ECD - Pregnancy outcomes of a birth cohort. Are adolescent mothers really at more risk?

ABSTRACT

This sub-study within the JAKIDS longitudinal cohort study compares medical and psychosocial outcomes of pregnancy in younger adolescent mothers (<16 years), older adolescent mothers (16–19 years) and adult mothers (>19 years) in Jamaica. Participants were recruited from July to September 2011 and included 9521 mother–infant dyads; mean maternal age 26.0 years (SD 6.8). Adolescent mothers represented 19.1% (n = 1822) of the sample – 1704 older adolescent mothers (17.9%) and 118 younger adolescent mothers (1.2%). Participants completed interviewer-administered questionnaires regarding their sexual and reproductive health history, feelings about the current pregnancy, and presence of anxious and depressive symptoms. Data on delivery and perinatal and neonatal outcomes were extracted from hospital charts. Younger adolescent mothers were more likely to deliver preterm (p < 0.001) and low birth weight infants (p < 0.001) than older adolescent and adult mothers. Younger adolescent mothers had lower levels of antenatal anxiety regarding the pregnancy and its outcome (p < 0.001) while prevalence of elevated depressive symptoms antenatally (EPDS ≥11) was similar across age groups. Older adolescent mothers with significant depressive symptoms had increased odds of preterm delivery. These findings call for close antenatal monitoring of younger adolescent mothers and highlight the need for psychological services for all mothers.

KEYWORDS:

• Adolescent pregnancy
• neonatal outcome
• Jamaican
• maternal
• psychosocial factors

Introduction

Annually, just over 10% of the 135 million live births worldwide are to adolescent women aged 15 to 19 years old (UNFPA, 2013). Although this reflects a decline from 65 births per 1000 in 1990 to 47 births per 1000 adolescent women in 2015 (United Nations Department of Economic and Social Affairs, Population Division, 2017), adolescent pregnancies still represent a serious public health concern. The Latin America and Caribbean region has the highest proportion of adolescent births (16%) to total births globally with an adolescent fertility rate of 67 per 1000 women aged 15–19 years old (United Nations, 2017). One in five children in Jamaica are born to adolescent girls (Maharaj et al., 2009) with Jamaica’s adolescent fertility rate of 57 per 1000 births to women aged 15–19 years old, which is much higher than that of high-income countries (US 21; UK 14; Canada 10 per 1000 births to 15–19 year olds) and most of its neighbouring countries in the English-speaking Caribbean (United Nations Population Division, World Population Prospects, 2018).

The contributing factors to the persistent challenge of adolescent pregnancy in the region may include sub-optimal sexual and reproductive health (SRH) knowledge among adolescents, the socio-cultural milieu which may enable gender-based violence, transactional sex and early age of onset of sexual activity (Caffe et al., 2017). Curbing this challenging trend has been identified by major NGOs as a key area to be addressed with calls for improved sexual and reproductive healthcare access for adolescents and increased access to age-disaggregated data (Caffe et al., 2017; Neal, Harvey et al., 2018).

Approximately three decades ago, a large prospective study in an urban hospital in the US, found adolescents (13–19 years) were at increased risk for low birth weight (LBW) and premature infants, even after controlling for antenatal care and maternal ethnicity (Leppert et al., 1986). Similar findings have been reported for low and middle-income countries (LMICs) in the last decade (Ganchimeg et al., 2014; Neal, Channon et al., 2018; Serunjogi et al., 2021; Thirukumar et al., 2020). The World Health Organization (WHO) reports complication in pregnancy and child birth is the second most common cause of death in female adolescents age 15–19 years old in LMICs (World Health Organization, 2016). Previous studies in India identified adolescent mothers as being at an increased risk of pregnancy-induced hypertension (PIH), preeclampsia and eclampsia, premature onset of labour, fetal deaths and premature delivery with increased neonatal morbidity (LBW, perinatal asphyxia, jaundice and respiratory distress syndrome) and mortality (Kumar et al., 2007; Mahavarkar et al., 2008). Younger adolescents (≤17 years) were deemed to be at higher risk than those between 18 and 19 years old. However, these findings were challenged when another team in India failed to identify an increase in PIH, intra-uterine growth restriction, preterm labour or postpartum heamorrhage (PPH) in adolescent mothers as compared to older mothers (women >19 years) (Sagili et al., 2012).

