BMJ Open Is Committed to Open Peer Review. As Part of This Commitment We Make the Peer Review History of Every Article We Publish Publicly Available

BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on September 30, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second and bronchoconstriction in young children living in Havana, Cuba: A prospective population-based study

ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-034192

Article Type: Original research

Date Submitted by the 10-Sep-2019 Author:

Complete List of Authors: Suarez-Medina, Ramon; INHEM Venero-Fernández, Silvia; INHEM Corona-Tamayo, Barbara; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Betancourt-López, Maria; INHEM Alvarez-Valdés, Vilma; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Sardiñas-Baez, Nieves; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Cristina, Carmona; Dirección Municipal de Salud Pública municipios Cerro

y Arroyo Naranjo http://bmjopen.bmj.com/ Loinaz-Gonzalez, Maria; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Verdecia-Pérez, Zunilda; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Britton, John; UK Center for Tobacco and Alcohol Studies, Division of Epidemiology and Public Health Fogarty, Andrew; University of Nottingham,

Keywords: PAEDIATRICS, Asthma < THORACIC MEDICINE, Cuba on September 30, 2021 by guest. Protected copyright.

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 26 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 6 7 8 9 I, the Submitting Author has the right to grant and does grant on behalf of all authors of the Work (as defined 10 in the below author licence), an exclusive licence and/or a non-exclusive licence for contributions from authors 11 who are: i) UK Crown employees; ii) where BMJ has agreed a CC-BY licence shall apply, and/or iii) in accordance 12 with the terms applicable for US Federal Government officers or employees acting as part of their official 13 duties; on a worldwide, perpetual, irrevocable, royalty-free basis to BMJ Publishing Group Ltd (“BMJ”) its 14 licensees and where the relevant Journal is co-owned by BMJ to the co-owners of the Journal, to publish the 15 Work in this journal and any other BMJ products and to exploit all rights, as set out in our licence. 16 17 The Submitting Author accepts and understands that any supply made under these terms is made by BMJ to 18 the Submitting Author Forunless you peer are acting as review an employee on behalf only of your employer or a postgraduate 19 student of an affiliated institution which is paying any applicable article publishing charge (“APC”) for Open 20 Access articles. Where the Submitting Author wishes to make the Work available on an Open Access basis (and 21 intends to pay the relevant APC), the terms of reuse of such Open Access shall be governed by a Creative 22 Commons licence – details of these licences and which Creative Commons licence will apply to this Work are set 23 out in our licence referred to above. 24 25 Other than as permitted in any relevant BMJ Author’s Self Archiving Policies, I confirm this Work has not been 26 accepted for publication elsewhere, is not being considered for publication elsewhere and does not duplicate 27 material already published. I confirm all authors consent to publication of this Work and authorise the granting 28 of this licence. 29 30 31 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 26

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second 1 and bronchoconstriction in young children living in Havana, Cuba: A prospective 2 3 population-based study

4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 * Ramon Suarez-Medina1 6 Silvia Josefina Venero-Fernández1 7 Vilma Alvarez-Valdés2 8 Nieves B Sardiñas-Baez2 9 Cristina Carmona2 10 2 11 María Luisa Loinaz-Gonzalez 12 Zunilda Verdecia-Pérez2 13 Barbara Corona-Tamayo2 14 María Antonia Betancourt-López1 15 John Britton3, 16 Andrew W Fogarty3 17 and the HINASIC (Historia Natural de la Sibilancia en Cuba/Natural History of Wheezing in Cuba) 18 For peer review only ++ 19 Study Group 20 21 1 Instituto Nacional de Higiene, Epidemiología y Microbiología, Infanta No 1158 e/ Llinás y Clavel, 22 Código Postal 10300, La Habana, Cuba. 23 2 Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo, La Habana, Cuba. 24 3 Nottingham Biomedical Research Unit, Division of Epidemiology and Public Health, University of 25 26 Nottingham, Clinical Sciences Building, City Hospital, Nottingham NG5 1PB, UK ++ 27 full list of collaborating researchers at the end of the manuscript 28 29 The corresponding author for the manuscript submission process is Dr Andrew Fogarty 30 ([email protected]) due to poor internet connections with Cuba. 31 32 (If accepted for publication) *Correspondence and requests for reprints to Dr. Ramón Suárez 33 34 Medina: 35 Address: Infanta No 1158 e/ Llinás y Clavel, Código Postal 10300, La Habana, Cuba. 36

Office phone: (537) 878 8479 http://bmjopen.bmj.com/ 37 Email: [email protected] 38 39 Key words: Cuba, asthma, children, wheeze 40 41 42 Word count: 3138 words 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 26 BMJ Open

Abstract (298 words) 1 2 3

4 Objectives; Asthma has not been extensively studied in low and middle-income countries, where BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 risk factors and access to treatment may differ from more affluent countries. we aimed to identify 6 7 the prevalence of asthma and local risk factors in Havana, Cuba. 8 9 10 11 Setting: Four municipalities in Havana, Cuba. 12 13 14 Participants: A population-based cohort study design of young children living in Havana, Cuba. 15 16 Children were recruited from primary care centres at age 12-15 months. 17 18 For peer review only 19 Primary and secondary outcome measures: Data on wheeze in the past 12 months, asthma 20 21 treatment and environmental exposures collected regularly until the age of six years, when Forced 22 23 Expiratory Volume in one second (FEV1) and reversibility to aerosolised salbutamol were also 24 measured. 25 26 27 28 Results: 1106 children provided data at the age of six years old. The prevalence of wheeze in the 29 30 previous 12 months was 422 (38%), and 331 (37%) of the study population had bronchodilatation 31 of 10% or more in FEV after administration of inhaled salbutamol. In the previous 12 months, 182 32 1 33 (16%) of the children had received inhaled corticosteroids, 416 (38%) salbutamol inhalers, and 34 35 283 (26%) a course of systemic steroids. 36 http://bmjopen.bmj.com/ 37 38 Wheeze in the first year and a family history of asthma were both positively associated with 39 40 bronchodilatation to inhaled salbutamol (+1.94%; 95% confidence intervals CI: +0.81 to +3.08, and 41 42 +1.85%; CI: +0.14 to +3.57 respectively), while paracetamol use in the first year was associated 43 with wheeze at six years (OR 1.64, 95% CI: 1.14 to 2.35). There were large differences in FEV , 44 1 on September 30, 2021 by guest. Protected copyright. 45 bronchodilatation and risk of wheeze across different geographical areas. 46 47 48 49 Conclusions: Asthma is common in young children living in Havana, and the high prevalence of 50 systemic steroids administrated is likely to reflect the underuse of regular inhaled corticosteroids. If 51 52 replicated in other comparable low and middle-income countries, this represents an important 53 54 global public health issue. 55 56 57 Keywords: infection, asthma, children, Cuba, lung function 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 26

STRENGTHS AND LIMITATIONS OF THIS STUDY 1 2  There have been many epidemiological studies of asthma in children who live in developed 3

4 countries, but few population-based studies that have used objective measures of asthma BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 6 in developing countries. 7  The prevalence of bronchoconstriction in young children living in Havana is high using both 8 9 subjective and objective measures of asthma. 10 11  These data are from one developing country and thus not generalisable to other nations 12 13 with different economies and health-care systems. 14  The high prevalence of administration of systemic steroids in this population-based sample 15 16 is public health concern, as it is likely to be avoidable, and may increase the risk of side- 17 18 effects in later life.For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 26 BMJ Open

Introduction 1 2 Asthma is a global disease that affects approximately 11% of six year-old children 1, but with 3 1 4 marked regional differences in prevalence . The aetiology of asthma is complex, and involves a BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 range of environmental exposures that are likely to have differential impacts at different ages over 6 7 the human lifetime 2-7. The early years of childhood is a particularly important period as this is 8 9 when the lungs and immune system are developing rapidly, and lung function in children is a key 10 8 11 determinant of health in adulthood . 12 13 14 This study was established as a consequence of concerns among public health specialists and 15 16 clinicians that asthma was becoming a large problem in Cuba. It aimed to determine the 17 prevalence of wheeze in young children living in Cuba, and to identify modifiable risk factors for 18 For peer review only 19 wheezing, reduced lung function and reversible bronchoconstriction. The role of infection in early 20 21 life in the development of allergic disease remains unclear 9. The main hypothesis of interest was 22 10 11 12 13 23 that infection with parasites , Helicobacter pylori , dengue , or systemic inflammation may 24 be associated with wheeze or bronchoconstriction. Exposure to environmental tobacco smoke and 25 26 paracetamol had previously been observed to be positively associated with wheeze 14 or atopic 27 28 dermatitis 15 symptoms respectively, and so the association of these exposures with wheeze and 29 30 bronchoconstriction were also studied. Finally, as growth from in utero onwards may also be 31 related to development of asthma and growth of the lungs 16, anthropometric measures from birth 32 33 onwards were also considered. The study design is a prospective population-based study of an 34 35 existing cohort of children followed from approximately one year of age for five years. 36 http://bmjopen.bmj.com/ 37 38 39 40 Methods 41 42 Study population and sample collection 43 The study population is a cohort of 1956 children aged 12 to 15 months who were randomly 44 on September 30, 2021 by guest. Protected copyright. 45 selected from four municipalities across Havana in 2010 and 2011 14 15 17. Data was collected by a 46 47 standardised questionnaire that was administered by a member of the study team at baseline and 48 49 subsequently at 2 years, 3 years and 5 years later. This included a number of health and lifestyle 50 questions that were answered by the parent or guardian and particular attention was paid to 51 52 exposure to environmental tobacco smoke. The child’s weight, height and mid-arm circumference 53 54 in both arms were collected at each study visit. Historical baseline data including birth weight and 55 height were collected from the primary care centre records. At each annual follow-up study the 56 57 participants’ guardians were asked if the child had received a medical diagnosis of dengue 58 59 infection in the past year (in Cuba all positive diagnoses of dengue infection are clinically 60 confirmed using a serological IgM assay) and a blood sample was collected from children to

