2.1. Design

Since the objective of this study was to explore HNC survivors’ views on PA and to compare these findings with objectively measured PA, a mixed methods approach was considered appropriate and the study was reported in accordance with the guideline for Good Reporting of a Mixed Methods Study (GRAMMS) [26]. A two-phase, parallel, mixed data analysis design was applied. Thus, in parallel, but independently, both qualitative and quantitative data were collected and analyzed. The qualitative and quantitative data were combined during the overall interpretation of the results [27]. Concepts from the ASE model served to explore subjective PA behavioral mechanisms on two levels: PA beliefs as described in interviews, and PA level as reported by using questionnaires. The data consisted of a predominantly qualitative analysis of semi-structured interviews framed within the ASE model, descriptive analysis of questionnaires regarding PA behavior, and actual measurements of PA with accelerometers, in a sample of individuals that were treated for HNC at the University Medical Center Groningen, The Netherlands. This study was a sub-study of a prospective observational study on PA in patients with head and neck cancer: the “PA in Patients with head and neck cancer” (PAP) study.

The Medical Ethical Committee of the University Medical Center Groningen (UMCG) approved the parent (PAP) study, under regulation of the Medical Research Involving Human Subjects Act (reference METc 2012/063). We obtained permission to add the semi-structured interviews to the original research protocol for the purpose of the current sub-study. The PAP study was registered in the Netherlands Trial Register, reference NTR4828. The primary outcome of the PAP study was PA as assessed by daily step count and PA level as measured by an accelerometer. Secondary outcomes included self-reported physical performance, self-reported PA level, and PA self-efficacy. The UMCG mainly serves patients from the north-east region of The Netherlands. Dutch-speaking adult patients with a new primary oral or oropharynx tumor to be treated with surgery with curative intent, with or without adjuvant radiotherapy or concomitant chemo-radiotherapy, were considered eligible. Patients with limited mobility or cognitive impairment prior to treatment were not included. Measurements for the PAP study were performed at three time points that were planned consistently with regular hospital visits: one or two days before surgery (T0), six to eight weeks after completion of HNC treatment (T1), and three months after completion of HNC treatment (T2).

All participants invited to participate in the interviews had to be included in the parent study, i.e., fulfilling the criteria used in the PAP study. The sample for the sub-study was recruited at T1, i.e., six to eight weeks after treatment. The sample of patients that were approached is a convenience sample consisting of all PAP participants who were scheduled for a T1 measurement in the spring of 2013 or in the spring of 2014. The interviews took place after completing the questionnaires and the measurements that were part of the parent study. The interviewers informed the participants about the purpose and nature of the semi-structured interviews by telephone and asked for additional consent prior to the interviews. The participants did not receive the results of the quantitative measurements prior to the interviews, and the interviewers were not present at the quantitative measurements or informed about the measurement results. Nine out of ten patients who were considered eligible agreed to participate. The patient who declined participation reported personal circumstances as the reason. Participants of neither the PAP study nor the current sub-study were financially compensated for their participation.