Mixed methods
We used a mixed methods approach with a convergent design. The quantitative and qualitative data were collected in parallel during the same time frame by the research group behind this article [22]. A subsample of interviewees was recruited from the Young & Active population that completed the KIDSCREEN 52 survey [23, 24]. The quantitative and qualitative sets of data were analysed separately, and then brought together for analysis and comparison during the discussion.
Study context and participants
The 12-week intervention study, Young & Active, was conducted with adolescents, age 13 – 14, from 2012 - 2014. School nurses in three counties from rural and urban districts of south-eastern Norway assisted in the recruitment to Young & Active. Following the recommended height and weight measurements in the 8th year [25], adolescents with age and gender adjusted BMI above 25 were invited to participate in the intervention. Adolescents involved in treatment programmes or other interventions were not eligible. A total of 84 adolescents volunteered to participate. As part of the Young & Active study, measurements of height and weight, a standard fitness test, and self-report instruments were individually completed at the adolescents’ schools. During the intervention, the adolescents were asked to register all weekly PA. This allowed them to follow their activity level on graphs and figures. They also kept a PA diary and could ask questions about PA if they wished. Once a week, they received online feedback and counselling from one of the researchers, an experienced physiotherapist. The counselling was based on principles from Motivational Interviewing (MI). Knowing that overweight or obesity, and adolescence in itself, potentially increases adolescents’ vulnerability, we considered it relevant to complement the study with qualitative interviews of a sample of the participants, following the post-intervention measurements and testing in the Young & Active intervention study [9, 21, 26, 27].
The qualitative data involved semi-structured interviews with a sample of 21 adolescents selected among the 84 participants in the Young & Active intervention. They were recruited to interviews by the researchers during the testing shortly after completion of the intervention. All adolescents were asked if they wanted to participate in the interviews. Based on a strategy of convenience sampling (14)[28], we accepted those who said yes to our request. Seven boys and 14 girls agreed to be interviewed. Distribution of gender, weight and cardiorespiratory fitness among the interviewees followed the main study [29]. After the interview sessions, we conducted a review of the data to consider if more informants were needed [30]. We concluded that 21 interviews was sufficient and that saturation was obtained, meaning no new themes were emerging [31].
In addition, data from a larger Norwegian HRQoL validation study involving 244 individuals, age 13–14, were applied. These data were extracted from a cross-sectional study, conducted across 20 randomly selected schools in Norway, involving 1,123 children, age 8–18 [32].
Measurement of health-related quality of life
HRQoL, in the Young & Active study, was measured using the Norwegian version of the KIDSCREEN-52 questionnaire [32]. The questionnaire covers ten subscales of social, mental and physical well-being, with five to seven sub-items rated by each individual using a five-point Likert scale [13, 33]. The scale indicates the frequency of certain behaviours or feelings (1=never to 5=always), or the intensity of an attitude (1=not at all to 5=extremely). The time frame refers to the previous week. The number of items in each dimension is illustrated in Table 1. The dimension scores in KIDSCREEN were converted into a 0–100-point scale, with 100 indicating the best HRQoL and 0 the worst. The KIDSCREEN instrument has been tested on 22,827 children aged 8–18 years across Europe and has shown satisfactory reliability and validity, both on healthy children and children with chronic health conditions [34]. In Norway, the instrument was tested and validated on 1,123 children, aged 8–18 years, showing a Cronbach's alpha value above 0.80 for all KIDSCREEN scales, which suggests good internal consistency reliability for the instrument [13, 27, 32].
Table 1 Overview of KIDSCREEN-52, a short description [35]
|
KIDSCREEN scales
|
No. of items
|
Short descriptions
|
1.
|
Physical well-being
|
5
|
Level of physical activity, energy, fitness and health
|
2.
|
Psychological well-being
|
6
|
Positive and negative emotions, life satisfaction and optimism
|
3.
|
Mood
|
7
|
Degree of depressive moods and stressful feelings
|
4.
|
Self-perception
|
5
|
Positive or negative bodily appearance, through questions about satisfaction with looks, clothes and accessories
|
5.
|
Autonomy
|
5
|
Freedom of choice and opportunities to create social and leisure time
|
6.
|
Parent relation and home life
|
6
|
Relationships with parents and home atmosphere
|
7.
|
Financial resources
|
3
|
Perceived quality of financial resources
|
8.
|
Social support and peers
|
6
|
Social relations with other adolescents
|
9.
|
School environment
|
6
|
Perceptions of cognitive capacity, feelings about school
|
10.
|
Social acceptance (bullying)
|
3
|
Aspects of feeling rejected by peers in school
|
Statistical analysis
We used IBM SPSS version 25 to conduct statistical analyses of the KIDSCREEN data. Continuous data were not normally distributed and were therefore described using median and range. Categorical data were presented as counts and percentages.
