Prevalence and impact of asthma among school-aged students in Lima, Peru

SUMMARY

SETTING AND OBJECTIVE:

The International Study of Asthma and Allergies in Childhood (ISAAC) reported asthma prevalence in Peru to be among the highest in the world. We assessed the prevalence and morbidity of self-reported asthma in an underserved, peri-urban area of Lima, Peru, with limited medical access.

DESIGN:

The study was conducted in the outskirts of Lima from March to May 2011. Five hundred children aged 6–18 years were selected through cluster sampling. Parents completed a modified version of the ISAAC questionnaire. Children underwent spirometry testing. Those with a forced expiratory volume in one second (FEV₁% predicted) ⩽80% were tested for reversibility using salbutamol.

RESULTS:

Of the 500 children selected, 71% participated. The prevalence of asthma was 16.7%. Asthma symptoms were associated with self-reported asthma (P < 0.001); 52.5% of children with asthma had ever used an inhaler (P < 0.001). However, 27.1% had never been to the doctor for respiratory problems (P < 0.001).

CONCLUSION:

We found a high prevalence of self-reported asthma and high morbidity related to asthma symptoms in the previous 12 months among the study cohort. Symptoms were poorly controlled due to limited availability of medication and access to medical services.

ASTHMA CAN CAUSE SIGNIFICANT morbidity among children and result in absence from school and poor performance due to fatigue, as well as hospitalization. The burden of asthma is highest in low-income countries, where more than 80% of deaths due to asthma occur.¹ An increased risk of asthma is associated with indoor/outdoor allergens, urban residence, respiratory infections, exercise, tobacco smoke, chemical irritants, diet, genetics, and a high body mass index.^2–4

The International Study of Asthma and Allergies in Childhood (ISAAC) implemented a simple questionnaire in multiple languages to assess asthma symptoms, allergic rhinitis, and atopy in children aged 6–7 years and 13–14 years in more than 50 countries.⁵ The self-reported prevalence of wheezing among 13–14 year-old children in Peru was 26%, one of the highest among all countries.⁵ Given the rapid urbanization occurring in Peru, studies using the ISAAC questionnaire have focused on the impact of traffic density and flow upon asthma development.^6–8 Few studies have used spirometry in conjunction with the ISAAC questionnaire, and even fewer have assessed morbidity related to asthma.

We assessed the prevalence and impact of asthma using a modified version of the ISAAC questionnaire and quantitative spirometry in a population of students aged 6–18 years in an impoverished periurban area of Lima, Peru.

STUDY POPULATION AND METHODS

Study site/population

The study was conducted between March and May 2011 in an impoverished area in the northern outskirts of Lima. The average temperature was 27.4°C, with an average humidity of 49.7%. Homes in this area consist of plywood, corrugated metal, and dirt floors. Water services were introduced only recently. Most families had access to electricity. There is no health post in the community and the nearest hospital is 40 min away by bus. About 1500 students aged 5–19 years attend the primary and secondary public school where the study took place.

We used cluster sampling to select and recruit students into the study (each classroom represented one cluster). Of 48 classrooms, 17 were selected randomly. All students in the selected classrooms were eligible to participate. Parents of students in these classrooms were sent a letter inviting them to an information session. At the session, parents were informed of the study, and those who wanted to enroll their child completed the written informed consent form, followed by a modified version of the ISAAC questionnaire. The study was explained to the students whose parents gave consent, and the students signed an assent form before participation. The study was coordinated and performed in agreement with the school administration and parent–teacher association.

All research procedures were approved by the Institutional Review Boards at the University of Washington (Seattle, WA, USA), Universidad Nacional Mayor de San Marcos (Lima), and Stanford University (Stanford, CA, USA).

Procedures

A modified version of the ISAAC questionnaire was administered by a trained investigator to each parent or guardian enrolling their child at the information session. The modified questionnaire included questions about asthma symptoms, respiratory symptoms, allergic rhinitis, atopy and medical history. It also included questions to assess for risk factors, including sex, pre-term birth, birth location, number of household members, parental employment and education, exposure to secondhand cigarette smoke, and socio-economic status (SES).

