Teenage pregnancy is a major public health and social problem the world over and its incidence is on the increase^1,2. It constitutes a significant public health hazard especially in a developing country like Ghana and is a significant contributor to the present high maternal and child morbidity and mortality. The pregnant teenagers are at increased risk of pregnancy-induced hypertension, anemia, obstructed labor and its related complications^3-5. They are also three times more likely to die because of the complications of pregnancy and delivery than those aged 20-24^6,7. The fetuses from such pregnancies are prone to be delivered preterm or small for gestational age and have an increased risk of perinatal death^3,5,6.

The nutritional and health status of a pregnant woman is an important determinant of growth and development of the fetus and child even after birth.

Dietary inadequacies during pregnancy and lack of economic resources contribute to a high neonatal morbidity and low birth weight⁸ even on provision of an ideal environment and nutritional inputs⁹. Maternal undernutrition leads to smaller placental size and with fewer cells available for transfer of oxygen and nutrients to the fetus, leads to lower birth weight¹⁰. In addition, the risk of having deficiencies of iodine, folic acid and iron, which are essential during pregnancy, has serious consequences for the fetus¹¹.

Monotonous and inadequate diets are known to contribute to the burden of malnutrition and micronutrient deficiencies especially in developing countries¹². Despite the many approaches used to combat micronutrient malnutrition in such poor settings the problem remains unabated. However, recommendations around the world have pointed to the use of food-based strategies as a most sustainable way in meeting micronutrient needs¹³ and especially during pregnancy. Dietary diversification, which is an important component of the food-based approach, is critical in ensuring sustainable diets that allows the population, and especially the vulnerable groups (mainly women in reproductive age and children under five years), meet their nutrient requirements.

In Ghana, the Northern and Central Regions have the highest burden of teenage pregnancies, with about 23% of girls aged 15-19 years who have either had a live birth or are pregnant with their first child as against the national average of 13%¹⁴. Generally, there is a disparity between urban teenagers and their rural counterparts; 11% of adolescents in urban areas have begun childbearing, compared with 16 percent of those in rural areas, representing a reduction in the urban-rural gap in teenage childbearing from 7 percent in urban areas and 22 percent in rural areas since 2003¹⁵. Food and nutrition security vulnerabilities are also very high in the Northern Region of Ghana¹⁶ and the effects are borne mostly by the vulnerable poor most of who are teenagers.

Despite the many negative nutritional, health, social, and demographic consequences of teenage pregnancy, sufficient attention has not been paid in the area of scientific research to the socio-cultural and physiological factors that affect the nutritional status of pregnant teenagers, which are critical for successful outcomes of their pregnancies.

An understanding of the determinants of nutritional status among pregnant teenagers in Tamale Metropolis will go a long way to offer opportunities for formulating public health policies that would engender better recognition of the social, clinical and nutritional needs of adolescence especially during pregnancy and childbirth, in order to address them adequately.

This study was therefore aimed at exploring the many factors including dietary, which are critical in determining the nutritional status and birth outcomes of teenagers in the Tamale Metropolis.

Study Design
The design adopted for this study was a cross-sectional one to collect information from pregnant teenagers at one point in time. Pregnant teenagers were selected at the antenatal care centers in 7 suburbs of Tamale Metropolis. These suburbs were Nyohini, Bilpiela, Vitting, Sagnarigu, Choggu, Tamale central and Taha/Kamina. The choice of these centers was made as wide as possible to target both rural and urban teenagers in the Metropolis.

Sampling
The research subjects were recruited purposefully at the antenatal care centers in the Metropolis. The criterion for recruitment was based on age of respondents. Therefore, pregnant women who were in their teens were eligible for recruitment. Ages were obtained from antenatal attendance cards which were further confirmed or verified during the interviews.

In the calculation of the sample size, a 10% of the statistically determined minimum value of 274 was added to account for none responses and sample attrition. Thus, a sample size of 299 pregnant teenagers was selected for the study.

Data Collection and Instruments
Data on food intake and socio-demographic characteristics of respondents as well as socio-demographic information of respondents' caretakers were obtained by interviews using semi-structured questionnaires. Data was also collected on other variables namely; micronutrient supplementation, gestational ages of pregnancies and diets/foods intakes from the various food groups in the population (via a food frequency questionnaire).

The food lists for drawing out the FFQ was obtained from a 24 h recall of a sample of 20 respondents who met the criteria for recruitment into the study. Those foods that were rarely consumed were removed from the food lists.

Data Analysis and Presentation
The level of food intake was analyzed using the Food Consumption Scores (FCS) indicator developed by the FAO and WFP¹⁷. Eight different food groups; cereals and tubers, pulses, vegetables, fruits, meat and fish, dairy products, sugar and fats and oil were used to calculate the FCS. This was done by multiplying the frequency of consumption (in days) by the respective weights based on their nutrient density to obtain their consumption scores. The scores for each respondent were then summed up to obtain a total FCS as in the formula below. Their levels of food intake were then categorized using the thresholds provided by FAO.