As the data did not follow normal distribution, non-parametric Mann-Whitney tests were used to examine the differences between the Young & Active sample and the reference sample on the ten dimensions of the KIDSCREEN scales.
In order to achieve an accurate estimate, bootstrapping was applied to the sample. When a sample is relatively small, this is a useful method of increasing the sample size without having to collect repeated samples from the population of interest [36]. P-values <0.05 were considered statistically significant and all tests were two-sided. As the study was considered exploratory, no correction for multiple testing was performed.
Qualitative interviews
The research group developed a semi-structured interview guide containing 11 open-ended questions. The interview guide ensured consistency and flexibility in the approach to enable the adolescents’ accounts to emerge. Probes were open and specific to the adolescents’ comments. The interview guide included questions that focused on the adolescents’ everyday lives. Examples of questions included; could you describe yourself to me, your everyday life, health, sleep, self-image, body image, life satisfaction, moods and emotions, family relations, leisure time activities, friendships, school, views on PA and perceptions of participation in Young & Active. The first author conducted all 21 individual interviews following the post-intervention measurements in Young & Active. Interviews took place during school hours at the adolescents’ schools with only the adolescent and the interviewer present. In our opinion all adolescents spoke freely, but some of them were less talkative that others and gave shorter responses. The longest interview took 70 minutes. All interviews were audio-recorded and transcribed verbatim by the first author.
Table 2 Overview of questions in the semi-structured interview guide
1. Family and everyday life
|
2. Friends
|
3. School
|
Whom do you live together with?
What does a typical day look like for you? What do you like to do in your spare time, and together with whom?
Meal patterns at home, and in school?
|
Do you have friends?
How often do you meet with friends, and what do you do together?
|
How do you experience school (courses, teachers, classmates)? Do you participate in physical education classes?
|
4. Self-image
|
5. Mood
|
6. Health
|
What do you think about yourself?
What do you think others think about you?
|
How would you describe your mood?
What is important for you to feel satisfied in your everyday life?
|
What are your thoughts about health?
How do you experience your own health?
|
7. Sleep
|
8. Activities
|
9. Physical activity
|
Do you sleep well? What is good sleep for you?
|
What kind of activities do you enjoy doing? Is there anything that prevents you from participating in those activities?
|
Could you tell me your thoughts about physical activity? What motivates you to be physically active? Is there anything that hinders you from being physically active?
|
10. Organized sports
|
11. Experiences with participation in the research project
|
|
Do you participate in any organized sports? How often? How do you experience participation? Do you participate for your own enjoyment?
|
What made you join the Young & Active intervention? Was anyone other than you involved in the decision? How did you experience the intervention?
|
|
Qualitative content analysis
The interview data were analysed using a qualitative directed content analysis [37]. The ten KIDSCREEN subscales were used as key concepts and theoretical basis (see Table 1).
The first author listened to the recordings and read through the transcripts multiple times, and coded the adolescents’ statements according to the ten subscales and definitions in KIDSCREEN 52. All statements were then summarized. The results were discussed and agreed upon in the research group.
Ethical considerations
The study was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (REK no. 2010/2978A). Parents gave their written consent for their children to participate in the Young & Active intervention study and in the qualitative research interviews after completion of the intervention. The adolescents also gave their written assent. Information about the study, confidentiality, potential hazards and benefits, the purpose of the study and the right to withdraw were emphasised, both in the written material and verbally to the adolescents. The anonymity and confidentiality of the participants and the material were ensured by using subject identifier codes instead of names. School nurses were aware of the study and were prepared to help adolescents who needed support in cases of personal difficulties.