Studies in LMICs have noted the impact of antenatal care on adolescent and older maternal outcomes. In Thailand, adolescent mothers were more likely to have had incomplete antenatal care, with a lower rate of delivery by C-section and PPH, but an increased likelihood of premature delivery and LBW infants when compared to older mothers (Narukhutrpichai et al., 2016; Watcharaseranee et al., 2006). In Malawi, LBW was significantly associated with the number of antenatal care visits, particularly among adolescents younger than 17 years (Brabin et al., 1998). The importance of antenatal care was similarly highlighted in a Jamaican sample of mothers, where there was no significant difference in neonatal outcome between adolescent and adult mothers once they received similarly consistent antenatal care. However, older mothers had significantly higher incidence of early pregnancy losses, while adolescents had higher prevalence of urinary tract infections and lower C-section rates (Young et al., 2007).

Adolescent pregnancy not only entails medical risks for mother and child but also has potential for negative psychosocial outcomes. Two local studies found evidence of negative psychosocial outcomes among a small subset of adolescent mothers, such as disruption in their education, exposure to violence, lack of parental support, psychological distress and suicidal ideation (Peart, 2020; Wilson-Mitchell et al., 2014). Studies have not always been consistent regarding the impact that mothers’ psychosocial health can have on neonatal outcomes (Mutambudzi et al., 2011) but evidence that significant maternal depression and anxiety may negatively impact birthweight and gestational age of infants is increasing (Khanghah et al., 2020; Loomans et al., 2013).

Although Jamaica has achieved a reduction in fertility rate of older adolescents in the past three decades, we have been less successful in this regard on younger adolescents who may be at higher risk (Gordon-Strachan et al., 2015; Serbanescu et al., 2010). While medical outcomes of adolescent mothers and their neonates have been explored in Jamaican adolescent mothers (Thame et al., 1999; Young et al., 2007), less has been investigated regarding maternal psychosocial factors (Peart, 2020; Wilson-Mitchell et al., 2014). This study set out to determine age-disaggregated differences in neonatal health outcomes and maternal medical and psychological outcomes – younger adolescent mothers (under 16 years) compared to older adolescent (16–19 years) and adult mothers (>19 years).

Materials and methods

This is a sub-study within the JAKIDS longitudinal cohort study, which was launched in 2011 (https://www.mona.uwi.edu/fms/jakids/). The JAKIDS is the second birth cohort study in Jamaica and aims to update our understanding of factors currently affecting Jamaican families. Mothers were recruited antenatally (n = 5204) by trained research assistants between July and September 2011 from public and private antenatal centres across Jamaica and were again contacted in the hospital after delivery (‘birth contact’), recruiting 87% of the total population of mothers nationally during the period (n = 9766). All study participants provided written informed consent, and all questionnaires were administered by interviewer, utilizing a standardized approach with care to ensure confidentiality of participants’ responses. Participants were given the option to refuse to answer any questions that made them uncomfortable.

Three questionnaires were administered during the antenatal period: (i) My Expectations of My Pregnancy, My Parenting and My Partner; (ii) My Life, My Home, My Community; (iii) My Health. A fourth questionnaire – My Pregnancy Labour and Delivery was administered post-delivery prior to discharge from the hospital (‘birth contact’). Trained interviewers asked questions in a consistent format, offering further clarification when necessary, and the instruments were validated for use in the Jamaican population (Bernard et al., 2018; Gray et al., 2018).

Medical outcomes

Data regarding perinatal outcomes were gathered from hospital charts and interviewer-administered questionnaires. These included antenatal maternal illnesses such as hypertensive disorders of pregnancy, diabetes mellitus, urinary tract infections and vaginal discharges. Neonatal data collected included gestational age, mode of delivery and birthweight.