The levels of food intake of the respondents were obtained with the food consumption scores. The score were used to categorize levels of intake as; acceptable, borderline and poor using the WFP/FAO thresholds as shown in the table below¹⁸. The frequency of respondents falling under each category was reported.

The dietary diversity of respondents was assessed using the Individual Dietary Diversity Score (IDDS). This score is defined as the total number of food groups consumed by an individual within a reference period (one day or a week). This indicator was used because it has been shown to have a considerably high accuracy in measuring dietary or nutritional adequacy among adolescents¹⁹. In line with the FAO guidelines for this indicator, the foods consumed by the respondents were grouped into one of twelve food groups²⁰. These groups are meats, fish, dairy products, eggs, cereals and roots, pulses and nuts, vitamin A rich vegetables, other vegetables, vitamin A rich fruits, other fruits, fats and oils and miscellaneous foods. These foods are grouped based on their nutrient qualities.

These two indicators of diet adequacy and diversity or quality were analyzed together for central tendencies such as means, mode and medians and measures of spread such as standard deviation. The confidence intervals were calculated at 95%. Spearman’s correlations and linear regression were also carried out to test the relationship between the FCS and IDDS. The chisquare test was carried out on the categorical variables to establish any associations between food intake and dietary diversity. The data was analyzed using SPSS (version 16.0)

Ethical Approval
This study received prior ethical approval from the Institutional Review Board of the School of Medicine and Health Sciences of the University for Development Studies. Local approval was also sought from each of the heads of the various health centers visited. Informed consent was also obtained from each respondent.

Socio-demographic Characteristics of the Pregnant Teenagers
The socio-demographic characteristics of the respondents are shown in Table 1 below. Majority of the respondents (71.6%) were nineteen years old whilst a few were under 19 years old. Over half of the respondents (55.3%) had no formal education, very few (5.1%) had just primary education, about a third (31.1%) had JHS education and a few (8.5%) had SHS education. With regards to marital status, majority (64.2%) were married whiles 34.5% were never married. Only a few of the respondents were living with their partners.

Table 1: FAO Thresholds for Food Consumption Scores (FCS)

Socio-Demographic Characteristics of the Caretakers of Pregnant Teenagers
Majority of the respondents (83.7%) had their caretakers being males. Most of the caretakers (63.5%) were between the ages of 26 and 35 years, a large proportion was married (80.6%) and had no formal education (70.3%) at all. For those with some form of formal education, majority had the most basic form of formal education (Table 2).

Table 2: Socio-Demographic Characteristics of the Teenagers and their Caretakers

The Socio-Economic Status of the Caretakers
Most of the caretakers (87.6%) were informally employed whiles just about 3.4% and 8.3% were formally employed and unemployed respectively (Table 3).

Table 3: Occupation of Caretakers

More than half (59.9%) of the respondent were being taken care of by their husbands, and more than a third (35%) by their parents. The others were taken care of by other relations as shown in (Table 4).

Table 4: Relationship of the Teenagers with their Caretakers

The Levels of Dietary Intake of the Pregnant Teenagers
Weekly Pattern of Dietary Intake
Staples (cereals, grains roots and tubers) were the highly consumed food group with a mean weekly frequency of consumption of 5.0 days. This was followed by fat and oils and sugars which had their mean weekly frequency of consumptions to be about 5 and 4 days respectively. The least consumed food groups were vegetables and fruits which both had equal mean weekly frequency of consumption (2 days). This is further shown in (Table 5).

Table 5: Descriptive Statistics for weekly food consumption and the Composite FCS

Levels of Food Intake
The mean composite FCS was 42.95. With respect to the levels of food intake based on the composite FCS, about half (51.7%) of respondents had acceptable levels whiles 27.2% and 21.1% were borderline and poor levels respectively (Figure 1).

Figure 1: Levels of food intake

The Dietary Quality of the Pregnant Teenagers
For the purpose of dietary diversity or quality, the staples again appeared to be highly consumed with a weekly mean of 6 days per week. Meats, pulses and nuts, other vegetables, vitamin A rich vegetables, vitamin A rich fruits and other fruits had comparable levels of consumption with a weekly mean of 3 days. Fish was the least consumed food group with a weekly mean of 1 day. The details are presented in (Table 6).

Table 6: Descriptive Statistics for Dietary Diversity

Individual Dietary Diversity Scores (IDDS)
The mean IDDS was 10.0 with a standard. The minimum IDD score of 7 was obtained by 10.2% of respondents and the maximum IDD score of 12 was obtained by 19.5% of respondents. Most of the respondents (32.7%) had Individual Dietary Diversity Scores of 11 (Figure 2)

Figure 2: Frequency of Individual Dietary Diversity Scores (IDDS)

The Relationship between Socio-Demographic Characteristics and Dietary İntake
None of the socio-demographic variables of the teenagers had any significant association with their dietary intake (both the levels of food intake and IDDS. Among all the socio-demographic variables of the caretakers, their gender, relationship with the respondents and occupation had significant associations with the level of food intake. However, none of the variables were associated with the IDDS.