For the examination of the children, height and weight were measured and spirometry performed according to criteria set by the American Thoracic Society for acceptability and reproducibility,⁹ using a portable, battery-operated Easy One spirometer (NDD Medical Technologies, Zurich, Switzerland). Students undergoing spirometry were healthy on the day spirometry was performed; if ill, the child was rescheduled for spirometry on a different day. Students unable to cooperate with spirometry were excluded from the study. Those with a baseline forced expiratory volume in one second (FEV₁% predicted) of ⩽80% predicted were treated with two puffs of inhaled salbutamol (100 μg per puff via a metered dose inhaler and spacer), and testing was repeated. Due to the lack of adequate facilities to respond to asthma exacerbations, participants were not challenged with exercise stress testing or hypertonic saline.

Data analysis

Frequencies and proportions were used to assess the prevalence of demographic characteristics and self-reported respiratory symptoms. ‘Asthma’ was defined as having wheezed in the previous 12 months. For FEV₁, we used the Student’s t-test to assess the difference in mean FEV₁ levels according to demographic characteristics and reported respiratory symptoms. For self-reported asthma, the Pearson’s χ² test and Fisher’s exact test for categorical variables were used, whereas the Student’s t-test was used to assess the mean difference related to asthma in continuous variables. Generalized linear models were used to obtain the risk ratios associated with demographic characteristics and asthma symptoms.^10,11 For self-reported asthma, we conducted multivariate analyses to assess the association between symptoms and self-reported asthma adjusted for sex, having been born at home, and mother’s education level. All tests were two-tailed; statistical significance was defined as P < 0.05. All analyses were conducted in Stata 11.2 (StataCorp, College Station, TX, USA).

RESULTS

A total of 356 of 500 (71.2%) students from 17 classrooms were enrolled for the study (Table 1). Two students were excluded because they were unable to perform spirometry, and one child left the school before spirometry was performed. One hundred and fifty (42.1%) participants were male. The mean age was 11.0 years (standard deviation [SD] 3.0), and mean body mass index (BMI) was 19.6 (SD 3.4). Three hundred and seventeen students (93%) were full-term births and 72 (21.1%) had been born at home. The average household size was 5.1 persons (SD 1.6).

Table 1.

Association between demographic characteristics, respiratory symptoms and asthma in students aged 6–18 years, Lima, Peru

Characteristic	With asthma (n = 59) %	Without asthma (n = 297) %	P value*
Demographic
Age, years			0.4
6–12	79.3	74.6
13–18	20.6	25.4
Male	69.5	55.6	0.05‡
BMI ⩾ 19.6 kg/m2†	45.8	42.4	0.6
Full-term birth	92.7	93.0	0.9
Born at home	12.3	22.8	0.075
Number of household members ⩾ 5	56.9	60.8	0.6
Mother employed	33.3	25.4	0.2
Mother's highest education level			0.03‡
None	5.2	6.0
Primary	29.3	36.5
Secondary	51.7	53.7
Technical or university	13.8	3.9
Father's highest education level			0.1
None	3.6	4.0
Primary	23.6	25.4
Secondary	56.4	64.1
Technical or university	16.4	6.5
Mean monthly income, USD§	$92.87	$107.09	0.4
Lives with a smoker	3.4	8.8	0.2
Symptoms
At any time
Used an inhaler	52.5	18.0	<0.001‡
Visited a doctor for respiratory problems	72.9	38.9	<0.001‡
Hospitalized for respiratory problems	15.3	9.1	0.2
Told child has asthma or bronchospasm	59.3	18.2	<0.001‡
In the previous 12 months
Woke up at night due to wheeze	58.6	2.4	<0.001‡
Could not speak more than 1–2 words between breaths due to wheeze	23.7	1.4	<0.001‡
Wheezed during or after exercise	49.2	12.5	<0.001‡
Dry cough at night (other than cough associated with infections or colds)	78.0	39.9	<0.001‡
Any of above symptoms in the previous 12 months	100.0	52.9	<0.001‡

^*Pearson's χ² test and Fisher's exact test for dichotomous variables, and Student’s t-test for continuous variables.

