Child health, nutrition and gut microbiota development during the first two years of life; study protocol of a prospective cohort study from the Khyber Pakhtunkhwa, Pakistan

Objectives

The primary objective of the CHAMP study is to characterize gut microbiota development and longitudinal growth in infants during the first two years of life. This will be done by assessing the microbial diversity, relative abundance of the dominant microbiota, and interindividual variations in microbial diversity and infant growth patterns. The secondary objective of the study is to investigate the impact of (i) sociodemographic characteristics, (ii) mode of delivery, (iii) maternal factors such as diet and anti-natal and post-natal care, (iv) child dietary intake and feeding practices, and (v) nutritional status of the child on gut microbiota composition at different time points from birth until the child is two years old.

Operational definitions

Study design and setting

The CHAMP study was a population-based, prospective cohort study involving mother-infant pairs (dyads) from District Swat, Pakistan. Swat is located at 34°46′58″ N and 72°21′43″ N in Khyber Pakhtunkhwa province of Pakistan and is home to more than 2.3 million people.³¹ Because of its geographic location in the Hindukush–Himalayan region, most of the land is occupied by mountains and thick forests. The area is one of the most marginalized and neglected regions of the province, with livestock, agriculture, horticulture, and tourism as the primary sources of income and livelihood of the local population.³² Swat is among the most vulnerable districts in Pakistan and is prone to climate change. Since 2010, the district has been hit by at least three major floods that have significantly disrupted essential services, such as healthcare, sanitation, and access to safe drinking water.³³ As a result, the majority of the population, especially children, are at a high risk of malnutrition and its deleterious consequences.

Administratively, the district is divided into seven administrative units known as Tehsil Municipal Administration units (TMAs). Tehsil Matta was chosen as the study area because it is the second most populated tehsil in Swat district, with the highest number of rural and remote communities. Following a preliminary survey, UC Beha was selected as the study site for several reasons. It is located at a high altitude with the longest adjoining borders with district Dir (Upper), and most of the population lives in agropastoral communities with very limited access to education and healthcare facilities. In winter, access to the area is limited by heavy snowfall and landslides.

Sampling technique

District swats are divided into seven administrative units known as Tehsil Municipal Administration units (TMAs). Of these, tehsil Matta was chosen as the study area because it is the second most populated tehsil in Swat district, located at a high altitude with the longest adjoining borders with Dir district (Upper), which has the highest number of rural and remote communities. The Matta tehsil is further divided into union ouncils, which are the primary administrative institutions in Pakistan. In rural areas, union councils are referred to as village councils. Of the 13 village councils in the sampling area, three union councils were randomly selected as the primary sampling unit. Subsequently, the researchers visited the primary health centers and designated vaccination centers within the chosen village councils. Newborn children were identified using the Extended Program for Immunization (EPI) register. Potential participants were appraoched with the help of Lady Health Workers (LHWs), community-based health workers linked to local health facilities. All families in the area were registered with a designated LHW who regularly visited them. Important information, such as the child’s name, father’s name, date of birth, and address, was recorded, and a list of eligible study participants was prepared. Once recruited, each participant will be followed up for two years. The study follows ethical principles involving human subjects as outlined in the declaration of Helsinki. Ethical approval for this study was granted by the Ethics Board of Khyber Medical University, Pakistan ref no DIR/KMU-EB/BR/001-03 dated 11/01/2024.

Participants selection

The study population included 70 mother-infant pairs (dyads), according to the following criteria:

Inclusion criteria

The exclusion criteria are:

Ethical approval for the study was granted by the Research Ethics Board of Khyber Medical University. This study was also registered at www.clinicaltrial.gov (NCT05793294).

Recruitment

In the local (Pashtun) culture, it is not possible to directly contact a potential female study participant without prior consent from the husband/male head of the household. Therefore, at each study site, an initial information session was held with the male parents/guardians with the help of elected local representatives and elders of the community. During the session, complete information about the study objectives, eligibility criteria, and data collection process was presented and explained by the team leader in the local (Pashto) language. A participant information sheet containing all the study details and procedures in an easy-to-understand, local (Pashto), and Urdu (national language of Pakistan) were also provided to the parents during the session. All queries were also answered. Following the session, eligible parents/guardians were invited by trained research assistants to enroll their babies in the study. Once they agreed, the parents were asked to sign a written informed consent form in their preferred language.