Socio-Demographic Background of the Teenagers and Their Caretakers
About half of the pregnant teenagers were in the third trimester of their pregnancy which could result from the fact that the study was carried out at ANC outlets where pregnant women in developing countries are known to normally book late¹¹. The high proportion of those without education could be due to the fact that most of the respondents were coming from the surrounding rural communities in the metropolis where formal education is usually a challenge. This result is consistent with other studies that have reported the level of education to be negatively associated to the prevalence of teenage pregnancy^11,15. About twothirds of respondents (64.2%) were married, which is also consistent with several research findings including the UNICEF-sponsored International Planned Parenthood Federation (IPPF) report on child marriage²¹.

As shown in Table 1, many of the caretakers were between their late 20s to early 30s and were mostly males (83.7%) because most of the teenagers were married and were living with their husbands. Again, many of the caretakers (70.3%) had no education which further confirms the findings that education of the spouses and not only that of the teenager could have a negative effect on the occurrence of teenage pregnancies since those who are not in school tend to marry earlier than their literate counterparts²². Almost all the caretakers (87.0%) were in informal employment (Table 3).

Dietary Intake of the Pregnant Teenagers
Among all the eight food groups, which were used to estimate the levels of food intake with the Food Consumption Scores (FCS), staples (cereal grains, roots and tubers) were the highly consumed food group with a mean weekly frequency of 6 days. This was followed by Fats and sugars, with 5 and 4 days per week respectively as their mean weekly frequencies of intake, were the next mostly consumed food groups. The least consumed food groups were vegetables and fruits (Table 5). This pattern is consistent with dietary patterns of populations in developing countries which are mainly composed of staples and energy dense food groups (fats and oil) and a limited quantity of animal products and fresh fruits and vegetables^23,24. This presents a risk of deficiencies or excesses of several micro or macronutrients²⁵. The low consumption of fruits and vegetables as well as meat and fish presents a nutritional concern especially with respect to micronutrient adequacy of their diet, as these nutrients are essential during pregnancy especially among teenagers.

The mean Composite Food Consumption Score (CFCS) was 42.9, which is at the extreme lower end for the acceptable level of intake. Despite this lower mean CFCS, about half of the teenagers (51.7%) had an acceptable level of food intake based on the FAO/WFP CFCS thresholds. Not much difference was observed in the proportions at borderline (27.2%) and unacceptable (21.1%) levels of food intake (Figure 1).

Dietary Quality of the Pregnant Teenagers
With regard to the quality of diet which was measured with the IDDS, it was observed that out of a total of 12 food groups, the respondents consumed averagely from 10 different food groups (Table 6). This means that generally the respondents had good quality diet because previous studies have reported dietary diversity as an important element of a high quality diet^26,27. They therefore have a greater probability of meeting their energy and micronutrient requirements as Dietary Diversity Score have been reported by many studies to be correlated positively with energy and micronutrient intake^28-32.

The Effects of the Socio-Demographic and Economic Factors on the Dietary Intake
Among all the socio-demographic variables of the teenagers, none except the term of their pregnancies, had significant associations with their level of food intake (Figure 2 and Table 7). This association could be explained by the fact that these teenagers are mainly dependent on their caretakers and as such obtained their food from them. Those in their second trimester had highest proportion of those with acceptable intake. This could be due to the aversions and other non-diet friendly symptoms of early pregnancy such as nausea and appetite loss that often affect food intake during the first trimester³³. Also, within this period, the teenager is still in a transition from her normal life to life during pregnancy and as such the nutritional support offered to her due to the pregnancy is yet to start^11,34,35.

Table 7: Significant associations of some socio-demographic variables with level of food intake

The significant associations observed between gender of the caretakers and their relationship with the respondents (Table 7) could be due to the fact most of them were married and thus were staying with their husbands. Occupational status that was used as a measure of economic status of the caretakers was also significantly associated with levels of food intake. This is consistent with the age-old literature that occupation is associated positively with economic food acquisition or accessibility of households or individuals^36,37.

Even though the overall food consumption and dietary diversity or quality of the pregnant teenagers was acceptably high, a considerable proportion still fell within both the poor and borderline food consumption categories intake. The food consumption was observed to be significantly associated with the caretakers’ sociodemographic factors such as gender, relationship with the teenager and occupation.

These findings highlight the need consider the caretakers or spouses of pregnant teenagers in designing intervention programs and policies to improve the dietary intake of pregnant teenagers for the improvement and maintenance of their nutritional status during and after pregnancy. These programs and policies will go a long way to reduce the contribution of teenage pregnancy on maternal and infant mortalities and morbidities in the metropolis and the country and beyond.