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 26

measure circulating eosinophil levels. This sample was stored at -20 degrees Centigrade and 1 2 subsequently defrosted and analysed for dengue immunoglobulin G (IgG) serology to generate an 3 18 17 4 antibody index , serum immunoglobulin E (IgE) , highly sensitive C- Reactive Protein (hsCRP, BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 SpinReact, Spain 19), toxoplasmosis immunoglobulin G antibodies (IgG) 20 and toxocariasis IgG 6 7 antibodies (DRG Instruments, Germany). A faecal sample was also collected at each review and 8 9 stored at minus 20C, and later examined for Helicobacter pylori using the faecal antigen test 10 11 (SpinReact, Spain) and intestinal parasites using the Kato-Katz test (Campiñas Medical COMI, 12 Brazil). 13 14 15 16 Lung function 17 18 Forced expiratory volumeFor in one peersecond (FEV review1) and Forced onlyVital Capacity were measured in 19 accordance with ATS/ERS criteria 21 using spirometers (CareFusion Micro I) calibrated each day 20 21 to allow for local climatic change. The best value of FEV1 within a threshold of repeatability of 22 23 200ml was used as the final value. Aerosolised salbutamol (300 g) was then administered via a 24 25 spacer and after 15 minutes lung function was measured again to quantify airway reversibility. In 26 children who provided a post-bronchodilator FEV that was less than the baseline value, they were 27 1 28 considered as having no reversibility to bronchodilator as this was likely to be due to fatigue. 29 30 31 32 Allergen skin prick testing 33 Skin prick testing was used to determine allergy to mite, cat, grass, cockroach, fungus, mosquitos, 34 35 wheat and soy (allergens from Diater, Argentina except mite allergen from Biocen, Cuba). For 36 http://bmjopen.bmj.com/ 37 each test a drop of allergen solution was placed on the skin and a lancet used to break the skin. 38 After 15 minutes, the skin wheal was measured at its maximum diameter, and also 39 40 perpendicularly, and a mean value generated. The final skin prick test result was calculated by 41 42 subtracting the saline result from the allergen. A value of ≥3mm was used to define a positive 43 44 atopic result for each allergen, and atopy was defined as any positive skin prick test. on September 30, 2021 by guest. Protected copyright. 45 46 47 Statistical analysis 48 49 The main outcome variables were FEV1, percent increase in FEV1 after to inhaled salbutamol and 50 51 wheeze in the past 12 months. The main exposure variables were grouped into three categories: 52 1. Prior exposures: wheeze in the first year of life, family history of asthma, nursery 53 54 attendance, birth weight, birth height, duration of breastfeeding, blood IgE and eosinophils 55 56 at one year old; faecal Helicobacter pylori antigen at two and three years old; blood hsCRP, 57 dengue IgG serology, eosinophils, toxoplasmosis serology, IgE at three years old, and any 58 59 prior medical diagnosis of dengue infection. 60

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 26 BMJ Open

2. Cross-section exposures: number of smokers living in the home, current weight, current 1 2 height, mean arm circumference, municipality of residence. 3

4 3. Biomarkers of current infection and inflammation: Helicobacter pylori stool antigen, BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Toxoplasmosis IgG serology, dengue IgG serology, blood hsCRP, eosinophils, IgE, 6 7 toxocariasis serology and atopy. Less than 2% of children has current gastro-intestinal 8 9 parasite infection and these data were not analysed further. 10 11 12 Statistical analysis used linear and logistic regression adjusting for sex and age in months as a 13 14 priori confounding factors, and also adjusted for clustering by municipality of residence. All 15 16 analyses used Stata 14 statistical software (Texas, USA). 17 18 For peer review only 19 Patient and Public Involvement 20 21 The study was designed as a consequence of concerns from the Cuban public health and clinical 22 23 communities about asthma morbidity. The patients were not involved in the design of the study 24 and patients did not receive a copy of the results. We thank the patients and their families for their 25 26 participation. 27 28 29 30 Ethics approval 31 The study was approved by Ethics Committees in the Instituto Nacional de Higiene, Epidemiología 32 33 y Microbiología, Cuba and at the University of Nottingham, UK. 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 Results 41 42 Data were available for 1106 children, of whom 422 (38%) had reported wheeze in the past year 43 (Figure 1). Wheeze in the first year of life was reported in 514 (46%) current participants and a 44 on September 30, 2021 by guest. Protected copyright. 45 prevalence of 42% (358 children) for those who did not participate in the study at the age of six 46 47 years (p=0.055). Nine hundred and thirteen (83%) children provided lung function data, and of 48 49 these 903 (99%) children had their reversibility to salbutamol measured. The mean FEV1 was 50 1.13L (standard deviation 0.31), and 331 (37%) had an increase in FEV1 of more than 10% after 51 52 administration of aerosolised salbutamol (Figure 2). Two hundred and eighty-three (26%) of the 53 54 current study population were reported to have received systemic steroids in the previous 12 55 months (Table 1). 56 57 58 59 60

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 26

Risk factors for wheeze in the past 12 months at six years old 1 2 The analysis of the association of exposures with wheeze in the past 12 months is presented in 3

4 Table 2. Both wheeze (odds ratio OR 1.89; 95%CI: 1.65 to 2.16) and paracetamol use (OR 1.64, BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 95% CI: 1.14 to 2.35) in the first year of life along with a family history of asthma (OR 1.66; 95%CI: 6 7 1.40 to 1.97) were associated with wheeze at six years. A positive Helicobacter pylori faecal 8 9 antigen test at age 2 years was negatively associated with wheeze in the past 12 months (OR 10 11 0.57; 95%CI: 0.40 to 0.82), but this association was not observed for Helicobacter pylori at the age 12 of three years or six years. The number of smokers in the household was a strong risk factor for 13 14 wheeze in the past 12 months (p<0.001 for trend), with homes with two or more smokers having 15 16 an odds ratio of the child having wheeze in the past 12 months of 2.08 (95% CI: 1.71 to 2.54) 17 compared to those homes with no smokers. The municipality of residence was associated with 18 For peer review only 19 wheeze in the past 12 months (p=0.04, chi2 test), with children living in Cerro municipality having 20 21 the highest risk of wheeze (OR 1.72 compared to Arroyo Naranjo; 95%CI: +1.61 to 1.84). These 22 23 differences were not substantially modified by adjusting for the number of smokers in the home. 24 25 26 Risk factors for decreased FEV1 at six years old 27 28 The analysis of the association of a number of exposures with FEV1 is presented in Table 3. A 29 30 number of measures of somatic growth were positively associated with FEV1 at six years. These 31 were birth height (+14ml per cm height at birth; 95% confidence intervals CI: +6 to +23), current 32 33 height (+8ml per cm; 95%CI: +2 to +14), current weight (+13ml per Kg; 95%CI: +8 to +17), and 34 35 current mean arm circumference (+20ml per cm; 95%CI: +14 to +25). The number of smokers 36 http://bmjopen.bmj.com/ 37 living in the home was not associated with lung function, but the municipality of residence was 38 strongly associated with current FEV1 (p<0.001, ANOVA test), with children living in La Lisa 39 40 having the lowest lung function (-100ml compared to Arroyo Naranjo, 95%CI: -127 to -73). These 41 42 differences were not substantially modified by adjusting for the number of smokers in the home. 43 No measures of infection or inflammation were associated with lung function. 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 Risk factors for bronchodilatation after inhaled salbutamol at six years old 48 49 The association of exposures with change in FEV1 from baseline after aerosolised salbutamol was 50 administered is presented in Table 4. Any wheeze in the first year of life was positively associated 51 52 with bronchodilatation (+1.94%; 95%CI: +0.81 to +3.08) as was a family history of asthma (+1.85; 53 54 95%CI: +0.14 to +3.57), but there was no relation with wheeze in the past 12 months (+3.61; 55 95%CI: -5.80 to 13.02). Children with a higher birth weight had a lower risk of bronchodilatation 56 57 (–2.67%, 95% CI: -4.49 to -0.84). Higher IgE at the age of one year was associated with a higher 58 59 risk of bronchodilatation (+1.68%; 95%CI: +0.54 to +2.82), but not IgE at age of three years or six 60 years.

7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 26 BMJ Open

1 2 The numbers of smokers living in the child’s home was not associated with bronchodilation, but 3

4 the municipality of residence was again a strong determinant of current bronchodilatation BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 (p=0.002, ANOVA test), with children living in La Lisa having the highest increase in FEV after 6 1 7 administration of inhaled salbutamol at six years old (+6.24% compared to Arroyo Naranjo; 8 9 95%CI: +5.56 to +6.91). These differences were not substantially modified by adjusting for the 10 11 number of smokers in the home. 12 13 14 15 16 Discussion 17 The most striking observation from our cohort study of young children living in Havana is that the 18 For peer review only 19 prevalence of wheeze in these children was high, with 38% reporting wheeze in the past 12 20 21 months. Similarly, there was a high prevalence of bronchoconstriction using the objective measure 22 23 of lung function reversibility to inhaled salbutamol, with 37% of the study population having an 24 increase in FEV of 10% or more. Over a quarter of these six-year old children have received 25 1 26 systemic steroids in the previous 12 months, suggesting that asthma control was suboptimal. The 27 28 main consistent association is that municipality of residence is a strong risk factor for wheeze, low 29 30 lung function and bronchoconstriction. A history of wheeze in early life, exposure to paracetamol in 31 the first year of life and current environmental tobacco smoke were risk factors for wheeze, 32 33 anthropometric measures were positively associated with higher FEV1, and wheeze in the first 34 35 year of life and lower birth weight were associated with untreated bronchconstriction. 36 http://bmjopen.bmj.com/ 37 38 Strengths and limitations of the data 39 40 This is the first cohort study of young children living in Cuba that has collected extensive life 41 42 course data to evaluate risk factors for asthma using objective measures of lung function and 43 reversibility along with laboratory measures of exposure to infection. The strengths of our data 44 on September 30, 2021 by guest. Protected copyright. 45 include the prospective nature of the data collected with exposures and outcome measures that 46 47 were selected to permit exploration of risk factors for asthma specifically within an urban Cuban 48 49 environment, where the lifestyle, environment and microbial exposures are very different to more 50 temperate, developed countries. The measurement of lung function is a challenge in young 51 52 children, yet measurements were obtained in 82% of eligible children. The availability of lung 53 54 function measurements along with reversibility to inhaled salbutamol is a particular strength of 55 these data, as it provides an objective measure that may reflect different aspects of lung health 56 57 compared to self-reported symptoms. 58 59 60

8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 26

Our dataset does have some limitations. We initially recruited 1956 children to the cohort at the 1 2 age of one year, and approximately five years later were able to collect data on 1106 (57%) of 3

4 these. This was mainly a consequence of the recent changes in Cuba that have made it easier for BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 the population more mobile, and hence many children left the policlinic from where they were 6 7 initially recruited when their family moved their residence. There was only a small difference in the 8 9 prevalence of wheeze in the first year of life between those who provided data at the age of 6 10 11 years (46%) and those who did not (42%), so loss to follow-up is unlikely to constitute a major 12 source of bias. 13 14 15 16 Risk factors for wheeze in the previous 12 months at six years old 17 Wheeze in the first year of life is associated with wheeze in later childhood 22, and may reflect the 18 For peer review only 19 establishment of the asthmatic phenotype in susceptible children. The observation that 20 21 paracetamol consumption in early life is associated with wheeze in later life 23 has been noted 22 23 before, and our data support these observations. However, they do not demonstrate a causality, 24 as an alternative explanation is reverse causality with the administration of more paracetamol to 25 26 children who are more susceptible to infections and wheeze 24. 27 28 29 30 The observation that Helicobacter pylori at the age of two years, but not three years old or six 31 years old is associated with lower risk of wheeze at six years old is an interesting observation, that 32 33 would be consistent with the hypothesis that age of exposure to infection is important. Data from 34 35 other countries has demonstrated the Helicobacter is associated with lower rates of allergic 36 11 25 26 http://bmjopen.bmj.com/ 37 disease but not wheeze in children , but not in adults . It is striking however that our 38 prevalence of Helicobacter pylori infection was very low at less than 7% in the first six years of life, 39 40 suggesting that this may be less important in Cuba compared to other countries 11. 41 42 43 From a public health perspective, exposure to cigarette smoking and the municipality of residence 44 on September 30, 2021 by guest. Protected copyright. 45 are the most important exposures for wheeze at the age of six years. Second-hand cigarette 46 47 smoke exposure is well recognised as a risk factor for wheeze in children 5, and it is a concern that 48 49 51% of the homes in our study population contained at least one smoker. The risk of wheeze was 50 higher in the municipalities of Cerro and La Lisa, which suggest that there may be an 51 52 environmental factor that predisposes to asthma symptoms such as higher levels of air pollution 27- 53 29 54 . 55 56 57 Risk factors for modified FEV1 at six years old 58 59 The fact that children living in La Lisa municipality already have lower lung function at the age of 60 six years is a major public health concern, and suggests that an exposure associated with living in

