Adolescent pregnancy in Sao Tome and Principe: a cross-sectional hospital-based study

Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 https://doi.org/10.1186/s12884-022-04632-z RESEARCH Adolescent pregnancy inSao Tome andPrincipe: across-sectional hospital-based study Alexandra Vasconcelos1*, Nelson Bandeira2, Swasilanne Sousa2, Filomena Pereira1 and Maria do Ceu Machado3 Abstract Background: Pregnancy starts early in Sao Tome and Principe (STP) and rates of adolescent pregnancy increased 16% in recent years reaching a 27.3% prevalence. This study aimed to understand the pregnant adolescents’ characteristics and factors associated to early childbearing in STP. Methods: A cross-sectional hospital-based study was undertaken in Hospital Dr. Ayres de Menezes between 2016 and 2018 with a randomly selected total sample size of 518 mothers. Mothers’ clinical records and interviews were used to collect relevant data. The results among adolescent girls 19 years of age and younger (n=104) were compared to adult mothers (n=414). A subgroup analysis of adolescent pregnant girls was also conducted. Statistically significance was considered at a p-value ≤0.05. Data were analysed using SPSS software. Results: The study revealed that 20.1% were adolescent mothers. Pregnancy at a very early age (≤15) was experienced by 7.7%. The characteristics founded to be positively associated with adolescent pregnancy were: 1) being single (OR 0.39, 95% CI=0.2–0.6, p≤0.001); 2) having a relationship with the baby´s father for a period of less than one year (OR 0.16, 95% CI=0.09-0.3, p≤0.001); 3) lack of the baby´s father support (OR 0.41, 95% CI=0.2–0.7, p=0.002); 4) not using a contraceptive method (OR 0.33, 95% CI=0.2–0.5, p≤0.001), and 5) inappropriate knowledge concerning the identification of the newborn’s danger signs (OR 15.7, 95% CI= 9–26, p≤0.001). Comparing pregnancy at very early age (≤15) to late (>18 and ≤19) adolescents, main differences were that previous contraceptives were not used at all in girls ≤15 years compared to 9.8% of late childbearing subgroup. Conclusions: Unfavourable factors linked to adolescent pregnancies were absence of a contraceptive method, getting pregnant in the early first months of one relationship and to be single. Gap age difference between adolescents’ partners, polygamous sexual relationships, previous abortion and having already other living children were also identified. Adolescents also had inappropriate knowledge of the identification of the newborns’ danger signs. Before being sexually active, adolescents critically need sexual and reproductive health information provided by a healthy community and through school programmes on sexual education. Schools should promote girl’s empowerment and awareness and, at the same time, reinforce boy’s role in fatherhood and shared responsibilities. The government should work on the prevention of early sexual initiation, as well as on improving family planning programmes to protect © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Open Access *Correspondence: [email protected] 1 Unidade de Clínica Tropical - Global Health and Tropical Medicine (GHTM), Instituto de Higiene e Medicina Tropical (IHMT), Universidade NOVA de Lisboa, Lisbon, Portugal Full list of author information is available at the end of the article Page 2 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 Background Adolescence is a critical phase for a successful transition to adulthood, that is threatened by becoming pregnant under 19 years old as it can greatly alter young women´s life prospects and those of their children [1, 2]. Adolescent pregnancies are a global problem as babies born to adolescent girls account for 11% of all births, with 95% occurring in developing countries [3]. Worldwide, adolescent pregnancies are frequently unplanned, unwanted, and out of wedlock that blocks girls’ education and their social relationships due to stigmatisation and loss of self-esteem, compromising the adolescents’ economic independence [1, 3–7]. Consequently, early motherhood impacts not only girls’ mental and physical health, but it also promotes an intergenerational perpetuation of a poverty trap and cycle [1, 2]. Thus, helping adolescent girls to avoid facing childbearing can have far-reaching benefits for them, their children, and