Seeking safety and empathy: Adolescent health seeking behavior during pregnancy and early motherhood in central Uganda

Introduction

Maternal mortality remains one of the most daunting public health problems in resource limited settings, and reduction in maternal mortality is number five of the Millennium Development Goals (MDGs) (Mavalankar & Rosenfield, 2005). This goal will not be met unless all countries create national maternal and child health services with universal access (WHO, 2005). Pregnancy and childbirth are still the leading cause of death and disease in women of reproductive age in low income countries. Improving maternal health is inextricably linked with promotion of gender equality and women's empowerment (MDG-3) (Shetty, 2006). This can be achieved through policies and programs which build women's capabilities, improve their access to economic and political opportunity, and guarantee their safety. Educating girls improves use of health services, reduces gender inequality and empowers women (Filippi et al., 2006; Grown, Gupta, & Pande, 2005). Long-term and sustained improvements in women's health require rectification of the inequities and disadvantages that women and girls face in education and economic opportunity. Gender equity and social transformation are likely to be achieved when men recognize that the lives of men and women are interdependent and that the empowerment of women benefits everyone (Shaw, 2006). Therefore, MDGs three and five ought to be promoted alongside.

Early sexual activities among adolescents (10–19 years) might lead to unwanted early pregnancies, which often result in pregnancy and/or delivery complications (Gispert & Falk, 1976; Hoyer, 1998). Other problems include complications arising from abortion and related psychosocial problems. In Uganda, the national adolescent health policy aims to streamline adolescent health concerns into the national development process in order to improve quality of life and standard of living (MoH, 2004a). The specific objectives of the policy include providing reference guidelines for addressing adolescent concerns and promoting dissemination of information concerning adolescent health. This would stimulate positive changes among individuals, community leaders and service providers. In the adolescent health policy it is stated that the proportion of mothers below 20 years who deliver in health facilities should increase from 48 to 80 percent and that pregnant school girls should continue school after delivery (MoH, 2004a). This is in line with the Ministry of Health vision to achieve the target in the Poverty Eradication Action Plan (PEAP) to reduce maternal mortality ratio by 30 percent (from 505 to 354/100,000 live births) by 2015. The implementation strategy includes scaling up adolescent friendly services and increasing links with the community through Village Health Team mechanism (MoH, 2004b). It is perceived that this would in the long run improve reproductive health in Uganda.

It is therefore important that adolescent health seeking behavior during pregnancy through delivery and early motherhood receive specific attention. Studies in Africa show that despite availability of services, many women including adolescents' book antenatal care (ANC) late in pregnancy and many attend only once thereby limiting the potential impact of quality of care (Westaway et al., 1998). A study in West Africa associated poor ANC attendance with more abortions unsafe and poor obstetric performance (Jimoh, 2003). In South Africa, obtaining the ANC card was mainly viewed important for enabling the mother in labor access to a public health facility for labor and delivery (Myer & Harrison, 2003). A number of factors like knowledge of the role of ANC, perceived health needs, nurse-patient relationship, economics and transport influenced attendance of ANC (Abrahams, Jewkes, & Mvo, 2001; Richter, 2000). A study done in Uganda indicates that interventions providing adolescent friendly services improved utilization of reproductive health services (Mbonye, 2003).

The theoretical approach advanced by Kroeger (1983) that guided this study emphasizes that perceived morbidity (e.g. pregnancy related morbidities) interacts with a complex network of explanatory variables (characteristics of: subject, disorder and service) that influence choice and use of health services. Additional factors such as the continuing process of cultural change, which include change of illness concepts and health behavior also influence the choice of health care (Kroeger, 1983). Within this theoretical approach, demographic and contextual factors are viewed as essential influences in which the pregnant adolescents' transition to motherhood occur (Kaiser, 2004). This study builds on existing work we did earlier. Therefore, some of the psychosocial factors influencing health seeking behavior and transition to motherhood specific to adolescents addressed in this study emerged from our previous study (Atuyambe et al., 2005).

Despite all these insights, little is known about reasons why adolescents in Uganda choose certain health services during pregnancy and delivery. The purpose of this qualitative study was to explore adolescent health seeking behavior during pregnancy and early motherhood in order to contribute to health policy formulation and improved access to health care. This would most likely have an impact on the reduction of morbidity and mortality among adolescent mothers and their newborns.