Psychosocial maternal factors

Details of mothers’ SRH history, feelings regarding the pregnancy, and presence of depressive symptoms were collected. Question scales were adopted from the Avon longitudinal study of parents and children (ALSPAC) cohort study (Boyd et al., 2013). A 10-question scale enquired mothers’ level of worry regarding the pregnancy and pregnancy outcome from which a summative antenatal anxiety score was generated – a higher score indicating a higher level of maternal worry. A seven-question scale enquired maternal self-care regarding healthy eating habits, exercising, attending antenatal care visits, rest, care regarding their whereabouts and practicing careful sexual activity. Responses were used to calculate a summative antenatal self-care score with higher scores indicating better self-care. A paternal emotional support score was generated as a sum of mothers’ responses to 17 questions regarding the level of affection displayed between herself and the father of the baby, and the closeness of the relationship. A higher score indicated more emotional connection with and support from the baby’s father. The Edinburgh post-natal depression scale (EPDS, Kozinszky & Dudas, 2015) explored symptoms of depression during the antenatal period and post-natal period. This study utilized the antenatal reports with mothers classified as ‘high risk for depression’ if their EPDS was ≥11. This cut-point was used to optimize combined sensitivity and specificity of identifying clinically significant depressive symptoms (Levis et al., 2020). At the ‘Birth Contact’, mothers’ perception of the level of antenatal care received during pregnancy was assessed using four questions (given good advice, received good care, had enough time for discussion, satisfied with care) and a summative antenatal satisfaction score was generated with higher scores representing greater maternal satisfaction with their care.

Insufficient data on socioeconomic status (SES) were available for this sub-study as adolescents were unable to provide information regarding overall household income; with participants’ educational attainment and occupation being a function of age, it would therefore be difficult to use for meaningful comparison.

Statistical analyses

Data were analysed using SPSS version 22. The data were disaggregated into three distinct age groups for comparison – younger adolescent mothers (<16 years), older adolescent mothers (16–19 years) and adult mothers (>19 years). Descriptive analyses were done regarding maternal and neonatal outcomes. Pearson chi-square analyses explored associations between categories of maternal age and categorical outcome variables. Analysis of variance and regression analyses further explored associations with continuous and categorical variables. Statistical significance was considered as a p value <0.05.

Ethical approval

Ethical approval was granted by the Ethics Committee for the Ministry of Health – Jamaica (approval #198; 2011) and the University of the West Indies (approval # ECP 122 10/11). All participants provided written informed consent.

Results

Data were collected on 9650 mother–infant dyads. Maternal age was missing for 129 dyads; hereafter analyses will be presented on n = 9521 mother–infant dyads. The overall mean maternal age was 26.0 (SD 6.8; range 12–52) years. Eighty percent of the sample were adult mothers (n = 7699; 80.9%), and the remainder were older adolescent mothers (n = 1704; 17.9%) and younger adolescent mothers (n = 118; 1.2%).

Antenatal medical complications are displayed in Table 1. Adolescent mothers had fewer medical complications than adult mothers with lower prevalence of proteinuria (6.8% vs 11.5%; p < 0.001), elevated blood pressure (8.9% vs 13.7%; p < 0.001), preeclampsia (0.1% vs 1.3%); p < 0.001), antepartum (1.3% vs 4.5%; p < 0.001) and post-partum haemorrhage (3.6% vs 15.8%; p < 0.001). Adolescents had higher prevalence of vaginal discharge (30.6% vs 26.6%; p < 0.05) but no significant difference was noted in urinary tract infections or sexually transmitted infections by maternal age.

Table 1. Medical complications by maternal age.

Complications	Total n (%)	Adolescent mothers ≤19 years n (%)	Adult mothers >19 years n (%)	p value
Proteinuria ***	394 (10)	53 (6.8)	341 (11.5)	0.000
Glycosuria	112 (3)	16 (2.0)	96 (3.2)	0.098
High blood pressure***	478 (12.7)	70 (8.9)	408 (13.7)	0.000
Preeclampsia***	101 (1.1)	2 (0.1)	99 (1.3)	0.000
Diabetes***	76 (2.0)	2 (0.3)	74 (2.5)	0.000
Urinary tract infection (UTI)	285 (7.6)	53 (6.8)	232 (7.8)	0.363
Vaginal discharge*	1058 (27.4)	247 (30.6)	811 (26.6)	0.024
Sexually transmitted infection (STI)	26 (0.7)	4 (0.5)	22 (0.7)	0.632
Antepartum haemorrhage***	368 (3.9)	24 (1.3)	344 (4.5)	0.000
Postpartum haemorrhage***	1282 (13.5)	65 (3.6)	1217 (15.8)	0.000