societies [1, 3–7]. Recognizing the adolescents’ vulnerabilities is the first step in approaching this problem. In Africa, according to Kassa etal. 2018 meta-analysis [8], girls were more likely to start childbearing if they were from rural residence, married, not educated, whose parents had no education, and from families that lack parent to child communication on sexual reproductive health issues. Similar findings were also published by Ahinkorah etal. 2021 multi-country analysis [9], except for the finding that girls who lived in rural areas had lower odds of first adolescent pregnancy [9]. Delaying childbearing requires focusing on its major underlying factors: timing of first sex and ineffective contraceptive use, usually associated to girls’ inadequate sexual knowledge; low risk perceptions; ambivalence towards sex; lack of power to negotiate safer sex options and to deal with older partner´s persuasive techniques towards sex [10–13]. Moreover, adolescent girls face a higher risk for unsafe abortion and to be a second time young mothers [10, 14, 15]. Pregnancies complications are more frequent, and babies of adolescent mothers are at a higher risk of being born premature, dying in the neonatal period [16] and facing malnutrition, low mental and physical development, inappropriate social connection with parents and poor education [11, 17, 18]. In general, these risks are lower for adolescents in their late teens but those who give birth before age 15 are at much higher risk [10, 19]. Additionally, the younger the girl the less likely is for her to attend antenatal healthcare as coverage among younger adolescents are lower when compared to older adolescents (18-19 years old) [20] (https:// data. unicef. org/ topic/ childhealth/ adole scenthealth/). In Africa, this problem is even more hazardous as 33% of girls are giving birth before the age of 18 and 3.5% even before the age of 15 years old [3, 21–23]. Furthermore, it is foreseen that the percentage of adolescent pregnancies will increase globally by 2030, particularly in the Sub-Saharan African countries [22]. These can be due to girls’ low economic status, poor living arrangements, early school drop-out, inappropriate knowledge of sexual and reproductive health issues, absence of family planning, lack of sexuality education in schools and low employment attainment [8]. This calls for efforts to address the sexual and reproductive health problems of adolescent girls, included for the first time in the Sustainable Development Goals (SDGs) Countdown to 2030 [24, 25]. First step should start in the empowerment of girls attending school, preventing their drop-out and giving them access to sexual education with the necessary skills to prevent pregnancy [8]. The Democratic Republic of Sao Tome and Principe (STP), twoislands in the Gulf of Guinea, has one of the highest rates of teenage pregnancies in sub-Saharan Africa, estimated at about 27% [26, 27]. Pregnancy starts early in STP and adolescent pregnancy rate in the past years increased by 16% (22.8% in 2009 to 27.3% in 2014). With a population of about 200.000, STP has a high adolescent fertility rate of 94.8 births per 1,000 women aged 15 to 19 years compared for example to 68 in South Africa or 61 in Sudan [23, 27–30]. Although, adolescent pregnancy is recognized as a major public health issue in the country and the government seeks to be able to reduce it to a rate of 15% [28, 31–36] (https:// w w w. un. org/ en/ de vel opment/ desa/ popul ation/ publi catio ns/ pdf/ popfa cts/ PopFa cts_ 2019-1. pdf), it has not been yet properly studied [21, 31–33, 37]. This study therefore sought to investigate the differences between adolescents and older women in relation to selected sociodemographic and family characteristics, obstetric history, and knowledge of the identification of the newborns’ danger signs. We also aimed to explore the eventual differences of “very early childbearing” defined as birth before age ≤15, early childbearing (birth between age >16 and ≤17) and late them from pregnancy with special focus for the very early adolescent girls. None of these goals can be achieved if the government doesn’t, simultaneously, improve educational and economic opportunities for girls. Keywords: Adolescent pregnancy, Sao Tome and Principe, Family planning, Newborns’ danger