^†Between 13.6 and 38.2 kg/m².

^‡Statistically significant.

^§Monthly income ranged from 0 to USD 613.

BMI = body mass index; USD = US dollar.

A total of 148 (41.8%) reported having ever wheezed, whereas 59 (16.7%) reported wheezing in the previous 12 months. Eighty-nine participants (25.1%) had been told they had asthma or bronchospasm by a health professional. In the previous 12 months, 41 (11.6%) had woken up at night due to wheezing, 18 (5.1%) had wheezed so severely that they could not speak more than 1–2 words between breaths, 66 (18.5%) had had wheezing during or after exercise, and 164 (46.2%) had had a dry cough at night.

Asthma was associated with male sex (P = 0.05), symptoms of asthma in the previous 12 months (P < 0.001), and having ever used an inhaler or been to a doctor for respiratory problems (P < 0.001) (Table 1). Respiratory symptoms were more common among students with asthma than in those without asthma (P < 0.001 for all comparisons). In the previous 12 months, 58.6% of students with asthma had had night-time symptoms, 23.7% had had respiratory distress, 29.2% had had symptoms during exercise, and 78% had had night-time cough. Only 52.5% of those with asthma had used an inhaler, and 27.1% had never been to the doctor for respiratory problems.

The mean FEV₁ was 101.1% (SD 16.0). A total of 13 (3.7%) had FEV₁ measurements of ⩽80%. Of these 13, six (46.2%) had an FEV₁ increase of .12% after salbutamol treatment. Male sex (P = 0.02), age 13–18 years (P < 0.001), BMI ⩾ 19.6 (P = 0.003), being born at home (P < 0.001), and having a father with less than secondary-school education (P = 0.03) were associated with a higher FEV₁ (data not shown). There were no differences in FEV₁ measurements among those who had ever had asthma symptoms or symptoms in the previous 12 months.

Table 2 reports the prevalence ratios (PRs) between respiratory symptoms and asthma in a multivariate analysis after adjusting for sex, birth at home, and mother’s education level. Students with asthma were 3.6 times (95% confidence interval [CI] 2.4–5.3) more likely to have used an inhaler, 3.1 times (95%CI 1.7–5.7) more likely to have been to the doctor for respiratory problems, 9.6 times (95%CI 6.7–13.8) more likely to have woken up at night due to wheeze, 5.2 times (95%CI 3.5–7.5) more likely to have respiratory distress due to wheeze, and 4.0 times (95%CI 2.8–5.9) more likely to have symptoms during or after exercise.

Table 2.

Association between respiratory symptoms and asthma among students aged 6-18 years, Lima, Peru

Characteristic	Unadjusted PR (95%CI)	Adjusted* PR (95%CI)
At any time
Used an inhaler	3.6 (2.4–5.3)	3.6 (2.4–5.3)
Visited doctor for respiratory problems	3.4 (1.8–6.1)	3.1 (1.7–5.7)
Hospitalized for respiratory problems	1.6 (0.8–3.3)	1.6 (0.7–3.8)
Told child has asthma or bronchospasm	4.4 (2.9–6.5)	4.1 (2.9–5.9)
In the previous 12 months
Woke up at night due to wheezing	10.8 (8.1–14.4)	9.6 (6.7–13.8)
Could not speak more than 1–2 words between breaths due to wheezing	5.8 (4.3–7.8)	5.2 (3.5–7.5)
Wheezed during or after exercise	4.2 (2.7–6.6)	4.0 (2.8–5.9)
Dry cough at night (other than cough associated with infections or colds)	4.1 (1.9–8.8)	3.9 (1.9–8.3)

^*Adjusted for sex, birth at home, and mother's education level.

PR = prevalence ratio; CI = confidence interval.