*p < 0.05; **p < 0.01; ***p < 0.001

Labour and delivery outcomes are displayed in Table 2. Significantly more adult mothers delivered by C-section (p < 0.001), with a significantly greater proportion delivered by elective C-section (adult mothers n = 1087, 14.4%; older adolescent (n = 113; 6.8%) and younger adolescent mother (n = 4; 3.4%) p < 0.001). A significantly greater proportion of younger adolescent mothers delivered their infants preterm (<37 completed weeks’ gestation) (younger adolescent mothers: n = 31; 27.9%; older adolescent mothers: n = 296; 18.0%; and adult mothers: n = 1103; 14.8%; p < 0.001). Younger adolescent mothers were significantly more likely to deliver an infant with LBW (<2500 g) compared to other mothers (younger adolescent mothers; n = 27, 23.9%; older adolescent mothers: n = 222, 13.4% and adult mothers: n = 785, 10.4%; p < 0.001). The neonatal mortality rate was highest in younger adolescent mothers (33.9 per 1000 live births) in comparison to older adolescent (10.0 per 1000 live births) and adult mothers (14.5 per 1000 live births) (p = 0.06). All neonatal deaths (n = 4) for younger adolescent mothers were secondary to prematurity. The cause of death was explicitly recorded for 12 of 17 deaths in older adolescent mothers with 75% deaths (n = 9) due to prematurity. Sixty-eight neonatal deaths to adult mothers had the cause of death recorded, with 76% deaths (n = 52) due to prematurity.

Table 2. Labour and delivery outcomes by maternal age.

Outcome	Total n (%)	Younger adolescent mothers <16 years n (%)	Older adolescent mothers 16–19 years n (%)	Adult mothers >19 years n (%)	p value
Mode of delivery (n = 9343)					0.000
Vaginal	7485 (80.1)	107 (92.2)	1469 (87.9)	5909 (78.2)
C-section	1858 (19.9)	9 (7.8)	202 (12.1)	1647 (21.8)*
Emergency C/S	654 (7.0)	5 (4.3)	89 (5.3)	113 (6.8)
Elective C/S	1087 (14.4)	4 (3.4)	113 (6.8)	1087 (14.4)*
Prematurity (n = 9196)					0.000
Preterm	1430 (15.6)	31 (27.9)***	296 (18.0)	1103 (14.8)
Term	7766 (84.4)	80 (72.1)	1348 (82.0)	6338 (85.2)
Multiple gestation (n = 9440)	126 (1.3)	3 (2.6)	15 (0.9)	108 (1.4)	0.117
Birthweight (n = 9337)					0.000
<2500	1034 (11.1)	27 (23.9)***	222 (13.4)	785 (10.4)
≥2500	8303 (88.9)	86 (76.1)	1439 (86.6)	6778 (89.6)
Infant survival (n = 9521)					0.062
Alive	9388 (98.6)	114 (96.6)	1687 (99.0)	7587 (98.5)
Dead	133 (1.4)	4 (3.4)	17 (1.0)	112 (1.5)
Neonatal mortality rate, per 1000 live births	14.0	33.9	10.0	14.5	0.06

*p < 0.05; **p < 0.01; ***p < 0.001

Mothers’ sexual and reproductive history

Younger adolescent mothers reported a lower median age (14 years, IQR 1) of first sexual intercourse in comparison to older adolescent and adult mothers (16 years, IQR 1; p < 0.001). Younger adolescents were also more likely to report using contraception at the time of getting pregnant in comparison to other mothers (p < 0.01), with only one younger adolescent mother reporting she was trying to get pregnant at the time of conception.