signs Page 3 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 childbearing (birth age >18 and ≤19) in comparison to the older women group trying to figure out main teen’s vulnerabilities. We have previously studied these adolescent girls regarding their adverse outcomes and founded that foetal distress and the need for performance of neonatal resuscitation manoeuvres were the main adverse perinatal outcomes imputable to these adolescent births. We believe that understanding the characteristics of pregnant adolescents and risk factors that promote early childbearing will support STP policy makers in order to design pragmatic interventional programs to reduce the countries overwhelming high rates of adolescent pregnancies and achieving key SDG [34]. Material andmethods Study design This is a cross-sectional analytical hospital-based study conducted between July 2016 to November 2018. Study context The study was located at Hospital Dr. Ayres de Menezes (HAM) in Sao Tome city, the capital of STP, a rapidly growing city, with 64% of the population living in the capital [33]. HAM is the only hospital in the island, with a maternity that delivers around 4700 babies annually, around 82,4% of all the deliveries in the country [32]. Comprehensive Emergency Obstetric and Neonatal Care Services are available at HAM. There are 6 health centres with maternities in the rural area. Study population andsampling Mothers were randomly selected by chance and the study was done in different months to avoid seasonal interference and effects by means of confounding variables by guaranteeing a sample with few biases. Two groups were considered, over or under and equal to 19 years of age. Of the total 518 pregnancies, 104 were in women aged ≤19 years (adolescent group) and 414 pregnancies in women more than 19 years of age (older childbearer group). For some variables (Tables1, 2 and 4) further analyses with three adolescents age subgroups were used for the purpose of finding eventual differences of “very early childbearing” as birth before age ≤15 (n= 8), early childbearing for birth between age >16 and ≤17 (n=45) and late childbearing for birth age >18 and ≤19 (n=51) in comparison to the older women group. Eligibility criteria All women admitted to the hospital for delivery with a gestational age of 24 weeks or more were eligible to be enrolled in the study. Those who gave birth outside the hospital but were admitted for postnatal were also included in the study. Data collection A face-to-face interview by the main investigator was conducted with the mothers before discharge to gather information regarding socio-demographic characteristics, antenatal care/obstetric history, and their knowledge of identification of newborns’ danger signs (signs and symptoms of disease). The survey was built in the tool QuickTapSurvey© and mothers’ answers were collected and entered to this app. Data regarding obstetric and antenatal care were also checked and collected from medical records and antenatal care pregnancy booklet into the app survey. Variables included inthestudy Sociodemographic characteristics included: (i) residence, (ii) standard of living, (iii) marital status, (iv) living arrangements, (v) mother’s education level, (vi) mother’s employment status, (vii) baby´s father age, (viii) father´s education level, (ix) father´s employment status and (x) father´s living with other woman/houses according to pregnant women´s answer during the interview. Residence was grouped into urban and rural, urban residence considered for women living at the capital city (Água Grande) and rural in all other districts. Standard of living was categorized according to household deprived or not of electricity, sanitation, drinking water, cooking fuel and type of housing as Sustainable Development Goal guidelines [38]. Marital status considered the two categories yes (married/union) or no if single/never married. Time of the union/marriage was also assessed. Mother’s and father’s education level was categorized as either elementary only or secondary level. Taking into consideration that not attending secondary school, usually beginning from the age of 12 years old, is considered to have an inappropriate educational level. Employment status were collected and classified as no if “not employed” or yes if “employed” (lists