Participants, funder and public involvement:

Community members, funder (NIH Pakistan) or the participants’ parents were not directly involved in developing the study design, conduct, and outcome measures. The study protocol was reviewed and approved by the Office of Research, Innovation, and Commercialization (ORIC) Khyber Medical University and National Institute of Health (NIH), Pakistan. Healthcare professionals (LHWs) involved in the study had close liaison with the community and acted as a bridge between the researchers and participants. They also contributed to the design of the data collection questionnaires, informed consent, and helped ensure confidentiality. However, the individual data were not reported to the participants.

Study procedures

Figure 1 provides an overview of the overall study flow.

Figure 1. Flowchart of the study.

Data collection

We will collect different data from both mother and infant at baseline and at each time point during the follow-up period. The data collection in this project used an Android version of the KoBoCollect/Toolbox as a data collection instrument. To collect the demographic and socioeconomic information of the study participants, an interviewer administered a structured questionnaire will be used. The questionnaire included information such as name, age, sex, household information, and socioeconomic status. Details about ante-natal and post-natal care will also be recorded using a structured, validated questionnaire adapted from the National Nutrition Survey Pakistan questionnaire.³⁴ These data were recorded only at baseline, that is, at the start of the study.

The Dietary Quality Questionnaire (DQQ) was used to assess the dietary intake of mothers. This method was chosen because, in Pakistan, there is no validated food frequency questionnaire to assess dietary intake. The DQQ is a standardized and easy-to-use tool to collect food group consumption data required for calculating diet quality indicators, such as Minimum Dietary Diversity for Women (MDD-W). The questionnaire has been validated and implemented across more than 100 countries including Pakistan.³⁵ The DQQ is a list-based method consisting of yes/no responses to 29 food groups consumed by the respondent during the previous day and night (24 hours). The procedure typically requires approximately 5–7 minutes to complete. Information regarding feeding practices of the children will be collected using validated, semi-structured questionnaires based on Infant and Young Child Feeding Practices (IYCF) indicators developed by the Technical Expert Advisory Group on Nutrition Monitoring (TEAM) of the WHO.³⁶ This set of simple, validated, and reliable indicators is widely used to assess IYCF practices across the globe, especially in low- and middle-income countries. The IYCF indicators were assessed by interviewing the mothers or the main caregivers. The mothers will also be interviewed to collect data about the child’s medical history, diarrhea and respiratory tract infection, hospitalization, and use of antibiotics using a structured questionnaire. DQQ, IYCF, and child health record data will be collected at baseline and at each time point during follow-up.

Anthropometric assessment

At baseline and at each time point, the following anthropometric assessment will be performed on each child/infant and their mothers using standard methods.³⁷

Height/length: Recumbent length will be measured using an infantometer. Before measuring, the measuring board will be placed on a hard, flat surface such as a ground or sturdy table. The mother was asked to place the child gently on the board such that the infant’s head is aligned against the headboard while an assistant straightens the infant body and feet. When the child is position is correct, the researcher moves the foot board firmly against the child’s heels. The length of the child is recorded to the nearest 0.1 cm. The measurement will be repeated twice, and the average of the two measurements is recorded as the final length of the child. Mothers’ height will be recorded using a stadiometer following standard protocols. Briefly, they will be asked to remove shoes and gently stand on stadiometer. To ensure accurate measurement, the mothers will be instructed to stand up against the board, ensuring the Frankfurt plan position in which the back of the head, shoulder blades and buttocks and heels touch the back of the stadiometer. Once ready, the researcher will gently and firmly slide the measuring board’s moveable headpiece down until it touches the crown of the person’s head (compresses the hair). The procedure is repeated twice and the average of the two measurements is recorded as the height of the mother.

Weight: To record the weight of the child, a specialized electronic scale (Secca, UK) will be used. Before weighing the child, the parent will be instructed/helped to remove shoes, socks and heavy clothing except dry clean diapers or underpants from the child. When ready, the child is placed on his/her back on the center of the scale pan with the help of a trained researcher. We wait until the infant is still in the pan and the digital display is no more changing. The weight of the child is recorded to the nearest 0.01 kg. The weight of the mother is measured in kilograms using a calibrated electronic weighing scale. The mothers will be asked to remove shoes, jewellery and extra clothing (dupatta/chaddar) and stand calmly on the scale. The weight will be recorded to the nearest 0.01 kg twice and the average of the two measurement is considered the final weight of the mother.

