The study was conducted in public secondary schools of Harar town, Ethiopia. According to the 2007 population projection by the Ethiopian Central Statistical Agency (CSA), the estimated total population of Harar town was 122,000. Of this, the adolescent population is estimated to be 99,368 (49,727 male adolescents, 49,641 female adolescents) (12). According to the Harar Town Education Bureau, the town has seven public secondary schools covering grades 9–12. The study was conducted in four of these schools: Aboker, Harar Senior, Shekib, and Jegnoch high schools. The study was conducted between 31 August and 19 September 2023.

The study employed an institutional-based cross-sectional study design. The source population consisted of all secondary school adolescents aged 14–19 years, including both male and female adolescents. Adolescents aged younger than 14 years were excluded from the study because they were less likely to fit the study’s variables, such as sexual behavior.

The sample size was determined using a single population proportion formula, [(Za/2)2(p)(1−p)]/d2, with the following assumptions, as there was no previous related evidence: 95% confidence interval (CI), 4% margin of error (d), 5% non-response rate, and 50% of the proportion (p) of SRHL among high school adolescents. The calculated sample size was 600. Considering 1.5 for the design effect and a 5% non-response rate, the final sample size was 945.

Simple random sampling was used to select the high schools, whereas a multistage random sampling technique was employed to select a representative sample of students. The number of study participants from each school, grade, and section (grades 9–12) was calculated proportionally to ensure representative allocation. The number of study participants from each school was calculated proportionally using the equation nf = (no × n1)/N, where nf is the number of participants from each school, no is the calculated sample size, n1 is the total number of students in each school, and N is the total number of students from the four (sampled) schools. On the other hand, the number of participants from each grade (section) was calculated using the equation mf = (mo × m1)/M, where mf is the number of study participants from each grade (section) (grades 9–12), mo is the number of study participants from each school, m1 is the total number of students in each grade, and mf is the total number of students in each school. Then, the required number of study participants was selected using simple random sampling from each section, with the sample size allocated to each section determined by Probability Proportional to Size (PPS). Hence, after the study participants were allocated to each section using PPS, simple random sampling methods were applied to recruit the required number of participants from each school and section (Figure 1). Of 945 students who were requested to bring their guardians’ signed informed consent, 8 students did not. A discussion with the remaining 937 students was held regarding the overall study and voluntary participation. Students were given the opportunity to raise any issues during the discussion and were also invited to questions before, during, and after data collection. After the discussion, data collection from the 937 students began immediately. Of the students, 909 completed the questionnaire while 28 returned incomplete questionnaires and were excluded from the analysis.

Pretested, validated, self-administered, and structured questionnaires were used to collect the data. The sexual characteristics and literacy tools were adapted from previously published literature and contextualized to local realities (13, 14). The tools were closed, open-ended, and had three parts: sociodemographic characteristics, sexual and reproductive characteristics, and SRHL. The SRHL part has 31 questions containing a 5-point Likert scale (very difficult, difficult, medium, easy, and very easy), with minimum and maximum scores of 31 and 155 points, respectively. It contains accessibility and ability to understand and identify appropriate reproductive health information concepts. In addition, it concerned reproductive health, including sexually transmitted infection, unwanted pregnancy, and family planning. The Likert scales were in the range of 0–4. Each question had a scoring scale reflecting the ability to access, understand, appraise, and apply SRH information from the scale “very difficult” = 0 to “very easy” = 4. The questionnaire was originally developed in English, translated into local languages (Afan Oromo and Amharic), and then translated back into English by language experts to check for language consistency and accuracy. The Amharic and Afan Oromo questionnaires were then used to collect the data. Participation was fully voluntary and oral consent was obtained from the participants as well as written informed consent from their guardians.

In this study, in addition to the Likert scale, both categorical and continuous variables were used. All 31 SRHL questions were based on the Likert scale while age was the only continuous variable. On the other hand, the following categorical variables were used: residence, religion, language, educational level, with whom the adolescents were living, parents’ occupation, access to Internet, sources of SRH, attending SRH at school, ever used SRH services, ever counseled on SRH, ever learnt about puberty, ever experienced sexual intercourse, ever used contraceptives, know the cause of cervical cancer, know the availability of the HPV vaccine, know where HIV/AIDS tests can be offered, ever get tested for HIV/AIDS, and know where to receive SRH service. All types of variables were collected.

The dependent variable was reproductive health literacy, which was computed from 31 SRHL questions, whereas social demographic characteristics, including age, sex, marital status, source of reproductive health information (Internet, media use, family and peer influences, mass media, education level), and health status of the adolescents (including current health status, doing physical exercise, having disability, having chronic illness, affordability for healthcare, and alcohol consumption) were independent variables.

A validated structured data collection tool was used to collect the data. The questionnaire was originally developed in English. It was then translated into local languages and subsequently translated back into English to check the appropriateness of translation. The Amharic questionnaire (translated from English) was used to collect the data. Four data collectors and two supervisors were allocated. Two days of training were given to the data collectors and supervisors on the objectives of the study, the contents of the data collection tools and procedures, and the participants’ autonomy. The questionnaires were pretested on 5% of the sample in Abadir High School before the actual data collection period to check the reliability and validity of the data collection tools, and question sequences were amended before the final data collection. The high school used for the pretest was excluded from the present study. If there was any ambiguity, the data collectors provided clarification during the data collection. The data collectors, supervisors, and principal investigators reviewed and checked the completed questionnaires for completeness, accuracy, and consistency on a daily basis. From the calculated sample size, 36 questionnaires were incomplete and excluded from the analysis.