DISCUSSION

Based on the modified version of the ISAAC questionnaire, asthma prevalence in this group of 6–18-year-old students was 16.7%, as defined by wheeze in the previous 12 months. Prevalence based on self-reported asthma was 25.1%. These rates are comparable with other studies in Lima. Since 2000, self-reported prevalence of asthma among 13–14 year olds in Lima using the modified version of the ISAAC questionnaire ranged from 10.1% to 26%.^8,12,13 A more recent study reported a prevalence of asthma of 10.1% among 13–15 year olds in Lima.^8,14 In Lima, studies on younger children are scarce. A survey of 8–10 year olds reported a prevalence of 20.7% in participants with a similar SES when rates were also reported by care givers.¹⁴ This value is much closer to the 16.7% detected in our study.

There were few significant differences in FEV₁ results between students with and those without asthma. Spirometry is useful for assessing reversible airway function, but many subjects with asthma may have normal spirometry when not having an acute exacerbation. We used spirometry in conjunction with a questionnaire on asthma symptoms to assess for differences in FEV₁ among those with and those without asthma symptoms. In our study, no differences were found between those with and those without asthma. However, due to a lack of facilities to respond to asthma exacerbations, we did not challenge participants with exercise stress testing or hypertonic saline, as has been done in other studies.¹⁴ This would have been useful, as the study was performed during the summer season, when symptoms would be less frequent.

Although smoking and exposure to secondhand smoke are risk factors for asthma, there were no differences in exposure between asthma patients and those who did not have asthma.¹⁵ However, the number of students exposed to secondhand smoke in our study was low.

In urban Latin America, poverty has been reported to be a risk factor for asthma.¹⁶ This phenomenon has been shown across industrialised and developing countries when underprivileged populations were compared with more affluent populations. Our study cohort was a fairly homogeneous population with an overall low SES, which may account for why we did not detect a difference. However, maternal and paternal education can also serve as proxy for SES. In our study, mothers of participants without asthma were more likely to have finished primary and secondary school.

The ISAAC questionnaire focuses on determining the prevalence of asthma symptoms. Using a modified version of the ISAAC questionnaire in conjunction with additional questions, we found increased morbidity in students with asthma. Those with asthma were more likely to wake up at night due to wheeze, and to have respiratory distress, respiratory symptoms during exercise, and dry cough at night. These symptoms result in an increased number of days missed from school, poor grades and eventually lower SES, resulting in long-term poor health outcomes. Studies have reported increased absences among children with asthma.^17,18 A review assessing the association between asthma, school attendance and academic outcomes also reported higher rates of absenteeism.¹⁹ Elucidating specific factors among our population could lead to interventions targeting absenteeism, which is especially important given the limited access to health care. In those with asthma, 47.5% had never used an inhaler and 27.1% had never seen a doctor for respiratory problems. In addition, we found that morbidity associated with asthma was high, asthma was frequently not being treated, and symptoms were usually poorly controlled.

The limitations of our study include the fact that self-reported asthma symptoms may have been misinterpreted by parents, given the limited knowledge of health care in the area. More specifically, parents may have had a limited understanding of the symptoms of asthma, such as wheezing, which may have led to inaccurate answers to the survey questions. Finally, as only 73% of the subjects selected through cluster sampling were enrolled in the study, selection bias was possible.

Future studies comparing our population with a population from a more affluent area in Lima, as well as with a rural population with less development and exposure to outdoor pollutants, could elucidate the additional risk factors associated with SES, pollution, traffic-flow density and urbanization. The environment in our study areas was dry and sandy compared with that in central Lima, and there are likely differences between the allergens these students would have been exposed to. Given the high burden of dogs, cats, chickens, and ducks in these living environments, there may have been differences in specific allergens in the home compared with school environments of those with and without asthma.

CONCLUSIONS

Our study detected a high self-reported prevalence of asthma among 6–18-year-old children in Lima. In these students, the morbidity associated with asthma was high (many not having an inhaler or been evaluated by a doctor) and symptoms were poorly controlled. In addition, our study represented a marginalized population where access to medical staff and treatments was severely limited. Identification of asthma and interventions to control symptoms could have substantial beneficial effects upon the educational trajectory of students.