Assisted Birth Delivery In Ghana: Evidence From Ghana Demographic And Health Survey 1988 - 2014

Background: Assistance during birth delivery is an essential reproductive care given to women during pregnancy. Even with the best possible antenatal care, any delivery can become a complicated one; therefore, skilled assistance is essential to safe delivery care. Proven health care interventions such as skilled assistance during childbirth can prevent or manage these complications. Methods: The study used secondary data from all the six rounds of the Ghana Demographic and Health Survey (GDHS). Bivariate, multivariate and binary logistic regression models were applied to examine the association between some background characteristics and assisted birth delivery. Results: The results showed that birth delivery is generally assisted by Nurses/Midwives over the period under review. The proportion of women who were not assisted during delivery declined over the period. Similarly, there was a decline in the percentage of women who were assisted by traditional birth attendants. In addition, it was observed that older women were more likely to be assisted by doctors during delivery as compared with younger women. Women with secondary education were more likely to utilize the services of nurses/midwives during delivery compared with those with no education. Women from the rural areas were more likely to receive delivery assistance from Traditional Birth Attendants compared with those from urban areas. Conclusion: These ndings heighten that future successful interventions for skilled birth attendants should prioritise the identied groups who were more likely to utilize the services of TBAs.

In the management of pregnancy and child delivery, TBAs frequently use their bare hands to examine the vagina and apply herbal medicines to the vulva or vagina to ensure the health of the growing fetus and safe delivery. However, some of these practices might cause genital infections including pelvic sepsis [8] which is one of the major causes of infertility, menstrual disorders and ectopic pregnancies [9]. Moreover, TBAs are responsible for management of family planning, getting opposite sex for a woman who is giving birth to the same sex, managing temporary impotence and infertility using traditional remedies [6].
Most obstetric complications that occur around the time of delivery cannot be predicted. Therefore, it is important that all pregnant women have access to the skilled attendant. Skilled attendance at delivery is advocated as the "single most important factor in preventing maternal deaths" [10]. The proportion of births attended to by skilled health personnel has been embedded in the Sustainable Development Goal (SDG) 3 target 3.1 -reducing the global maternal mortality ratio to less than 70 per 100,000 [11].
Strategies to improve maternal health are constrained by inadequate skilled birth attendants and underfunding of maternal and child health [12]. Other factors obstructing progress toward improving maternal health include barriers to accessing healthcare, inadequate essential supplies, lack of emergency transport systems and poor referral services [13].
The risk of women dying from pregnancy-related causes during their lifetime is 1 in 40 in Africa as compared with 1 in 3,300 in Europe and 1 in 190 globally [14]. An insigni cant number of the deaths (1%) occur in the developed countries as a result of availability and accessibility of quality of care by pregnant women in the developed countries. This quality care includes skilled attendant at the delivery point [12].
The government of Ghana introduced a maternal care exemption fee policy in 2007 to eliminate the nancial burden of access to maternal health services [15], enhance easy access to facilities, train more midwives, upgrade health assistant schools into midwifery schools, offer Community-based Health Planning and Services (CHPS) Program and train Community Health O cers (CHOs) as midwives to address the gap in skilled attendance [16,17,18,13,12].

Sources of Data
This study used six datasets from Ghana Demographic and Health Survey (GDHS) conducted in 1988, 1993, 1998, 2003, 2008 and 2014. The GDHS is a nationwide survey that is carried out every ve years since it began. The survey gathers information on fertility, family planning, infant and child mortality, maternal and child health as well as nutrition. The information on maternal and child health include Antenatal Care and delivery care. This provides adequate data for monitoring the population and health situation in Ghana.

Dependent Variable
The study equated assisted delivery at birth to an outcome variable. This variable was generated from the response to the question that assessed "the type of person that assisted women in the delivery of the child". Responses were grouped as women who were assisted by a doctor, nurse/ midwife, community health o cer, traditional birth attendants, relatives, friends, no one, and other.

Independent Variables
The study made use of nine independent variables, these are; maternal age, educational level, residence, marital status, wealth status, parity, occupation, ethnicity and region. Maternal age was categorized into 15-19, 20-24, 25-29, 30-34, 35-39, 40-44 and 45-49. The educational level was classi ed into women having No education, Primary education, Secondary education and Higher education. The type of residence was coded as Rural or Urban settlement.
Marital status was recoded into Married, never married, widowed, divorced and Cohabitation. The wealth status was grouped in ve (5) quintiles. These are Poorest, Poorer, Middle, Richer, and Richest. Parity was coded from a question that assessed the number of children a woman had ever given birth to prior to the current pregnancy. Responses were grouped into Null birth, One birth, Two births, Three births and Four or more births. The occupation was grouped into Not working, Managerial, Clerical, sales, agricultural, services and Manual workers. In addition, Ethnicity was classi ed as Akan (Asante, Akwapim, Fante and other), Ga-Adangbe, Ewe, Guan, Mole-Dagbani, Grussi, Gruma, Mande, and Others. The recoding of variables was done based on the literature reviewed.

Data Analysis
The data was analyzed using STATA version 13 with the use of both inferential and descriptive statistics. Some variables were recoded based on literature to suit the objectives of this study after which tables and graph were used for the presentation of the results. The results were weighted with the available sample weight factor (v005) within the GDHS data. Univariate, bivariate, multivariate and binary logistic regression analysis were employed.
Binary logistic was used to assess the effect of the independent variables on a speci c outcome variable. It is based on the assumption that the outcome variable is dichotomous in nature and also has no outliers. The outcome variables are captured in dichotomous variables and so helped to explain their associations with independent variables.