Methodology

Focus group discussions (FGDs) and key informant (KI) interviews were used to collect data. Focus groups are a qualitative technique that allows a small group of participants to discuss issues led by a moderator using a discussion guide (Lakshman et al., 2000; Patton, 2002). In addition, key informant interviews were also conducted. Key informants are people who, because of their position or experience, have knowledge on the topic of investigation (Newman, 2003).

This study was carried out in Wakiso district, central Uganda between March and May 2005. Although this district neighbors Kampala city, the capital of Uganda, it is mainly rural (UBOS, 2002). It has poor road infrastructure in some areas and this makes it difficult for the population to access health services. The district has a population of nearly one million with about one third being adolescents (UBOS, 2002).

Findings

Socio demographic characteristics of FGD participants

One third of the FGD participants were below 17 years of age, and majority (46%) had only completed upper primary school of education (the first 7 years in school). Sixty two percent were married or staying with partner as if married but none of them had professional employment as described in Table 1.

Table 1. Demographic characteristics of FGD participants.

Variable	Number n = 92	Percent
Age
16	11	12.0
17	17	18.5
18	33	35.9
19	31	33.7
Education
None	3	3.3
*P1-P4	2	2.2
P5-P7	42	45.7
**S1-S2	23	25.0
S3-S4	20	21.7
S5-S6	2	2.2
Marital status
Married	57	62.0
Not married	21	22.8
Separated	4	4.3
Occupation
None	42	45.7
Housewife	18	19.6
Peasant farmer	16	17.4
Petty trade	12	13.0
Other	4	4.3

• *P means Primary. P1 to P7 are the first 7 years in school for children in the Ugandan education system. **S means Secondary. S1 to S6 are the 6 years of school after primary education.

Our findings are presented according to the two themes that emerged: “feeling exposed and powerless” and “seeking safety and empathy”. The following six categories were identified: “dilemma of becoming an adolescent mother”, “lack of decision making power”, “cultural practices and beliefs about birth”, “expectations and experiences”, “transport, a key determinant to health seeking” and “dealing with constraints”. Quotations are used to illustrate and validate the categories. The themes, categories and codes are presented in Fig. 1.

Feeling exposed and powerless

Adolescents who considered themselves too young felt exposed as they gained weight due to pregnancy. They were in dilemma as they felt ashamed to meet their peers and feared to visit health facilities. Moreover, as the men dominated the decision making process, adolescents felt powerless as they lacked adequate financial and social support.

The dilemma of becoming an adolescent mother

Adolescent girls who conceive tend to live with a feeling that they are too young to manage the pregnancy. Additionally, they often feel that they might not get adequate support to go through the pregnancy and deliver healthy babies. Commonly, adolescent girls lacked stable relations with the father of the baby. In nearly all the FGDs and key informants, it was mentioned that the boys/men responsible for the pregnancy denied paternity.

The father of this child after making me pregnant denied it. So my mothers' relatives took care of me and I started living with my grand mother. …the baby's father lives in same village but does not give any support. (FGD Adolescent with child not married, Namayumba HC IV).

Paternity of children was contested by men/boys responsible for the pregnancy. This was mainly faced by the unmarried adolescents. Some adolescents could not tell who the father of the child was. This came about as a result of having unprotected sexual encounters with different men in the same time period. In addition, these men/boys were either too young or did not have a strong financial backup to take up family responsibility. This led to material and financial vulnerability, hence poor access to ANC and delivery services. This situation also led to a feeling of regret as they were not ready for having a child at this young age.

Adolescents were ashamed of being met by their peers who were still in school. They therefore preferred not to be seen until they had delivered. They found it convenient to avoid public places like health units and preferred to seek care at the traditional birth attendants' (TBAs) place where there was confidentiality and a limited possibility to meet many people.

Those who become pregnant while still in school fear to go for health care at the health units. They fear to get ashamed or meet their own colleagues in schools. Such girls would before pregnancy have been proud and calling themselves virgins, so they find that they cannot stand all that shame, so they decide to keep at home. (FGD married pregnant adolescent Namayumba HC IV).

It was emphasized that nearly all unmarried adolescents who become pregnant were not prepared for pregnancy and motherhood. Knowledge about material and psychosocial demands of motherhood were lacking and essential needs for both mother and baby like clothing, food, soap, advice and counseling were not available. Married adolescent received support from spouses in contrast to those who were not married.

Most men make the girls pregnant and runaway. You suffer with the pregnancy till you deliver without a single cloth or even the soap to bathe. …you can be without any one to advice about what is needed when one is pregnant. (FGD Adolescent not married, Tikalu HC III).

