This is, to our knowledge, the first study to explore how adolescents and their parents experience having the adolescent's CD detected through mass screening, as well as their opinions about future screening efforts.
The mixed-method study design allowed for an in-depth understanding of adolescents' and parents' reasoning, as well as quantified measures of the representativeness of some of the findings in the study population. Focus group discussions build on group interaction and can facilitate sharing experiences, especially when attempting to elicit children's views . The emergent design of our study implied that the analysis from the first focus group discussions helped to steer the discussions towards a deeper exploration of details in the following groups. Preliminary analysis of focus group discussion data guided the triangulation of data collection methods by suggesting a need for including narratives, as well as short-answer questions in follow-up questionnaires.
A limitation of our study is that we did not gain approval from the ethical review board to invite those who declined participation in the screening and those whose serological marker results were positive but who declined further clinical investigation, a group likely to be more negative towards mass screening than our informants. Another limitation is that our study does not allow for exploration of differences between experiences of receiving a screening-detected CD diagnosis in a research study compared to a general screening program or being diagnosed due to symptoms. It is likely that the context of being invited to a screening study influenced both decision-making and reactions to the diagnosis in a way that routine screening offered within the healthcare system may not. The quantified measures of the proportions of individual narratives supporting different categories should of course be interpreted with caution, since the informants might support a certain view without having expressed it in the narratives. However, these figures do provide information concerning what the informants spontaneously brought up in their accounts.
We found that the reasons adolescents and parents participated in the screening involved a feeling of duty to contribute to research as well as a means of handling the risk they were introduced to. Their idea that participation would lead to greater benefits for everyone most likely reflects societal norms to help others. These findings are in line with results found in a previous qualitative ETICS sub-study involving adolescents who had not yet received their test results . However, in that study the adolescents felt involved, well informed and perceived that they had a general understanding about CD. Contrary to this, we found that some adolescents in retrospect felt not having participated in the decision to take part in the screening. They also felt that they had not been provided with enough information about the possible consequences of their participation and expressed a need for improved information before the test. These results imply that it is probably first when diagnosed that the consequences are fully incorporated and detailed information is sought. Nevertheless, this underlines the importance of true informed consent, which is especially challenging when approaching minors and their guardians. Our findings emphasize the importance of involving children and adolescents in the information and consent process, even if the final decision is made by their guardians. However, the information provided when inviting persons to CD screening programs must be balanced to avoid unnecessary anxiety while waiting for the test results, since the majority of those participating in the screening will not be diagnosed with the disease.
Being surprised, feeling angry and sad, or even questioning if the test results were correct characterized the initial reaction to the diagnosis. Later these feelings developed into a belief in the benefits of becoming aware. The emotional responses of denial, anger, and later acceptance correspond to stages of grief first described by Elisabeth Kübler-Ross . As most adolescents and parents had no suspicion about the disease, they felt unprepared. Lack of sufficient preparation for receiving "bad news" has also been observed in studies of other screening-detected diagnoses [33, 34]. Having a moderate amount of concern prior to a stressful event (proactive coping) is suggested to facilitate the psychological adaptation once the event occurs . As a screening-detected diagnosis is often unexpected, and the transition from being healthy to becoming a patient is abrupt, this probably influences the psychological adaptation process negatively. Healthcare providers therefore need to be sensitive to what screening-detected patients actually know about their disease, and how they cope with the diagnosis, in order to provide adequate support and follow-up care. However, in our study we also found that for some informants, mainly parents, the diagnosis actually decreased anxiety and was perceived as a relief, since it provided an explanation for the symptoms that were experienced. These results are supported by other studies showing that among screening-detected CD cases true symptomatic cases are also found [15, 36]. Overall, boys and girls seemed to have similar reactions towards receiving the diagnosis. However, we found that mothers, compared to fathers, were more prone to express feelings of guilt for not previously having suspected what their child was suffering from. This gender difference in parental coping has been observed in other studies [37, 38].
The screening seemed acceptable to most of the adolescents and their parents. They felt grateful for becoming aware, based on experiences of improved health after initiated treatment or an adaptation of the risks of what otherwise could have happened. In fact, those who were asymptomatic before diagnosis, and their parents, seemed even more prone to advocate screening, as it would be the only way to identify their disease. We also observed that the individual doctor's view of risk strongly influenced the informants since the risks mentioned differed between study sites. The informants' accounts reflect how the screening itself produced concerns, resulting in an increased risk awareness during the screening process. Implementation of screening programs in society as a means of preventing complications could be seen as part of the modern risk society described by Beck . In our study, some informants had ambivalent feelings about the personal benefits. According to Beck this can be seen as a consequence of the modern risk society where increased awareness of risk is followed by an increased capability to question the suggested risks. It is often people themselves who start to reflect on risks, since it become harder to rely on scientists to do this for them.
A common argument against CD screening is that strict compliance with a gluten-free diet is hard to achieve in symptomatic CD patients, and even harder in those presumed to be asymptomatic [18, 40]. Indeed, we found that experiencing health improvements after initiating the gluten-free diet functioned as a motivator for adherence. However, even if the adolescent's health had not improved, most adolescents and parents expressed that avoiding future health complications motivated dietary compliance. In addition, adhering to the diet was perceived as more important than following health promotion messages aimed at society as a whole, since CD was perceived as more of a personal threat.
Both adolescents and parents welcomed future CD screening but thought it should be implemented earlier than in the present screening. Previous studies have shown that compliance with a gluten-free diet among adolescents is higher when CD is diagnosed earlier in life due to symptoms . However, a crucial factor in deciding the optimal age (or ages) involves obtaining a balance between identifying the majority of cases, and not having to repeat the screening many times. It is now evident that CD can develop at any age; however, the age distribution for disease initiation (sero-conversion) is not yet known. Thus, multi-country birth cohort studies with repeated blood-sampling for analysis of CD serological markers are needed, and such studies are underway .
In our study, some informants questioned the benefits of the CD screening and asked for more scientific evidence about the consequences of exposing the population to a mass screening. However, most adolescents and parents responded that if there are means to find the disease, then everyone should have the right to be offered the test. Screening to identify undiagnosed CD was seen as beneficial for health outcomes, but the diagnosis was also important in itself. The latter finding relates to the ethical considerations introduced by genetic screening tests that are increasingly advocated, where preventive measures are sometimes not available . If information about a diagnosis (or carrier status) is alone viewed as a right and as sufficient for advocating screening, some of the currently accepted principles for screening will be ruled out. Thus, the challenge with mass screening is that the presumed "right to know" must be balanced against the principle of non-maleficence, and the autonomy of individuals with a right to choose not to know.
Before implementing a mass screening for CD we need to ensure that the potential benefits outweigh the harm (and costs) both for involved individuals and society. Many of the suggested mass screening criteria are fulfilled for CD, but not yet all of them. Further research is needed to evaluate the extent to which treatment of screening-detected CD reduces long-term negative health consequences. In addition, health economic evaluations are needed specifically for CD mass screening, but also allowing for comparison with other public health interventions. We have taken a first step to increase the understanding of CD screening acceptability from the perspective of newly diagnosed adolescents and their parents in a Swedish context. We believe that some of our results are transferable to screenings in other settings, and for other diseases. Nevertheless, additional studies on the acceptability of CD screening are needed that involve other age groups and cultural settings, as well as those individuals found to have normal and false positive results in a screening.