9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 26 BMJ Open

this area may negatively impact on lung growth. La Lisa had by far the highest increase in FEV1 1 2 after administration of aerosolised salbutamol, suggesting that bronchoconstriction is contributing 3

4 to the low baseline FEV1 and that a component of the apparent low lung function is reversible. La BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Lisa is located inland and while there is no obvious sources of excessive industrial pollution 29, and 6 7 we speculate that this may be associated with higher levels of air pollution as a consequence of 8 9 the relative absence of sea winds compared to the other municipalities. Weight, height and mean 10 11 arm circumference were all positively associated with FEV1 at six years of age which is likely to be 12 simply because they are all measures of somatic growth. 13 14 15 16 Risk factors for bronchodilatation after inhaled salbutamol at six years old 17 Wheeze in the first year of life was associated with bronchodilatation after the administration of 18 For peer review only 19 inhaled salbutamol at six years. This suggests that symptoms of asthma in early life are predictive 20 21 of untreated bronchoconstriction five years later, and is consistent with the hypothesis that the 22 22 23 asthmatic phenotype can be observed to persist from early life onwards . The observation that 24 birth weight is inversely associated with bronchoconstriction provides objective evidence to 25 26 support the earlier epidemiological evidence that adults with higher birth weights have a lower risk 27 28 of having a diagnosis of asthma 16. The observation that a family history asthma is associated with 29 30 bronchoconstriction also wheeze is consistent with the knowledge that there is a genetic 31 component to asthma 30. 32 33 34 35 Self-reported wheeze and bronchoconstriction 36 http://bmjopen.bmj.com/ 37 One interesting observation from our data is that the wheeze in the past 12 months is not 38 associated with either FEV1, or the change in FEV1 after administration of inhaled bronchodilator. 39 40 Although it is possible that children who have had wheeze subsequently received asthma 41 42 treatment, this lack of association between asthma symptoms and objective measures of asthma 43 has been described previously 31. This supports that concept that wheeze and reversibility of FEV 44 1 on September 30, 2021 by guest. Protected copyright. 45 after administration of aerosolised salbutamol may measure different aspects of lung development 46 47 and health. 48 49 50 Public health implications of these data 51 52 The current study was designed in response to clinical concerns that asthma was becoming a 53 54 public health problem in Havana. These data support that hypothesis, and in particular identify that 55 living in certain municipalities may result in higher levels of currently unknown exposures that 56 57 require public health intervention. It is possible that environmental air pollution may drive some of 58 59 these geographical differences, and that studies of air pollution are urgently required in Havana. 60 The prevalence of exposure to second-hand tobacco smoke in children living in Havana remains

10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 26

high, and this is an area where a combination of public health interventions including taxation, 1 2 legislation that enforces bans of smoking in the workplace and public places, restrictions on 3

4 tobacco product advertising, and helping individual smokers quit smoking are known to be BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 effective. 6 7 8 9 Summary 10 11 Asthma is common in young children living in Havana, and the high prevalence of the use of 12 systemic steroids probably reflects the underuse of regular inhaled corticosteroid prophylaxis 13 14 treatments leading to the requirement for rescue treatment for wheezing. As societies urbanise, 15 16 environmental air pollution may increase from a variety of sources. Cuba’s economy has struggled 17 as a consequence of a historical and political events 32, and it remains under an economic 18 For peer review only 19 embargo from the USA that has negatively impacted on healthcare 33. This makes providing 20 21 regular inhaled corticosteroids to all children who need them challenging. However, other low and 22 23 middle-income countries also have difficult economic and environmental circumstances, and if 24 these observations are replicated elsewhere, then this represents an important global public health 25 26 issue. 27 28 29 30 31 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 26 BMJ Open

Acknowledgements 1 2 We would like to thank the children and their parents and guardians for participating in this study. 3

4 We would like to thank all of the staff at the municipalities and policlinics who helped in many ways BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 with data collection and sample analysis. 6 7 8 9 Contributors: The study was conceived and designed by RSM, SJVF, AWF and JB. Data were 10 11 collected by RSM, SJVF, VAV, NBSB, CC, MLLG, ZVP, BCT and MABL. Data analysis was 12 conducted by RSM and AWF. RSM, SVF and AWF wrote the first draft of the manuscript. All 13 14 authors critically reviewed the manuscript, provided intellectual content and approved the final 15 16 version prior to submission for publication. 17 18 For peer review only 19 Funding: This study was funded by The Wellcome Trust, The University of Nottingham, 20 21 Nottingham University Hospitals Charity, NIHR Nottingham Biomedical Research Centre and a 22 23 BMA Award (James Trust). 24 25 26 Competing interests: None declared. 27 28 29 30 Data availability statement: Cuban regulations do not allow dissemination of national datasets. 31 However, statistical analyses can be performed upon reasonable requests to the corresponding 32 33 author. 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 26

References 1 2 1. Pearce N, Ait-Khaled N, Beasley R, et al. Worldwide trends in the prevalence of asthma 3

4 symptoms: phase III of the international study of asthma and allergies in childhood (ISAAC). BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Thorax 2007;62:758-66. 6 7 2. Jartti T, Gern J. Role of viral infections in the development and exacerbation of asthma in 8 9 children. J Allergy Clin Immunology 2017;140:895-906. 10 11 3. Edwards M, Strong K, Cameron A, et al. Viral infections in allergy and immunology: How allergic 12 inflammation influences viral infections and illness. J Allergy Clin Immunology 13 14 2017;140:909-20. 15 16 4. Prescott S. Effects of early cigarette smoke exposure on early immune development and 17 respiratory disease. Paediatric respiratory reviews 2008;9:3-10. 18 For peer review only 19 5. Lewis S, Antoniak M, Venn A, et al. Secondhand smoke, dietary fruit intake, road traffic 20 21 exposures, and the prevalence of asthma: A cross-sectional study in young children. Am J 22 23 Epidem 2005;161:406-11. 24 6. Litonjua A, Gold D. Early-life exposures and later lung function. Am J Resp Crit Care Med 25 26 2016;193:110-11. 27 28 7. Schultz E, Hallberg J, Bellander T, et al. Early-life exposure to traffic-related air pollution and 29 30 lung function in adolescence. Am J Resp Crit Care Med 2016;193:171-77. 31 8. Sly P, Bush A. From the cradle to the grave: The early-life origins of chronic obstructive 32 33 pulmonary disease. Am J Resp Crit Care Med 2016;193 34 35 9. Postma D, Weiss S, den Berge M, et al. Revisiting the dutch hypothesis. J Allergy Clin 36 http://bmjopen.bmj.com/ 37 Immunology 2015;136:521-29. 38 10. Feary J, Britton J, Leonardi-Bee J. Atopy and current intestinal parasite infection: a systematic 39 40 review and meta-analysis. Allergy 2010;66:569-77. 41 42 11. Amberbir A, Medhin G, Abegaz W, et al. Exposure to Helicobacter pylori infection in early 43 childhood and the risk of allergic disease and atopic sensitisation: a longitudinal birth cohort 44 on September 30, 2021 by guest. Protected copyright. 45 study. Clin Exp Allergy 2014;44:563-71. 46 47 12. Mabalirajan U, Kadhiravan T, Sharma S, et al. TH2 immune response in patients with dengue 48 49 during defervescence: preliminary evidence. Am J Trop Med Hygiene 2005;72:783-85. 50 13. Fogarty A, Jones S, Britton J, et al. Systemic inflammation and decline in lung function in a 51 52 general population: a prospective study. Thorax 2007;62:515-20. 53 54 14. Venero-Fernandez S, Suarez Medina R, Mora Faife E, et al. Risk factors for wheezing in 55 infants born in Cuba. Quarterly Journal Medicine 2013;106:1023-29. 56 57 15. Suarez-Medina R, Venero-Fernandez S, Mora Faife E, et al. Risk factors for eczema in infants 58 59 born in Cuba. BMC Dermatology 2014;14:6. 60

13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 26 BMJ Open

16. Shaheen S, Sterne J, Montgomery S, et al. Birth weight, body mass index and asthma in 1 2 young adults. Thorax 1999;54:396-402. 3

4 17. Fundora Hernández H, Suarez-Medina R, Venero Fernandez S, et al. What are the main BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 environmental exposures associated with elevated IgE in Cuban infants? A population- 6 7 based study. Tropical Medicine and International Health 2014;19:545-54. 8 9 18. Vircell dengue ELISA IgG. http://wwwperamedcom/peramed/docs/G1018_ENpdf (accessed 10 11 14/4/2016) 12 19. Venero-Fernandez S, Fundora-Hernández H, Batista-Gutierrez L, et al. The association of low 13 14 birth weight with serum C reactive protein in three year old children living in Cuba: A 15 16 population-based prospective study. Am J Human Biol 2016:DOI 10.1002/ajhb.22936. 17 20. Garcia F, Venero Fernandez S, Fundora Hernández H, et al. Seroprevalencia de anticuerpos 18 For peer review only 19 IgG anti-Toxoplasma Gondii en infantes de La Habana. Parasitaria 2015;73:(in press). 20 21 21. Miller M, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Resp J 22 23 2005;26:319-38. 24 22. Martinez F, Wright A, Taussig L, et al. Asthma and wheezing in the first six years of life. New 25 26 Eng J Med 1995;332:133-38. 27 28 23. Beasley R, Clayton T, Crane J, et al. Association between paracetamol use in infancy and 29 30 childhood, and risk of asthma, rhinoconjunctivitis, and eczema in children aged 6-7 years: 31 analysis from Phase Three of the ISAAC progranmme. Lancet 2008;372:1039-48. 32 33 24. Rusconi F; Gagliardi L; Galassi C; Forastiere F; Brunetti L; La Grutta S; Piffer S; Talassi F; 34 35 SIDRIA-2 Collaborative Group. Paracetamol and antibiotics in childhood and subsequent 36 http://bmjopen.bmj.com/ 37 development of wheezing/asthma: association or causation? Int J Epidemiol 2011;40:662- 38 67. 39 40 25. Amberbir A, Medhin G, Erku W, et al. Effects of Helicobacter pylori, geohelminth infection and 41 42 selected commensal bacteria on the risk of allergic disease and sensitization in 3 year old 43 Ethiopian children. Clinical Exp Allergy 2011;41:1422-30. 44 on September 30, 2021 by guest. Protected copyright. 45 26. Fullerton D, Britton J, Lewis S, et al. Helicobacter pylori and lung function, asthma, atopy and 46 47 allergic disease - A population-based cross-sectional study. Int J Epidemiol 2009;38:419- 48 49 26. 50 27. Wu W, Jin Y, Carlsten C. Inflammatory health effects of indoor and outdoor particular matter. J 51 52 Allergy Clin Immunology 2018;141:833-44. 53 54 28. Gordon S, Bruce N, Grigg J, et al. Respiratory risks from household air pollution in low and 55 middle income countries. Lancet Respiratory Medicine 2014;2:823-60. 56 57 29. Cuellar-Luna L, Puerto-Rodriguez A, Maldonado-Cantillo G, et al. Distribucion espacial de 58 59 fuents fijas contaminantes y su impact en salud, provincia La Habana (Cuba). Hig Sanid 60 Ambient 2013;13:968-74.

