Study design and study population: This was a cross-sectional study that explored the knowledge and practices of hygiene among SIYP in southwest, Nigeria. A community engagement program (CEP) on WASH was implemented in December 2018, to access the SIYP community and foster rapport between the research team, field workers and study participants. This approach was adopted to facilitate and retain participation of SIYP for a one-year project that commenced after the CEP program. The WASH CEP consisted of health talks on personal hygiene (hand-washing, sanitation, oral hygiene, and menstrual hygiene) and street hygiene. Data were collected from recruited SIYP before the WASH CEP began. Study participants were adolescents and young people, aged 10 to 24 years, who met the criteria for being SIYP (those who spend most of their time on the street and return home at night, or who continually live and sleep on the street).
Sample size: Sample size determination for the project was guided by Turner , who recommended estimates to derive sample size for surveys on orphaned and vulnerable children (OVC) in homeless situations. Considering the unavailability of data to generate prevalence rate of OVC in the proposed study environment, the suggested minimum sample size of 800 to 1000 for this study. A total of 845 (452 males and 393 females) SIYP were interviewed during the WASH CEP.
Study Instruments: The WASH questionnaire had sections on socio-demographic characteristics; personal hygiene (collection and storage, handwashing, sanitation and waste disposal); knowledge about hygienic practices; and oral hygiene. Each section was adopted from validated sources. The tool used to collect information on personal hygiene was adapted from the UNHCR 2018 WASH knowledge, attitude and practice survey questionnaire . The tool used to collect data on oral hygiene has been used in past studies in Nigeria . A national WASH consultant who is a community health expert in sanitarian and public health undertook the judgment quantification of the instruments used for the study. The judgement quantification is a part of the instrument content validation process . A pilot was then conducted to test the questions for ease of understanding and language clarity. The survey took place between the first and third weeks of December 2018.
Socio-demographic data collected were age, sex, marital status, school attendance and state of residence. Age was designated as age at last birthday and was stratified into three groups: 10-14 years, 15-19 years, and 20-24 years. Marital status was designated either married or not married; those living with their spouse were considered married. History of school attendance was designated either ‘yes’ or ‘no;’ a ‘yes’ response indicated that the person had had some level of formal education.
Information on oral hygiene practices was collected with a 5-item questionnaire that sought information on the frequency of tooth cleaning, use of fluoride-containing toothpaste, use of dental floss, consumption of sugar between meals, and frequency of dental check-ups. The responses to the questions ranged from four to seven alternatives. The acceptable responses were brushing more than once a day, use of fluoridated toothpaste always or almost always, flossing at least once a day, eating sugary snacks between main meals less frequently than once a day, and attending a dental check-up within the last year. Only three of the five items were used to measure good oral hygiene practice. The expected practices were: brushing teeth more than once a day, used fluoridated toothpaste always or almost always, and ate sugary snacks between main meals less often than once a day . The responses were given a score of 1 and otherwise 0. A possible score range of oral hygiene practices was 0-3. Respondents were considered to have good oral hygiene practice  if they had 2 to 3 of these oral health practices: Those with 0 to 1 practice were considered to have poor oral hygiene practice.
Information on oral hygiene knowledge: The questions on oral hygiene practices were adapted to inquire about respondents’ oral hygiene knowledge. Respondents were asked about their knowledge of the frequency of tooth cleaning, use of fluoride-containing toothpaste, use of dental floss, consumption of sugar between meals, and frequency of dental check-ups. Correct responses to the questions were assigned ‘1’ if positive and ‘0’ if otherwise. After that, an index was computed for the variables used to measure oral hygiene knowledge with a possible score ranging from 0 to 5. The mean score cut-off point was 2. Respondents who answered yes to 3 and more of the 5 questions reflected good oral hygiene knowledge. Responses fewer than these three correct responses were categorized as poor oral hygiene knowledge . The questions on oral hygiene knowledge were asked after the questions on oral hygiene practices.
Information on personal hygiene: Questions on personal hygiene consisted of four: water collection and storage, handwashing, sanitation and waste disposal. The section on water collection and storage had 5-item, closed-ended questions that explored responses about source of drinking water (improved and non-improved); and water storage (if had containers for collecting and storing water; if the container for collecting and storing water is protected; frequency of cleaning drinking and storage water containers and how the containers were cleaned). Only those who were positive to have good ‘water collection’ and ‘storage practice’ were assigned ‘1’ indicating good water collection and storage practice, while others were assigned ‘0’.
For handwashing practices, a 4-item close-ended questionnaire was used to elicit information on washing hand with soap, washing hands before eating, washing hands after toilet, and washing hands soon after getting home. Responses to the questions on handwashing practices were coded as no - ‘0’and yes – ‘1’. The possible score range was 0 to 4. Only respondents who answered positive to the four questions were assumed to have good handwashing practices.
A 4-item questionnaire was used to elicit information on sanitation. It obtained information on the usual place for defecation (improved facility and non-improved facility); access to handwashing facility at a place of defecation and access to water and soap to wash hands after defecation. The responses to the questions on sanitation were coded as ‘0’ – no and ‘1’ – yes and were thereafter merged. Score for good sanitation practice ranged from 0 to 4. Any score less than 4 was categorized as poor sanitation.
A 3-item questionnaire was used to elicit information on the management of waste. It inquired about methods for disposal of waste and the use of bare hands to collect waste. The three questions were scored to arrive at a composite score for good waste disposal practice. The score for good waste disposal was 3. Any score less than 3 was categorized as poor waste management.
Study procedures: The study participants were recruited through respondent-driven sampling . The first seeds for the respondent-driven sampling were identified in areas where SIYP cluster. Clusters were identified through mapping conducted by the research team along with officials of the State Ministry of Health. Two high-volume clusters each in Lagos and Osun States were identified: the cluster sites in Lagos State were Bariga and Ajah; the sites in Osun State were Oke-Baale and Olaiya/Sabo.
Two cards with identification numbers were given to the first ten seeds in these cluster areas after they had been interviewed to recruit friends or peers, which some SIYP accomplished. The recruitment effort of seeds was complemented by direct recruitment by the research team, who also visited the cluster areas.
All recruited study participants gave consent before administration of the questionnaire by research team members. The questionnaire was administered in a place the respondent felt most comfortable to respond to questions. At the end of the interview, participants were given lottery tickets that enabled them to enter a raffle draw during a sideshow at the end of the survey. The values of the draw prize were $0.40 to $2.25.
Variables: The outcome variable for this study was ‘oral hygiene practices,’ measured as either good or poor. The explanatory variables were good or poor water collection and storage, handwashing, sanitation and waste disposal practices. Other explanatory variables included background information of respondents: sex, age, state of residence, school attendance, and marital status.
Data analysis: Data were collected electronically with open data kit, an online/offline platform for the collection and management of data. After that, data were downloaded from the online server for analysis purposes. The statistical software STATA 15.1 was used for data analysis. Univariate analysis was conducted to describe the proportion of respondents for the outcome, explanatory and background variables. Bivariate analysis was conducted with Pearson chi-square (or Fisher’s exact t-test where appropriate) to determine the associations between the outcome variables and the explanatory and background information variables. Inferential analysis was conducted to identify risk indicators of good oral hygiene. Binary logistic regression, guided by two models, was conducted. The first model tested the association between the outcome and explanatory variables, whereas the second model tested the associations between outcome and explanatory variables after adjusting for confounders (background information). Statistical significance was considered a p-value less or equal to 0.05.