The present study focuses on self-reported allergy-like conditions among adolescents, in particular food hypersensitivity. The results show that as many as 64% of the adolescents reported allergy-like conditions, of which nearly one third reported food hypersensitivity. In most cases the allergy-like conditions were complex, i.e. included hypersensitivity to multiple offending substances and/or allergic diseases.
Compared with the prevalence for Swedish children 13–14 years of age, shown in an ISAAC-study , the adolescents (13–21 years) in the present study reported about the same prevalence of asthma/wheezing (15% and 16%, respectively) and of eczema (17% and 20%, respectively), but a higher prevalence of rhino-conjunctivitis (26% and 38%, respectively). The differences in prevalence of rhino-conjunctivitis between these studies might to some extent be explained by the fact that different age groups were sampled. Furthermore, the referred ISAAC-study was performed at least six years earlier than the present one. During this time span the allergy problem has been a growing concern among both children and adults and the rates may have risen [1, 2].
The adolescents with allergy-like conditions generally showed significantly lower HRQL than adolescents without such conditions. Previous studies have shown that doctor-diagnosed asthma, eczema, and rhinitis have negative impacts on HRQL [10–18]. In the present study we have shown that those who reported that they suffered from these allergic diseases scored low on SF-36, regardless whether the diseases were verified by medical expertise or not. Furthermore, also the adolescents who reported hypersensitivity without having such diseases, scored low on SF-36, especially on the scales concerning mental health and emotional-social functioning. This is in accordance with previous findings in children with doctor-diagnosed peanut allergy [20, 21]. Living with constant vigilance, uncertainties and risks of adverse reactions, is likely to influence HRQL in adolescents in a negative way.
The question of co-morbidity has been discussed , as it is not always possible to grasp what component(s) of a complex allergy-like condition affects the HRQL. Furthermore, one has to consider the possibility that the HRQL-deteriorations in these adolescents may not be a direct effect of their allergy-like conditions, but related to an overall poorer general state of health. The presence of poor social, mental or somatic health may increase the perception of allergy-like conditions. Still, irrespective of what the underlying causes are, it is evident that adolescents who experience allergy-like conditions also experience HRQL-deterioration.
In most of the comparisons between different subgroups of adolescents with and without allergy-like conditions, the SF-36 health scale for physical functioning (PF) showed no significant difference. This is noteworthy, as the physical parameter often is in focus when health-care professionals assess an individual's state of health. However, it has been previously shown that in patients with asthma/wheezing, the link between lung function and HRQL is weak [11, 12] and the results of the present study indicate that the link between physical parameters and the HRQL may be weak also in other kinds of allergy-like conditions. Hypersensitivity may perhaps be considered as a practical, emotional and psychosocial health problem – not primarily a physical. It has been shown, however, that HRQL-deterioration among peanut allergic children is related to anxiety and fear for adverse reactions [20, 21]. Moreover, the adolescents with a non-severe chronic allergic disease may be well adapted to the disease, physiologically and/or psychologically, so that the disease as such has no significant impact on their physical quality of life.
In the present study we show that in adolescence, significantly more females than males experienced not just asthma/wheezing but also eczema/rash, rhino-conjunctivitis, and hypersensitivity to food, dust/mite and nickel. The females presented more complex allergy-like conditions compared with the males. In addition, females with allergy-like conditions showed more severe HRQL-deterioration compared with the males. It is known that female gender among adults implies a larger report of burden of health problems in general  and SF-36 Swedish normative data show that females 15–19 years of age score lower compared with males in this age group . Thus, the gender differences with respect to allergy-like conditions were in accordance with a known pattern within the health-and-gender-field. Biological, hormonal and socio-cultural explanations to gender differences in asthma and allergy have been discussed [24, 31, 32] as well as possible gender biased diagnostic practices [33–35] such as underdiagnosis of females due to gender differences in disease severity. Further research in this area is needed so that health-care professionals, school personnel, relatives and friends can improve their care and support given to both females and males suffering from allergy-like conditions.