of jobs were merged) regardless of the type of job. Obstetric and antenatal characteristics included: (i) obstetric history, (ii) unplanned pregnancy, (iii) previous contraceptive use, (iv) family planning after current pregnancy, (v) antenatal care visits. Antenatal care (ANC) consisting of less than 4 of expected consultations during pregnancy was considered as inadequate, between 4-7 as adequate and 8 or more visits as complete. Gravidity categorized as primigravida (1), multigravida (2-4) and grandmultigravida (5 or above). Page 4 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 Table 1 Socio-demographic characteristics for different adolescent subgroups and total adolescent pregnant girls in comparison to the older women group Characteristics Pregnancies ≤19 years old Pregnancies >20 years old very early childbearing (≤15) n= 8 (%) early childbearing (≥16 and ≤17) n=45 (%) late childbearing (≥18 and ≤19) n=51 (%) Total adolescents n= 104 (%) n=414 (%) Residence rural 6 (75%) 30 (67%) 27 (52.9%) 63 (60.6%) 213 (51.4%) urban 1 (12.5%) 15 (33%) 22 (43.2%) 38 (36.5%) 194 (46.9%) missings 1 (12.5%) - 2 (3.9%) 3 (2.9%) 7 (1.7%) OR and 95% CI 0.1 (0.02-1.5) 0.5 (0.29-1.06) 0.8 (0.48-1.59) 0.66 (0.4-1.03) p value 0.11 0.07 0.66 0.07 Mother´s education level elementary only 5 (62.5%) 20 (44%) 19 (37.3%) 44 (42.3%) 160 (38.6%) secondary 3 (37.5%) 25 (56%) 32 (62.7%) 60 (57.9%) 254 (61.4%) OR and 95% CI 2.8 (0.66-11.9) 1.4 (0.75-2.63) 0.9 (0.5-1.8) 1.3 (0.9-2) p value 0.16 0.29 0.99 0.19 Employed no 5 (62.5%) 31 (69%) 32 (64%) 68 (65.4%) 262 (63.3%) yes - 1 (2%) 7 (14%) 8 (7.7%) 134 (32.4%) missings - - 1 (6.3%) 1 (0.9%) 9 (2.2%) OR and 95% CI - - - 0.17 (0.8-0.37) p value - - - ≤0.001* Still studying yes 3 (37.5%) 13 (29%) 11 (22%) 27 (26%) 9 (2.2%) OR and 95% CI - - - 15.8 (7-35) p value - - - ≤0.001** Married no 6 (75%) 14 (31%) 13 (25.5%) 33 (31.7%) 59 (14.3%) yes 2 (25%) 31 (69%) 37 (72.5%) 70 (67.3%) 336 (81.2%) missings - - 1 (2%) 1 (1%) 19 (4.6%) OR and 95% CI 0.05 (0.01-0.2) 0.34 (0.16-0.7) 0.46 (0.23-0.9) 0.39 (0.2-0.6) p value ≤0.001 0.003 0.03 ≤0.001*** Babies’ father support yes 6 (75%) 34 (76%) 44 (86.3%) 84 (80.7%) 373 (90.1%) no 2 (25%) 11 (24%) 7 (13.7%) 20 (19.2%) 41 (9.9%) OR and 95% CI 0.3 (0.06-1.7) 0.34 (0.1-0.7) 0.7 (0.29-1.6) 0.41 (0.2-0.7) p value 0.19 0.005 0.42 0.002**** Father´s age <20 years old 2 (25%) 11 (24%) 4 (9.1%) 17 (16.3%) 2 (0.5%) 20-29 5 (62.5%) 27 (60%) 33 (75%) 65 (62.5%) 124 (30%) 30-39 - 3 (7%) 7 (15.9%) 10 (9.6%) 161 (38.9%) 40-49 - - - - 64 (15.5%) >49 - - - - 25 (6%) missing 1 (9.2%) 4 (9%) 7 (13.7%) 12 (11.5%) 38 (9.2%) OR and 95% CI - - - - p value - - - - Father´s education level elementary only 3 (37.5%) 9 (20%) 16 (31.4%) 28 (26.9%) 129 (31.2%) secondary 3 (37.5%) 15 (33%) 19 (37.2%) 37 (35.6%) 151 (36.5%) Page 5 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 In each survey, women were asked about previous contraception use and about their future desire to have a contraception method after the pregnancy including the type of contraceptive method they wish to use. We considered as traditional contraceptives the following methods: rhythm, calendar, withdrawal, lactation amenorrhea and “other traditional” methods. As modern contraceptives, we included intrauterine devices and systems, subdermal implants, oral contraceptives, injectable, diaphragms and cervical caps, condoms (male and female), patch, emergency contraception and sterilization. Mother´s knowledge of identification of newborns’ danger signs: was assessed during the interview as “yes” or “no” answer grouped and organized according to the definition presented by the WHO and the Young Infants Clinical Signs study group [39, 40] to the following questions: “Do you think the following are danger signs in the newborn?” (i) convulsions/spasms/rigidity; (ii) difficult/fast breathing; (iii) very small baby (less than 2.5 kg) or losing weight in the first weeks of life; (iv) lethargy/ unconsciousness; (v) fever; that is high body temperature that makes the newborn body hot. A score of 3 or more “yes”answers was considered as appropriate knowledge and less than 2 “yes” answers as inappropriate. Locally appropriate and known terms in the appropriate cultural area were used for clinical simplicity purpose and mother´s understanding. Data management andstatistical