Perspectives of adolescents, parents, and teachers on barriers and facilitators of physical activity among school-age adolescents: a qualitative analysis

To investigate social-ecological factors influencing adolescent PA, we collected qualitative data from participants in Taza city, a medium-sized city (207,984 inhabitants) in north-central Morocco [39], from February to July 2016. Specifically, semi-structured focus groups (FG) were asked about their perspectives and experiences related to factors that influence adolescent PA. Because triangulation of data sources guarantees the credibility of results from qualitative studies [24, 40], we included three types of participants: adolescents, parents, and teachers. The study was approved by the Faculty of Medicine and Pharmacy of Casablanca Research Ethics Committee and by the National Control Commission for the Protection of Personal Data (A-RS-193-2015) and authorized by The Provincial Directorate of Education Ministry in Taza. Informed consent was obtained from all parents and teachers. Adolescent consent forms were signed by their parents, but an additional verbal consent was also given by adolescent participants. Participant privacy was protected through anonymous and voluntary participation.

This qualitative study is part of a larger underway research project [41], which has a major goal of developing and implementing an intervention that supports a healthy lifestyle among adolescents. To reach this goal, potential factors influencing risk behaviors (e.g., PA and nutrition) were first identified through a mixed methods approach involving a quantitative study [42].

Participants were recruited from two middle schools (disadvantaged and advantaged per socioeconomic level) that participated in the previous quantitative study [42]. From each school, 14- to 16-year-old students were invited to participate in FG discussions. This age group of students was selected because their curriculum includes topics related to health risk behaviors. Teachers were of the three school subjects that are concerned with health risk behaviors: “Physical Education,” “Life and Earth Sciences,” and “Islamic Education.”

Participant recruitment was significantly facilitated by directors and the general surveyor of schools. Adolescent, parent, and teacher participants first received an information letter specific for their group type. The letter provided a comprehensive explanation of study objectives and processes and contained an informed consent to sign. Second, after the participant informed consent was received, researchers, participants, and school officials selected an appropriate date and setting to conduct FGs. Unlike quantitative studies, the number of participants was not previously calculated, and sampling continued until data saturation was reached.

We chose the FG method because it encourages interactive discussions between participants [43]. Discussions were conducted by a moderator and an observer who are already trained to conduct FG discussions. The moderator managed the discussions with participants while the observer noted nonverbal cues (e.g., mimics, gestures, silence) and group interactions that seemed useful to understanding nonverbal participation. All FG discussions were conducted in the Arabic language, the mother tongue of the participants and interviewers.

The FG discussions followed a semi-structured interview that began with an introduction, followed by an opening question, and then open-ended questions that served to map further discussions to identify barriers and facilitators of adolescent PA (Table 1). Our FG guide, which was developed using the social-ecological framework as a theoretical framework [15], was pilot tested with a convenient sample of each type of participant and refined before definitive use. The total number of FGs was determined by the saturation principle.

FGs were homogeneous regarding sex to overcome possible bias and to provide an opportunity for possible gender-specific factors to emerge [29]. In both schools, FG discussions were conducted in a suitable room where participants felt most comfortable and which allows a good quality audio recording. Participants also gave permission to record and use their anonymous quotes in research publications.

Audio tapes from the discussion were fully transcribed and organized into separate datasets for adolescents, parents, and teachers. Relevant field notes such as nonverbal cues were also included in the transcribing process. All datasets and notes were analyzed by the coders using thematic analysis method of Braun and Clarke, which is a method for identifying, analyzing, and reporting themes within data [44]. Themes are patterns which capture something interesting about the data in relation to the research question [44]. Coders followed one-by-one the different steps proposed by Braun and Clarke: become familiar with the data, generate initial codes, search for themes, review themes, define themes, and write-up [44]. Coders started reading and re-reading the interview transcripts to become familiar with the data, took notes, and jot down early impressions. Then, open coding was used; that means we did not have preset codes but developed and modified the codes as we worked through the coding process. In addition, the coding process included consideration of our research question. So, this was an inductive-deductive thematic analysis. The next step was to search for themes; once the initial codes were generated, we started to categorize and combine them to form overarching themes. Themes were predominately at the semantic level, i.e., they are identified within the explicit or surface meanings of the data. After developing the initial themes, they were continually reviewed, created, and discarded using an iterative process. Thus, we constantly moved back and forth between the selected extracts from the data and the entire dataset to check if the themes make sense and account for all the coded extracts and the entire data set to assess the applicability of themes. Finally, each theme was defined, labeled, and analyzed. A thematic map was also established; it shows the organization of themes according to different levels of the social-ecological model and possible interactions between themes.

