements for all subscales: emotional wellbeing (W = 13118; p<0.001), functional wellbeing (W = 16155.5; p<0.001), physical wellbeing (W = 8885.5; p<0.001), and social and family context (W = −1840; p = 0.037). PLOS  began after authorization by the ethics committee corresponding to each hospital. The study population included patients with any type of cancer (oncological and/or hematological malignances), of both sexes, older than 18 years, who agreed to participate in the study, and who showed anxiety, muscle tension, sleeping difficulties, sadness, and/or anxiety attacks. Patients were excluded if they showed severe cognitive or physical impairment, were unable to understand or reproduce the relaxation technique used in the intervention, or were in a terminal condition with a prognosis of imminent death. Although no adverse effects have been reported following use of this technique, it is important to highlight that such techniques should not be considered a substitute for medical treatment. Patients suffering from hallucinations, delirium or other psychotic symptoms were also excluded from this study, as the exercises used in the intervention may lead to potentially unpleasant extracorporeal sensations in those patients. Patient recruitment was conducted in the oncology units of participating hospitals via posters, informative flyers, and information provided to the health professionals caring for the patients (i.e., oncologists, nurses, and psychologists). In total, 272 patients from the oncological services of the participating hospitals satisfied all eligibility criteria and agreed to participate. Six patients (2%) did not practice the technique at home, and were excluded from the analysis (Fig 1).

Data collection
Data were collected using the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire and an ad hoc data collection notebook. The FACT-G comprises 27 items featuring general questions divided into four quality of life domains: physical wellbeing, social/ family wellbeing, emotional wellbeing, and functional wellbeing [30]. Item scores range from 0-4 points. The total score ranges from 0-108 points, with higher scores indicating better quality of life. The FACT-G is considered appropriate for use with patients suffering from any type of cancer [31]. A systematic review found that the FACT-G total and subscale scores had excellent reliability, with Cronbach's alpha values ranging from 0.71-0.88 [32]. In the present study we used the validated Spanish version of the FACT-G [33]. Information collected in the data collection notebook included: 1) sociodemographic and medical characteristics (medical center, age, gender, marital status, children, and educational level); 2), oncological process (cancer diagnosis, cancer therapy-chemotherapy, radiotherapy, hormone therapy, biological therapy, and surgery-side effects of cancer treatment, cancer pain, and analgesic use); 3) mental health issues (use of anxiolytics/hypnotics/antidepressants, comorbid psychiatric diagnoses, psychiatric-psychological treatment, and use of relaxation techniques); and 4) other variables related to the intervention, such as symptoms that motivated participation in the study and questions including "Have you practiced the technique at home?" and "How many times do you practice the technique in a week?" This was followed by weekly telephone calls to each patient over a 1-month period.

Intervention
All participants received an initial guided session to learn abbreviated progressive muscle relaxation training, following Bernstein and Borkovec [34]. This technique consists of contraction and subsequent relaxation of all muscle groups sequentially. In a sitting position and with their eyes closed, participants were instructed to contract and relax the muscles of their hands, forearms, face, neck, shoulders, abdomen, and lower limbs in turn. During implementation of the technique, patients were recommended to perform normal breathing. The initial sessions were conducted individually or in groups, according to the patient's condition. To unify criteria and reduce possible inter-examiner bias, researchers who conducted the sessions were fully trained regarding the study selection criteria, information provided to participants, data collection procedures, and application of the technique. All researchers were instructed in, and received written guidance about, conducting the relaxation session. The main researcher was present at the first treatment session at all hospital centers to homogenize all aspects of the intervention. A pilot test was performed with the first 30 participating patients. Patients performed the technique in a sitting position, and the sessions were conducted in rooms furnished with armchairs, cushions, pleasant lighting, and a quiet environment. Each session lasted approximately 60 minutes and was divided into four parts: 1) explanation of the characteristics of the abbreviated progressive muscle relaxation training [34]; 2) application of a relaxation session; 3) answering any questions; and 4) data collection using the self-administered FACT-G and data collection notebook. At the end of the session, patients were provided with information about the intervention, including a brief description of the session based on text and images to support them in performing the technique at their own homes.  08/01/2014) [35]. This trial is registered with ISRCTN, under registration number 81335752. Clinical trial registration was delayed as this study was classified by the Spanish Agency of Medicine as an Observational Study No Epa, and therefore should not be included in the Spanish Registry of Clinical Studies. In addition, the study sponsor considered this work as a behavioral intervention rather than a clinical trial because of the lack of drugs, biologics, or devices. The authors confirm that there are no ongoing or related trials for this intervention. All procedures were conducted according to the Declaration of Helsinki [36]. All study participants provided written informed consent after they received appropriate information regarding the study aims, potential benefits and possible risks. Data were treated anonymously and confidentially according to the Spanish Personal Data Protection Act [37].

Sample size calculation
The sample size was calculated using EPIDATA version 4.1. As the inclusion of different hospitals was gradual throughout the study, we estimated the sample size based on an infinite population-based sample. With a 95% confidence level, an expected proportion of 20% of anxiety disorders in the cancer population [6,7], and a maximum error of estimation of 5%, the estimated sample size was 246 patients. An expected loss rate of 5% was assumed; therefore, the final estimated sample size was 259 patients.

