This randomized, single-blinded clinical trial was conducted at Burie and Debre Elias towns’ primary schools, Northwest Ethiopia from March to May, 2019 and included school-aged children aged 6 to 14 years. The study was approved by the Ethical and Review Committee of School of Biomedical and Laboratory Sciences, College of Medicine and Health Sciences, University of Gondar, Ethiopia. This trial is retrospectively registered in www.pactr.org, number PACTR201911466695052 on November 26, 2019.
Prior to participant enrolment, all parents/ legal guardians’ of the children were informed about the objective, purpose, study procedures, and the potential risk and benefits of participating in the study. Parents/ legal guardians who agreed their child to be included in the study were asked to sign a written informed consent. Verbal assent was also sought from each participant. Moreover, for those parents/legal guardians who were unable to read and write, they were asked to give thumbprint after having been read the full informed consent form by a data collector.
Intervention, trial medication, and outcome measures
This randomized trial was carried out in two treatment arms: (i) single dose (500mg) and (ii) a multiple dose mebendazole (100 mg twice a day for three consecutive days). The multiple dose of mebendazole (WORMIN tab) was commercially obtained from private pharmacy in the local market, while the single dose mebendazole (Vermox®) was provided by the local coordinator office of the deworming program. The CR against Hookworm and ERR for determining the changes in infection intensity were the main outcome measures after 14-21 days following dosing.
Eligibility criteria and Sample size
Eligible for inclusion were all Hookworm-positive children with a signed informed consent and who did not have additional health problems (based on medical history, physical examination, vital signs) other than Hookworm were randomized and allocated to one treatment arm. The following exclusion criteria were also applied: children who received any form of anthelminthic treatment within the past 30 days, had diarrhea at the time of the first sampling, had a hemoglobin level <8g/dl, experienced severe concurrent medical condition, had any known history of allergic reaction to mebendazole, and infected with other parasitic infection.
The desired sample size was determined by using WHO guideline(21) with the following assumptions: the local prevalence of Hookworm infections was assumed to be 50% due to lack of recent finding and a total of 100 Hookworm infected individual (50 in each of the two treatment arms) would be needed to detect differences in the CR following different treatment arms for the cure of Hookworm infections with 80% power using a 2-sided statistical test with alpha-level of 0.05. Moreover, by considering the potential loss to follow-up; 20% as a non-response rate was added. Finally, 300 school aged children was screened for Hookworm infections microscopically to get the minimum required sample size.
Data collection and Laboratory procedures
Study participants had responded to a short questionnaire investigating demographic and other health-related issues using the WHO drug efficacy assessment from. A specific identification number was given to each participant. Then, each participant received a sterile stool container labeled with his/her unique identification number and was asked to provide approximately 10mg of fresh stool. All children were well informed to avoid any contamination of the sample. Following this, samples were immediately transported to the nearby health center laboratory in Debre Elias and Burie hospital laboratory.
The McMaster concentration technique, which is the standard reference method for evaluating drug efficacy in Veterinary Parasitology and has recently been evaluated for human helminthes, was used for this study(21,22). Laboratory quality control was performed through re-reading 10% of the slides of each laboratory technician by an expert microscopist. Only one stool sample was collected from each participant in both baseline and follow-up survey. To ensure the objectivity and avoid the risk of prejudgment on the treatment response, the laboratory technicians were blinded to the dose allocation, the hypothesis and objective of the study.
Upon completion of all the baseline parasitological and participant information survey, all children who were found to be positive for Hookworm infection were subjected to a physical and clinical examination by senior health officer. Height was measured with a standard meter (to the nearest 0.1 cm), and weight with an electronic balance (to the nearest 0.1 kg). Haemoglobin levels were measured in capillary blood using the finger-prick method (HemoCue®301).
Finally, eligible Hookworm positive children were randomly assigned either to the single (500mg) or multiple dose regimen of mebendazole (100 mg twice a day for three consecutive days) arms of the study with 1:1 ratio. Randomized of the children were not based on their age, sex or by any means of personal parameter before randomization. Using simple randomization lottery techniques, 55 eligible children were categorized into the single and 54 into multiple dose mebendazole arms.
A slice of biscuit was given to each eligible child before drug administration. Participants who were randomized in single dose arm were asked to take the drug in front of their parents and a public health officer at school. After administering the drug, children were monitored for 3 to 4 hours for the occurrence of any vomiting and other adverse events following treatment. In the case of children who were randomized into multiple dose arm; parents/guardians were asked to take home the remaining tablets in a sealed envelope and were instructed on how to administer the drugs. They were instructed to give the drug twice a day (every morning and evening for three days), avoid skipping/doubling any dose, follow strictly their child up to the end of treatment and reminded not to drink alcohol. Participants/parents/guardians were requested to report any medical discomfort following treatment to the investigators or the nearby health extension worker.
Follow-up data collection
Each treated children were revisited after 14-21 days of drug administration and asked to provide one stool sample for the second time. At this time point, children were also asked about the occurrence of vomiting and diarrhea following drug administration.
A participant who vomited within 4 hours after drug administration or a participant with diarrhea was excluded for the final analysis. The same laboratory procedures took place at the follow-up. Children who remained infected with Hookworm and other STH were treated with albendazole (400mg) at the end of the study.
Data was entered to Epi-data software to check data completeness and clearance, and then
transferred to SPSS version-23 for statistical analysis. All analyses were performed on a per-protocol basis. Only children with complete data sets were included in the analysis to determine the treatment efficacy. The baseline characteristics of the study participants are summarized using frequencies, mean and standard deviation. Infection intensity with Hookworm were grouped in to light, moderate and heavy infections, according to WHO guideline (23). Cure rate and ERR were used to assess the efficacy of the drug based on the following mathematical calculation. Cure rate was assumed to be the proportion of individual hosts positive for parasites who become parasitologically negative after treatment (24). Whereas egg reduction rate was the arithmetic mean egg count at baseline of the treatment group minus mean egg count at the end of treatment period divided by the mean egg count at the baseline and express it in percent (25).
ERR=100% × 1- arithmetic mean (post-intervention FEC)
Arithmetic mean (pre-intervention FEC)
Confidence intervals for ERR were calculated using bootstrap re-sampling method with 5000 iterations. An independent t-test was used to compare group means, whereas CRs were compared by calculated Odds Ratio (OR) using logistic regression. For all statistical analyses a P-value of 0.05 was considered as the limit for statistical significance.