This is a cross-sectional descriptive study. We included children over six years old of eight schools in two cities of Colombia (Medellin and Cartagena). Through medical records and questionnaires based on the ISAAC study, the ARIA guidelines and the GINA guidelines (www.ginasthma.org), we obtained information on the presence or absence of respiratory diseases and other chronic diseases, pharmacological treatment and severity of symptoms. Data were collected by the investigators from all families who agreed to participate, however, we used the Charlson Comorbidity index,14 to exclude patients with severe chronic comorbidities that could overestimate the severity of symptoms for causes other than the diseases studied and might influence the interpretation of the results.

Data on school performance in the previous school year was obtained from the academic history of each child who participated in the study. For the evaluation of the primary outcome (clinical impact of respiratory diseases in school performance), we used as measurement variables the number of absent days, the average grade of students in Mathematics and Spanish during the last academic year. Mathematics and Spanish have the same academic curriculum among the schools, so they were selected to evaluate school performance. Moreover, they are academic areas that use similar objective evaluative parameters for their grading in all schools. To present the academic score in the eight schools, we unified the results using a semi-quantitative scale: >80% (Excellent performance according to the scale of each school), 60–79% (average), and <60 (insufficient). For school absent days we took into account only those related to health issues that could be directly or indirectly related to the diseases we studied. We excluded absent days due to vacations, permits for non-medical procedures, etc.

To evaluate the secondary outcome (clinical impact of pharmacotherapy in school performance), we performed a stratified analysis among patients with respiratory diseases (asthma and/or rhinitis), according to whether the patient was receiving pharmacotherapy and if so, whether they had good adherence or not. We classify as good adherence patients who received treatment at least five days per week according to medical recommendations taking into account previous data.12

All parents signed an informed consent (with the child's consent) authorizing the review of the academic data kept by the institution, the medical records and also agreed to complete the questionnaire. The ethics committee of the University of Cartagena evaluated and granted permission to carry out the present study.

For the primary analysis, we divided the children between symptomatic and asymptomatic according to the presence or absence of symptoms of asthma or rhinitis. For the secondary endpoint we divided the group of symptomatic into two sub-groups between adherents and non-adherents. In Colombia the evaluation of Mathematics and Spanish subjects have the same objectives, and the same parameters are used in all schools, so we do not expect that variables in the system of qualification could influence the results. However, we conducted intergroup analysis in each school to identify other factors that could influence school scores that were different from the study objectives.

A sample size calculation was performed based on the probabilities difference technique (x2 test) with Yates continuity correction by the Fleiss formula. A Type I error of 0.05 was chosen; a Type II error of 0.20; a confidence level of 95%; we estimated a proportion of cases of rhinitis of 30% and of asthma of 12% based on previous studies carried out in this area; with an expected number of two controls per case. Therefore, a minimum sample size was obtained for the cases of 312 patients and for the controls of 628 patients.

The analyses were performed using the SPSS software version 21 for Windows using the Wilcoxon U test for intergroup analysis, Mann Whitney for intragroup and the T test for mean differences in independent samples. A p<0.05 was considered statistically significant. For analysis of multiple comparisons, we used the Dunn test. The proportions were analyzed using contingency tables and the Fisher exact test.

A total of 408 patients had asthma (n=112) or rhinitis (n=394); taking in consideration that 94% of patients with asthma also had rhinitis, we decided to treat the sample as a single group (Table 1) and it was divided into patients who were adherent or non-adherent to pharmacotherapy. According to ARIA guidelines classification, among patients with rhinitis 66% had “persistent moderate/severe rhinitis”, 30% “persistent mild rhinitis” and 4% “intermittent rhinitis”. According to ISAAC questionnaire 17% of patients had “persistent severe asthma”, 33% “persistent moderate asthma”, 23% “persistent mild asthma”, and 27% “intermittent asthma”. A total of 701 patients were included in the asymptomatic group.

From the symptomatic group, 97% had a previous diagnosis of asthma and or rhinitis during the school period that was evaluated; 3% were diagnosed during this study. 93% of patients had a pharmacological treatment according to clinical management guidelines, however, a high proportion of them did not have adherence to the therapy recommended by their doctor for asthma (35% vs. 72% p 0.02) or rhinitis (40% vs. 59% p<0.01).

The main cause of absenteeism was associated with acute illnesses like rhinitis exacerbation (39%), asthma exacerbation (18%) or medical control consultation (27%).

Fig. 2 presents absenteeism according to days/year by number of people in each group. In the symptomatic group, the frequency of absenteeism was higher in comparison to the asymptomatic group. Among the symptomatic group, we observed that those children receiving pharmacological treatment with good adherence, presented lower absenteeism due to allergic diseases compared to those who did not receive it continuously (day/year/patient: 0.8 vs. 3.1 p 0.02). Patients in the adherent group had no statistically significant differences in frequency of absenteeism compared to the asymptomatic group (Fig. 2).

Figure 2.

Absenteeism. Absenteeism calculated according to the number of absentee days per year by number of people in each group. *p<0.05, **p<0.01. Symptomatic group 408 patients with 516 absent days vs. asymptomatic group 701 patients with 1014 absent days.

(0.07MB).

Regarding school performance, in the symptomatic group it was observed that the number of students with insufficient score (<60%) for Mathematics and Spanish was higher than in the asymptomatic group (Fig. 3). There were no significant differences between the two areas or among the eight schools. Patients with adequate therapy and adherence had better performance than patients who did not receive continuous treatment and were similar to the asymptomatic group. There were no significant differences between patients with asthma or rhinitis.

Figure 3.

School performance. At the school level the performance was classified with a semi-quantitative scale. *p<0.05, **p<0.01.

(0.06MB).

When we analyzed the adherent and non-adherent groups, some risk factors were identified as possible causes for non-adherence (Table 2). The main non-modifiable factors were access to the healthcare system/cost of medication, and socioeconomic status. The principal modifiable factor was poly-medication.