Mid upper arm circumference (MUAC): For infant MUAC measurements, a non-stretchable, numbered and colored MUAC tape will be used. To prepare for the measurement, the infant mother will be asked to sit on a comfortable chair, place the child in her lap and remove any clothing covering the left hand of the child. The mid-point between the tip of the elbow and the shoulder is located by straightening the arm along the axis of the body and forearm at 90-degree forwards. Once the mid-point is marked with a pen, the arm is straightened and the MUAC tape is placed around the arm at the midpoint and the reading is taken to the nearest mm.

Head circumference: The head circumference of the child will be measured using a non-stretchable measuring tap around the most prominent part of head to the middle of the forehead while the hair and soft tissue is compressed. At least two measurements will be recorded and the average of the two will be taken.

Sample collection

Dry blood spot: Dry blood spots (DBS) are collected from both the mother and the infant at baseline and during the follow-up by skin puncture of the third or fourth finger of the non-writing hand.³⁸ Before DBS collection, the participants’ hands were first warmed, followed by anterograde massaging of the finger in the direction of blood flow towards the puncture site. The DBS collection site, that is, the palmar side of the tip of the distal phalanx, was cleaned with 70% isopropyl alcohol and punctured using a single-use lacet. The first drop of blood was wiped off with a gauze pad, and subsequent drops were transferred to the marked circles on the surface of the filter paper without touching the surface. After drying, the DBS papers were transferred to a zip-lock bag and stored at -80 °C until further processing in the KMU main lab.

Stool: Stool samples are collected from the diapers of infants at each time point. For this purpose, the parents were provided diapers, a specimen collection jar, disposable gloves, a zip lock bag, and an instruction sheet one day before sample collection. The parents were instructed to regularly check their child’s diaper and remove it immediately after the child passed the stool. Stool samples were transferred immediately to the specimen container, placed in a zip-lock bag, and handed over to the research assistant within 4 h of sample collection. The research assistant transferred 400 mg of stool sample into 2 mL screw-top tubes prefilled with DNA shield reagent (Zymo Research USA) and sent it to the main KMU lab at room temperature, where it was stored at -80 °C.

Laboratory analysis

Hematological aassessment

Dry blood samples will be used to assess the biomarkers of infections, micronutrient deficiencies, and inflammation. For this purpose, commercially available chemiluminscence-based Q-plex Aray kits (Quansys Biosciences) will be used following the manufacturer’s instructions. This assay reliably detects and quantifies histidine-rich protein II (biomarker of malarial infections), C-reactive protein, alpha-1-acid glycoprotein (biomarkers of inflammation), ferritin and soluble transferrin receptor (biomarker of iron deficiency), retinol binding protein (biomarker of vitamin A deficiency), and thyroglobulin (biomarker of iodine deficiency). ³⁹

DNA extraction and 16S rRNA sequencing

Bacterial genomic DNA will be extracted from fecal samples (200 mg) stored at -80 °C using a ZymoBIOMICS DNA Miniprep Kit (Zymo Research, Irvine, CA, USA), following the manufacturer’s instructions. Each sample will be quantified using Denovix and quality was checked by PCR of the 16S rRNA gene target and agarose gel electrophoresis. For the amplicon PCR step, 12.5 ng of purified bacterial genomic DNA will be used to amplify the V3-V4 target region according to the manufacturer’s protocol. The amplified samples were then subjected to agarose gel electrophoresis to separate the DNA samples from the primer dimers. Each DNA band was extracted and purified using the QIAquick gel extraction kit (Cat. no. 28705; QIAGEN, Germany). After purification, the concentration of each amplicon PCR will be quantified using Nanodrop. Next, library preparation was performed via indexing PCR and the reaction was subsequently cleaned up and quantified, following the Illumina kit’s guidelines. The concentrations of all samples were then standardized to a minimum of 2.5 nmol. One microliter of DNA from each sample was pooled and sent for high-throughput sequencing on the Illumina Miseq^® platform (Illumina, San Diego, California, USA) following the manufacturer’s instructions.