Psychosocial characteristics of pregnancy

Older adolescents (n = 652; 84.3%) and adult mothers (n = 2696; 87.1%) were significantly more likely to still be in a relationship with the father of the child when compared to younger adolescent mothers (n = 23; 54.8%) (p < 0.001). The mean paternal age of fathers as reported by mothers was 31.4 (SD 8.4; range 14–80) years, with fathers in adolescent pregnancies having significantly lower mean (SD) ages (younger adolescent mothers: 21.3 (5.2) years, older adolescent mothers: 25.1 (6.1) years, adult mothers: 32.9 (8.17) years; p < 0.001). Older adolescent mothers were more likely to report sub-optimal attendance at antenatal clinic visits; 60 attended visits sometimes or less frequently (n = 60; 8.0%) compared to younger adolescent mothers (n = 2; 4.9%) and older mothers (n = 148; 4.9%) (p < 0.01). There was no significant difference in overall mean antenatal self-care scores among mothers (Table 3).

Table 3. Social characteristics of pregnancy by maternal age.

	Total	Younger adolescent mothers <16 years	Older adolescent mothers 16–19 years	Adult mothers >19 years	p value
Median age of first sexual intercourse, years (interquartile range)	16 (2)	14 (1)***	16 (1)	16 (1)	0.000
n (%)^a
Using contraception at the time of becoming pregnant (n = 3886)	609 (15.7)	8 (19.0)	100 (13.1)	501(16.3)	0.004
Trying to become pregnant	689 (17.6)	1(2.4)	69 (8.9)	619 (20.0)	0.000
Mother in relationship with father of baby (n = 3912)	3371 (86.2)	23 (54.8)	652 (84.3)	2696 (87.1)	0.000
Attend antenatal clinic (n = 3803)					0.004
Sometimes or less often	210 (5.5)	2 (4.9)	60 (8.0)	148 (4.9)
Often to almost always	3593 (94.5)	39 (95.1)	690 (92.0)**	2864 (95.1)
Lives with parents in their home (n = 3902)	1275 (32.7)	30 (71.4)***	421 (54.7)	824 (26.7)	0.000
Antenatal self-care mean (SD) (n = 3917)	22.3 (4.6)	22.3 (4.6)	22.3 (4.4)	22.5 (5.1)	0.967
Age, father of baby (mean, SD) (n = 9114)	31.4 (8.4)	21.3 (5.2)***	25.1 (6.1)	32.9 (8.2)	0.000

*p < 0.05; **p < 0.01; ***p < 0.001

^apercent (%) within age category

Maternal emotional factors are shown in Table 4. Significantly more younger adolescent mothers reported being unhappy when they were pregnant (n = 20; 52.6%) compared to older adolescent mothers (n = 187; 26.2%) and adult mothers (n = 513; 17.45, p < 0.001) This proportion decreased over the course of the pregnancy to 19% (n = 8) of younger adolescent mothers reporting they were not happy with the pregnancy (p < 0.001). Antenatal worry represented by the antenatal anxiety score was significantly lower in younger adolescent mothers (mean (SD) younger adolescent mothers 17.4 (5.6); older adolescent mothers 20.2 (5.3); adult mothers 19.4 (5.4); p < 0.001). No significant difference was noted in mothers’ perception of paternal emotional support or depressive symptoms by maternal age category.

Table 4. Maternal emotional factors by maternal age.

