The study was conducted in the Hula District, which is one of the 37 districts located in the Sidama Region, situated 96 km south of Hawassa. The Hula District is geographically divided into 2 urban and 17 rural kebeles or wards. According to the data from the 2022 Gregorian calendar, the estimated total population in the district was 103,359, with females accounting for 52,403. Within the female population, approximately 10,246 were teenagers aged 13–19 years. To cater to the healthcare needs of the district, 1 primary hospital, 3 health centers, and 17 health posts are available. In rural areas, health centers and health posts offer primary curative and preventive care, with health workers providing services such as antenatal care, delivery care, and family planning (FP) programs. Similarly, in urban settings, health centers, health posts, and hospitals fulfill the same responsibilities, ensuring access to essential healthcare services. The Hula health office maintains a workforce of 54 health extension workers serving the community.

A community-based cross-sectional study was conducted in the Hula District from 15 February to 15 March 2023. The source population for this study included all female teenagers aged 13–19 years residing in the Hula District. From this source population, a study population was selected by random sampling of the kebeles in the district. The inclusion criteria consisted of teenagers aged 13–19 years who had been living in the selected kebeles for at least 6 months during the study period. Informed consent was obtained from both the teenage girls and their guardians. However, those who were seriously ill or unwilling to participate were excluded from the study.

To determine the sample size, the single population proportion formula was used, with the following assumptions: a 95% confidence level, a 5% margin of error, and an estimated teenage pregnancy proportion of 28.6% (15) based on data in Wogedi District. A 10% non-response rate and a design effect of 1.5 were also considered. As a result, the final sample size was determined to be 518 participants.

The study employed a multistage sampling technique. Initially, all kebeles in the Hula District were stratified into urban and rural areas. Six kebeles were then randomly selected to represent approximately 30% of the teenage population in the district. The estimated number of teenagers aged 13–19 years from these six kebeles was 2,569. The sample size was proportionally allocated to each selected kebele. Health post records were used to create a sampling frame, which included all registered women aged 13–19 years in each selected kebele. The calculated sample sizes of 92, 95, 88, 85, 80, and 78 were allocated proportionally to the respective kebeles: Luda, Gase, Bochessa, Odolakura, Chalbesa, and Hanikomolicha.

To select study participants, a starting point was determined using a lottery method from the sampling frame. Then, simple random sampling techniques were applied within each kebele until the desired sample size was reached. In cases where eligible teenagers were not available in the household, data collectors made repeated visits within the study period, leveraging the labeling and mapping of households that had been done prior to data collection. In addition, when there were multiple eligible teenagers in a household, the interviewee was selected using a lottery method.

Data collection involved face-to-face interviews conducted by six trained nurses (diploma holders) using a structured, interviewer-administered questionnaire and by reading close-ended questions for study participants. The questionnaire was developed based on a comprehensive review of documents, guidelines, and existing literature on similar topics. It covered various aspects such as sociodemographic characteristics, sexual and reproductive characteristics, and individual and peer-level factors. The questionnaire was initially prepared in English, and later translated into the local language, Sidamu Afoo, then back-translated to English to ensure consistency. The translations were performed by experts who were competent in both English and Sidamu Afoo.

Data quality control measures were implemented to ensure the accuracy and reliability of the collected data. A properly designed and pre-tested data collection instrument was used. Data collectors and supervisors received training on data collection procedures, the objectives of the study, and the use of data collection tools. A pre-test was conducted in two kebeles, involving 5% of the sample size, to identify any inconsistencies or ambiguities in the questionnaire. Any identified issues were addressed and incorporated into the final version of the questionnaire. Data completeness was checked daily by the supervisor and weekly by the principal investigator to identify and address any missing values.

The primary study outcome was teenage pregnancy, which was determined based on self-reported pregnancy status and confirmed by health professionals (16). Teenage pregnancy was defined as any pregnancy occurring between the ages of 10 and 19 years. The health professionals typically used structured interviews to gather data on self-reported pregnancies to confirm the pregnancies. Hence, pregnancy history was assessed by inquiring whether respondents had ever been pregnant, encompassing pregnancies resulting in live births, stillbirths, or abortions. Adolescents ranged from 10 to 19 years old, encompassing young adolescents (10–14 years old) and older adolescents (15–19 years old). The independent variables included sociodemographic and economic factors (such as age, residence, religion, education level, income level, marital status, parental education, parent occupation, and parent separation), reproductive characteristics (age at first marriage, premarital sex, early age at first sexual encounter, multiple sexual partners, ever had pregnancy, age at first pregnancy, forced marriage, pregnancy planning and desire, knowledge of contraceptives, awareness of where family planning methods are provided, utilization of family planning methods, knowledge of fertile periods for pregnancy, age at menarche, and communication with parents on sexual issues), and individual and peer-level factors (drug/substance use, number of sexual partners, and peer pressure).