14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 26

30. Shrine N, Portelli MA, John C, et al. Moderate-to-severe asthma in individuals of European 1 2 ancestry: a genome-wide association study. The Lancet Respiratory medicine 3

4 2019;7(1):20-34. doi: 10.1016/S2213-2600(18)30389-8 [published Online First: 2018/12/16] BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 31. MacKenney J, Oyarzun M, Diaz P, et al. Prevalence of asthma, atopy and bronchial 6 7 hyperresponsiveness and their interrelation in a semi-rural area of Chile. Int J Tuberculosis 8 9 & Lung Dis 2005;9:1288-93. 10 11 32. Gott R. Cuba. New Haven, USA: Yale University Press 2010. 12 33. Barry M. Effect of the US embargo and economic decline on health in Cuba. Ann Intern Med 13 14 2000;132:151-54. 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 26 BMJ Open

Figure 1. Flow diagram of participants and data collection 1 2 3

4 Figure 1 goes here BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 Figure 2. Histogram of percent increase in Forced Expiratory Volume in one second in 32 study population (N=903 children) 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 Figure 2 goes here 53 54 55 56 57 58 59 60

16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 26

Table 1. Description of study population 1 2 3 Total (N=1106) Provided FEV1 (N=913)

4 Male sex (%) 575 (52) 473 (52) BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Mean age, months (range) 74 (63 to 83) 74 (64 to 83) 6 Mean FEV1, L, (sd) - 1.13 (0.31) 7 Mean bronchodilation after - 13.4 (20.1) 8 9 salbutamol, % (sd) N=903 10 Wheeze in the past 12 months 422 (38) 356 (39) 11 (%) 12 Received inhaled steroids in 13 the 12 months before (%): 14 Year 1 89 (8) 68 (7) 15 16 Year 2 176 (17) N=1051 153 (18) N=869 17 Year 3 203 (18) 163 (18) 18 Year 4 For peer review- only - 19 Year 5 182 (16) 150 (16) 20 Received salbutamol inhaler 21 in the previous 12 months (%): 22 Year 5 416 (38) 345 (38) 23 24 Received IV or oral steroids in 25 the 12 months before (%): 26 Year 1 295 (27) 244 (27) 27 Year 2 418 (40) N=1051 345 (40) N=869 28 Year 3 407 (37) 333 (36) 29 Year 4 - - 30 31 Year 5 283 (26) 240 (26) 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 26 BMJ Open

Table 2. Association of exposures with wheeze in past 12 months. 1 Total number = 1106 No (%, sd) Odds ratio of wheeze 2 3 (95% CI)

4 Prior exposures BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Any wheeze in 1st year of life 514 (46) 1.89 (1.65 to 2.16) 6 Family history of asthma 614 (56) 1.66 (1.40 to 1.97) 7 Attendance at nursery 159 (14) 0.84 (0.57 to 1.24) 8 Paracetamol in 1st year of life 256 (23) 1.64 (1.14 to 2.35) 9 10 Mean birth weight, N=, Kg (sd) N=1104 3.31 (0.51) 0.87 (0.75 to 1.01) 11 Mean birth height, cm (sd) 50.2 (2.4) 0.95 (0.89 to 1.00) 12 Breastfeeding  4months 608 (55) 0.83 (0.66 to 1.03) 13 Age = 1 year 14 Mean log IgE, N=885 +3.38 (1.47) 1.06 (0.88 to 1.27) 15 Mean log eosinophils, N=856 -2.11(1.05) 0.93 (0.77 to 1.13) 16 17 Age = 2 years 18 Helicobacter stool +ve,For N=1067 peer review40 (4) only0.57 (0.40 to 0.82) 19 Age = 3 years 20 Mean log CRP, sd, N=986 -2.07 (2.70) 1.01 (0.95 to 1.07) 21 Log dengue IgG, N= 865 -0.44 (2.00) 0.99 (0.96 to 1.02) 22 Helicobacter stool +ve, N=951 58 (6) 1.01 (0.64 to 1.60) 23 Mean log eosinophils, N=1039 -1.77 (1.23) 0.91 (0.86 to 0.96) 24 25 Toxoplasmosis serology +ve, N=966 565 (58) 1.15 (0.82 to 1.64) 26 Mean log IgE, N=986 3.70 (1.53) 1.14 (0.98 to 1.31) 27 Medical diagnosis of dengue infection 93 (8) 1.40 (0.91 to 2.14) 28 Current exposures 29 Male Sex 575 (52) 1.35 (1.00 to 1.82) 30 Age (months), (range) 74 (63 to 83) 0.97 (0.96 to 0.99) 31 32 Mean current weight, Kg, N=1062 23.3 (11 to 51) 1.01 (0.97 to 1.06) 33 Current height, cm, N=1057 119 (7) 1.00 (0.97 to 1.03) 34 Mean arm circumference, cm N=1062 18.3 (2.4) 1.02 (0.95 to 1.08) 35 Number of current smokers in house 36 0 546 (49) 0 http://bmjopen.bmj.com/ 37 1 322 (29) 1.61 (1.22 to 2.12) 38 238 (22) 2.08 (1.71 to 2.54) 39  2 40 Municipality of residence pTREND<0.001 41 Arroyo Naranjo 455 (41) 0 42 Cerro 139 (13) 1.72 (1.61 to 1.84) 43 Habana del Este 307 (28) 0.86 (0.84 to 0.88) 44 La Lisa 205 (19) 1.24 (1.21 to 1.27) on September 30, 2021 by guest. Protected copyright. 45 p=0.04 46 47 Current infection and blood assays 48 Helicobacter stool antigen +ve, N=756 11 (1) 1.45 (0.63 to 3.33) 49 Toxoplasmosis +ve, N=759 487 (64) 0.97 (0.70 to 1.34) 50 Mean log dengue IgG serology, N=758 1.67 (1.53) 0.95 (0.90 to 1.01) 51 Log CRP, N=757 -0.83 (1.41) 0.99 (0.86 to 1.15) 52 Mean log eosinophils, N=868 -1.62 (1.45) 1.02 (0.90 to 1.15) 53 Mean log IgE, N=759 4.51 (0.97) 1.27 (1.15 to 1.41) 54 55 Toxocariasis +ve, N=673 64 (9) 0.96 (0.70 to 1.33) 56 Any atopy* (N=857) 330 (39) 0.88 (0.64 to 1.21) 57 58 59 60

18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 26

Table 3. Association of exposures with Forced Expiratory Volume in one second. 1 Total number = 913 No (%, sd) FEV ml (95% CI) 2 1, 3 Prior exposures st 4 Any wheeze in 1 year of life 423 (46) -17 (-45 to +12) BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Family history of asthma 517 (57) -18 (-79 to +43) 6 Attendance at nursery 141 (15) -49 (-155 to +57) 7 Paracetamol in 1st year of life 216 (24) +14 (-25 to +52) 8 Mean birth weight, N=, Kg (sd) N=911 3.32 (0.53) +60 (-12 to +131) 9 10 Mean birth height, cm (sd) 50 (2) +14 (+6 to +23) 11 Breastfeeding  4months 508 (56) +14 (-71 to +98) 12 Age = 1 year 13 Mean log IgE, N=460 +3.44 (1.39) -18 (-50 to +13) 14 Mean log eosinophils, N=454 -2.14 (1.04) -14 (-52 to +25) 15 Age = 2 years 16 17 Helicobacter stool +ve, N=593 29 (5) -72 (-284 to +140) 18 Age = 3 yearsFor peer review only 19 Mean log CRP, sd, N=622 -2.13 (2.71) +1 (-12 to +14) 20 Log dengue IgG, N= 554 -0.43 (2.01) +5 (-11 to +21) 21 Helicobacter stool +ve, N=784 52 (7) -42 (-110 to +25) 22 Mean log eosinophils, N=659 -1.78 (1.26) +21 (-3 to +45) 23 Toxoplasmosis serology +ve, N=614 366 (60) +50 (-29 to +128) 24 25 Mean log IgE, N=622 +3.68 (1.56) +6 (-21 to +34) 26 Medical diagnosis of dengue infection 78 (9) -24 (-59 to +11) 27 Current exposures 28 Male Sex 473 (52) +47 (-11 to +106) 29 Age (months), (range) 74 (64 to 83) +7 (-8 to +23) 30 Wheeze in past 12 months 356 (39) -60 (-174 to +55) 31 32 Mean weight, Kg, N=903 (range) 23.5 (11 to 51) +13 (+8 to +17) 33 Current height, cm, N=903 (range) 119 (90 to 175) +8 (+2 to +14) 34 Mean arm circumference, cm N=903 18.4 (2.4) +20 (+14 to +25) 35 Number of current smokers in house 36 0 452 (50) 0 http://bmjopen.bmj.com/ 37 1 264 (29) +5 (-11 to +22) 38 197 (22) -32 (-109 to +46) 39  2 40 Municipality of residence 41 Arroyo Naranjo 375 (41) 0 42 Cerro 120 (13) +61 (+23 to +99) 43 Habana del Este 251 (27) +45 (+38 to +53) 44 La Lisa 167 (18) -100 (-127 to -73) on September 30, 2021 by guest. Protected copyright. 45 P<0.001 46 47 Current infection and blood assays 48 Helicobacter stool antigen +ve, N=691 10 (1) -9 (-231 to +213) 49 Toxoplasmosis +ve, N=694 441 (64) +3 (-52 to +58) 50 Mean log dengue IgG serology, N=693 +1.64 (1.55) -4 (-21 to +14) 51 Log CRP, N=692 +1.48 (5.73) -8 (-32 to +17) 52 Mean log eosinophils, N=791 -1.60 (1.47) -7 (-43 to +29) 53 Mean log IgE, N=694 +4.52 (0.97) -15 (-98 to +67) 54 55 Toxocariasis +ve, N=673 60 (9) -3 (-128 to +121) 56 Any atopy* (N=826) 316 (38) +8 (-59 to +76) 57 adjusted for sex, age in months and clustering by municipality 58 *defined as any allergen skin prick test > 3mm larger than saline control 59 60 FEV1 = Forced Expiratory Volume in one second CRP = C Reactive Protein, sd= standard deviation

19 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 26 BMJ Open

Table 4. Association of exposures with bronchodilatation after inhaled salbutamol 1 Total number = 903 No (%, sd) % change in FEV 2 1 3 (95% CI)

