Introduction

In recent years, the term health-related quality of life (HRQL) is being positioned as an indicator of physical and mental wellbeing since childhood and adolescence ¹. This concept emerges as a useful indicator of the aggregate state of health, which bears in mind the individual's self-perception as a requirement in the health evaluation ². Currently, the HRQL is defined as the "subjective perception, influenced by the current state of health, of the capacity to perform activities deemed important for the individual"³ and authors, like Ramírez-Vélez et al. ⁴, propose that monitoring of HRQL in different populations will permit population identification of altered physical and/or mental health as an input to guide policies or interventions to improve their collective health.

To estimate this important indicator of the state of health, generic and specific instruments have been developed to measure child and adolescent HRQL starting at 8 years of age through self-report questionnaires ⁴. Among the most used, the Health Related Quality of Life Questionnaire for Children and Young People and their Parents (KIDSCREEN-27) ⁵, the Child Health and Illness Profile (CHQ) ⁶, the Autoquestionnaire Qualité de Vie-Enfant-Imagé (AUQUEI) ⁷, the Pediatric Quality of Life Inventory (PedsQL) ⁸, and the EQ-5D-Y ^9,10; all used and validated in Hispanic pediatric population. The EQ-5D-Y proxy is a generic and practical questionnaire, of easy comprehension and application, which provides values in various dimensions of self-perceived health, as well as a visual scale that represents the current state of health ^9,10. The EQ-5D-Y proxy questionnaire, originally developed by Wille et al. ⁹, was adapted into Spanish by Olivares et al. ¹¹, and recently a "proxy" version was published by Gusi et al.¹⁰, on Spanish pediatric and young population.

Previously, substantial differences have been shown in the self-perception of HRQL in Hispanic, European, and American children and adolescents, thus, suggesting the need to have ethnic-specific values that describe and characterize this important health indicator ^5-10. It has also been described that the young Latin American population has particular characteristics in its growth, development, and physical health, a product of the inbreeding of European, Amerindian, and African ancestors, making it difficult to establish a clear differentiation between the influence of the environmental and genetic factors and the self-perception of health ^12,13.

In spite of increased studies on HRQL, few investigations are aimed at the Latin population, specifically within the epidemiological and educational context of Colombia ¹⁴. Thereby, the objective of this study was to describe the perceived quality of life in children and adolescents with the Spanish version of the EQ-5D-Y proxy, to later use it in the Colombian school context. A second objective was the comparison between the EQ-5D-Y proxy values observed in this study and international reports.

Materials and Methods

Study type and population

We performed cross-sectional analyses of baseline data from participants in The FUPRECOL study (In Spanish Asociación de la Fuerza Prensil con Manifestaciones Tempranas de Riesgo Cardiovascular en Niños y Adolescentes Colombianos), which focused on the associations between fitness, health and non-communicable diseases. We have recently published a complete description of The FUPRECOL Study design, methods, and primary outcomes for our current cohort, residing in the metropolitan area of the District of Bogotá, Colombia (2,480 masl). The population reference used 546,000 registration records for 2013, provided by the District's Secretary of Education. This calculation required using the equation for finite samples:

Where, N= 546,000; (e) Precision= 1.5%; α= 0.05, 95% confidence interval. The sample size calculated was of 4,235. The sampling was conducted through convenience in order of arrival to the data collection point. To diminish bias because it was a non-probabilistic sample, sampling weight - a posteriori - was assigned to each participant, calculated from the stratification by age groups (±1 year). This kept in mind that "N" is the population size and "n" is the sample size, whose inclusion probabilities were π_i= n/N and the weighted samplings ω_i= n/N. The study excluded students with clinical diagnosis of physical, sensory, and intellectual disability, non-communicable diseases, like type 1 or 2 diabetes, cardiovascular disease, autoimmune disease, cancer, pregnancy, alcohol or drug abuse, and - in general - pathologies not directly related to nutrition. Effective exclusion was carried out a posteriori, without the participants' knowledge; therein, respecting their dignity.