analysis The data were secured in a confidential and private location. Participants were referred to by identification numbers and the informed consent forms were kept separate from the questionnaires. Both could only be linked by a coding sheet available only to the investigators. The software used for sample calculation was Raosoft (http:// www. raoso ft. com/ sampl esize. html), but this value was supported by pass software (https:// www. ncss. com/ softw are/ pass/). There was a mean of 4540 HAM deliveries/ year within the study period. The sample was calculated based on a minimum sample of 10% of the population validated by the sample calculation software, which placed the right dimension between 355 (95%) and 579 (99%) confidence. It was possible to collect 518, which gave some comfort at this level. Differences in the demographic and obstetric characteristics between the adolescent and adult mother were assessed using the Pearson’s chi-squared test (χ2). During data analysis, missing values were treated as missing. The associations between maternal age were categorized as adolescent mothers (1) vs others (0). For each outcome with a statistically significant difference at the first level, further analysis was developed using mothers´ age as a continuous variable with adjustments made simultaneously for father´s support and contraceptive use independently of their statistical significance. For statistical purpose, maternal age was kept as a continuous variable in the models. Mean age of the mothers was 26.59 years (sd 7.1), with a quasi-normal distribution, just underrepresented at the very young mother’s level. Two step logistic regression models were developed to assess the power of age as a predictor. The models were adjusted for marital status, education and antenatal care, the probabilities of event were recorded for each case and the plots obtained with the respective 95% CI for Table 1 (continued) Characteristics Pregnancies ≤19 years old Pregnancies >20 years old very early childbearing (≤15) n= 8 (%) early childbearing (≥16 and ≤17) n=45 (%) late childbearing (≥18 and ≤19) n=51 (%) Total adolescents n= 104 (%) n=414 (%) doesn´t know/missingɫ2 (25%) 21 (47%) 16 (31.4%) 39 (37.5%) 134 (32.4%) OR and 95% CI - - - - p value - - - - Abbreviations: OR Odds Ratio, CI Confidence interval * The adolescent group reported to have a job in 7.7% compared to a higher proportion of 32.4% for the older counterpart with a significant statistical difference (OR 0.17, 95% CI=0.8-0.37, p≤ 0.001). ** There was a higher proportion of adolescent mothers who were students at the time they got pregnant, with a statistically significant difference (OR 15.8, 95% CI=7-35, p≤0.001). ***67.3% of the teenage girls were married and 31.7% were single compared to 81.2% and 14.3% in the older women group, which was statistically significant (OR 0.39, 95% CI=0.2-0.6, p≤0.001). **** The adolescent´s pregnancy was not supported/followed by the baby´s father in 20% of the cases compared to 9.9% of the older women group, the difference being statistically significant (OR=0.41, 95% CI=0.2–0.7, p=0.002). ɫ Most adolescents didn´t answer because they were not aware of the fathers’ education level achieved Page 6 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 Table 2 Obstetric history, ANC and family planning for the different adolescent subgroups in comparison to the older women group Characteristics Pregnancies ≤19 years old Pregnancies >20 years old very early childbearing (≤15) n= 8 (%) early childbearing (≥16 and ≤17) n=45 (%) late childbearing (≥18 and ≤19) n=51 (%) Total adolescents n= 104 (%) n=414 (%) Previous contraceptive use yes 0 (0%) 2 (4%) 5 (9.8%) 7 (6.7%) 119 (28.7%) no 7 (87.5%) 36 (80%) 41 (80.3%) 84 (80.7%) 217 (52.4%) missings#1 (12.5%) 7 (16%) 5 (9.8%) 13 (12.6%) 78 (18.8%) OR and 95% CI - 0.14 (0.03-0.59) 0.24 (0.08-0.6) 0.33 (0.2-0.5) p value - 0.007 0.007 ≤0.001* Pregnancy was planned yes 0 (0%) 5 (11%) 14 (27.4%) 19 (18.3%) 102 (24.6%) no 5 (62.5%) 30 (67%) 25 (49%) 60 (57.7%) 231 (55.8%) missings#3 (37.5%) 10 (22%) 12 (23.5%) 25 (24%) 81 (19.6%) OR and 95% CI - 0.39 (0.1-1.0) 1.2 (0.6-2.5) - p value - 0.06 0.49 - Gravidity 1 7 (87.5%) 39 (87%) 27 (52.9%) 73 (70.2%) 54 (13%) 2 1 (12.5%) 5 (11%) 19 (37.3%) 25 (24%) 78 (18.8%) 3 or more 0 (0%) 1 (2%) 5 (9.8%) 6 (5.8%) 