Lack of decision making power

Adolescent choice of health care during pregnancy was heavily influenced by partners or parents who could provide financial support. Results from both FGDs and KIs indicated that financial support was needed for transport and maternity requirements (delivery kit) for the mother and the newborn. In most cases, adolescents seemed to lack access to these requirements.

When I was pregnant what prevented me from seeking health care was lack of transport money because my legs were a problem. I used to live far away in the hills and I could not ask anyone to take me on a bicycle because I would be asked for money. So I decided to rely on my grandmother's traditional herbs. (FGD Unmarried adolescents, Namayumba HC IV).

The results showed that social factors influenced adolescents' possibility to seek health care in relation to their pregnancy. The choice of health care was mostly influenced by family members, partners or people the adolescent girl lived with during the pregnancy. The level of awareness among adolescents regarding health care was also found to be poor. At times family members said that there was no difference between traditional and western medicine. Some made herbal preparations in the home as indicated below:

Ok, the elders can convince you not to go to health facilities. …they tell you that they will cook for you traditional herbs, they also tell you there is no difference between western and traditional medicine. (FGD unmarried Pregnant adolescent, Wakiso HC IV).

Elders and grandmothers tended to refer to their own experiences of pregnancy and childbirth in order to convince pregnant adolescents that they did not need to seek biomedical health care. The quotes below highlight this view:

I have ever witnessed it, the adolescent was pregnant, she was our neighbor and the grandmother asked her that ‘how come we did not go to the hospital during our days. Didn't we deliver?’ and yet the adolescent really wanted to go to the hospital because she was in pain. (FGD Married adolescent with child Namayumba HC IV).

Old women can mislead a young pregnant girl and tell her that they had ten children in their life without going to health units for antenatal care. So such misinformation make some of us not go for antenatal care to the health facilities. (FGD married pregnant adolescent Buwambo HC IV).

Gender power dynamics play an important role in the choice of health care. Spouses of adolescents were reported to have absolute power in health care decision making of the adolescent mothers.

Your partner decides for you where to go. For example he can tell you to go to the government hospital and in case you get other complications he then decides to give you money to go to a better health centre. (FGD unmarried pregnant adolescent, Ndejje HC IV).

The father has absolute power over me and the baby. He can decide that today; don't take the baby back to hospital. You as a woman you have no way of opposing him because he has all the powers … he is the one who pays. (FGD, Married adolescent with child Tikalu HC III).

Seeking safety and empathy

In seeking safety, adolescent mothers' expectations were largely not met as they experienced lack of compassion at health facilities. They therefore resorted to the use of herbal remedies and were entangled in cultural beliefs. Improved transport system was apparent to improve their safety. Attempts to deal with these constraints thereby improving on safety of the mother and the baby were made.

Cultural practices and beliefs about birth

Some Ugandan cultural practices for childbirth include placenta rites. Like adult mothers, adolescents have a strong traditional belief surrounding the placenta. To be able to access the placenta was considered an important factor that influenced the choice of place for delivery. Often in public health facilities, adolescents were not given the placenta because it was considered biological/medical waste and was incinerated or thrown into a placenta pit.

For us the Baganda we undergo a traditional practice we call “OKUFUGIKA”2– The practice will make the child who stays on earth to be a clever person. But if they [health workers] throw the ‘second child’ [placenta] into a latrine the child who stays on earth become dull and not intelligent. (FGD married pregnant adolescent Namayumba HC IV).

On the other hand, TBAs were said to give the placenta to the mother who took it to perform some traditional rituals as indicated below:

If you deliver with the help of elderly women, they wrap the placenta so well. This helps the Childs' life and that of the mother. (FGD adolescents with child, Wakiso HC IV).

One of the practices also believed to be therapeutic is the use of ‘Emumbwa’3, which reinforces health seeking from TBAs and other herbalists. In some cases adolescents were able to distinguish a condition suitable for traditional medicine from conditions suitable for western medicine. A married adolescent described her experience in the following way:

I used clay [Emumbwa] when I was pregnant because my grandmother used to make it but when I saw that I had fever I then told my uncle and he gave me some drugs [tablets] to take. (FGD married adolescent with a child Namayumba HC IV).

In addition, some herbs were widely used as they were believed to be helpful in ensuring safe delivery. They were said to make the pelvic bones flexible ‘okumenya’4 making the delivery process easy and less painful. The route of administering herbal medicine was to bathe or smear the body.