Results

Of the 6,950 individuals asked to participate in the study, 6,599 were of school age: 3,245 males (9 and 12 years of age) and 3,354 were adolescents (13 and 17 years of age), who filled out the EQ-5D-Y proxy survey (94.9% rate of participation). Of the population surveyed, 41.7% (n= 2,751) were males and 58.3% (n= 3,848) women (mean age 12.7 ±2.3 years) (Table 1).

Table 2 presents the frequency of the problems indicated by Colombian males and adolescents, according to age and gender. In general, males show better quality of life than women in the majority of dimensions, especially in the dimensions "pain/discomfort" and "feeling sad/worried or unhappy" (p= 0.45).

Tables 3 and ⁴ show the frequency of the problems indicated during childhood and adolescence in males and females, respectively. In general, children (9 to 12 years of age) have a lower percentage of problems in all the dimensions than the adolescents (13 to 17 years of age). Dimensions related to "pain/discomfort" and "feeling sad/worried or unhappy" indicated more problems, especially in the group of adolescent females (p <0.41).

The results of the global evaluation of the state of health indicated in the VAS in Colombian children and adolescents are shown in Table 5. Both in males and females, higher values of self-perception of health were observed in the group of children 9-12 years of age (median 95, CI range 85-100) against the group of adolescents (median 90, CI range 80-99).

Table 6 compares the results from this study (dimensions and VAS) of the EQ-5D-Y proxy version with data from other international studies from Spain ^10,15,16, Australia ¹⁷, Germany ¹⁶, Italy ¹⁶, South Africa ¹⁶, and Sweden ¹⁶. In the "mobility" dimension, the scores for this work were above those reported in Germany ¹⁶ and South Africa ¹⁶. The "self-care" dimension showed higher values than the works from Spain ^10,15,16, Italy ¹⁶, and South Africa ¹⁶. The "habitual activities" dimension had similar behavior when compared to school-age children from Spain ^10,15,16, Germany ¹⁶, Italy ¹⁶, and South Africa ¹⁶. Similarly, the dimensions of "pain/discomfort" and "feeling sad, worried or unhappy" showed values above those indicated by children and adolescents from Germany ¹⁶, Italy ¹⁶, South Africa ¹⁶, and Sweden ¹⁶.

Lastly, the reliability analysis shows internal consistency results (Cronbach's α of 0.61) for the "mobility" dimension; (0.66) in "self-care"; (0.62) for "performing habitual activities"; (0.43) in "pain/discomfort"; and (0.69) in the dimension of "feeling sad/worried or unhappy". The Cronbach's α for the total EQ-5D-Y proxy questionnaire was found at 0.64.

Discussion

The study had the participation of 6,599 children and adolescents of both genders, attending basic primary and basic secondary education in 24 public institutions belonging to several localities of Bogotá, Colombia. To our knowledge, this is the first report publishing the use of the EQ-5D-Y proxy in students in Colombia. Regarding the HRQL, males report better quality of life than females in most categories - especially in the dimensions of "pain/discomfort" and "feeling sad/worried or unhappy". These results are similar to those obtained by Quiceno and Vinaccia ¹⁸ in 686 children and adolescents from Bogotá Colombia with the "KIDSCREEN-52" childhood HRQL instrument. These authors state that males perceive better levels of HRQL than females.

Most of the HRQL values within the school context that applied the EQ-5D-Y proxy version have shown results similar to those in this study ^15-17. The findings in this study agree with the work published for Spanish children and adolescents ^10,15,16, as well as with that found in other research of the same type ^17,18. Low prevalence of serious problems was also found reported in the different dimensions of the EQ-5D-Y, results that coincide in infant/child population from European countries ^10,15-17, South Africa ¹⁶, or Australia ¹⁸.