282 (68.1%) OR and 95% CI 17.7 (9-26) p value ≤0.001** Abortion/ miscarriage 0 7 (87.5%) 41 (91%) 41 (82%) 89 (85.5%) 274 (66.2%) 1 1 (12.5%) 4 (9%) 9 (18%) 14 (13.5%) 105 (25.4%) 2 0 (0%) 0 (0%) 0 (0%) 0 22 (5.3%) >3 0 (0%) 0 (0%) 0 (0%) 0 13 (3.1%) OR and 95% CI - p value - Living children 0 8 (100%) 43 (96%) 35 (68.6%) 86 (82.7%) 74 (17.9%) 1 0 (0%) 2 (4%) 13 (25.5%) 15 (14.4%) 89 (21.5%) 2 0 (0%) 0 (0%) 3 (5.9%) 3 (2.9%) 97 (23.4%) >3 0 (0%) 0 (0%) 0 (0%) 0 154 (37.2%) OR and 95% CI - p value - 1st attendance ANC <12th 3 (37.5%) 18 (40%) 34 (66.6%) 55 (52.9%) 218 (52.7%) >12th 3 (37.5%) 22 (49%) 12 37 (35.6%) 136 (32.8%) missings 2 (25%) 5 (11%) 5 (9.8%) 12 (11.5%) 60 (14.5%) OR and 95% CI 0.8 (0.17-4.3) 0.7 (0.3-1.4) 2.47 (1.2-4.9) - p value 0.87 0.39 0.01 - ANC at least once 104 (100%) 413 (99.7%) <4 visits 2 (25%) 8 (18%) 5 (9.8%) 15 (14.4%) 55 (13.3%) 4-7 visits 4 (50%) 21 (47%) 20 (39.2%) 45 (43.3%) 196 (47.3%) >8 visits 2 (25%) 16 (36%) 26 (51%) 44 (42.3%) 163 (39.4%) OR and 95% CI - p value - Page 7 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 each age point. Regarding confounding variables, multivariate logistic regressions were performed in two stages - first only with the variable under study (mother’s age) and afterwards with the inclusion of control variables to understand whether the identified effect occurred via others (education, partner and antenatal care). Statistical significance was defined as p<0.05. Data were entered in QuickTapSurvey (©2010-2021 Formstack), and the dataset exported to Excel for cleaning and further analysis using the Statistical Package for the Social Sciences for Windows, version 25.0 (IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp.). Ethics approval andconsent toparticipate Ethical clearance was obtained from the Democratic Republic of Sao Tome and Principe Ministry of Health and Hospital Dr.Ayres de Menezes. Written informed consent was obtained from all participants (or their parent or legal guardian in the case of teenage under 16) after the purpose of the research was explained orally by the investigator. This was done only after delivery to reduce coercion due to labour pains. Participation in the survey was voluntary, as participants could decline to participate at any time during the study. All methods were performed in accordance with the relevant guidelines and regulations in practice. Results Socio‑demographic andfamily characteristics A total of 518 pregnant women were enrolled. Among the enrolled women, 20.1% (CI – 16,3% to 23%) were adolescents, with 104 teenage pregnancies (≤19 years old) and 414 deliveries in older women. The mean age was 17.42, median 17, with a minimum age 14 years. Regarding childbearing subgroup ages (Fig.1), 7.7% (8) were very early childbearing adolescents (birth age ≤15), 43.3% (45) were early (≥16 and ≤17) and 49% (51) were late (≥18 and ≤19). The older women group mean age was 28.9 years, median of 28 with a maximum age of 43 years. Regarding the adult group, the mother´s mean age at first pregnancy was 20.1 (±3.76 SD) years (minimum 13, maximum 36). Socio-demographic characteristics for the different adolescent subgroups in comparison to the older women group are described in Table1. Marriage, living arrangements andbaby´s father support The duration of the relationship/union or marriage was shorter for the younger ages: 59.5% (62/104) of those were in the union less than one year compared to 19.7% (81/414) of the older counterpart, this was statistically significant (OR 0.16, 95% CI 0.09-0.3, p ≤ 0.001). In terms of existence and acknowledgment of a “polygamous” marital informality, 11.6% (12/104) of the adolescent girls reported they knew that the baby´s father was also living with another woman compared to 20.6% (85/414) in the older group, although it should be noticed that 48.5% (50/104) of the adolescent girls and 36.1% (149/414) of the older women group preferred not to answer to this question. Concerning living arrangements, 68% (71/104) of the adolescents were living with their husbands/boyfriend and 28.7% with others: 13.4% (14/104) with the pregnant woman´s mother and 15.3% (16/104) with other relatives; Table 2 (continued) Characteristics Pregnancies ≤19 years old Pregnancies >20 years old very early childbearing (≤15) n= 8 (%) early childbearing (≥16 