Impact of Haemophilus influenzae type b combination vaccination on asthma symptoms and pneumonia in 5-year-old children in rural Bangladesh: a longitudinal study and comparison with a previous cross-sectional study

The aim of the present study is to determine if Hib vaccination contributed to the decrease of asthma and wheezing through pneumonia prevention.

We conducted this case–control study on the risk of Hib combination vaccination and history of pneumonia for asthma symptoms among 5-year-old children in Matlab. We also compared the results of the present study with those of our previous study (described below) to determine whether Hib combination vaccination contributed to decrease pneumonia prevalence [9], the result of which was partially unpublished. We finally compared death from respiratory infections including pneumonia among children under 5 years of age in the area before and after the introduction of Hib combination vaccination using reports of health and demographic surveillance system (HDSS) run by icddr,b, formerly the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), to determine the impact of Hib combination vaccination on pneumonia related mortality [18‐24].

Matlab is a low-lying riverine area, where the principal occupations are farming and fishing. Since 1966, the HDSS, which consists of regular cross-sectional censuses and the longitudinal registration of vital events, has been maintained in the area by the icddr,b [24]. The population of the HDSS area, which encompasses 142 villages, was approximately 220,000 in 2001 and 230,000 in 2014 when the present study was initially planned. A maternal and child health, and family-planning program serve approximately half of the population of the HDSS area (the icddr,b service area, which covers 67 villages). A record-keeping system (RKS) records all instances of immunization in the service area.

The present study was conducted from December 2015 to October 2016, and planned to include 1800 children aged 5 years who were randomly selected from all 67 villages of the HDSS service area. A total of 1658 children ultimately participated in the study. The International Study of Asthma and Allergies in Childhood questionnaire was used to identify wheezing. One hundred and forty-five children had experienced wheezing during the previous 12 months and 1513 children had not. The 145 children who experienced wheezing during the previous 12 months were placed in the “wheezing” group and 1513 children who did not experience wheezing during the previous 12 months were placed in the “non-wheezing” group. The participants of this study were able to receive Hib combination vaccine since it was introduced in the area in 2009. We compared these children with the participants of our former study conducted in 2001. The total population of our 2001 study consisted of 1705 children, who were all children aged 5 years from 51 villages that were themselves randomly selected from the service area [9]. Data on deaths from respiratory infections including pneumonia were obtained from the Matlab HDSS Scientific Reports published by the icddr,b [18‐24].

The study protocol was approved by the Ethical Review Committee of the icddr,b (PR-15054). The Ethics Committee of Tokyo Kasei University (Sayama H27-09), and The University of Tokyo (11018 and 2020180NI) approved the study. The protocol of our 2001 study was approved by the Ethical Review Committee of the International Centre for Diarrhoeal Disease Research, Bangladesh (2000-038). As the studies involved human subjects, the ethical principles of the Declaration of Helsinki were followed. Written informed consent was obtained from the legal guardians of all participants.

The procedures used for data collection in the present study have been described elsewhere [10]. In brief, trained local field-research assistants visited the homes of the children and collected information using a semi-structured, pre-tested questionnaire adopted from the International Study of Asthma and Allergies in Childhood questionnaire [25]. Wheezing was defined as any episode of wheezing or whistling in the chest in the 12 months preceding the interview. Children who answered “No,” to the above question were placed in the “non-wheezing” group. Information was also collected regarding family history of allergy, SES, and environmental factors. Information on the participants’ history of pneumonia was retrieved from Matlab Hospital clinical records. Children with suspected pneumonia had been referred to Matlab Hospital from the CHRWs and the diagnosis of pneumonia was based on the WHO guidelines for the management of common illnesses with limited resources [26].

The procedures used for data collection in the 2001 study have also been previously described [9]. In 2001, Episodes of pneumonia was obtained from the RKS of Matlab HDSS. Pneumonia was recorded in the RKS by CHRWs during surveillance (performed every 2 weeks), based on statements from mothers regarding increased respiratory rates with or without chest indrawing.

Information on deaths from respiratory infections including pneumonia among children under 5 years of age and the mid-year population in the area were obtained from HDSS Scientific Report Nos. 74, 82, 90, 103, 109, 121, and 138 published by the icddr,b [18‐24]. The number of deaths from respiratory infections that include acute respiratory infection, pneumonia and influenza at < 1 year old and at 1–4 years old in the years 1991, 1996, 2001, 2006, 2009, 2011, and 2016, in the icddr,b service area of Matlab HDSS, as recorded in the Matlab HDSS Scientific Reports, were compared to find out the decrease in mortality. Since the 5-year-old participants of the 2001 study were < 1 year of age in 1996, and the 5-year-old participants of the 2016 study were < 1 year of age in 2011, the data for 1996 and 2011 are included here. Similarly, the data for 2006 and 2008 are included to compare the pre and post effect Hib combination vaccine on children of < 1 year of age and 1–4 years of age, respectively.

This study was initially planned to determine the impact of the national deworming program on wheezing [9]. Therefore, the sample size was determined based on the levels of anti-Ascaris IgE. We calculated the sample size based on the assumption that at least 16% of children aged 60–71 months would have wheezing [27]. Given 80% power and a 5% significance level, 209 children in each group were required to detect a difference of 1.8 U_A/mL (SD, 1.5) to 1.4 U_A/mL (SD, 1.4) in the values of serum anti-Ascaris IgE levels between the wheezing and never-wheezing groups of children. Thus, we needed to recruit 240 children for the children with and without wheezing each, assuming a 15% refusal rate (including absences). To obtain the required number of children for the wheezing group, we needed to approach 1800 individuals, assuming a 20% loss due to absences, refusal, and other reasons.