4 Prior exposures BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Any wheeze in 1st year of life 417 (46) +1.94 (+0.81 to +3.08) 6 Family history of asthma 510 (56) +1.85 (+0.14 to +3.57) 7 Attendance at nursery 138 (15) -3.21 (-9.10 to +2.68) 8 Paracetamol in 1st year of life 213 (24) -0.67 (-4.68 to +3.35) 9 10 Mean birth weight, Kg (sd) N=901 3.32 (0.53) -2.67 (-4.49 to -0.84) 11 Mean birth height, cm 50 (2) +0.06 (-0.56 to +0.69) 12 Breastfeeding  4months 503 (56) +0.61 (-2.87 to +4.09) 13 Age = 1 year 14 Mean log IgE, N=455 +3.42 (1.39) +1.68 (+0.54 to +2.82) 15 Mean log eosinophils, N=449 -2.13 (1.04) -0.16 (-2.54 to +2.21) 16 17 Age = 2 years 18 Helicobacter stool +ve, ForN=586 peer review28 (5) only-6.41 (-14.94 to +2.10) 19 Age = 3 years 20 Mean log CRP, sd, N=615 -2.12 (2.71) -0.28 (-0.88 to +0.32) 21 Log dengue IgG, N= 550 -0.43 (2.01) -0.38 (-1.50 to +0.74) 22 Helicobacter stool +ve, N=776 51 (6) -1.40 (-7.30 to +4.51) 23 Mean log eosinophils, N=653 -1.78 (1.26) -0.39 (-4.45 to +3.68) 24 25 Toxoplasmosis serology +ve, N=607 362 (60) -0.45 (-7.06 to +6.16) 26 Mean log IgE, N=615 +3.68 (1.57) +0.23 (-1.81 to +2.27) 27 Medical diagnosis of dengue infection 78 (9) -0.58 (-8.55 to +7.38) 28 Current exposures 29 Male Sex 466 (52) +2.22 (-0.28 to +4.71) 30 Age (months), (range) 74 (64 to 83) -0.13 (-0.47 to +0.20) 31 32 Wheeze in past 12 months 353 (39) +3.61 (-5.80 to +13.02) 33 Mean current weight, Kg, N=893 (range) 24 (11 to 51) -0.23 (-0.54 to +0.08) 34 Current height, cm, N=893 119 (90 to 175) -0.08 (-0.50 to +0.34) 35 Mean arm circumference, cm N=893 18.4 (2.4) -0.14 (-0.81 to +0.54) 36 Number of current smokers in house http://bmjopen.bmj.com/ 37 0 446 (49) 0 38 1 263 (29) -0.87 (-5.89 to +4.14) 39 40  2 194 (21) +2.93 (-3.80 to +9.65) 41 Municipality of residence 42 Arroyo Naranjo 368 (41) 0 43 Cerro 119 (13) -1.34 (-2.56 to -0.11) 44 Habana del Este 249 (28) -0.07 (-0.46 to +0.31) on September 30, 2021 by guest. Protected copyright. 45 La Lisa 167 (18) +6.24 (+5.56 to +6.91) 46 p=0.002 47 48 Current infection and blood assays 49 Helicobacter stool antigen +ve, N=684 10 (1) +8.89 (-9.17 to +26.96) 50 Toxoplasmosis +ve, N=687 438 (64) -0.47 (-3.34 to +2.39) 51 Mean log dengue IgG serology, N=686 +1.64 (1.56) +0.30 (-1.44 to +2.03) 52 Log CRP, N=685 -0.87 (1.40) +0.37 (-1.64 to +2.38) 53 Mean log eosinophils, N=782 -161 1.47) +0.99 (-0.48 to+2.46) 54 55 Mean log IgE, N=686 +4.52 (0.98) +0.64 (-0.18 to +1.45) 56 Toxocariasis +ve, N=666 58 (9) -1.03 (-5.29 to +3.23) 57 Any atopy* (N=816) 314 (38) -0.47 (-8.61 to +7.68) 58 adjusted for sex, age in months and clustering by municipality 59 60 *defined as any allergen skin prick test > 3mm larger than saline control FEV1 = Forced Expiratory Volume in one second, CRP = C Reactive Protein sd= standard deviation 20 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

BMJ Open Page 22 of 26

1 Year 1. 2 3 N = 1956. Wheeze = 45% 4 5 6 7 8 Year 2. 9 10 N = 1701. Wheeze = 60%

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 Year 3. 16 N = 1543. Wheeze = 58% 17 18

19 on September 30, 2021 by guest. Protected copyright. 20 21 22 23 24 25 26 27 28 29 Year 5. 30 31 N = 1106. Wheeze = 38% 32 33 34 35 36 37 38 39 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

Page 23 of 26 BMJ Open

1 2 37% of population with bronchodilatation greater than 10% FEV1 3 4 5 6 7 8 9 10

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 16 17 18

19 on September 30, 2021 by guest. Protected copyright. 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

BMJ Open Page 24 of 26

1 Year 1. 2 3 N = 1956. Wheeze = 45% 4 5 6 7 8 Year 2. 9 10 N = 1701. Wheeze = 60%

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 Year 3. 16 N = 1543. Wheeze = 58% 17 18

Page 25 of 26 BMJ Open

1 2 37% of population with bronchodilatation greater than 10% FEV1 3 4 5 6 7 8 9 10

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 16 17 18

1 2 STROBE Statement 3

4 Item BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 No Recommendation 6 P1/2 Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the 7 abstract 8 (b) Provide in the abstract an informative and balanced summary of what was done 9 10 and what was found 11 Introduction 12 P4 2 Explain the scientific background and rationale for the investigation being reported 13 14 Background/rationale 15 P4 Objectives 3 State specific objectives, including any prespecified hypotheses 16 Methods 17 18 P4 Study design For4 Present peer key elements review of study design earlyonly in the paper 19 P4 Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, 20 exposure, follow-up, and data collection 21 22 P4 Participants 6 (a) Give the eligibility criteria, and the sources and methods of selection of 23 participants 24 P5 Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and 25 effect modifiers. Give diagnostic criteria, if applicable 26 27 P5 Data sources/ 8* For each variable of interest, give sources of data and details of methods of 28 measurement assessment (measurement). Describe comparability of assessment methods if there 29 is more than one group 30 P6 Bias 9 Describe any efforts to address potential sources of bias 31 32 P4 Study size 10 Explain how the study size was arrived at 33 P5 Quantitative 11 Explain how quantitative variables were handled in the analyses. If applicable, 34 variables describe which groupings were chosen and why 35 P5 Statistical methods 12 (a) Describe all statistical methods, including those used to control for confounding

36 http://bmjopen.bmj.com/ 37 (b) Describe any methods used to examine subgroups and interactions 38 (c) Explain how missing data were addressed 39 (d) If applicable, describe analytical methods taking account of sampling strategy 40 41 (e) Describe any sensitivity analyses 42 Results 43 P16 Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially 44 on September 30, 2021 by guest. Protected copyright. 45 eligible, examined for eligibility, confirmed eligible, included in the study, 46 completing follow-up, and analysed 47 (b) Give reasons for non-participation at each stage 48 49 (c) Consider use of a flow diagram 50 P18 Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and 51 information on exposures and potential confounders 52 (b) Indicate number of participants with missing data for each variable of interest 53 54 P18-20 Outcome data 15* Report numbers of outcome events or summary measures 55 P18-20 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and 56 their precision (eg, 95% confidence interval). Make clear which confounders were 57 adjusted for and why they were included 58 59 (b) Report category boundaries when continuous variables were categorized 60 (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 26 BMJ Open

1 2 P18-20 Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and 3 sensitivity analyses 4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Discussion 6 P8 Key results 18 Summarise key results with reference to study objectives 7 P8 Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or 8 9 imprecision. Discuss both direction and magnitude of any potential bias 10 P9 Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, 11 multiplicity of analyses, results from similar studies, and other relevant evidence 12 13 P10 Generalisability 21 Discuss the generalisability (external validity) of the study results 14 Other information 15 P12 Funding 22 Give the source of funding and the role of the funders for the present study and, if 16 17 applicable, for the original study on which the present article is based 18 For peer review only 19 *Give information separately for exposed and unexposed groups. 20 21 22 Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and 23 published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely 24 available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at 25 http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is 26 27 available at www.strobe-statement.org. 28 29 30 31 32 33 34 35

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second and bronchoconstriction in young children living in Havana, Cuba: A population-based cohort study

ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-034192.R1

Article Type: Original research

Date Submitted by the 30-Dec-2019 Author:

Complete List of Authors: Suarez-Medina, Ramon; INHEM Venero-Fernández, Silvia; INHEM Alvarez-Valdés, Vilma; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Sardiñas-Baez, Nieves; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Cristina, Carmona; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Loinaz-Gonzalez, Maria; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo

Verdecia-Pérez, Zunilda; Dirección Municipal de Salud Pública municipios http://bmjopen.bmj.com/ Cerro y Arroyo Naranjo Corona-Tamayo, Barbara; Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo Betancourt-López, Maria; INHEM Britton, John; UK Center for Tobacco and Alcohol Studies, Division of Epidemiology and Public Health Fogarty, Andrew; University of Nottingham,

Primary Subject Respiratory medicine on September 30, 2021 by guest. Protected copyright. Heading:

Secondary Subject Heading: Global health, Paediatrics

Keywords: Asthma < THORACIC MEDICINE, Cuba, PAEDIATRICS

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 24 BMJ Open

1 2 3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 24

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second 1 and bronchoconstriction in young children living in Havana, Cuba: A population-based 2 3 cohort study

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 24 BMJ Open

Abstract (298 words) 1 2 3

4 Objectives; Asthma has not been extensively studied in low and middle-income countries, where BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 risk factors and access to treatment may differ from more affluent countries. we aimed to identify 6 7 the prevalence of asthma and local risk factors in Havana, Cuba. 8 9 10 11 Setting: Four municipalities in Havana, Cuba. 12 13 14 Participants: A population-based cohort study design of young children living in Havana, Cuba. 15 16 Children were recruited from primary care centres at age 12-15 months. 17 18 For peer review only 19 Primary and secondary outcome measures: Data on wheeze in the past 12 months, asthma 20 21 treatment and environmental exposures collected regularly until the age of six years, when Forced 22 23 Expiratory Volume in one second (FEV1) and reversibility to aerosolised salbutamol were also 24 measured. 25 26 27 28 Results: 1106 children provided data at the age of six years old. The prevalence of wheeze in the 29 30 previous 12 months was 422 (38%), and 294 (33%) of the study population had bronchodilatation 31 of 12% or more in FEV after administration of inhaled salbutamol. In the previous 12 months, 182 32 1 33 (16%) of the children had received inhaled corticosteroids, 416 (38%) salbutamol inhalers, and 34 35 283 (26%) a course of systemic steroids. 36 http://bmjopen.bmj.com/ 37 38 Wheeze in the first year and a family history of asthma were both positively associated with 39 40 bronchodilatation to inhaled salbutamol (+1.94%; 95% confidence intervals CI: +0.81 to +3.08, and 41 42 +1.85%; CI: +0.14 to +3.57 respectively), while paracetamol use in the first year was associated 43 with wheeze at six years (OR 1.64, 95% CI: 1.14 to 2.35). There were large differences in FEV , 44 1 on September 30, 2021 by guest. Protected copyright. 45 bronchodilatation and risk of wheeze across different geographical areas. 46 47 48 49 Conclusions: Asthma is common in young children living in Havana, and the high prevalence of 50 systemic steroids administrated is likely to reflect the underuse of regular inhaled corticosteroids. If 51 52 replicated in other comparable low and middle-income countries, this represents an important 53 54 global public health issue. 55 56 57 Keywords: infection, asthma, children, Cuba, lung function 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 24

