Zinc is a vital micronutrient in humans and is essential for protein synthesis, cell growth, and differentiation and thus is important for functioning of the immune system [1]. Mild to moderate zinc deficiency is common in several developing countries, including India, because the commonly consumed staple foods have low zinc contents and are rich in phytates, which inhibit the absorption and utilization of zinc [2]. Strong evidence for a causal relationship between zinc deficiency and childhood infections has come from randomized controlled trials of zinc supplementation [3].

This was a community-based, randomized, double-blind, parallel-arm placebo-controlled trial, conducted from 1st January 2011 to 15th January 2012. We included all infants 6-11 months of age residing in Gokulpuri, an urban resettlement colony in North East District of Delhi, India, who were likely to stay till the completion of the study. Gokulpuri has a predominantly migrant population of about 23000, the majority belonging to the middle and lower socioeconomic strata. To achieve the final sample size, additional children were recruited from adjacent area of Gangavihar which has a similar population.

The study was approved by the Institutional Ethical Committee of Maulana Azad Medical College and Associated Hospitals, New Delhi.

We hypothesized that zinc supplementation for 2 weeks will reduce the incidence of ARIs in subsequent months. We excluded childred receiving zinc supplement in the past 3 months, those who were severely malnourished, known immuno-deficient or on steroid therapy, severely ill children requiring hospitalization, and children of families likely to migrate from the study area. A house-to-house survey was done at the beginning of the study to identify and recruit the eligible infants. The study purpose was explained and an informed consent was obtained from parents of all infants before they were recruited. The recruitment was done during first two weeks of January and July followed by subsequent five months of follow-up.

Intervention: The liquid preparations were prepared by Abyss Pharma, Delhi. Each 5 mL of the preparation contained placebo (syrup base) or zinc (20 mg elemental zinc as zinc sulfate). The syrups were of similar color (orange), taste (orange flavored), and consistency, and were packaged in similar bottles. We randomized the treatment allocation by simple randomization using computer generated random numbers (Excel 2010). The bottles were labeled with serial numbers after randomization in the Department of Community Medicine, MAMC, without the knowledge of the field investigator. The field investigator and parents were blinded to the treatment allocation till the end of follow-up period. The mothers received the bottles with labeled serial numbers and names. The field investigator administered the first dose of the intervention at the time of recruitment and advised the mother to give 5 mL of syrup (using standard 5 mL plastic spoon) daily to the infant for the remaining 13 days. Subsequently visits were made on the 7th and the 14th day to ensure compliance. In case the syrup had not been given regularly, a maximum of one week was given to complete the dosages. We collected data for any possible side effects as reported by the caregivers during these visits. To ensure that the child did not receive additional doses of zinc, we provided mothers with identity cards indicating the study title and that the infant was participating in the study. These cards were to be produced whenever the child was taken to any medical practitioner.

Follow-up for ARIs began at the 15th day post- intervention. Each child was followed up fortnightly (± 3) days and the follow-up continued till 5 months after completion of zinc/placebo supplementation. At each follow-up, mother/caregiver was asked about history of ARIs during the previous 15 days. Recovery from an ARI episode was considered when the last day of ARIs was followed by a 72-hour ARI-free period [10]. Subsequent episodes were considered to be new ARIs episodes.

Statistical Analysis: The data were collected and checked for accuracy on a daily basis and entered in SPSS version 16. The incidence was expressed as episodes per child per year. The counts were expressed by means and standard deviation. Difference between means was tested using t-test, for normally distributed data or Mann Whitney U test, for skewed data.

Generalized Estimating Equations (GEE) were used to obtain an incident rate ratio with 95% confidence intervals, in order to compare month-wise number of episodes and duration of ARIs using Poisson log linear distribution, by intention to treat analysis. The exchangeable working correlation matrix was selected for all the outcomes. We included all children who had taken at least two doses of the intervention for the analyses. The follow-up visits for which the infant outcomes were not available were imputed using the worst case (2 episodes of ARI) and best case scenarios (no episodes). As it did not change the study results, the missing data were excluded from the final analysis. We decided to adjust the incident rate ratio for covariates which appeared to be different at baseline in the two groups. We also decided to compare the month-wise mean episodes of ARIs in the two groups.