Seventeen per cent of the adolescents, of whom the most part were tested during school age, reported positive allergy test results. It can be assumed that there were some additional adolescents with doctor-diagnosed allergy but without verifying test results. Thus, the total of adolescents doctor-diagnosed as having allergy may well be more than 17%. However, in the present study the focus was on self-reported (perceived or verified) allergy-like conditions. The mean SF-36 scores did not differ whether the adolescents reported objectively verified allergy or not. The lack of difference in HRQL was not surprising as a diagnostic test that verifies allergy says nothing about the individuals experience or the severity of the allergic condition.
Most food allergies are something that children outgrow  and adverse reactions to food should consequently be a problem that decreases with age. The prevalence of food allergy is estimated to 5–8% in children and 1–2% in adults . However, figures of perceived reactions to food may well be over 20% . In the present study, 19% of the adolescents did report adverse reactions to food. This high figure may be explained by the fact that the reactions were self-perceived and not necessarily doctor-diagnosed and that the figure includes all kinds of food hypersensitivity regardless of the mechanisms behind the adverse reactions. The existing diagnostic methods for food hypersensitivity are not sufficient and the underlying mechanisms of perceived food hypersensitivity are not always known . Nevertheless, it is noteworthy that almost every fifth adolescent perceive herself or himself as food hypersensitive and subsequently avoid certain food items.
Food items constituted a considerable part of the offending substances reported in this study and up to 41% of all adolescents with hypersensitivity specified at least one food item as an offending substance. Professional counselling and diagnostic procedures may to some extent be able to help the adolescents to reduce their food avoidance. Yet, this kind of perceived allergy-like condition – regardless of what the underlying mechanisms were – was evidently associated with HRQL-deterioration and the adolescents' experiences deserve sincere attention. Previous research show that food allergy in a child implies disruption in daily life and HRQL-deterioration for both the child and the family [19–21].
A pattern emerged in this group of food hypersensitive adolescents, showing a great deal of hypersensitivity to pollen, rhino-conjunctivitis and OAS-like symptoms, which is in accordance with the well known cross-reactivity of pollen and food allergens . Significantly more males than females reported additional hypersensitivity to furred animals. To our knowledge this has not been reported before but may be associated with the higher atopy prevalence among males . The results also pointed to a possible gender difference in what offending food items females and males, respectively, experience. Further research is needed to elucidate such a phenomenon.
More than half of the adolescents with food hypersensitivity reported positive allergy tests, but as a consequence of how the questions in the survey were asked, it is not known if they were verified as allergic specifically to food. However, a vast majority of the food hypersensitive adolescents did undoubtedly suffer from complex allergy or allergy-like conditions, which included multiple types of hypersensitivity and/or allergic diseases. This should require competence in health-care when trying to tackle the adolescents' multifaceted problems, including food hypersensitivity.
The extensive number of adolescents who participated in the present survey, together with a very high answer rate, makes the results reliable. However, the sample of adolescents used in this study may limit the generalizability as the socio-demographic distribution in this particular municipality was slightly above in socio-economics and slightly below in number of immigrants compared with the country as a whole. Additional studies are warranted to confirm the results.
The results of the present study emanate exclusively from adolescents' statements. There is always a risk that the respondents involved do not correctly remember things that were asked for in a questionnaire. However, our main interest was in adolescents' experiences of allergy-like conditions during the past twelve months and daily functioning during the past four weeks. Remembering correctly over a longer period of time was only of importance in questions about allergy testing. It seems likely that the adolescents would remember such events as skin prick tests or blood tests during their school age, although tests carried out in early childhood might have been unknown or forgotten.
The magnitude of allergic diseases was measured by means of ISAAC-questions. The ISAAC questionnaire, which asks for events during the past 12 months, has been used in many countries all over the world and for many years. Its results constitute basis for international comparisons and it can be considered well validated. However, hypersensitivity items are not included in the ISAAC questionnaire. The questions in the present study about hypersensitivity were developed specifically for the present study and a pilot test was performed. Only lexical adjustments were needed.
It could be discussed if the fact that the questionnaires asked for events during two distinct periods: 12 months (allergy-like conditions/ISAAC-questions) and 4 weeks (daily functioning/SF-36) might have biased the present results. It is also possible that an allergy-specific HRQL measure instrument would give another picture of the physical scale in relation to allergy-like conditions.