STRENGTHS AND LIMITATIONS OF THIS STUDY 1 2  There have been many epidemiological studies of asthma in children who live in developed 3

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 24 BMJ Open

Introduction 1 2 Asthma is a global disease that affects approximately 11% of six year-old children 1, but with 3 1 4 marked regional differences in prevalence . The aetiology of asthma is complex, and involves a BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 range of environmental exposures that are likely to have differential impacts at different ages over 6 7 the human lifetime 2-7. The early years of childhood is a particularly important period as this is 8 9 when the lungs and immune system are developing rapidly, and lung function in children is a key 10 8 11 determinant of health in adulthood . 12 13 14 This study was established as a consequence of concerns among public health specialists and 15 16 clinicians that asthma was becoming a large problem in Cuba. It aimed to determine the 17 prevalence of wheeze in young children living in Cuba, and to identify modifiable risk factors for 18 For peer review only 19 wheezing, reduced lung function and reversible bronchoconstriction. The role of infection in early 20 21 life in the development of allergic disease remains unclear 9. The main hypothesis of interest was 22 10 11 12 13 23 that infection with parasites , Helicobacter pylori , dengue , or systemic inflammation may 24 be associated with wheeze or bronchoconstriction. Exposure to environmental tobacco smoke and 25 26 paracetamol had previously been observed to be positively associated with wheeze 14 or atopic 27 28 dermatitis 15 symptoms respectively, and so the association of these exposures with wheeze and 29 30 bronchoconstriction were also studied. Finally, as growth from in utero onwards may also be 31 related to development of asthma and growth of the lungs 16, anthropometric measures from birth 32 33 onwards were also considered. The study design is a prospective population-based study of an 34 35 existing cohort of children followed from approximately one year of age for five years. 36 http://bmjopen.bmj.com/ 37 38 39 40 Methods 41 42 Study population and sample collection 43 The study population is a cohort of 1956 children aged 12 to 15 months who were randomly 44 on September 30, 2021 by guest. Protected copyright. 45 selected from the general population from four municipalities across Havana in 2010 and 2011 14 46 47 15 17. The response rate of those who were eligible to participate initially was 96% 14. Data was 48 49 collected by a standardised questionnaire that was administered by a member of the study team at 50 baseline and subsequently at 2 years, 3 years and 5 years later. This included a number of health 51 52 and lifestyle questions that were answered by the parent or guardian and particular attention was 53 54 paid to parental/guardian reported wheeze in the past 12 months using the methodology 55 developed for the ISAAC epidemiological studies of asthma 18, use of asthma medication in the 56 57 past 12 months and exposure to environmental tobacco smoke. The child’s weight, height and 58 59 mid-arm circumference in both arms were collected at each study visit. Historical baseline data 60 including birth weight and height were collected from the primary care centre records. At each

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 24

annual follow-up study the participants’ guardians were asked if the child had received a medical 1 2 diagnosis of dengue infection in the past year (in Cuba all positive diagnoses of dengue infection 3

4 are clinically confirmed using a serological IgM assay) and a blood sample was collected from BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 children to measure circulating eosinophil levels. This sample was stored at -20 degrees 6 7 Centigrade and subsequently defrosted and analysed for dengue immunoglobulin G (IgG) 8 9 serology to generate an antibody index 19, serum immunoglobulin E (IgE) 17, highly sensitive C- 10 20 11 Reactive Protein (hsCRP, SpinReact, Spain ), toxoplasmosis immunoglobulin G antibodies (IgG) 12 21 and toxocariasis IgG antibodies (DRG Instruments, Germany). A faecal sample was also 13 14 collected at each review and stored at minus 20C, and later examined for Helicobacter pylori 15 16 using the faecal antigen test (SpinReact, Spain) and intestinal parasites using the Kato-Katz test 17 18 (Campiñas Medical COMI,For Brazil). peer review only 19 20 21 Lung function 22 23 Forced expiratory volume in one second (FEV1) and Forced Vital Capacity were measured in 24 22 25 accordance with ATS/ERS criteria using spirometers (CareFusion Micro I) calibrated each day 26 to allow for local climatic change. The best value of FEV1 within a threshold of repeatability of 27 28 200ml was used as the final value. Aerosolised salbutamol (300 g) was then administered via a 29 30 spacer and after 15 minutes lung function was measured again to quantify airway reversibility. In 31 32 children who provided a post-bronchodilator FEV1 that was less than the baseline value, they were 33 considered as having no reversibility to bronchodilator as this was likely to be due to fatigue. 34 35 36 http://bmjopen.bmj.com/ 37 Allergen skin prick testing 38 Skin prick testing was used to determine allergy to mite, cat, grass, cockroach, fungus, mosquitos, 39 40 wheat and soy (allergens from Diater, Argentina except mite allergen from Biocen, Cuba). For 41 42 each test a drop of allergen solution was placed on the skin and a lancet used to break the skin. 43 44 After 15 minutes, the skin wheal was measured at its maximum diameter, and also on September 30, 2021 by guest. Protected copyright. 45 perpendicularly, and a mean value generated. The final skin prick test result was calculated by 46 47 subtracting the saline result from the allergen. A value of ≥3mm was used to define a positive 48 49 atopic result for each allergen, and atopy was defined as any positive skin prick test. 50 51 52 Statistical analysis 53 54 The main outcome variables were FEV1, percent increase in FEV1 after to inhaled salbutamol and 55 56 wheeze in the past 12 months. The main exposure variables were grouped into three categories: 57 1. Prior exposures: wheeze in the first year of life, family history of asthma, nursery 58 59 attendance, birth weight, birth height, duration of breastfeeding, blood IgE and eosinophils 60 at one year old; faecal Helicobacter pylori antigen at two and three years old; blood hsCRP,

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 24 BMJ Open

dengue IgG serology, eosinophils, toxoplasmosis serology, IgE at three years old, and any 1 2 prior medical diagnosis of dengue infection. 3

4 2. Cross-section exposures: number of smokers living in the home, current weight, current BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 height, mean arm circumference, municipality of residence. 6 7 3. Biomarkers of current infection and inflammation: Helicobacter pylori stool antigen, 8 9 Toxoplasmosis IgG serology, dengue IgG serology, blood hsCRP, eosinophils, IgE, 10 11 toxocariasis serology and atopy. Less than 2% of children has current gastro-intestinal 12 parasite infection and these data were not analysed further. 13 14 15 16 Statistical analysis used linear and logistic regression adjusting for sex and age in months as a 17 priori confounding factors, and also adjusted for clustering by municipality of residence. All 18 For peer review only 19 analyses used Stata 14 statistical software (Texas, USA). 20 21 22 23 Patient and Public Involvement 24 The study was designed as a consequence of concerns from the Cuban public health and clinical 25 26 communities about asthma morbidity. The patients were not involved in the design of the study 27 28 and patients did not receive a copy of the results. We thank the patients and their families for their 29 30 participation. 31 32 33 Ethics approval 34 35 The study was approved by Ethics Committees in the Instituto Nacional de Higiene, Epidemiología 36 http://bmjopen.bmj.com/ 37 y Microbiología, Cuba and at the University of Nottingham, UK. This involved parental consent on 38 the behalf of the child. 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 Results 46 47 Data were available for 1106 children, of whom 422 (38%) had reported wheeze in the past year 48 49 (Figure 1). Wheeze in the first year of life was reported in 514 (46%) current participants, while 50 there was a prevalence of wheeze in the first year of life of 42% (358 children) for those who did 51 52 not participate in the study at the age of six years (p=0.055). Nine hundred and thirteen (83%) 53 54 children provided lung function data, and of these 903 (99%) children had their reversibility to 55 salbutamol measured. The mean FEV was 1.13L (standard deviation 0.31), and 294 (33%) had 56 1 57 an increase in FEV1 of more than 12% after administration of aerosolised salbutamol (Figure 2). 58 59 Two hundred and eighty-three (26%) of the current study population were reported to have 60 received systemic steroids in the previous 12 months (Table 1).

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 24

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Risk factors for wheeze in the past 12 months at six years old 6 7 The analysis of the association of exposures with wheeze in the past 12 months is presented in 8 9 Table 2. Both wheeze (odds ratio OR 1.89; 95%CI: 1.65 to 2.16) and paracetamol use (OR 1.64, 10 11 95% CI: 1.14 to 2.35) in the first year of life along with a family history of asthma (OR 1.66; 95%CI: 12 1.40 to 1.97) were associated with wheeze at six years. A positive Helicobacter pylori faecal 13 14 antigen test at age 2 years was negatively associated with wheeze in the past 12 months (OR 15 16 0.57; 95%CI: 0.40 to 0.82), but this association was not observed for Helicobacter pylori at the age 17 of three years or six years. The number of smokers in the household was a strong risk factor for 18 For peer review only 19 wheeze in the past 12 months (p<0.001 for trend), with homes with two or more smokers having 20 21 an odds ratio of the child having wheeze in the past 12 months of 2.08 (95% CI: 1.71 to 2.54) 22 23 compared to those homes with no smokers. The municipality of residence was associated with 24 wheeze in the past 12 months (p=0.04, chi2 test), with children living in Cerro municipality having 25 26 the highest risk of wheeze (OR 1.72 compared to Arroyo Naranjo; 95%CI: +1.61 to 1.84). These 27 28 differences were not substantially modified by adjusting for the number of smokers in the home. 29 30 31 Risk factors for decreased FEV at six years old 32 1 33 The analysis of the association of a number of exposures with FEV1 is presented in Table 3. A 34 35 number of measures of somatic growth were positively associated with FEV1 at six years. These 36 http://bmjopen.bmj.com/ 37 were birth height (+14ml per cm height at birth; 95% confidence intervals CI: +6 to +23), current 38 height (+8ml per cm; 95%CI: +2 to +14), current weight (+13ml per Kg; 95%CI: +8 to +17), and 39 40 current mean arm circumference (+20ml per cm; 95%CI: +14 to +25). The number of smokers 41 42 living in the home was not associated with lung function, but the municipality of residence was 43 strongly associated with current FEV (p<0.001, ANOVA test), with children living in La Lisa 44 1 on September 30, 2021 by guest. Protected copyright. 45 having the lowest lung function (-100ml compared to Arroyo Naranjo, 95%CI: -127 to -73). These 46 47 differences were not substantially modified by adjusting for the number of smokers in the home. 48 49 No measures of infection or inflammation were associated with lung function. 50 51 52 Risk factors for bronchodilatation after inhaled salbutamol at six years old 53 54 The association of exposures with change in FEV1 from baseline after aerosolised salbutamol was 55 administered is presented in Table 4. Any wheeze in the first year of life was positively associated 56 57 with bronchodilatation (+1.94%; 95%CI: +0.81 to +3.08) as was a family history of asthma (+1.85; 58 59 95%CI: +0.14 to +3.57), but there was no relation with wheeze in the past 12 months (+3.61; 60 95%CI: -5.80 to 13.02). Children with a higher birth weight had a lower risk of bronchodilatation