From a total of 3155 households identified during the house to house survey, we assessed 272 infants for eligibility and all were recruited (Fig. 1). Infants in both the groups shared similar baseline characteristics (Table I). Seven families which migrated during the study period also shared similar baseline characteristics. Final analyses included 134 infants in the zinc group and 124 in the placebo group, who had completed the study. The mean number of follow-ups was 10 in each group. A total of 19 infants (13.5%) in zinc group and 26 infants (20%) in placebo group were given additional one week to complete the intervention as they were found to be initially non-compliant. The parents of infants who refused to be the part of the study even after this period were excluded from the analysis as refusals.

TABLE I Baseline Characteristics of the Study Participants

Fig. 1 Trial flow.

Out of the total 862 episodes observed, 424 episodes occurred in the zinc group and 438 in the placebo group, accounting for an incidence of 7.84 and 8.70 per child- year, respectively, at the end of 5 months (Table II). GEE regression model showed that there was a non-significant reduction of 9% (Adjusted IRR 0.91, 95% CI 0.81-1.02) in episodes of ARIs in the zinc group as compared to the placebo group.

TABLE II	Effect of Zinc Supplementation on Incidence and Duration of ARIs in the Study Participants

When types of ARIs were analyzed separately (Table II), we found a non-significant increase of 1% in the episodes of AURIs (Adjusted IRR 1.01, 95% CI: 0.89-1.14). However a significant decrease of 62% in the episodes of ALRIs (Adjusted IRR 0.38, 95% CI: 0.26-0.36) was observed in the zinc group.

Zinc supplementation led to a significant reduction of 15% (Adjusted RR 0.85, 95% CI: 0.78-0.94) in days with ARIs. There was also a significant reduction of 12% in duration per episode of ARIs (Adjusted RR 0.88, 95% CI: 0.78-0.94) observed in the zinc group (Table II). A subgroup analysis on wasted and stunted infants showed similar effects of prophylactic zinc (Web Table I).

After the second month the episodes were almost similar in the two groups. Zinc prophylaxis significantly reduced the incidence of ALRI for all months of follow-up (Web Table II).

Reported side effects were diarrhea, vomiting and constipation. The percentage of children reporting these were 9%, 10.4% and 1.5%, respectively in the zinc group and 7.3%, 4.8% and 0%, respectively in the placebo group; the difference was non-significant. A death due to diarrhea was reported in the zinc group three months after recruitment. Verbal autopsy revealed severe dehydration due to non-administration of oral rehydration solution or home available fluids.

We observed a significant reduction in duration of all forms of ARIs after 14 days of zinc supplementation (20 mg/day). A significant decrease of 62% in the episodes of ALRIs was observed.

The major limitation of this study is that serum zinc levels were not done to assess the deficiency and the subsequent effect on serum zinc levels. Nevertheless, previous studies in similar populations of Delhi have shown that zinc deficiency is prevalent to the extent of 73.3% for values less than 10.4 µmol/L and 33.8% for values less than 9 µmol/L [15]. Moreover, in our study the proportion of stunted infants was more than 20%, which suggests an elevated risk of zinc deficiency since stunting is a proxy indicator of zinc deficiency in population studies [16]. Thus the results of this study may be extrapolated to similar zinc deficient populations only.

This trial on short course prophylactic zinc supplementation for 2 weeks in infants of 6-11 months has shown to cause a large decrease the incidence and duration of ALRIs in subsequent 5 months. Zinc prophylaxis in zinc deficient populations may significantly decrease morbidity due to ALRIs. The results of this study have important cost- and operational- implications as short-course prophylaxis of zinc in adequate dose might be more feasible than continuous therapy.

Contributors: AM: formulated the research question and contributed to study design, carried out the field investigations and initial data analyses, and wrote the initial draft; DKT: contributed to study design, and methodology and revised the manuscript; ND: supervised the data collection, analyzed the data, and interpreted the findings; KR; supervised the clinical data collection and management, and contributed to methodology. all authors approved the final manuscript.

Funding: Indian Council of Medical Research, Department of Health Research (Ministry of Health and Family Welfare) Government of India. Ref No. 3/2/2011/PG-thesis-MPD-10.

Competing interests: None stated.