Journal of Rehabilitation Medicine 51-7 | Page 64

Physical activity recommendations for patients with heart failure attitudes and experiences is recognized (12–13). Re- commendations by healthcare professionals have also been shown to be associated with higher levels of PA (14). Patient adherence is, in turn, multidimensional, and depends on motivations and barriers related to patient-specific factors, the healthcare system, so- cioeconomic factors and factors associated with the prescribed therapy and the specific condition (8). One factor associated with lower levels of adherence with treatment recommendations is female sex (8). Female sex is also associated with lower levels of PA, both in the healthy population and in patients with HF specifically, which is directly related to adverse health effects (8, 15). The reasons for this discrepancy have been studied in young people and, when examining differences in PA between the sexes in general, women report less self-efficacy (confidence in successfully attaining a desired behaviour), less social support and less motivation to exercise (15). However, in contrast to the vast amount of research on sex differences in HF, little HF-specific research has been published in this particular area. Furthermore, in 1 of the few HF- specific studies on this topic, women were found to report higher levels of motivation than men, although no difference in the level of PA was found (16). The aim of this study was to investigate the percep- tions that healthcare providers may have with regard to sex differences in PA, motivations and barriers, and whether adaptations in care based on sex differences might be meaningful. METHODS Study design This was a qualitative interview study conducted at 2 univer- sity hospitals in Israel: Beilinson Hospital in Petah Tikva and Soroka Hospital in Beer Sheva. Interviews were held during February to May 2017. Sampling Participants (healthcare providers) were recruited for the study using purposeful sampling, with the aim of including partici- pants of different sexes and ages, participants with different professions involved in HF care, and participants working at 2 different hospitals. As we were unfamiliar with the hospitals and their employees, we were introduced to healthcare professionals involved in HF care by cardiologists working at the hospitals. They were then invited to participate in the study in person, by e-mail or by phone call. Ethical considerations The Institutional Ethics Board of the Rabin Medical Center wai- ved interview studies in healthcare personnel. All participants provided written informed consent. They were also informed that they could end the interview if desired, that their confiden- tiality would be assured, and that data would not be shared in 533 a recognizable form with others. Furthermore, by not sharing the characteristics and names of the participants, no quotations in this paper can be traced back to the participant. Procedure and interviews Through literature review, a semi-structured interview guide was developed building on the general aims and research questions of the study. This guide comprised 2 parts addressing: (i) dif- ferences and similarities in male and female HF patients with regard to PA (e.g. “Do you think that male and female heart failure patients might have different reasons to be active?”) and (ii) perceptions on individualization of care based on patient’s sex (for example “Do you adapt your practice to account for sex differences in PA/capacity”). Pilot interviews were held and the interview guide was revised before conducting a total of 12 interviews, after which no new information came from the interviews and the data were considered saturated. For convenience, most interviews were conducted at the office of the participant and the duration of the interviews ranged from 30 min to 1 h, depending on time restrictions. Twelve health- care providers were interviewed, including 5 cardiologists, 2 residents, 3 nurses and 2 physiotherapists. The male to female ratio of the sample was 1:1 and ages ranged from approximately 30 to 65 years. The interviews were conducted in English by a (female) medical student (EC) (3 rd year) with limited experience of HF management and attended by another (male) medical student (AJ), who observed and assisted the first interviewer when needed. The interviewers had no prior relationship with the study participants. Data analysis The interviews were performed and transcribed concurrently. The transcriptions were carried out, read through and summa- rized by the first author (EC). The data were analysed using qualitative content analysis, according to Graneheim & Lundman (17). Each transcribed interview was considered a unit of analysis. The units were analysed one at a time, starting with repeated reading in order to gain an overview of the content. The interviews were then divided into meaning units, e.g. extracts from the unit of analysis that describe the same central meaning. Meaning units were then extracted, put into a table in a Word document and further condensed. Condensed meaning units were then gathered in a single collective document and sorted into groups, which in turn were given codes. All codes were subsequently read through and analysed for similarities and differences. They were then sorted into categories, i.e. a grouping that one imposes on the coded segments, in order to reduce the number of different pieces of data in our analysis (“What is in the data?”). To improve reliability, a discussion was held at this stage with a second researcher (TJ), who is an experienced researcher with an HF nursing background. Categories were revised, and codes re-sorted. Based on this information, themes were derived, and these were considered a higher level of categorization, used to identify a major element of our entire content analysis (“What is this about?”) (18–20). This was done with continuous feed- back from the second researcher. Finally, an overarching theme addressing the quintessence of the data was developed. Several measures were taken in order to improve the trust- worthiness of the analysis: • To improve credibility, participants were chosen through purposeful sampling, thus providing diverse insights into the subject being researched. J Rehabil Med 51, 2019