To ensure transparency and reliability, all transcripts were coded thoroughly by two researchers (the first and the second author of this paper) independently. Regular meetings with the research team were conducted to discuss and agree on any outstanding questions. The analysis team discussed their coding and interpretation of the transcripts in detail, and any coding differences were discussed and resolved to refine codes and identify key themes from the data. We translated into English the quotes used to illustrate themes and subthemes. No software was used in the data analysis.

This theme refers to internal motivations and restrictions toward adolescent PA. It comprises three subthemes: enjoyment and competition, perceived health benefits, and perceived limiting factors.

This theme could be both a barrier and a facilitator of PA, depending on the level of adolescent awareness regarding PA. Awareness included having knowledge of adequate PA duration and intensity and the benefits on health (e.g., physical and psychological health, disease prevention). Participants stated that media, family, and school play a role in raising PA awareness, although some participants argued that awareness played only a modest role. Parents and teachers highlighted that young people were often disinterested toward advice from others.

This theme comprised two subthemes, both being barriers to PA: (1) overloaded curriculum and (2) time mismanagement and competing leisure sedentary activities.

Active adolescents had support from family members, which was also confirmed by parents and teachers. Parental support included encouragement and incentive to exercise, enrolment in sport clubs especially at an early age, and parental modeling. In addition, a family atmosphere without problems or conflicts supported adolescent PA. Support from friends was also stated by participants as a facilitator of adolescent PA. In addition, some adolescents practiced PA just to meet and spend time with friends.

Many participants highlighted that lack of support from parents, family, or friends has a negative influence on adolescent participation in PA. Lacking PA awareness in the family, not having friends, having friends who are inactive, or having friends with other interests or negative interests (e.g., substance use) keep adolescents away from practicing PA.

Some participants expressed that role models could also include successful national sports figures who may inspire adolescents to engage in sports.

As reported previously [25‐29, 31], positive expectations in PA, such as having fun, spending time with friends, and breaks to daily routine, were important facilitators of adolescent PA. Similarly, all participants considered that health benefits, weight management, and preserving a particular physical appearance were important reasons to engage in PA. These factors were also identified in previous studies [25‐28, 31]. In literature, competition is often considered as a barrier to adolescents’ PA especially among girls and those lacking skills or physical conditions [25, 26, 28, 46]. In our study, competition was perceived as a facilitator for PA by some participants, which were almost exclusively boys. A likely explanation for this result may be that these boys have good skills or physical conditions, which encourage them to enjoy physical activities of a competitive nature. Self-defense, against an aggressor for example, was a reason for some adolescents (especially girls) to engage in self-defense sports, expressing feeling unsafe within their neighborhoods. In line with a widely known custom among boys but unusual in the literature, many boys stated that they play football to win money. Hence, organizing sports competitions with prizes to the winning teams could encourage PA in boys.

Similar to previous studies, adolescents with physical disability [47] and/or with poor physical condition and skills [25, 27, 28, 31, 46] engage in less PA than their normal peers. In a systematic review, Carlon et al. found that youth with cerebral palsy participated in 13 to 53% less PA than their peers and were more sedentary [47]. Long-term negative health consequences of inactivity among disabled youth are therefore more likely, which can in turn reduce their life expectancy [47]. Therefore, intervention programs that seek to increase the PA level among this category of population are needed. Interventions can target some modifiable physical, psychological, and environmental correlates identified in previous studies [48]. However, it was concluded in two recent systematic reviews [49, 50] that the effectiveness of this kind of interventions was not always sure, and future studies are required before giving recommendations for such interventions. In the literature, physical disability was rarely reported as a barrier, especially in relevant systematic reviews of quantitative studies [16‐18, 20]; this may be because questions related to disabilities were not included in questionnaires measuring PA. Thus, considerations of physical disability in the development of such tools could prevent possible biases in the estimation of PA level.