They also [healers] give us herbal medicine, which is administered through bathing it mixed with water [in form of herbal bath]. It is important to take herbal bath when pregnant…. this medicine is got from the traditional healers. It refreshes body and helps to regain energy. (FGD Pregnant married adolescent Wakiso HC IV).

Some traditional medicine was perceived to be effective for specific conditions after delivery.

Sometimes health problems can best be treated by herbal medicine like some abdominal pains which we experience after delivery, our elders advise us to take some herbs. (FGD Married pregnant adolescents Wakiso HC IV).

Expectations and experiences

Respondents indicated that there was a relatively high degree of laxity among health workers with regard to patient care at the health unit. This discouraged adolescents to seek ANC and delivery services. Discussions showed that some health workers had a ‘don't care attitude’ and that they were sometimes rude and abusive to patients. These practices tend to de-motivate adolescents from utilizing public health units as described below.

They don't care for patients, for example when you go in the morning they will ask you “at your home don't you sleep”. When you go at lunch time, they will ask you whether at your place you don't take lunch. And when you go for treatment in the evening, they will tell you they have closed up. (FGD Married adolescent with child, Tikalu HC III).

Some adolescent mothers in labor had experienced physical violence by health workers. Besides, the behavior of the health staff sometimes made clients fear to express themselves or explain their health conditions as indicated below:

There is a health worker who is rough, when you get labour pains, and you tell her, she will just slap you. Where do you get the energy to push? They really slapped me from here. They told me to climb on the bed, I did not have the energy what she did she slapped me. (FGD Married adolescent with child Tikalu HC III).

There are health workers who are naturally rough, even when they are talking. And yet he is supposed to handle me very well, and I tell him my problems. But because he has shouted at me, I may fear to tell him my problem. (FGD Married adolescent with child, Tikalu HC III).

Good hygiene or lack of it at public health facilities appeared to influence the health seeking behavior. In some cases facilities were reported to be dirty with a bad smell, which made adolescents detest the services. They were worried that they might acquire infections from the health units.

No one would like to deliver in health units, which are dirty. Some maternity units have a bad smell; have dirty mattresses and blankets with lice. So these make us uncomfortable to deliver from, a mother can even get an infection from dirty beddings in such places. (FGD married pregnant adolescents Wakiso HC III).

This was reechoed by a key informant that emphasized the absence of basic needs like water, which could compromise hygiene and increase the risk of infection and spread of communicable diseases.

Although we have these water tanks here water is not accessible to patients. This water is rationed for conducting deliveries and cleaning the floor. …it is a problem keeping oneself [the mother] clean after delivery. No bathing. (KI Wakiso HC IV).

Perceived good quality of care served as an attraction when seeking health care at public health facilities. Good quality of care was said to be found in some units in the district. The key elements considered were consistency of good service, empathy from health workers, assurance of access to the unit day and night, and less delays on arrival. There was, however, a gap between the real situation at health units and the information prevailing in the community [rumors].

I had heard rumors that this health centre was dirty but when I came to deliver the beds were so clean and the floor was clean. One could sit down on the floor but before I came I was scared by people that it was dirty and I would acquire diseases. The health workers were caring and kept checking on me, I was counseled. (FGD married adolescent with a child Wakiso HC IV).

Transport, a key determinant to health seeking

The mode of transport to the health units posed a threat to pregnant adolescents' health. The common mode of transport in the district is by bicycle or motorcycles. In some areas the homesteads are far from regular routes of commuter taxis. The commonly used type of transport is uncomfortable to expecting mothers. Bicycles were described as either inappropriate or too slow to address the urgent needs of adolescents in labor. The remaining option was then home delivery or TBAs.

Due to labor pains, they [adolescents] cannot even sit on a bicycle. TBAs try to assist them but end up getting complications because they failed to reach the health centers. (FGD married adolescent with child, Namayumba HC IV).

The referral system is an important component in the maternal and child survival strategies. Key informants indicated that they received patients who required referral but lack of transport delayed care and this could lead to complications and even death.

You might receive a pregnant adolescent due in labor that needs an operation. This necessitates a doctor whom we do not have and therefore need to be referred. You know we even do not have an ambulance at this facility. (KI Health worker Tikalu HC III).

There is one who died from there due to delay in transport. She went to the TBA to deliver. She was told that her situation was a hospital case. So she died on the way to hospital. (FGD married adolescent with child, Tikalu III).

Dealing with constraints

In search for wellbeing, adolescents face situations that require a secure referral system in case of emergencies, financial security, clean and safe delivery places and competent health staff. Respondents mentioned that these elements were critical in influencing health seeking behavior.