7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 24 BMJ Open

(–2.67%, 95% CI: -4.49 to -0.84). IgE at the age of one year was positively associated with a 1 2 higher risk of bronchodilatation (+1.68%; 95%CI: +0.54 to +2.82), but not IgE at age of three years 3

4 or six years. BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 6 7 The numbers of smokers living in the child’s home was not associated with bronchodilation, but 8 9 the municipality of residence was again a strong determinant of current bronchodilatation 10 11 (p=0.002, ANOVA test), with children living in La Lisa having the highest increase in FEV1 after 12 administration of inhaled salbutamol at six years old (+6.24% compared to Arroyo Naranjo; 13 14 95%CI: +5.56 to +6.91). These differences were not substantially modified by adjusting for the 15 16 number of smokers in the home. 17 18 For peer review only 19 20 21 Discussion 22 23 The most striking observation from our cohort study of young children living in Havana is that the 24 prevalence of wheeze in these children was high, with 38% reporting wheeze in the past 12 25 26 months. Similarly, there was a high prevalence of bronchoconstriction using the objective measure 27 28 of lung function reversibility to inhaled salbutamol, with 33% of the study population having an 29 30 increase in FEV1 of 12% or more. Over a quarter of these six-year old children have received 31 systemic steroids in the previous 12 months, suggesting that asthma control was suboptimal. The 32 33 main consistent association is that municipality of residence is a strong risk factor for wheeze, low 34 35 lung function and bronchoconstriction. A history of wheeze in early life, exposure to paracetamol in 36 http://bmjopen.bmj.com/ 37 the first year of life and current environmental tobacco smoke were risk factors for wheeze, 38 anthropometric measures were positively associated with higher FEV1, and wheeze in the first 39 40 year of life and lower birth weight were associated with untreated bronchconstriction. 41 42 43 Strengths and limitations of the data 44 on September 30, 2021 by guest. Protected copyright. 45 This is the first cohort study of young children living in Cuba that has collected extensive life 46 47 course data to evaluate risk factors for asthma using objective measures of lung function and 48 49 reversibility along with laboratory measures of exposure to infection. The strengths of our data 50 include the prospective nature of the data collected with exposures and outcome measures that 51 52 were selected to permit exploration of risk factors for asthma specifically within an urban Cuban 53 54 environment, where the lifestyle, environment and microbial exposures are very different to more 55 temperate, developed countries. The measurement of lung function is a challenge in young 56 57 children, yet measurements were obtained in 82% of eligible children. The availability of lung 58 59 function measurements along with reversibility to inhaled salbutamol is a particular strength of 60

8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 24

these data, as it provides an objective measure that may reflect different aspects of lung health 1 2 compared to self-reported symptoms. 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Our dataset does have some limitations. We initially recruited 1956 children to the cohort at the 6 7 age of one year with a response rate of 96% eligible children, and approximately five years later 8 9 were able to collect data on 1106 (57%) of these. This was mainly a consequence of the recent 10 11 changes in Cuba that have made it easier for the population more mobile, and hence many 12 children left the policlinic from where they were initially recruited when their family moved their 13 14 residence. There was only a small difference in the prevalence of wheeze in the first year of life 15 16 between those who provided data at the age of 6 years (46%) and those who did not (42%), so 17 loss to follow-up is unlikely to constitute a major source of bias. Collecting data on parental or self- 18 For peer review only 19 reported wheeze using questionnaires is challenging, and although we used the optimal 20 21 methodology from the ISAAC international studies of allergic disease 18 it is possible that some of 22 23 the cases of wheeze were not asthma. Similarly, parental reported asthma medication use may be 24 prone to recall bias, although as acute asthma in children is an upsetting family event it is likely to 25 26 be remembered accurately although the time period possibly less so. This included a number of 27 28 health and lifestyle questions that were answered by the parent or guardian and particular 29 30 attention was paid to parental/guardian reported wheeze in the past 12 months using the 31 methodology developed for the ISAAC epidemiological studies of asthma 18, use of asthma 32 33 medication in the past 12 months and exposure to environmental tobacco smoke. There are no 34 35 validated lung function normal values in Cuban children aged 6 years old and as a consequence 36 http://bmjopen.bmj.com/ 37 we were unable to generate reliable percent predicted values of the lung function in our study 38 population. 39 40 41 42 Like many cohorts, our study began as a cross-sectional population-based that aimed to explore 43 causes of the high prevalence of asthma in Habana, Cuba. Further funding permitted this to 44 on September 30, 2021 by guest. Protected copyright. 45 become a prospective study, with a particular emphasis on infections found in the tropics. As a 46 47 consequence we have tested a variety of analyses and present the results in their entirety. While 48 49 we recognise that it is possible to generate a number of papers analysing different hypotheses 50 from these data, this approach has the advantage of presenting a complete body of work that 51 52 hopefully can inform researchers who are interested in this area. 53 54 55 Risk factors for wheeze in the previous 12 months at six years old 56 57 Wheeze in the first year of life is associated with wheeze in later childhood 23, and may reflect the 58 59 establishment of the asthmatic phenotype in susceptible children. The observation that 60 paracetamol consumption in early life is associated with wheeze in later life 24 has been noted

9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 24 BMJ Open

before, and our data support these observations. However, they do not demonstrate a causality, 1 2 as an alternative explanation is reverse causality with the administration of more paracetamol to 3 25 4 children who are more susceptible to infections and wheeze . BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 6 7 The observation that Helicobacter pylori at the age of two years, but not three years old or six 8 9 years old is associated with lower risk of wheeze at six years old is an interesting observation, that 10 11 would be consistent with the hypothesis that age of exposure to infection is important. Data from 12 other countries has demonstrated the Helicobacter is associated with lower rates of allergic 13 14 disease but not wheeze in children 11 26, but not in adults 27. It is striking however that our 15 16 prevalence of Helicobacter pylori infection was very low at less than 7% in the first six years of life, 17 suggesting that this may be less important in Cuba compared to other countries 11. 18 For peer review only 19 20 21 From a public health perspective, exposure to cigarette smoking and the municipality of residence 22 23 are the most important exposures for wheeze at the age of six years. Second-hand cigarette 24 smoke exposure is well recognised as a risk factor for wheeze in children 5, and it is a concern that 25 26 51% of the homes in our study population contained at least one smoker. The risk of wheeze was 27 28 higher in the municipalities of Cerro and La Lisa, which suggest that there may be an 29 28- 30 environmental factor that predisposes to asthma symptoms such as higher levels of air pollution 31 30. 32 33 34 35 Risk factors for modified FEV1 at six years old 36 http://bmjopen.bmj.com/ 37 The fact that children living in La Lisa municipality already have lower lung function at the age of 38 six years is a major public health concern, and suggests that an exposure associated with living in 39 40 this area may negatively impact on lung growth. La Lisa had by far the highest increase in FEV1 41 42 after administration of aerosolised salbutamol, suggesting that bronchoconstriction is contributing 43 to the low baseline FEV and that a component of the apparent low lung function is reversible. La 44 1 on September 30, 2021 by guest. Protected copyright. 45 Lisa is located inland and while there is no obvious sources of excessive industrial pollution 30, and 46 47 we speculate that this may be associated with higher levels of air pollution as a consequence of 48 49 the relative absence of sea winds compared to the other municipalities. Weight, height and mean 50 arm circumference were all positively associated with FEV1 at six years of age which is likely to be 51 52 simply because they are all measures of somatic growth. 53 54 55 Risk factors for bronchodilatation after inhaled salbutamol at six years old 56 57 Wheeze in the first year of life was associated with bronchodilatation after the administration of 58 59 inhaled salbutamol at six years. This suggests that symptoms of asthma in early life are predictive 60 of untreated bronchoconstriction five years later, and is consistent with the hypothesis that the

10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 24

asthmatic phenotype can be observed to persist from early life onwards 23. The observation that 1 2 birth weight is inversely associated with bronchoconstriction provides objective evidence to 3

4 support the earlier epidemiological evidence that adults with higher birth weights have a lower risk BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 of having a diagnosis of asthma 16. The observation that a family history asthma is associated with 6 7 bronchoconstriction also wheeze is consistent with the knowledge that there is a genetic 8 9 component to asthma 31. 10 11 12 Self-reported wheeze and bronchoconstriction 13 14 One interesting observation from our data is that the wheeze in the past 12 months is not 15 16 associated with either FEV1, or the change in FEV1 after administration of inhaled bronchodilator. 17 Although it is possible that children who have had wheeze subsequently received asthma 18 For peer review only 19 treatment, this lack of association between asthma symptoms and objective measures of asthma 20 32 21 has been described previously . This supports that concept that wheeze and reversibility of FEV1 22 23 after administration of aerosolised salbutamol may measure different aspects of lung development 24 and health. 25 26 27 28 Public health implications of these data 29 30 The current study was designed in response to clinical concerns that asthma was becoming a 31 public health problem in Havana. These data support that hypothesis, and in particular identify that 32 33 living in certain municipalities may result in higher levels of currently unknown exposures that 34 35 require public health intervention. It is possible that environmental air pollution may drive some of 36 http://bmjopen.bmj.com/ 37 these geographical differences, and that studies of air pollution are urgently required in Havana. 38 The prevalence of exposure to second-hand tobacco smoke in children living in Havana remains 39 40 high, and this is an area where a combination of public health interventions including taxation, 41 42 legislation that enforces bans of smoking in the workplace and public places, restrictions on 43 tobacco product advertising, and helping individual smokers quit smoking are known to be 44 on September 30, 2021 by guest. Protected copyright. 45 effective. We are exploring these hypotheses further by doing further objective measurements of 46 47 particulate air pollution within the study population. 48 49 50 Summary 51 52 Asthma is common in young children living in Havana, and the high prevalence of the use of 53 54 systemic steroids probably reflects the underuse of regular inhaled corticosteroid prophylaxis 55 treatments leading to the requirement for rescue treatment for wheezing. As societies urbanise, 56 57 environmental air pollution may increase from a variety of sources. Cuba’s economy has struggled 58 59 as a consequence of a historical and political events 33, and it remains under an economic 60 embargo from the USA that has negatively impacted on healthcare 34. This makes providing

11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 24 BMJ Open

regular inhaled corticosteroids to all children who need them challenging. However, other low and 1 2 middle-income countries also have difficult economic and environmental circumstances, and if 3

4 these observations are replicated elsewhere, then this represents an important global public health BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 issue. 6 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 24

Acknowledgements 1 2 We would like to thank the children and their parents and guardians for participating in this study. 3

4 We would like to thank all of the staff at the municipalities and policlinics who helped in many ways BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 with data collection and sample analysis. 6 7 8 9 Contributors: The study was conceived and designed by RSM, SVF, AWF and JB. Data were 10 11 collected by RSM, SVF, VAV, NBSB, CC, MLLG, ZVP, BCT and MABL. Data analysis was 12 conducted by RSM and AWF. RSM, SVF and AWF wrote the first draft of the manuscript. All 13 14 authors critically reviewed the manuscript, provided intellectual content and approved the final 15 16 version prior to submission for publication. 17 18 For peer review only 19 Funding: This study was funded by The Wellcome Trust, The University of Nottingham, 20 21 Nottingham University Hospitals Charity, NIHR Nottingham Biomedical Research Centre and a 22 23 BMA Award (James Trust). 24 25 26 Competing interests: None declared. 27 28 29 30 Data sharing statement: Cuban regulations do not allow dissemination of national datasets. 31 However, statistical analyses can be performed upon reasonable requests to the corresponding 32 33 author. 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 24 BMJ Open