Data analysis
Statistical analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC, USA). The level of statistical significance was set at 0.05. A descriptive univariate analysis to calculate the total score and percentage for each category was performed for categorical variables. Basic descriptive statistics and the Wilcoxon signed-rank test (W) were performed for quantitative variables. For the bivariate analyses, we used the Kruskal-Wallis (KW), chi-square (χ2) or likelihood ratio chi-square (G 2 ) (when there are more than 25% of cells with expected counts less than 5) tests for contingency tables, including variables with low expected counts.
We also performed multiple correspondence analyses [38] to detect groups of patients with similar progressions based on FACT-G responses. To perform these analyses, we considered that in each questionnaire item there was a negative, constant, or positive progression according to baseline and final scores. Based on this evolution, we considered the active variables as: GP1, GP6, GS3, GS5, GS6, GS7, GE1, GE2, GE3, GE4, GE5, GF1, GF4, GF6, and GF7 (a minimum of 5% of patients in each type of evolution) (see S1 Table).
From these analyses, we obtained 21 factors which summarized all changes in the sample (see S2 Table). We also performed a classification defining four clusters of patients with similar responses (S1 and S2 Figs).
After obtaining the four patient groups, we analyzed which outcomes (among the various FACT-G items) had a different proportion with regard to the proportion of the sample (hypergeometric distribution). Finally, a bivariate analysis of the variation rate of the FACT-G was performed for each cluster of patients. Table 1 describes the sociodemographic and medical characteristics of participating patients for the total sample, and stratified by clusters. No significant differences in sociodemographic features were observed among clusters, except for having children (G 2 = 89; p = 0.031).

Results
More than 90% of patients in each cluster had received chemotherapy, with no significant differences among the clusters (G 2 = 4.44; p = 0.218). However, significant differences were     Table 3 shows the rate score of the FACT-G, stratified by subscales and clusters. Statistically significant differences were detected for the change rate score of the FACT-G according to cluster (KW = 51.84; p<0.001), with the lowest rate score in cluster 3. According to the overall scores, we detected a greater change rate score in cluster 1 (0.22±0.12) and cluster 2 (0.23±0.16), and lower rate scores in cluster 4 (0.16±0.13) and cluster 3 (0.09±0.09) (KW = 47; p<0.001). Further, we observed that the mean initial scores for clusters 1 and 2 were the lowest in the ensemble of data. S3 Table displays the progression of the FACT-G scores and statistical significance of over-represented characteristics in the clusters.

Discussion
The present study indicates that patients with cancer with symptoms of anxiety who received a protocol of abbreviated progressive muscle relaxation training [34] improved their perceived quality of life, as measured by the FACT-G. Patients who performed the technique experienced an increase in overall quality of life as well as in emotional, functional, and physical wellbeing. Our findings are consistent with previous reports that concluded CAM techniques (including relaxation techniques) can improve the quality of life of patients with cancer [11,[18][19][20][21][22][23][24][25].
Beard et al. [22] investigated the effects of reiki and muscle relaxation in 54 patients with prostate cancer. Patients received two reiki sessions plus one muscle relaxation session per week for 8 consecutive weeks. The latter was also recommended to be practiced daily at home. Although those authors did not find differences in overall FACT-G scores, statistically significant improvements were detected in emotional wellbeing. Interestingly, in our study, the FACT-G emotional wellbeing subscale showed the greatest improvement. Andersen et al. [18] analyzed the effects of a 6-week program of exercises and muscle relaxation in 213 patients undergoing chemotherapy treatment. In that study, the combination of high and low intensity physical exercises and relaxation exercises did not produce significant changes in FACT-G scores [18]. Isa et al. [23,24] studied the effect of progressive muscle relaxation on anxiety in a group of 155 patients with prostate cancer. They found statistically significant improvements in quality of life at 4 and 6 months, measured with the 36-item Short Form Health Survey.
Another study on mindfulness also demonstrated improvements in quality of life, particularly in emotional wellbeing [39]. Ü lger et al. [20] concluded that yoga practiced for 8 weeks by patients with cancer significantly improved energy levels, pain, emotional levels, sleep, social adaptation, and social skills. Another study involving yoga, in which women surviving breast cancer performed yoga five times a week for 6 months, showed improved quality of life and decreased abdominal perimeter [25]. Finally, Bar Sela et al. [11] published a study on complementary techniques (including relaxation) involving 163 oncological patients undergoing active treatment, and found significant improvements in quality of life (measured with the EORTC QLQ-C30 questionnaire) and in symptoms such as nausea, pain, and insomnia.
The strengths of our study include the inclusion of a large sample from different hospitals (multicenter study) and thorough training of research staff in performing the technique. However, potential limitations should also be considered. First, the lack of a control group represents the main limitation of this study, and does not allow determination that the results obtained were exclusively due to our intervention. However, a control/placebo group including patients with cancer was difficult to include from an ethical perspective. Second, recruitment for this study took place in hospitals via informative flyers, posters, and direct information provided by health professionals caring for the patients. Therefore, we do not know the total number of prospective participants who were informed of the study or the number of those who did not have access to this information. Third, we only evaluated short-term effects of the intervention and cannot determine long-term effects. Nevertheless, as several cancer types included in the present study have high mortality, longer term information may be difficult to obtain in future studies.

Conclusions
Correct learning and regular use of progressive muscle relaxation techniques in the abbreviated version as described by Bernstein and Borkovec [34] contributes to short-term improvements in the perceived quality of life of patients with cancer.