Data processing and analysis involved cleaning, coding, and entry of the collected data into Epi Data version 4.4.1 (Jens, Denmark). The data were then analyzed using SPSS version 25 (IBM, Chicago, IL, USA). Descriptive analysis was performed to summarize the data, and bivariable logistic regression analysis was conducted to identify any statistical associations between each independent variable and the dependent variable. Variables with a p-value <0.25 in the bivariable analysis were considered for inclusion in the multivariable logistic regression analysis. Multicollinearity was checked, and the Hosmer–Lemeshow test was used to assess the model's fitness. Adjusted odds ratios (AOR) with 95% confidence intervals (CI) were calculated to determine the presence and strength of associations between predictors and the outcome variable. Statistical significance was considered for p-values <0.05.

The study was approved by the Institutional Review Board (IRB) of the Hawassa College of Health Sciences. Ethical clearance was obtained from Hawassa College of Health Sciences under reference number IRB/225/15. Permission to conduct the study was granted in writing by the Hawassa College of Health Sciences and the Hula District health office. The Sidama Eastern Zone Health Bureau Research Ethics Committee approved verbal consent with reference number REC/35/15. The participants were extensively informed about the study's objectives, and verbal consent was obtained from the guardians of teenagers aged 13–17 and from each participant aged 18–19. Stringent measures were taken to ensure the confidentiality and anonymity of all participants, and they were explicitly informed of their right to refuse or terminate the interview at any point. The methods employed adhered to standardized principles, procedures, as well as relevant guidelines and regulations.

A total of 518 teenagers participated in this study, with replacement, resulting in a response rate of 100%. The study revealed that the participants' ages ranged from 13 to 19 years, with a mean age of 16.2 (±1.8) and a standard deviation of 1.8. The majority of the respondents (42.10%) were between the ages of 16–17. More than half of the participants (53.30%) resided in urban areas. The majority of the respondents (84.90%) identified as Protestants. In terms of marital status, the majority of the teenagers (84.50%) were single or had never been married. In addition, 68% of the teenagers attended primary school. A significant percentage (78.80%) of the respondents lived with their parents. Regarding income, approximately 45.20% of the participants earned less than 1,000 Birr per month (Table 1).

The study results revealed that out of the total respondents, 340 (65.60%) of the individuals had engaged in sexual intercourse, and 91(17.60%) of them had experienced teenage pregnancy (Figure 1). Furthermore, it was found that 131 (38.6%) of the respondents initiated sexual activity before the age of 15 years. In addition, 17 (18.70%) respondents experienced their first pregnancy before age 18 years. Among the majority, 58 (63.70%) respondents had unplanned pregnancies. Approximately half, 253 (48.80%), of the respondents had heard about FP. Among those who were aware of FP, more than half, 143 (56.50%), of the respondents were unaware of any location where FP methods were available. Furthermore, 168 (49.40%) of the respondents had never utilized family planning methods. The primary reason for the non-use of FP methods among 82 (67.80%) teenagers was attributed to never engaging in sexual intercourse (Figure 2). Only 18 (3.50%) of the teenagers were aware of the fertile period during the menstrual cycle. Moreover, over two-thirds, 366 (70.70%), of the teenagers did not communicate with their parents about sexual matters. It was found that more than one-third, 34 individuals (42.50%), of the married teenagers had experienced forced marriages. Lastly, only 60 (11.60%) teenagers reported having married or pregnant peers (Table 2).

The main strength of this study is that it is a community-based study. Another strength is the random sampling that permits to have a representative sample of the study source. However, the study had some limitations. Teenage pregnancy is often underreported by survey participants due to the stigma surrounding early sexual activity and pregnancy during adolescence. This underreporting was partly from the sensitivity and personal nature of the questions asked about sexual behaviors, which can lead to a social desirability bias. In addition, it can be challenging to accurately assess certain health risk factors associated with early teenage pregnancy. Finally, teenagers may either underreport or overreport their exact ages. Therefore, when interpreting the findings, caution should be taken.