References 1 2 1. Pearce N, Ait-Khaled N, Beasley R, et al. Worldwide trends in the prevalence of asthma symptoms: phase 3 III of the international study of asthma and allergies in childhood (ISAAC). Thorax 2007;62:758-66. 4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 2. Jartti T, Gern J. Role of viral infections in the development and exacerbation of asthma in children. J 6 Allergy Clin Immunology 2017;140:895-906. 7 3. Edwards M, Strong K, Cameron A, et al. Viral infections in allergy and immunology: How allergic 8 inflammation influences viral infections and illness. J Allergy Clin Immunology 2017;140:909-20. 9 4. Prescott S. Effects of early cigarette smoke exposure on early immune development and respiratory 10 11 disease. Paediatric respiratory reviews 2008;9:3-10. 12 5. Lewis S, Antoniak M, Venn A, et al. Secondhand smoke, dietary fruit intake, road traffic exposures, and 13 the prevalence of asthma: A cross-sectional study in young children. Am J Epidem 2005;161:406-11. 14 6. Litonjua A, Gold D. Early-life exposures and later lung function. Am J Resp Crit Care Med 2016;193:110- 15 16 11. 17 7. Schultz E, Hallberg J, Bellander T, et al. Early-life exposure to traffic-related air pollution and lung 18 function in adolescence.For Am peer J Resp Crit Carereview Med 2016;193:171-77. only 19 8. Sly P, Bush A. From the cradle to the grave: The early-life origins of chronic obstructive pulmonary 20 disease. Am J Resp Crit Care Med 2016;193 21 22 9. Postma D, Weiss S, den Berge M, et al. Revisiting the dutch hypothesis. J Allergy Clin Immunology 23 2015;136:521-29. 24 10. Feary J, Britton J, Leonardi-Bee J. Atopy and current intestinal parasite infection: a systematic review 25 and meta-analysis. Allergy 2010;66:569-77. 26 27 11. Amberbir A, Medhin G, Abegaz W, et al. Exposure to Helicobacter pylori infection in early childhood 28 and the risk of allergic disease and atopic sensitisation: a longitudinal birth cohort study. Clin Exp 29 Allergy 2014;44:563-71. 30 12. Mabalirajan U, Kadhiravan T, Sharma S, et al. TH2 immune response in patients with dengue during 31 defervescence: preliminary evidence. Am J Trop Med Hygiene 2005;72:783-85. 32 33 13. Fogarty A, Jones S, Britton J, et al. Systemic inflammation and decline in lung function in a general 34 population: a prospective study. Thorax 2007;62:515-20. 35 14. Venero-Fernandez S, Suarez Medina R, Mora Faife E, et al. Risk factors for wheezing in infants born in 36 Cuba. Quarterly Journal Medicine 2013;106:1023-29. http://bmjopen.bmj.com/ 37 15. Suarez-Medina R, Venero-Fernandez S, Mora Faife E, et al. Risk factors for eczema in infants born in 38 39 Cuba. BMC Dermatology 2014;14:6. 40 16. Shaheen S, Sterne J, Montgomery S, et al. Birth weight, body mass index and asthma in young adults. 41 Thorax 1999;54:396-402. 42 17. Fundora Hernández H, Suarez-Medina R, Venero Fernandez S, et al. What are the main environmental 43 44 exposures associated with elevated IgE in Cuban infants? A population-based study. Tropical on September 30, 2021 by guest. Protected copyright. 45 Medicine and International Health 2014;19:545-54. 46 18. Committee TISoAaAiCIS. Worldwide variations in the prevalence of asthma symptoms: the 47 International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Eur Resp J 48 1998;12:315-35. 49 50 19. Vircell dengue ELISA IgG. http://wwwperamedcom/peramed/docs/G1018_ENpdf (accessed 14/4/2016) 51 20. Venero-Fernandez S, Fundora-Hernández H, Batista-Gutierrez L, et al. The association of low birth 52 weight with serum C reactive protein in three year old children living in Cuba: A population-based 53 prospective study. Am J Human Biol 2016:DOI 10.1002/ajhb.22936. 54 55 21. Garcia F, Venero Fernandez S, Fundora Hernández H, et al. Seroprevalencia de anticuerpos IgG anti- 56 Toxoplasma Gondii en infantes de La Habana. Parasitaria 2015;73:(in press). 57 22. Miller M, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Resp J 2005;26:319-38. 58 23. Martinez F, Wright A, Taussig L, et al. Asthma and wheezing in the first six years of life. New Eng J Med 59 1995;332:133-38. 60

14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 24

24. Beasley R, Clayton T, Crane J, et al. Association between paracetamol use in infancy and childhood, and 1 2 risk of asthma, rhinoconjunctivitis, and eczema in children aged 6-7 years: analysis from Phase 3 Three of the ISAAC progranmme. Lancet 2008;372:1039-48.

4 25. Rusconi F; Gagliardi L; Galassi C; Forastiere F; Brunetti L; La Grutta S; Piffer S; Talassi F; SIDRIA-2 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 Collaborative Group. Paracetamol and antibiotics in childhood and subsequent development of 6 wheezing/asthma: association or causation? Int J Epidemiol 2011;40:662-67. 7 8 26. Amberbir A, Medhin G, Erku W, et al. Effects of Helicobacter pylori, geohelminth infection and selected 9 commensal bacteria on the risk of allergic disease and sensitization in 3 year old Ethiopian children. 10 Clinical Exp Allergy 2011;41:1422-30. 11 27. Fullerton D, Britton J, Lewis S, et al. Helicobacter pylori and lung function, asthma, atopy and allergic 12 13 disease - A population-based cross-sectional study. Int J Epidemiol 2009;38:419-26. 14 28. Wu W, Jin Y, Carlsten C. Inflammatory health effects of indoor and outdoor particular matter. J Allergy 15 Clin Immunology 2018;141:833-44. 16 29. Gordon S, Bruce N, Grigg J, et al. Respiratory risks from household air pollution in low and middle 17 income countries. Lancet Respiratory Medicine 2014;2:823-60. 18 For peer review only 19 30. Cuellar-Luna L, Puerto-Rodriguez A, Maldonado-Cantillo G, et al. Distribucion espacial de fuents fijas 20 contaminantes y su impact en salud, provincia La Habana (Cuba). Hig Sanid Ambient 2013;13:968- 21 74. 22 31. Shrine N, Portelli MA, John C, et al. Moderate-to-severe asthma in individuals of European ancestry: a 23 24 genome-wide association study. The Lancet Respiratory medicine 2019;7(1):20-34. doi: 25 10.1016/S2213-2600(18)30389-8 [published Online First: 2018/12/16] 26 32. MacKenney J, Oyarzun M, Diaz P, et al. Prevalence of asthma, atopy and bronchial hyperresponsiveness 27 and their interrelation in a semi-rural area of Chile. Int J Tuberculosis & Lung Dis 2005;9:1288-93. 28 33. Gott R. Cuba. New Haven, USA: Yale University Press 2010. 29 30 34. Barry M. Effect of the US embargo and economic decline on health in Cuba. Ann Intern Med 31 2000;132:151-54. 32 33 34 35 36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 24 BMJ Open

Figure 1. Flow diagram of participants and data collection 1 2 3

16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 24

Table 1. Description of study population 1 2 3 Total (N=1106) Provided FEV1 (N=913)

17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 24 BMJ Open

Table 2. Association of exposures with wheeze in past 12 months. 1 Total number = 1106 No (%, sd) Odds ratio of wheeze 2 3 (95% CI)

18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 24

19 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 24 BMJ Open

Table 4. Association of exposures with bronchodilatation after inhaled salbutamol 1 Total number = 903 No (%, sd) % change in FEV 2 1 3 (95% CI)

BMJ Open Page 22 of 24

1 Year 1. 2 3 N = 1956. Wheeze = 45% 4 5 6 7 8 Year 2. 9 10 N = 1701. Wheeze = 60%

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 Year 3. 16 N = 1543. Wheeze = 58% 17 18

Page 23 of 24 BMJ Open

1 2 3 4 5 6 33% of population with bronchodilatation greater than 12% FEV1 7 8 9 10

11 http://bmjopen.bmj.com/ 12 For peer review only 13 14 15 16 17 18

1 2 STROBE Statement 3

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 24 BMJ Open

36 http://bmjopen.bmj.com/ 37 38 39 40 41 42 43 44 on September 30, 2021 by guest. Protected copyright. 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second and bronchoconstriction in young children living in Havana, Cuba: A population-based cohort study

ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-034192.R2

Article Type: Original research

Date Submitted by the 31-Jan-2020 Author:

Primary Subject Respiratory medicine on September 30, 2021 by guest. Protected copyright. Heading:

Secondary Subject Heading: Global health, Paediatrics

Keywords: Asthma < THORACIC MEDICINE, Cuba, PAEDIATRICS

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 24 BMJ Open

1 2 3

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 24

Prevalence and risk factors for wheeze, decreased forced expiratory volume in one second 1 and bronchoconstriction in young children living in Havana, Cuba: A population-based 2 3 cohort study

4 BMJ Open: first published as 10.1136/bmjopen-2019-034192 on 22 April 2020. Downloaded from 5 * Ramon Suarez-Medina1 6 Silvia Josefina Venero-Fernández1 7 Vilma Alvarez-Valdés2 8 Nieves B Sardiñas-Baez2 9 Cristina Carmona2 10 2 11 María Luisa Loinaz-Gonzalez 12 Zunilda Verdecia-Pérez2 13 Barbara Corona-Tamayo2 14 María Antonia Betancourt-López1 15 John Britton3, 16 Andrew W Fogarty3 17 and the HINASIC (Historia Natural de la Sibilancia en Cuba/Natural History of Wheezing in Cuba) 18 For peer review only ++ 19 Study Group 20 21 1 Instituto Nacional de Higiene, Epidemiología y Microbiología, Infanta No 1158 e/ Llinás y Clavel, 22 Código Postal 10300, La Habana, Cuba. 23 2 Dirección Municipal de Salud Pública municipios Cerro y Arroyo Naranjo, La Habana, Cuba. 24 3 Nottingham NIHR Biomedical Research Unit, Division of Epidemiology and Public Health, 25 26 University of Nottingham, Clinical Sciences Building, City Hospital, Nottingham NG5 1PB, UK ++ 27 full list of collaborating researchers at the end of the manuscript 28 29 The corresponding author for the manuscript submission process is Dr Andrew Fogarty 30 ([email protected]) due to poor internet connections with Cuba. 31 32 (If accepted for publication) *Correspondence and requests for reprints to Dr. Ramón Suárez 33 34 Medina: 35 Address: Infanta No 1158 e/ Llinás y Clavel, Código Postal 10300, La Habana, Cuba. 36