and ≤17) n=45 (%) late childbearing (≥18 and ≤19) n=51 (%) Total adolescents n= 104 (%) n=414 (%) Want to do family planning yes 1 (12.5%) 34 (76%) 16 (31.3%) 51 (49%) 165 (39.9%) no 0 (0%) 1 (2%) 3 (5.9%) 4 (3.8%) 39 (9.4%) missings#7 (87.5%) 10 (22%) 32 (62.8%) 49 (47.1%) 194 (46.9%) OR and 95% CI 0.66 (0.4-1.03) p value 0.07 # as don’t know, doesn’t want to reply Abbreviations: ANC Antenatal care, OR Odds Ratio, CI Confidence interval. * Previous contraceptive use was stated in 6.7% of the teenage group and in 28.7% of the older counterpart, what is a significant statistically difference (OR =0.33, 95% CI = 0.2–0.5, p ≤0.001). ** The proportion of primigravidas among teenagers in relation to those aged >20 years was statistically significant (OR 15.7, 95% CI= 9-26, p≤0.001). Page 8 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 1.9% (2/104) adolescents mentioned that they were living alone. Three hundred and fifty-three/414 (85%) of the older women were living with their husbands/men, 4.5% (19/414) alone, 4.5% (19/414) with the pregnant woman´s mother, 4.1% (17/414) with other relatives and 2.2% (9/414) were living with a man who was not the baby´s father. The adolescent´s pregnancy was not supported/followed by the baby´s father (Table1) in 20% of the cases with a statistically significant difference (OR0.41, 95% CI 0.2–0.7, p=0.002) comparing to the older counterpart. However, after the model was adjusted for marital status, education and antenatal care, the effect of age disappeared totally, as the relevant variable becomes marital status (Fig.2). In the controlled model, marital status showed a OR 19, with a p=0.001. Obstetric history andantenatal care Gravidity, abortions, and number of living children are described in Table2. Previous contraceptive use andfuture family planning Previous contraceptive use is described in Tables2 and 3. Fig. 1 Adolescent ages and childbearing subgroup ages Fig. 2 Father´s support according to mother´s age. The line at 0.5 represents equal probability of father’s support during pregnancy Page 9 of 15 Vasconcelosetal. BMC Pregnancy and Childbirth (2022) 22:332 When model was adjusted for marital status, education, and antenatal care, the probability of using contraception rises with age, with a turning point at 34 years old (Figure3), meaning that in Sao Tome and Principe most young women do not assume birth control (Table2). Methods women stated that are willing to use to prevent future pregnancy are described in Table 3. Adolescent group most rated (66.7%) preference were for traditional methods (rhythm, calendar, withdrawal, lactation amenorrhea). In this study none of the group referred they would choose the use of a condom by their partner. To note that two older women had an urgent hysterectomy at the time of the current birth and other 7.9% (13/165) performed a sterilization during the caesarean intervention. Mothers´ knowledge ofidentification ofnewborns’ danger signs Mothers´ knowledge of identification of newborns’ danger signs was inappropriate in 70.2% (73/104) of the adolescent group and in 13% (54/414) of the older childbearers, being statistically significant (OR 15.7, 95% CI 9–26, p ≤0.001). Very early, early, andlate childbearing adolescents incomparison toolder pregnant women For some variables (Tables1, 2 and 4) age subgroups were used for the purpose of analysing eventual differences between “very early childbearing” for birth before age ≤15 (n= 8), early childbearing for birth between age >16 and ≤17 (n=45) and late childbearing for birth age >18 and ≤19 (n=51). Regarding standard of living (Table 4) almost all women in the study were considered as having a deprived Table 3 Future family planning: methods women are willing to use to prevent future pregnancy * Rhythm, Calendar, Withdrawal, Lactation amenorrhea Adolescent group n=51 (%) Older counterpart n= 165 (%) Oral contraceptives 8 (15.6%) 43 (26%) Injectables 6 (11.7%) 44 (26.7%) Intra uterine device 3 (5.8) 8 (4.8%) Sterilization 0 23 (14%) Condom (male) 0 0 Traditional methods* 34 (66.7%) 32 (19.4%) OR and 95% CI 0.66 (0.4-1.03) p value 0.07 Fig. 3 Previous contraceptive use according to women´s age. The model was adjusted for marital status, education, and antenatal care. The line at 0.5 represents equal probability of contraception before birth