The collected data were entered into Epi-Data version 4.6.02 and then exported to SPSS version 25 (IBM Corp.). Cronbach's alpha was calculated as 0.965, indicating the excellent internal consistency of the questionnaire. The data cleaning, completeness, and outer layers were checked. A descriptive statistical analysis was performed, and the outputs were summarized as proportions and frequency, whereas the literacy levels were determined from a total score that was standardized from 100. Scores in the range of 0–50 were considered inadequate SRHL, 50.1–66 slightly adequate SRHL, 66.1–84 adequate SRHL, and 84.1–100 excellent SRH. Inadequate and slightly adequate were considered limited SRHL, while adequate and excellent were considered optimum SRHL. Literacy status was then determined by two categories: limited and optimal SRH. In addition, once the relationship between independent and dependent variables were established, bivariate analysis or simple linear regression analysis was performed. Variables with a p-value >0.25 in simple linear regression were added to the multivariate linear regression model. Significance was determined at a maximum p-value of 0.05 and a 95% CI.

In this study, 412 (45.3%) boys and 497 (54.7%) girls responded to the questionnaires, giving a response rate of 96%. The remaining questionnaires were excluded from the analysis because they were incomplete. The mean ± SD age of the participants was 16.8 ± 0.04 years. A total of 548 (60.3%) respondents were aged 15–17 years and 60% of the adolescents were Muslim. Regarding educational status, 390 (42.9%) were in the ninth grade. Of the adolescents, 81% reported that they lived with their parents. Moreover, nearly half of respondents (430, 47.3%) had a family member working in health-related areas. Internet access was available for 625 (68.8%) adolescents (Table 1).

Table 2 shows the results for sexual and reproductive health among adolescents and related knowledge and behaviors. Most adolescents (31%) preferred to receive SRH information from health professionals. Of the adolescents, 64% reported attending school-based SRH education, and one-quarter of the adolescents thought that there should be additional SRH classes. Only 20% of the adolescents reported utilizing SRH services throughout their life. Among 909 adolescents, 370 (40%) had received counseling for SRH issues. Regarding sexual activity, 128 (14.08%) adolescents had prior sexual intercourse and 98 (10.8%) had used contraception.

Nearly half of the school-aged adolescents (420, 46.2%) did not know about the causes of cervical cancer and 497 (45.3%) did not know about vaccines against HPV. Only 262 (28.8%) adolescents had ever been tested for HIV. Of the adolescents, 60% agreed that first sex can lead to pregnancy. A total of 408 (45%) adolescents responded that having sex halfway between periods was most likely to lead to pregnancy. More than half the adolescents (531, 58.4%) responded that using a condom was not safe during sexual intercourse. The majority of the respondents (568, 62.5%) were aware of where to find SRH information (Table 3).

Approximately one-third (288, 31.7%) of the adolescents perceived their health status as worse than the majority when they compared it with those their peers. Within the past 6 months, 277 (30.5%) respondents had faced health problems. The majority of the adolescents (391, 43%) had not done any form of physical exercise >30 min, and 81% had not consumed alcohol within the past 12 months. Of the adolescents, 8% and 7% were disabled and had a chronic disease, respectively. The ability to pay for their own healthcare was found to be very difficult for 251 (27.6%) adolescents (Table 4).

The mean SRHL score was 55.7 ± 0.58, whereas the minimum and maximum SRH literacy scores were 19.4 and 96.8, respectively. Among the four measurements of SRHL, 362 (38.9%) participants had slightly adequate literacy, whereas only 57 (6.3%) had excellent SRHL (Figure 2). According to the final SRH measurements, 677 (74.5%) participants had limited SRHL and only 25% had optimal SRHL (Table 5).

Multiple linear regression was fit after the following assumptions had been checked: multicollinearity (variance inflection factor < 3), autocorrelation (Watson Durbin test = 1.99), linearity (indicated by a scatter plot), and normality (indicated by a histogram). Accordingly, with healthcare workers being the preferred sources of SRH information, SRHL increased by 6.42 (range 1.61–11.22), while if the source of SRH information was books then SRHL would increase by 6.57 (range 0.85–12.29). However, if the Internet was the source of SRH information, SRHL would decrease by 10.26 (range −20.01 to −0.5). On the other hand, if adolescents could easily pay for their healthcare, their SRHL level would increase 13.76 times (range 8.21–19.32) (Table 6).

SRH literacy: 0–50 = inadequate, 50.1–66 = slightly adequate, 66.1–84 = adequate, and 84.1–100 excellent. Limited SRH = the combination of inadequate and slightly adequate (0–66). Optimal SRH = the combination of adequate and excellent SRH literacy (66.1–100) (15).

Sexual health literacy: the ability to understand and apply information regarding sexual health, decreasing the risk of STIs, and providing various benefits beyond health.