There was an expressed wish for a good referral system that could assure safety in case of emergency. Among adolescents' worries was that they are prone to complications that might require referral.

I would like to deliver from health units where I am given good care, like quick attention or in case I fail to deliver, they give me ambulance vehicle to take me quickly to Mulago hospital [national referral hospital]. (FGD Unmarried pregnant adolescent Buwambo HC IV).

Most pregnant adolescents were said to visit the health units mainly to acquire an ANC card. Although their preferred choice of delivery was at the TBAs place they would want to be easily referred to the public health care system in case complications arose. These ANC cards gave a sense of security as indicated in the following quote:

People say that in case you have an emergency in the village, for example if you fail to deliver then you can come to the health centre and easily get quick treatment but if you do not have the antenatal card you are chased away. So it is a security measure. (FGDs Married adolescents with a child, Namayumba HC IV).

Financial considerations seemed to influence the health care seeking process. Money was needed to cover costs of drugs and supplies (gloves, registration book etc) and consultation (in case of private clinics). Extra money was needed in case of emergency or if the pregnant girl needed to be referred to another level of health care for better management.

The adolescents discussed their need for maternity dresses to accommodate their ‘expanding bodies’. Due to financial constraints, they could not afford these new clothes, which inhibited them from seeking ANC services. They feared to ‘look funny’ in very tight dresses or in their previous school uniforms. The discussions also indicated that if they were to deliver in public health units they would need ‘big shopping’ for their newborn babies. They feared to be laughed at by the peers and better-off women if they did not bring new clothes for the newborn, towel, baby coat and other gifts.

All these costs scared off adolescents who had no income but were dependant on parents, relatives, boyfriends and wellwishers. The traditional sector offered an attractive alternative care. Traditional birth attendants (TBAs) do not always require money in cash before they provide treatment. Adolescents were at times allowed to pay later when they got better and were able to pay. Traditional healers were said to be more socially and physically accessible. They live within the community, spend more time with the client, and were perceived to have good counseling skills.

Some go to deliver at the TBAs place because they do not have money for transport or pay at the private clinics. The TBAs do not charge much money and we do not need transport because they live within the community. (FGDs pregnant married adolescent, Wakiso HC IV).

The result showed that adolescents had different ways of dealing with monetary requirements related to pregnancy and childbirth and care. Pregnant adolescents were sometimes involved in mini trade like selling tomatoes, and subsistence agriculture (growing of cassava, vegetables etc). They also offered manual labor in the gardens to earn money.

Where I live my grandfather has much land with a forest, so I cut trees and burn charcoal for sell. I also grow cassava. If my child gets sick I can sell the cassava and get money to pay for his treatment. (FGD Adolescent with child, Namayumba HC IV).

It also emerged that adolescents got involved in commercial sex work (CSW) to earn money for survival as well as taking care of their babies.

We do not have jobs … have no income for survival and looking after our kids. … that is why most girls have ended up becoming pregnant early or turning out as commercial sex workers in Kampala. (FGD Married adolescents, Buwambo HC IV).

Key informants and FGDs recommended that the ‘girl child’ needs to be economically empowered in order to become selfreliant. This could start in small initiatives like gardening and rearing chicken.

In fact these mothers should be encouraged right at the age of 12 to start small projects like keeping chicken, growing vegetables like beans. This would earn the income to cope with the situation. … it is lack of money that encourages these adolescent girls to go to men. It is a major problem because they cannot afford to buy petroleum jellies, buy cloths. (KI Wakiso HC IV).

Adolescent mothers can make mats, baskets and knitting of table clothes which doesn't require bulks of money. (KI Kakiri HC III).

Also some pregnant adolescents try to rear chicken and other domestic animals which can be sold and earn money to meet those needs. … Some keep chicken which lay eggs that can be sold to buy soap. (FGD Married pregnant adolescents, Wakiso HC IV).

Discussion

The two themes ‘feeling exposed and powerless’ and ‘seeking safety and empathy’ seem to be very closely linked since powerlessness leads one to seek for safety and empathy. This study established that adolescent health seeking behavior (HSB) during pregnancy and early motherhood depends on the quest for safety and empathy from health care providers. Adolescent mothers feel ‘exposed and powerless’ due to the dilemma that they are not too young to conceive but too young to deliver safely and take on the motherhood role.