With respect to age, greater frequency was found in the dimensions of "pain/discomfort" and "feeling sad/worried or unhappy", with the EQ-5D-Y proxy questionnaire, consistently and in both genders as age increased. These findings agree with that described by Gusi et al. ¹⁰, in the study to validate the EQ-5D-Y in Spanish school-age children. It has been reported that children (9-12 years of age) have a better perception of their HRQL compared to adolescents (13-17 years of age), particularly in the dimensions referring to their perception on the school environment, as well as their physical and emotional wellbeing ¹⁹. For Guedes et al. ²⁰, the differences in health perception between children and adolescents may be due in part to the start of menarche and of the expected hormonal imbalance occurring in the female organism at these ages, reducing the opportunities women have to confront satisfactorily stressful events operating during this period of their lives. Although this difference of perception in children could be explained partially by the adolescent behavior, we could also inquire on the school's role and its specific contribution in this regard, as suggested by Quiceno and Vinaccia ²¹. Newman et al. ²², indicate that a lower health perception in adolescents can also be explained by "bullying", - a very frequent situation during the second half of adolescence-, an aspect that is quite related to the EQ-5D-Y proxy dimensions of "pain/discomfort" and "feeling sad/worried or unhappy".

In addition, the global assessment of the state of health indicated in the VAS of the study participants showed high values of self-perception of health, especially in the group of children from 9 to 12 years of age compared to the group of adolescents from 13 to 17 years of age, in both genders. In this sense, Hernan-Gómez et al. ²³, propose that during adolescence there may be a decrease of positive emotions and of vital satisfaction with a gradual increase of negative emotions that could affect the self-perception of health. This finding coincides with that reported by Urzúa ²⁴, who stated that female adolescents have the worst perception with respect to their HRQL, with the physical wellbeing and, particularly, their perception of themselves being the two dimensions with less favorable evaluation. This perception of adolescent women of themselves constitutes a warning signal for health authorities, given that it could be indicating the lack of satisfaction of adolescent women with their own bodies and, thus, a higher risk for the onset of eating pathologies ²⁵. In synthesis, the lower perception of HRQL during the adolescent stage could certainly be related to the complexity that characterizes this stage of development, especially in women from Bogotá, Colombia.

Regarding the administration of the EQ-5D-Y proxy instrument, "in paper and self-administered", it is practical and of easy application, suggesting the implementation of further studies in other Colombian regions and latitudes to analyze possible disparities with our results. Several authors have described life factors that are considered important for the HRQL, like family relationships and social support, general health, the functional state, and economic availability ^9,10,23-25.

Hence, we can state that, although new studies are needed to advance in establishing the instrument's transcultural equivalence, the results obtained constitute an important starting point to have an instrument to measure childhood HRQL useful for pediatrics and Colombian public health. The high ceiling effect (values close to 100%) of the participants may perhaps be explained in part because most of the children and adolescents were healthy and had no important health problems. For this purpose, it will be necessary to analyze the psychometric properties of the EQ-5D-Y proxy version in Colombian population.

The principal limitations of this study are those inherent to its transversal nature and type of sampling. For example, it is important to indicate that information equivalent to the components of HRQL was obtained via self-report. The self-report is a common procedure in studies with these characteristics, being the most feasible way to obtain data related to HRQL in broad populations ^2-9,23-25. Also, the transversal approach of the data could have limited identification of differences, without being able to formulate the likelihood of the existence of inverse causality. Also, socioeconomic level was not studied, although it has been described that this variable does not seem to modify substantially the perception on HRQL in children and adolescents from both genders, except for the dimension specifically related to the perception of the family's financial resources ²⁵. Among the strengths found is that the study included a large population sample and adjusted by population expansion factors of both genders, offering new perspectives on the self-perceived state of health of school-age children in Bogotá, Colombia, which should be kept in mind by those involved in the settings of planning, decision, and execution of health policies.

Conclusion

Favorable perception of HRQL is noted in the school-aged children evaluated. We expect that subsequent studies can establish interpretation standards in the general population and in different clinical populations, besides the study of their psychometric properties. In spite of its importance, methodological simplicity, and clinical use, the determination of HRQL is still not part of the protocols to evaluate the state of health of school-age children in Bogotá, Colombia.