	Total	Younger adolescent mothers <16 years^a	Older adolescent mothers 16–19 years^a	Adult mothers >19 years^a	p value
n (%)^a
Maternal feelings about pregnancy before delivery (n = 3699)					0.000
Happy	1046 (28.3)	0	127 (17.8)	919 (31.2)
Mixed feelings	1900 (51.4)	17 (44.7)	395 (55.2)	1488 (50.5)
Unhappy	720 (19.5)	20 (52.6)	187 (26.2)	513 (17.4)
Other	33 (0.9)	1 (2.6)	6 (8)	26 (0.9)
Maternal feelings about pregnancy post-delivery (n = 3871)					0.000
Happy from the start	1678 (43.3)	4 (9.5)	242 (31.4)	1432 (46.8)
Mixed feelings at first but now happy	1877 (48.5)	22 (52.4)	436 (56.5)	1419 (46.4)
Still mixed feelings	250 (6.5)	8 (19.0)	72 (9.3)	170 (5.6)
Mainly not happy	66 (1.7)	8 (19.0)	21 (2.7)	37 (1.2)
Mean (SD)
Current pregnancy anxiety score, mean (SD) (n = 3914)	19.5 (5.4)	17.4 (5.6)*	20.2 (5.3)	19.4 (5.4)	0.000
Paternal emotional support, mean (SD) (n = 3057)	54.7 (9.0)	55.3 (10.0)	54.2 (9.0)	54.8 (9.0)	0.363
Antenatal Edinburgh post-natal depression score (EPDS), mean (SD) (n = 3895)	7.09 (5.4)	7.62 (4.7)	7.3 (5.5)	7.03 (5.4)	0.342
‘At-risk’ for antenatal depression (EPDS ≥11) (n = 3895)	1008 (25.9)	11 (26.2)	210 (27.3)	787(25.5)	0.598

*p < 0.05; **p < 0.01; ***p < 0.001

^apercent (%) within age category

Binary logistic regression analyses were performed to assess the impact that paternal emotional support, antenatal depressive symptoms and attendance at antenatal care visits had on the incidence of premature delivery. No significant predictors of prematurity were identified for the group of mothers as a whole. However, older adolescent mothers were 48% less likely to deliver a term infant if they had significant depressive symptoms (EPDS ≥11, p < 0.05).

Discussion

In Jamaica, where one in five pregnancies are to adolescent mothers, an age-disaggregated review of the comparative outcomes of these pregnancies is useful to guide policy revision where necessary. The findings of our study corroborate with those of previous studies done locally on a sub-population of adolescent mothers, with adolescent mothers not at an increased risk for most maternal complications except they were significantly more likely to deliver LBW infants (Thame et al., 1999; Young et al., 2007). These findings differ, however, from other regional findings (Ganchimeg et al., 2014) and other LMICs (Kumar et al., 2007; Mahavarkar et al., 2008). Our findings of increased rates of preterm delivery in younger adolescent mothers, which was not previously reported by Young (2007), concur with regional and global findings in LMICs.

This study is novel in the local context, exploring the differences that exist in outcomes for the younger adolescent mother compared to the older adolescent with two critical outcomes – prematurity and LBW being significantly more prevalent in the younger adolescent. This underscores the need for the younger adolescent sub-population of mothers to be monitored very closely in a ‘high-risk’ clinic. The Ministry of Health and Wellness in Jamaica currently stipulates a minimum of six antenatal visits for all mothers (Ministry of Health and Wellness, Jamaica, 2020). An increase in this requirement may be warranted in keeping with the WHO’s recommendations (World Health Organization, 2017) of eight antenatal visits to achieve a positive pregnancy experience, particularly for younger adolescent mothers, in an effort to reduce maternal and neonatal morbidity and mortality.

This study adds to the literature by exploring the psychosocial well-being of adolescent mothers. Previous studies (Peart, 2020; Wilson-Mitchell et al., 2014) have explored these psychosocial factors by taking a qualitative approach; however, our study takes a quantitative approach, further extending the literature by giving a cross-sectional understanding. The majority of adolescent mothers, particularly younger adolescents, continued to live with their parents. Akella and Jordan (2015) indicated the importance of social support for optimal psychological well-being of adolescent mothers (Akella & Jordan, 2015). It is possible that the younger adolescents’ mothers own parental support may have contributed to their high attendance at antenatal clinics and overall antenatal self-care with no significant difference from that of adult mothers, unlike older adolescent mothers who attended antenatal visits less frequently, which may have been a result of less parental support.

Our findings that adolescent mothers initially lacked happiness was not unexpected, as adolescents in early pregnancy have to acknowledge and work through multiple potential challenges including telling their parents, school and partner about what would likely be seen as a mistimed pregnancy in most cases. However, the proportion of adolescent mothers reporting happiness increased by the end of the pregnancy. Younger adolescent mothers’ lower levels of worry regarding the pregnancy and its outcome are congruous with their cognitive developmental stage, limiting their ability to recognize and internalize the potential risks and complications.