The dilemma of becoming an adolescent mother emerged as a strong factor affecting HSB. In some instances the girls felt too young to become mothers or lacked stable relationships. The feeling of guilt, shame and blame hindered them from seeking health care in the formal western system. Privacy and confidentiality especially for adolescents is an important issue in health systems as shown in other studies (Izugbara & Ukwayi, 2003; Langhaug et al., 2003; Webb, 2000). TBAs were seen as offering both the privacy and confidentiality adolescents required, and this appears to attract adolescents to the TBAs.

In the Ugandan context, being able to trace lineage is very important. Our findings show that at times unmarried adolescent girls are not able to identify a person responsible for pregnancy which leads to paternity being contested. This implies that the girls would not know who to turn to for social and financial support for themselves as well as the baby. This influenced their choice of health seeking.

Poverty and lack of power and decision making influence adolescent health seeking practices. When there is not enough money for transport and other expected requirements the immediate resort is traditional medicine and use of TBAs. In Uganda, every 150 people have access to a traditional healer. On the other hand the doctor patient ratio is 1:1800. Besides, elders dissuade adolescents from visiting health facilities basing their recommendations on their past positive experiences with the traditional sector. There is a strong belief in traditional medicine and this affects health seeking and utilization of biomedical service.

Our study results indicate that the socioeconomic position for women in society affects their decision making. There is gender imbalance and men/boys have absolute power in the household which affects the choice of health care. This has also been observed in South Africa where males dominate in decision making processes, which affects women and child health (Varga, 2003). This calls for financial and psychosocial support for adolescent mothers (Varga, 2001).

Access to transport is an important factor when the pregnant adolescents seek health care. The most available means of transport were bicycles and motorcycles. This mode of transport was viewed as inappropriate for pregnant women and even worse for a woman in labor. Although this mode of transport appears to be cheap (0.5 – 4 US$), it is not affordable for most adolescents since they do not have any income. This makes it much more convenient for the adolescents to seek care at the most available community resource persons, the TBAs, for treatment of pregnancy ailments and delivery. Mode of transport has been documented elsewhere as a key determinant to health seeking maternal services especially in low income countries (Myer & Harrison, 2003). Ambulance services at community and health facilities could avert the high maternal morbidity and mortality (Lule, Tugumisirize, & Ndekha, 2000; Orach, 2000).

Cultural practices and beliefs surrounding birth affect HSB. Cultural prescriptions in relation to placenta rites ‘repelled’ adolescents from seeking delivery services at the health units but acted as an attraction to visit TBAs and herbalists. The ‘don't care attitude’ and the tendency to be rude to clients by health workers was discomforting to clients. This has also been documented in central Uganda where health workers were not empathetic to adolescents (Atuyambe et al., 2005). Furthermore, unhygienic public facilities were perceived to be a source of infection and therefore unsafe. This might also hinder health seeking in public facilities.

Adolescents' sense of safety is an important factor for health seeking. Presence of functional referral system was viewed as panacea in case complications occur during pregnancy and delivery. Pregnant adolescents visited health units to obtain an ANC card. This card acts as a security in that if one visits the unit she is better received especially in case of emergency.

Factors under the themes ‘feeling exposed and powerless’, and ‘seeking safety and empathy’ specific to adolescents are viewed as direct influences on transition to motherhood. These factors contribute to and or increase understanding of individual differences in adolescent motherhood as identified by other literature (Clemmens, 2003; Rogan et al., 1997; Wigert et al., 2006).

Limitations of this study are that participants were recruited from clinic settings only and none from the traditional sector. However, our earlier study which reviewed experiences of pregnant adolescent mothers had participants from both the community and clinic setting, and this gave us insight into the community perspective.

Conclusions and reflections

Although pregnant adolescents seek health care both in the modern and the traditional health sectors in order to get safety and empathy, our findings give an impression that they mostly utilize the traditional sector. Adolescent mothers felt exposed and powerless, and in a dilemma as they were ashamed to meet their peers and also feared to visit health facilities. They were disempowered in decision making because of their pregnancy. The situation was aggravated by the fact that men and family members dominated the decision making process including where the adolescent should seek health care. The key factors influencing HSB include physical and economic access, dilemma becoming an adolescent mother, and cultural practices and beliefs. The problem of transport in terms of road-network and money for transport featured prominently. Health care practices that contravene cultural practices and beliefs e.g. handling of placenta are prohibitive to health facility utilization. We suggest that maternal health services for adolescent girls are improved. This can be done by training health care providers in youth friendliness. Specifically, health care providers' attitudes towards young pregnant girls need to be improved. Besides, infrastructure such as roads needs to be improved.