Similar to a number of previous studies [27‐29, 46], our results showed that participation in PA decreases negative psychological characteristics such as low desire and importance, poor enthusiasm toward PA, and laziness, timidity, and fear of accidents. Thus, improvement of these mediators can positively affect adolescent PA level.

Smoking was perceived as a barrier to PA by participants of this study. This behavioral factor was not identified as a barrier in qualitative studies, although results from quantitative studies are inconclusive [16]. Furthermore, in a longitudinal cohort study, Audrain-McGovern et al. concluded that adolescent smoking uptake depended on the kind of PA; some types of PA are negatively associated with smoking uptake, whereas others are positively associated [51]. In studies supporting a negative association between PA and smoking, the underlying mechanisms are not yet understood. Thus, Audrain-McGovern et al. found that PA reward as one mechanism by which PA can reduce the likelihood of adolescent smoking uptake [52]. In the same sense, Verkooijen et al. conducted another study aimed at gaining a better understanding of this association and found that some motives like friendships and competition appeared to moderate the association between PA and smoking rather than others (e.g., weight management and self-esteem). This study recommends focusing on first motives and avoid the last ones when designing a smoking prevention program [53]. Finally, the relationship between PA and smoking is complex, further research is warranted to identify other mechanisms underlying this relationship to provide evidence-based smoking prevention targets [53].

Adolescents and parents were generally aware of the health benefits of PA. In line with previous studies [29, 54], this was considered as a facilitator of PA practice. Regarding adolescent PA guidelines, most adolescents and parents did not know that adolescents must accumulate at least 60 min of daily moderate to vigorous intensity PA to stay healthy. Being unaware of this threshold may lead to an overestimation of PA level among adolescents, believing themselves to be more active than they really are [55]. In the same way, parents who overestimate the PA level of their children may see no need to encourage them to increase their PA [55]. Improving awareness related to PA health benefits and recommendations among adolescents and their parents may encourage behavior changes in adolescents [55]. Furthermore, involving media, school, and family may be useful in raising PA awareness among adolescents because participants considered them to be the main awareness-raising sources of PA.

Our study results agree with previous findings that lack of time was a strong barrier to adolescent PA [25, 28, 29]. Participants linked the lack of time to school demands (e.g., overloaded curriculum, private lessons, homework, and giving higher priority to academic success), screen device use, time spent with friends, time mismanagement, and household tasks (among girls). Improving the adolescent’s time management skills should be targeted when setting up an intervention program. Data from the latest Global School-based Health Survey indicated that 32.9% of Moroccan adolescents spent 3 or more hours per day on sedentary activities, not including time in school or doing homework [33]. It is proven that sedentary behavior and PA are two different constructs and that they do not directly displace one another [56]. Therefore, efforts to encourage decreases in sedentary behavior time and increases in PA must go together [56].

Social support has been consistently associated with adolescent PA [21, 57]. This support is provided by different sources (parents, family, friends, and others) and in different ways, either tangibly (e.g., doing activity with, watching/supervision, transportation, payment of fees) or intangibly (e.g., encouragement and praise, discussing benefits, role models) [20, 30, 57]. Similar to previous studies [28, 29, 58], our findings showed that tangible and intangible social support provided by parents, family, friends, and sometimes by national sports leaders was a salient facilitator of adolescent PA. However, an unsupportive social environment was a barrier to PA. Results also indicated that not having friends or having friends who are physically inactive or with other interests and negative interests (e.g., drug use) may keep adolescents away from PA. Therefore, social support should be targeted in intervention programs aimed at increasing PA levels among adolescents.

