A quantitative, cross-sectional study was conducted among asymptomatic food handlers at HU from August 2015 to January 2016. HU is located in East Hararghe zone at a distance of 510 kms from Addis Ababa, 17 kms from Harar town, and almost 5 kms off the main road from the nearby Haramaya town. The University comprises three campuses: Main, Harar and Chiro campus and consists of fifteen cafeterias (9 in Main, 4 in Harar and 2 in Chiro campus). During the study period, 1274 individuals were serving as food handlers throughout HU cafeterias.

Study population consisted of asymptomatic food handlers working at HU cafeterias.

All asymptomatic food handlers who had a direct contact with foods and drinks were included in this study. Those participants who took antiparasitic drugs in the last three weeks or during data collection were excluded.

The sample size was calculated using single population proportion formula considering 95% confidence level (CL), 5% margin of error, and a 52.4% prevalence of IPIs [9]. An initial sample size was 384. After considering 10% non-response, the final sample size was determined to be 422.

Food handlers were stratified into two strata based on the ownership of cafeteria (HU and private cafeteria). Food handlers working at HU cafeterias were further stratified by their working place (Main, Harar and Chiro campus). Food handlers from private cafeterias were also stratified similarly. Then, the sample size was proportionally allocated based on the size of food handlers working in each cafeteria. Finally, a sample size of 422 from a total population size of 1274 was selected using systematic random sampling technique (Fig. 1). A payroll of food handlers was used for sampling frame.

Data related to socio-economic factors, food handling practices and other related factors were collected using a pretested structured questionnaire administered by trained nurses. The questionnaire was adopted from the World Health Organization (WHO) food safety checklist and literatures [4, 10, 11]. Data on fingernail trimming was recorded by a simple observation. After interviewing and observation, respondents were asked to bring a small amount of fresh stool sample (2–3 g/pea-size if formed, or 4 ml if watery diarrhea) into a clean, tight-lid sample container after orienting on how to collect the stool specimen by attending laboratory technologist. All specimens were immediately transported using cold box to the Parasitology Laboratory of the College of Health and Medical Sciences of HU for analysis.

Each stool sample was smeared and tested in triplicate. A saline and Lugol's direct smears were performed by emulsifying small amount of stool (0.25 mg) uniformly in a drop of normal saline (0.85% NaCl) on one end of a glass slide, and Lugol's iodine on the other end of the same slide. A saline direct stool smear was used for detection of motility of protozoan trophozoites, which were seen in semisolid/diarrheic specimens. Iodine direct smears showed the characteristic feature of the diagnostic stages in more details. Smear prepared from a sedimented stool was also examined to detect parasites that where too low to be seen in the direct wet smear and exclude false negative result in case of light infections, thus increasing the sensitivity of copromicroscopic techniques. Sedimentation technique was performed using Ridley modified formol ether sedimentation technique. In brief, 1 g of stool was placed in 15 ml conical tube containing 7 ml of 10% formol water, gently emulsified and sieved using 65 mm plastic strainer. The sieved sample was transferred into another conical tube containing 9 ml of 10% formol water and 3 ml of ethyl acetate and centrifuged at 3000 rpm. The supernatant was decanted by inverting the tube and the last drop allowed to sediment by gravity for 15 min. The sediment was then poured on slide, covered with cover glass (22 mm × 22 mm) and examined microscopically under low (10×) with the condenser iris closed sufficiently to give good contrast, and high (40×) objective lenses. Eggs and larvae of helminths, and cysts and trophozoites of protozoan were assessed by propagule size, shape, cell wall width and distinctive internal characteristics [12].

Asymptomatic infections defined by the absence of acute gastrointestinal symptoms such as abdominal discomfort, vomiting, nausea and diarrhea [13].

Food handler is a person performing under contractual agreement or permanent employee who handles, prepares, serves, or sells food and drink, or who comes in contact with eating or cooking utensils or other equipment used in the handling, preparation, service, or sale of food.

The questionnaire was initially prepared in English and translated into the local language (Amharic and Afan Oromo) by language experts and translated back into English by another expert and pretested on 5% asymptomatic food handlers working in Dire Dawa University, Eastern Ethiopia to ensure its consistency. The reliability of the questionnaire was validated using Cronbach's alpha, and the result was 0.80, indicating a high level of internal consistency [14]. Data collectors and supervisors were trained for two days on method of data collection, specimen collection and examination techniques. The stool specimen examination was conducted in triplicate by trained Medical Parasitologists. The final result was determined through inter-examiners agreement. Moreover, investigators were involved in the decision in case of disagreement. Completion, accuracy and clarity of the collected data were checked every day.

Data were double entered into epidemiological information software (EPI Info™ version 3.5.1) to ensure accuracy of the data. The data were cleaned and exported to statistical package for social sciences (SPSS) software version 20.0 (SPSS Inc., Chicago, IL) for further analysis. Descriptive statistics were calculated for all variables. Bivariate and multivariable logistic regression models were used to assess the association between independent variables and IPIs. Variables that had a p-value < 0.25 in bivariate analysis were run in multivariable logistic regression at 95% CL to determine independent predictors of the outcome. A p-value < 0.05 was considered to indicate statistical significance association.

This study included 417 asymptomatic food handlers (331 females and 86 males) with a response rate of 98.9%. The mean age of participants was 36.1 ± 8.7 standard deviation. Large proportion (39.3%) of food handlers were found in the age group of 31–40 years and had formal education (76.3%). Nearly half (47%) of the participants had more than 10 years of work experiences (Table 1).

Of 417 stool specimens, 25.2% (95% CI: 18.3, 29.6) were found to be positive for one or more parasite species: comprising protozoa (61%) and helminths (39%). The most prevalent parasite was E. histolytica/ dispar (46.7%) followed by A. lumbricoides (14.3%) and then G. lamblia (13.3%). Less frequent identified intestinal parasite spp. were T. trichuria, Schistosoma mansoni and Dientamoeba fragilis; accounted for 2.9% of the total isolates (Table 2). Mixed intestinal parasites (E. histolotica/dispar, G. lamblia and H. nana) were detected in 2.6% of the participants.

Among the total variables included in the bivariate logistic analysis, ten variables (age, educational status, years of service, monthly income, hand washing after touching body parts, use of the apron when cooking, hand washing after the use of the toilet, fingernail trimming, hand washing before eating foods and medical check-up) were associated with IPIs at a p-value < 0.25. These variables were then computed in a multivariable logistic regression analysis. The results of multivariable logistic analysis showed that not attending formal education [AOR: 2.13, 95% CI: 1.24, 3.67], earning of a monthly income of less than 45.7 USD [AOR: 3.86, 95% CI: 1.62, 9.20], lack of hand washing after the use of the toilet with soap [AOR: 2.43, 95% CI: 1.22, 4.86] and untrimmed fingernails [AOR: 3.31, 95% CI: 1.99, 5.49] were significantly associated with IPIs (Table 3).