Monitoring and evaluation strategy

The monitoring and evaluation of the entire project will be conducted at all levels. Data entry will be randomly rechecked for accuracy by re-interviews or re-visits to households participating in the study. The principal investigator oversees the data collectors and enumerators. Bimonthly assessments will be performed by the project team consisting of the project manager, the PI, and the co-PI. In addition, the KMU ORIC team will monitor the overall progress of the project through its own monitoring and evaluation system. Any discrepancy or deviance from the main objectives or lapse in the timeframe and data collection procedures will be handled in close coordination of the project team with the field data collector and support from KMU.

Participants retention strategies

A common issue in longitudinal cohort studies is systemic attrition, which can have a significant impact on the generalizability of the outcome of interest. Study participants may lose interest in the study and therefore decline their participation, especially in long-term cohort studies. Below is a list of the potential reasons for attrition in the current project and mitigation strategies.

Data analysis

All in sillico data processing and analysis will be performed on the IBU Linux Cluster for Rocky Linux 8 from the servers of the University of Bern on a local machine running Ubuntu 20.04.06 LTS (32GB, Intel^® Core™ i7-9750H CPU @ 2.60GHz × 12) and within R (4.4.1 2024-06-14) package environments in RStudio (2024.04.2). Sequence reads obtained for each sample were processed for quality using FastQC (v0.11.9) and MultiQC (version 1.11). Further quality checks, read trimming, chimera removal, denoising, and taxonomy assignment (using a trained Naive Bayes kmer classifier) will be performed directly from within the QIIME2 workflow (q2cli version 2024.5.0) using the included DADA2 software. Standard core metrics such as alpha diversity indices (Faith’s richness, Simpson’s evenness, and Shannon index) and beta diversity (Bray-Curtis, weighted UniFrac for the quantitative metrics; jaccard and unweighted UniFrac for the qualitative ones; Aitchison) will be computed with QIIME2, then explored in Qiime2View, and further processed and compared with Phyloseq and qiime2R packages in R. We will use the Emperor biplot plugin, Principal Coordinate Analysis (PCoA), and beta-group-significance plugin to test beta diversity significance and characterize the top bacterial genera that explain the microbiota distributions across the samples.

We will seek to identify any correlations between the collected metadata and microbiome composition, and identify any gut/oral microbiota trends associated with nutritional status and food diversity. We decorticate the variations in the observed microbiota composition across different groups, such as weight, height, age, and nutritional status. For alpha diversity comparisons, the groups’ significance was computed and tested using Kruskal-Wallis pairwise tests. We will also use the q2-longitudinal plugin to investigate longitudinal and paired sample comparisons, that is, to determine if or how samples change between the different time points and treatments/nutrients. All postprocessing and statistical analyses will be performed in R.

Weight for height, weight for age, and height for age z-score will be calculated using the WHO Anthroplus software and compared with microbiome composition using R.

Dissemination

The results of the study will be disseminated to the public and academia, including our research team and collaborators, scientists, and the broader research community via participation in scientific conferences, meetings, and publications in peer-reviewed journals.

Study status

The study began in 2024 and will continue until May 2026. As of October 2024, 70/70 (100%) were recruited. Data and samples were collected for the baseline (n=70) and first follow-up periods (3 months; n = 66).

Ethical considerations

The study follows ethical principles involving human subjects as outlined in the declaration of Helsinki. Ethical approval for this study was granted by the Ethics Board of Khyber Medical University, Pakistan ref no DIR/KMU-EB/BR/001-03 dated 11/01/2024.

Consent statement

In the local (Pashtun) culture, it is not possible to directly contact a potential female study participant without prior consent from the husband/male head of the household. Therefore, at each study site, an initial information session was held with the male parents/guardians with the help of elected local representatives and elders of the community. During the session, complete information about the study objectives, eligibility criteria, and data collection process was presented and explained by the team leader in the local (Pashto) language. A participant information sheet containing all the study details and procedures in an easy-to-understand, local (Pashto), and Urdu (national language of Pakistan) were also provided to the parents during the session. All queries were also answered. Following the session, eligible parents/guardians were invited by trained research assistants to enrol their babies in the study. Once they agreed, the parents were asked to sign a written informed consent form in their preferred language.