Girls are less active than boys, but the reasons behind this are not well understood [59, 60]. These gender differences were explained previously by non-modifiable variables, including a girl’s biology [61], and by some modifiable variables related to a girl’s psychology [60, 62, 63], social support [59, 60, 62], or cultural and physical environmental factors [60, 62]. Our use of single-gender FGs and involvement of parents and teachers proved useful in identifying barriers that led to inactivity in girls. Findings showed a variety of barriers mainly of a sociocultural aspect. Similar to most Arab countries [34, 64], girls received less social support compared with boys, as many families do not allow their girls to practice PA outdoors because of conservative cultural and religious norms. To preserve public modesty, many parents do not permit their girls to practice PA with sports dress, but with traditional dress, which is not suitable for PA [37, 64]. Some traditional long dress covering the whole body may hide women fatness and consequently reduce their motivation to do PA [36, 37, 64]. In addition, cultural expectations concerning women’s roles were considered as barriers to PA. Thus, like many countries in the world, Moroccan women were expected to care for the family and household and their exercise needs were afforded low priority [34, 35]. Participants reported that some sports were socially perceived to be valid only for boys, and there is also a lack of appropriate spaces specific for girls that are secure, safe, and well supervised. In a review on Arab adolescents, Obermeyer and associates indicated that some features of social context such as unequal gender norms can affect adolescent health [38]. Similarly, the recent Arab Human Development Report indicated that gender equality and women’s empowerment are more restricted in Arab regions than in other regions [65]. In brief, the identification of modifiable factors that explained gender differences in adolescent participation in PA is important since they can be targeted in interventions [62]. The mediators found in this study can be targeted to minimize the effects of gender and thus improve PA levels in girls. Nevertheless, objective measurements are required to ensure that they are relevant mediators because previous studies concluded that among a multitude of mediators examined, only some could explain the association between gender and PA [59, 62].

Various theoretical frameworks have been used to understand factors influencing PA. In the previous decades, the use of ecological models revealed numerous environmental factors that may affect adolescent PA levels. Identification of such factors is useful because young people are especially susceptible to environmental influences given that they have relatively little autonomy over their own behaviors [59]. Furthermore, many studies showed that interventions that affect environmental factors, along with other personal and interpersonal factors, are more likely to be effective [66]. This study showed that PE classes in school and recreational facilities outside school are two opportunities available for adolescents to practice PA.

Similar to previous studies [26, 27], our results highlighted that PE classes contributed to improving the overall PA level among adolescents. However, PE classes face several obstacles related mainly to infrastructure, equipment, schedules, and insufficient hourly volume. In such context, students, especially girls, apply for a certificate of exemption from PE classes. This negative impact of PE classes on PA levels in female adolescents was previously reported by other studies [25, 27]. Consequently, to improve adolescent PA, PE classes must be reinforced because it touches a high number of adolescents regardless of their social differences.

Outside school, participants stated that there is a lack of recreational facilities for many sports (e.g., jogging, cycling, basketball, tennis, swimming). The rare existing spaces, for football mainly, were often distant, insecure, unsupervised, not equipped, or expensive, often making these spaces impracticable. These issues were more pronounced among girls. According to the participants, a lack of opportunities outside the school was because addressing these issues was not a priority for city policymakers.

Participant recruitment in this study was limited to two middle schools in an urban area. Thus, it is not possible to generalize findings to a larger population. However, our goal was to generate a deep exploration of influencing factors of adolescent PA and not to generalize results. However, our high number of participants, as well as triangulation of participants, allowed us to obtain a more complete overview. Because of the qualitative nature of this research, results are not equal to objectively measured data. Thus, it was not possible to define which influencing factor was most significant for adolescent participation in PA. Furthermore, we did not give the percentages of participants reporting a theme or subtheme. As reported by Braun and Clarke [22], the issue raised most frequently or by a larger number of participants is not necessarily the most pertinent to answer the research question. Finally, the interactions between themes, as shown in the thematic map (Fig. 1), require an objective evaluation to prove their strength.