Study area description and period

The study was carried out in nineteen rural kebeles (the lowest administrative units in Ethiopia) of Aneded District, northwest Ethiopia, from March 1 to May 31, 2021. The district has three urban kebeles. In 2019/2020, the district had a projected population of 110,183 (53,989 males and 56,194 females) residents [31] and 4731 HHs. It had 4 health centers and 20 health posts. The district has two temperature-based agro-ecological zones: midlands (most of the kebeles) and lowlands. The district is largely characterized by good socioeconomic status, adequate and improved water supply (mainly public standpipes), good sanitation coverage, and moderate personal hygiene practice.

Study design and eligibility criteria

A community-based matched case–control study was used. Individual matching by sex and category matching by age (below 15, 15–49, and above 49 years) for cases and controls were employed. All individuals aged 1 year and above who were permanent residents (had lived for at least 6 months) in the rural kebeles of the district were included. Individuals who had other dermal illnesses or severe diseases except scabies were excluded from this study.

Sample size calculation

The sample size was estimated using a matched case–control study design calculation formula (32) by assuming a 95% confidence level, 90% power to observe exposure variability, an odds ratio (OR) of 1.75, a 1:3 case–control ratio, and considering that 36.3% of the control HHs had water accessibility for personal hygiene [33, 34]. In addition, a 10% nonresponse rate was considered for both the estimated sample size and the nonrespondent sample and computed using the formula (final sample size = \(n/110\%\)). As a result, a matched sample size of 190 pairs (190 cases and 570 controls) could be enrolled in the study.

Study population and sampling technique

A two-stage sampling technique was used to select the study subjects. In the first stage, five rural kebeles, including Enaskay, Mislewash, Wonganefasam, Daget, and Yewush (30% of the total rural kebeles), were included using the lottery method from the nineteen rural kebeles of the district. In the second stage, a house-to-house census/visit was performed to screen individuals having scabies infestation in the five kebeles. In each HH, all members were screened for scabies status. Then, sampling frames were created during house-to-house visit in each kebele. Finally, a proportional sample of matched controls (matched by sex and age) was taken from each kebele on the basis of the number of scabies cases. For selection of matched controls, simple random sampling was employed. Controls were selected from the same kebele where the cases were lived. When there was more than one scabies case or control in an HH, only one was selected randomly. Study subjects (scabies cases and controls) were included in the study based on the WHO case definition for scabies [35, 36].

Data collection and quality assurance

A consensus diagnostic criterion using the Delphi method was used to confirm scabies infestation in the study population. The Delphi method is the standardized and accepted diagnostic method for scabies [36]. Cases were individuals who had clinical signs and symptoms of scabies, with at least scabies burrows, typical lesions affecting male genitalia, or typical lesions in a typical distribution and two history features (Level B) [35, 36]. Controls were individuals who had no clinical signs or symptoms of scabies. Experienced public health professionals who were certified by the Ethiopian Federal Ministry of Health confirmed whether the individual had scabies using the Delphi method. One-day training was given for the public health professionals for the diagnosis of scabies.

A pretested structured questionnaire was used to collect sociodemographic, socioeconomic, behavioral, water supply, sanitation, hygiene/WaSH related, and health service coverage characteristics of the study participants. Face-to-face interviews and/or direct observations were used to gather data.

Questionnaire-based data were collected by Environmental Health professionals with a Bachelor of Science degree. Three-day training was given by the lead researcher for data collectors about house-to-house data collection. The questionnaire was pretested on 5% of the sample size in a rural kebele in Gozamin district. Strict follow-up was applied to optimize the quality of data through trained supervisors. If any scabies case or control was not present at the first visit in house-to-house data collection, the HH was revisited for the second time. House-to-house data collectors were blinded to the scabies status of the study subjects. The questionnaire was first developed in English and translated into the Amharic language, and only the open responses from the study subjects were translated to English by independent language professionals.

Variable measurements

Variables included in the study were defined as follows: 1) scabies infestation is the presence of itching with typical lesions on hands, interdigital, and/or genitalia and/or itching and close contact with an individual who had itching or typical lesions in a typical distribution [36]; 2) when an individual did not take a shower at least once in a week, it was defined as infrequent bathing; 3) when a person did not wash his/her clothes at least once in a week, the variable was measured as infrequent cloth washing; 4) when a person did not change his/her clothes at least once in a week, infrequent cloth changing was used as a definition; 5) daily water consumption of ≥ 20 L/capita/day (L/C/D) was considered as adequate water quantity; 6) disposal of HH solid wastes by using sacks, disposal pit, or composting in the compound and liquid wastes by using septic tank or soak pit in the compound was defined as proper waste management; and 7) receiving scabies-specific interventions was defined as receiving health interventions, including mass therapy, health education, training, and ongoing surveillance.

The HH wealth index for rural Ethiopia was constructed from several binary and ordinal variables adapted from the literature [37] and the Ethiopian Demographic and Health Survey [38]. Variables, including longer-run and shorter-run HH assets, latrine availability and type, housing characteristics, farmland size, ownership and number of domestic animals, and beekeeping, were used to compose the wealth index. Variables had values that ranged from 0 for the lowest to 1 for the highest value.

Data analysis

Data entry, coding, cleaning and verification were performed in EpiData Version 3.1 software (EpiData Association, Odense, Denmark). However, data analysis was performed in SPSS statistical software version 24 (IBM SPSS Statistics for Windows; NY, USA) and STATA version 14.0 (Statistical Software: College Station, TX 77845, USA). The median with IQR (interquartile range) was computed for continuous data. Frequencies and proportions were calculated for categorical data.

Principal component analysis (PCA) was run to describe HH wealth [37]. Principal component 1 was considered to rate the HHs into three categories. Cut-off values were determined for the three ranks. The original variables were standardized. Standardized variables were multiplied by principal component 1 and resulted in third variables. Ultimately, a composite variable, the HH wealth score, was derived by summing the third variables. By applying the cut-off values of principal component 1, a variable, the HH wealth index with three categories, was defined. The variable was categorized into tertiles, with the lower tertile indicating lower wealth.

Bivariable and multivariable conditional logistic regression models were fitted to detect predictors of scabies infestation. A bivariable matched analysis was performed, and variables having P < 0.25 along with variables having scientific meaning were included in the multivariable matched analysis. Adjusted matched odds ratios (AmORs) with 95% confidence intervals (CIs) and P-values were reported for the multivariable matched analysis. Variables at P < 0.05 were declared statistically significant predictors of scabies infestation.