Data were analyzed using IBM SPSS Statistics version 26 (IBM Japan, Tokyo, Japan). First, the prevalence of wheezing was calculated. An initial exploratory analysis was conducted to determine the distribution of the independent variables. After each variable had been subjected to a descriptive analysis between the children with and without wheezing, continuous variables (e.g., height) were compared using a t-test (if approximately normally distributed) or Mann–Whitney U test (if not normally distributed), and categorical variables were compared using a χ² test. Then, the odds ratios for wheezing, with or without adjustments for risk factors, were calculated using multiple logistic regression analysis, with wheezing status as the outcome variable. Then, we compared the reduction in the history of pneumonia from 2001 to 2016 across for each age at the time of pneumonia history, and this reduction between 2001 and 2016 was analyzed using the paired t-test in both the overall study population and the wheezing group. We also compared the history of pneumonia between the 2001 and 2016 study participants for each age at the time of pneumonia development by using the χ² test. We then defined the reduction rate of pneumonia (%) = (percentage of pneumonia cases in 2001 − percentage of pneumonia cases in 2016)/(percentage of pneumonia cases in 2001) times 100 (%), and made a graph to visualize it. The number of deaths from pneumonia among children aged 0 and 1–4 years during the period from 1991 to 2016 was analyzed using the χ² test to determine the impact of various measures, such as antibiotic use by the CHRWs, Vitamin A distribution, improved hospital care, improved referral, improved nutritional status, including Hib combination vaccination, on mortality from respiratory infections including pneumonia in this area.

In the present study, the prevalence of wheezing during the 12 months prior to the survey was 8.7% (95% confidence interval [CI] 7.4%, 10.1%). The characteristics of the study participants are presented in Table 1. Wheezing children received significantly less Hib combination vaccination than children without wheezing, while antibiotic use during the first 1 year of life, a history of pneumonia at 1, 2, 3 and 4 years of age, parental asthma and allergic rhinitis were significantly more common in children with wheezing than in children without wheezing. Hib combination vaccination was a protective factor against wheezing (OR: 0.50; p = 0.001), while antibiotic use during the first 1 year of life, a history of pneumonia at 1 year of age (OR: 2.86, p = 0.046), at 2 years of age (OR: 3.18; p = 0.030), at 3–4 years of age (OR: 3.69; p = 0.030), and parental asthma were risk factors for wheezing (Table 2). Crude and adjusted odds ratios of pneumonia history at < 1 year were not significant for wheezing.

The history of pneumonia decreased as the children grew older in the 2001 study participants in both the overall cohort and the wheezing group. In contrast, it peaked at the age of 2 years in the 2016 wheezing group, whereas the history seems almost the same across the years in the 2016 overall cohort (Table 3).

The pneumonia history significantly decreased from 2001 to 2016 in both the overall cohort and the wheezing group (paired t-test: p = 0.014 and p < 0.001, respectively). We found that in both the overall cohort and the wheezing group the pneumonia histories at the age of < 1 year, 1 year and 2 years significantly decreased in the 2016 (Fisher two-tailed exact test: p < 0.001), however, the pneumonia history did not significantly decrease at the age of 3–4 years in either of the overall study population or the wheezing group (p = 0.196 and 1.000, respectively) (Table 3).

Table 4 shows that the percentage of deaths from respiratory infections including pneumonia continued to decrease significantly from 1996 to 2016 aged < 1 year (trend analysis: p < 0.001) and at the age 1–4 years (p < 0.001).

The percentages of deaths from respiratory infections including pneumonia aged < 1 year decreased significantly from 1996 to 2011, when our study participants were < 1 year of age (†: Fisher two-tailed exact test: p < 0.001), and from 2001 to 2006 (‡: Fisher two-tailed exact test: p < 0.001), however, it did not differ significantly when compared between 2006 and 2011 (§: Fisher two-tailed exact test: p = 0.572). Similarly, the decrease in the children aged < 1 year was not significant before and after the implementation of the vaccination (||: Fisher two-tailed exact test: p = 0.535). The deaths from respiratory infections including pneumonia at < 1 year old seems to have declined significantly in the year of 2006 in the area (data not shown).

In contrast, the difference at 1–4 years was significant between 1991 and 1996 (¶: Fisher two-tailed exact test: p < 0.001) but not significant between 2001 and 2016 (††: Fisher two-tailed exact test: p = 0.292), when our study participants of the 2001 and 2016 studies were1–4 years of age. The decline in the mortality among older children occurred in the early 1990′s earlier than the children aged < 1 year old.

The reduction rate of pneumonia defined by the following formula (%) = (percentage of pneumonia cases in 2001 − percentage of pneumonia cases in 2016)/(percentage of pneumonia cases in 2001) times 100 (%) at the ages of 0, 1, 2, and 3–4 years were 87.9%, 88.0%, 71.9%, and 22.2%, respectively, in the overall cohort, and 89.1%, 76.1%, 49.4%, and 0%, respectively, in the wheezing group (Fig. 1). Namely, whereas the reduction rate aged < 1 year was nearly 90%, the reduction rates aged 2–4 years were nearly less than 70%. It declined when the participants grew older and this decline was faster in the wheezing group.