As many as one in four mothers had significant levels of depressive symptoms antenatally, significantly higher than that identified in high-income countries (Bennett et al., 2004) but similar to that found in other developing countries (Biratu & Haile, 2015). Our findings substantiate previous local findings indicating that pregnant adolescents report depressive symptoms including loneliness, hopelessness, guilt and attempted suicide (McFarlane et al., 2014). Of further concern, older adolescent mothers with significant depressive symptoms had an increased risk of delivering their infant prematurely. These findings indicate a critical need for psychological support for all mothers but even more so for our adolescent mothers, inclusive of psychoeducation regarding the pregnancy and motherhood and improved SRH knowledge aimed at reducing perinatal morbidity and mortality in this vulnerable sub-population. The Jamaican mental health system is resource limited with 1.6 psychiatrist and 0.7 psychologists per 100,000 persons in the population (WHO 2005). Further, there are only three child and adolescent psychiatrists and 15 psychologists who serve children and adolescents in 22 Child Guidance Clinic sites in the public health system, and these sites are not all open daily. In the face of these limitations, routine screening for antenatal and post-natal depression along with addressing the psychological effects of neonatal mortality should be utilized for mothers in this vulnerable sub-group, adolescent mothers, to ensure that those at most risk are identified early with a timely referral.

Only one in five younger adolescent mothers reported using contraception at the time of getting pregnant, and the average age of their sexual contact and father of the baby was 5 years their senior. This questions the presence of barriers to younger adolescents’ access to contraceptive methods, girls’ condom negotiation skills and likelihood of transactional or coerced sexual relations and highlights the need to explore these concerns in future studies. Caffe et al. (2017) highlighted the need for sustainable, evidence-based interventions focused on the most vulnerable groups to end the public health challenge of adolescent pregnancy. This study has determined the younger adolescent mother’s needs to be the focal point with interventions aimed at promoting abstinence, delaying the age of sexual initiation and ensuring close antenatal monitoring of adolescent mothers in addition to improving adolescent girls’ contraceptive knowledge and condom negotiation skills.

In summary, while some findings from this study corroborate previous findings, it has contributed new knowledge that inform the provision of adequate interventions that meet the needs of our pregnant adolescents. While it was previously thought that adolescent mothers were uniformly at risk for negative health outcomes, we have found a unique group of at-risk pregnant adolescents – the younger adolescent mother, under 16 years. Despite our findings of elevated depressive symptoms, we also found evidence of the significance of social support by way of continued cohabitation with their own parents and its positive influence on adolescent mothers’ optimal antenatal care and psychological well-being. Still, in addition to the medical concerns of this vulnerable population, the study uncovered other psychological challenges such as depressive symptoms and neonatal morbidity that adolescent mothers have to contend with and the urgency for psychological resources to be available for this population. That being said, the findings from this cohort study give the opportunity to explore the medium- and long-term outcomes of adolescent pregnancy in Jamaica as time unfolds.

Acknowledgments

We are grateful to all the families who took part in this study, the staff in health centres and hospitals throughout Jamaica for their help during recruitment, and the JA KIDS team, of interviewers, computer and laboratory technicians, clerical and administrative workers, research scientists, volunteers and managers at the University of the West Indies (Mona). The Inter-American Development Bank (Grant ref: ATN/JF-12312-JA; ATN/OC-14535-JA) and the University of the West Indies, Mona Campus provided core support for JA KIDS. Secondary data analysis at the Centre for Maternal and Newborn Health in the UK was funded through a Global Health Grant (Project number OPP1033805) from Bill and Melinda Gates Foundation and WHO. Additional support was provided by the World Bank, UNICEF, the CHASE Fund, the National Health Fund, Parenting Partners Caribbean, the University of Nevada – Las Vegas, the University of Texas Health Science Centre at Houston and Michigan State University and its partners. This publication is the work of the authors listed, who will serve as guarantors for the contents of this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Inter-American Development Bank [ATN/JF-12312-JA,ATN/OC-14535-JA].