To be eligible, patients had to be ≥18 years old, be diagnosed with cancer, be scheduled to receive chemotherapy for at least 9 weeks at Moffitt Cancer Center (MCC), have not received chemotherapy within the past 2 months, be medically cleared for engaging in moderate intensity exercise (including Eastern Cooperative Oncology Group performance status ≤ 2), and be fluent in English.
Study coordinators identified patients scheduled for consultation with a medical oncologist and then reviewed medical records of these patients to evaluate eligibility. For those eligible, clearance was sought from the oncologist confirming ability to engage in moderate intensity exercise. Coordinators approached eligible participants after the oncology consultation to explain the institutional review board-approved study, answer questions, and obtain written informed consent from those wishing to participate. After consenting and before the first infusion, participants completed a baseline (T1) questionnaire and were randomized using a computerized program. Randomization was stratified by gender, physical activity level (i.e., inactive, minimally active, or active), chemotherapy administration schedule (e.g., 21 or 28 days), and concurrent radiotherapy (i.e., yes or no).
Intervention materials were distributed by a doctoral-level psychologist whose adherence to standardized protocols for each of the four conditions was monitored by an interventionist trainer via live observation of 12% of meetings. Immediately following intervention delivery, coordinators administered a single item (‘How important do you think that it is that we make this program available to other chemotherapy patients?’; 0 = not at all important to 6 = extremely important)  to confirm that the four intervention conditions were perceived as equally credible forms of assistance. Approximately 1 week later, the interventionist telephoned participants to ask if they had reviewed the materials and address any questions. Those participants who indicated they had not reviewed the materials were encouraged to do so. Participants completed follow-up questionnaires approximately 6 (T2) and 12 weeks (T3) following their initial infusions. Questionnaires were administered to participants during scheduled clinic visits or were completed at home.
Usual care only participants had access to the full range of psychosocial services provided to MCC patients. These included opportunities to meet with a social worker and attend support groups free of charge. In addition, patients could be referred or self-refer to in-house psychiatrists and psychologists on a fee-for-service basis. Specific to this condition, patients also met with the interventionist for approximately 5 min before the first infusion, at which time they received the National Cancer Institute booklet titled ‘Chemotherapy and You’. This 62-page booklet includes information about chemotherapy and ways to manage side effects.
In addition to usual care, SM participants met with the interventionist for approximately 10 min before the first infusion. As in our previous study , they were provided with a 15-min video, 12-page booklet, and 30-min audio recording titled ‘Coping with Chemotherapy’. The video and booklet included information, demonstrations, and instructions regarding paced breathing , progressive muscle relaxation with guided imagery , and use of coping self-statements to manage stress . Comments by chemotherapy patients about benefits of using stress management techniques were interspersed throughout the video and booklet. Participants were instructed to follow directions for how to learn and practice the techniques and use them during chemotherapy. Directions included listening to the audio recording, which led participants through an abbreviated form of progressive muscle relaxation training.
In addition to usual care, EX participants met with the interventionist for approximately 10 min before the first infusion. They were provided with a 12-min video and 14-page booklet titled ‘Stepping Forward: A Guide to Exercise During Chemotherapy’ developed for this study. The video and booklet included information and instructions on engaging in regular exercise while undergoing chemotherapy. The emphasis was on walking because it is well suited to home-based exercise. For this reason, participants were also provided with electronic pedometers (Digi-Walker SW-651-04, New-Lifestyles Inc., Lees Summit, MO, USA) on the basis of evidence suggesting that self-monitoring with pedometers promotes greater physical activity [27-29]. Topics covered in the materials included warming up and cooling down, pulse and exertion monitoring, and use of the pedometer to monitor numbers of steps taken and exercise duration. Comments by chemotherapy patients about benefits of exercising were interspersed throughout the video and the booklet. Participants were advised to exercise 3–5 times per week for 20 to 30 min at approximately 50% to 75% of their estimated heart rate reserve, a goal consistent with exercise recommendations for cancer patients published prior to study initiation . The interventionist calculated and provided each participant an exercise heart rate ‘training zone’, based on age and resting heart rate, and demonstrated how to use pulse rate to monitor exercise intensity. Participants were also instructed how to monitor intensity using the Rating of Perceived Exertion (RPE) Scale  and were advised to achieve a level between light (RPE = 11) and somewhat hard (RPE = 13).
In addition to usual care, SMEX participants met with the interventionist for approximately 15 min before the first infusion. They were provided with a 20-min video and a 20-page booklet titled ‘Stepping Forward: A Guide to Stress Management and Exercise During Chemotherapy’ developed for this study that combined content from SM and EX videos and brochures. Participants also received the same audio recording provided to SM participants and pedometer provided to EX participants. Instructions for exercising and for learning, practicing, and using stress management techniques mirrored those in the SM and EX conditions.
Demographic data were obtained at baseline through the use of a standard self-report questionnaire. Medical information was obtained through review of patient records.
The acute (1-week) version of the Medical Outcomes Survey 36-item Short Form (SF-36)  served as the primary outcome measure. Analyses focused on the physical component summary (PCS) and the mental component summary (MCS) scores. The 20-item Center for Epidemiological Studies Depression (CES-D) Scale  and 21-item Beck Anxiety Inventory (BAI)  served as secondary outcome measures. Items on both instruments are keyed to the past week.
Exercise activity was assessed using the Leisure Score Index of the Godin Leisure-Time Exercise Questionnaire (LTEQ) , which assesses the frequency and duration of mild, moderate, and strenuous exercise in the past week. Responses are scored to yield an estimate of weekly metabolic equivalent tasks (METs). Stress reduction activity was assessed using the Stress Reduction Checklist (SRC) , which assesses the use of the stress reduction methods that comprise SM. Questions are worded so that participants in any study condition could complete the checklist.
To examine longitudinal changes in quality of life across the three time points, we applied random effects models by using SAS Proc Mixed (SAS Institute, Cary, NC, USA). The planned analyses focused on comparing each intervention separately with UCO. For each outcome, initial analysis consisted of three separate 2 (group comparison: UCO versus SM, UCO versus EX, and UCO versus SMEX) by 3 (time: T1, T2, and T3) random effects analyses. The main effects of group and time, and the group-by-time interaction, quadratic time effect, and quadratic-group-by-time interaction were evaluated. Quadratic terms were included to determine if there were nonlinear intervention effects (e.g., plateaus or decrements in quality of life following improvement). Of principal interest was the significance of group-by-time and quadratic-group-by-time interactions, which would indicate differential improvement in quality of life as a function of group. If more than one intervention condition performed better than UCO, then similar analyses were conducted comparing those interventions. Principal analyses were performed using data from 286 participants who completed all three assessments. To examine the potential effects of attrition on findings, we conducted additional analyses by using data from 351 participants who completed at least one follow-up assessment. A p-value of 0.05 (two-tailed) was used for statistical significance. The study was powered at .80 to detect an effect size of f = 0.25 for the interaction between an intervention condition versus UCO and time, assuming a per group sample size of 80 and alpha = 0.05. The study was also powered at 0.80 to detect an effect size of f = 0.14 for within group change across the three measurement points.