Background Characteristics Of Respondents
This section covers the socio-demographics of the respondents namely; age, residence, region, education, wealth status, marital status, occupation, parity and ethnicity as captured in the GDHS. The age of the respondents ranges from 15-49 years. The mean age of the respondents was 27 years in 1988 and 30 years in 2014. Table 3  With regards to the wealth status of the respondents, the proportion ranges from about 14 percent to about 25 percent (see Table 1 Table 1 indicates that the highest proportion of the respondents had four births or more. For example, about 49 percent of the respondents had four birth or more in 1988 and about 39 percent of the respondents had four births or more in 2014.  Table 2). Women from rural areas (OR = 0.50, CI = 0.40, 0.63), who have had four or more births (OR = 0.37, CI = 0.27, 0.52) and with having agriculture (OR = 0.63, CI = 0.45, 0.88) as their occupation have a lesser likelihood of being assisted by a doctor during delivery compared to those from urban areas, with one birth and those who are not working respectively (see Table 2).
It was also observed that women age 35 Table 2). Also, women with richest wealth status were seen to have a higher likelihood (OR = 1.72, CI = 1.42, 2.09) of being assisted by a nurse compared to those with poorest wealth status. The ndings also showed that women from rural areas (OR = 0.33, CI = 0.29, 0.38), with four and more births (OR = 0.46, CI = 0.37, 0.56) and those from mole Dagbani ethic group (OR = 0.68, CI = 0.56, 0.83) had a lesser odd of being assisted by a nurse compared to those from urban areas, with one birth and those from Akan ethnic group (see Table 2).  Table 2).

Discussion
The paper sought to examine the predictors of assisted birth delivery. Results showed that majority of the respondents over the years were assisted by nurses/midwives during birth delivery. The proportion of births assisted by nurses during child delivery increased over the years. Similar result was found in Garu Tempane District of Ghana where there was increase in the proportion of women who utilized the service of a skilled attendants such as nurses during birth delivery [19]. Hossain et al., (2020) also con rmed this in his study at Bangladesh. The study concluded that women receiving assistance from nurses increased from 69 percent in 1994 to 73 percent in 1999 [20].
Again, [21] posited similar ndings in their study. The study concluded that women receiving care from skilled attendants have had an upward trend over the years, increasing from about 25 percent in 1996 to about 58 percent in 2011. There was a reduction in the proportion of birth assisted by Traditional Birth Attendants over the years (see Fig. 1). This nding goes contrary to the nding of [22]. They found out that more than half of the births occurred in the home with the assistance of TBAs. This might be as a result of the distance to the health facility or the cost of delivering in a health facility. The increased in the proportion of birth assisted by nurses can be as a result of the introduction of the CHPS compound in the country. This brought health services to the doorsteps of the citizenry, especially, those in the rural areas.
This section sought to examine the proportion of delivery that was assisted by doctors from 1988 to 2014. It was observed that the proportion had increased in 2014. This con rms the study by [23], that there has been an increase in birth delivery assisted by doctors in developing regions. Similarly, [24] posited that utilization of skilled birth attendants such as doctors was low from 1988 but increased sharply and levelled off after 2003. However, [25] in their cross-sectional study conducted among women who had ever given birth in the three years prior to the study concluded about 79 percent of respondents were assisted by skilled attendants such as doctors. This might be as a result of the introduction of the free maternal health care policy in the country.
The result also shows that older women were four times more likely to utilize the services of doctors compared with younger women. Similar results were found by the work conducted [7]. He concluded that younger women do not utilize the services of doctors because they do not have power in decision making in their homes even though it affects their own health. [26,27] gave different opinion in their study conducted in rural Tanzania that younger mothers utilized the services of doctors during birth delivery [27]. This might be because of the fear of dying from pregnancy if not attended to by a skilled worker. This result is in line with the Conceptual framework that, the predisposing factors such as age has an association with delivery assisted by doctors [28]. Women working in the clerical sector were twice more likely to utilize the services of doctors during delivery compared with those not working. This is consistent with the study conducted at Bangladesh by [29] which found that women who worked for cash went to quali ed medical personnel such as doctors for assistance during birth delivery. [17] in their cross sectional descriptive study at Ga East Municipality came out with similar ndings.
It shows clearly that respondents with higher education were four times more likely to be assisted by doctors during birth delivery compared with those with no education (OR = 4.54, p < 0.001). A cross sectional study conducted at Makueini, Kenya concluded that mothers with higher education were assisted by skilled attendants such as doctors during birth delivery [30]. Also, study conducted at Bangladesh posited that the odds of utilizing the services of skilled attendants such as doctors increased with mother's education [31]. Mpembeni et al., (2007) in a study conducted at Southern Tanzania concluded that years spent in school showed a signi cant association with seeking skilled care such as doctors during birth delivery [32]. Women with more schooling years had the highest proportion of deliveries attended by doctors.
Women with a child or less were more likely to utilize the services of doctors during birth delivery. Chakraborty et al. (2003) posited similar nding that the percentage of women who sought care from a doctor decreased from 29 percent in women with no pregnancies to about 26 percent in mothers with one to four births [29]. Also, [33] a rm that Skilled Delivery Attendance was higher among women who were primigravida. Respondents from the rural areas were less likely to utilize the services of doctors during birth delivery. This con rms the study conducted in Ghana from 1988-2008 by [24]. They concluded that there is a percentage gap between women in the rural and urban areas in accessing the services of skilled attendants such as doctors during delivery.
This section sought to analyze birth delivery assisted by nurses/midwives from 1988 to 2014. The results show that proportion of birth delivery assisted by nurses/midwives has increased over the years. This con rms the cross-sectional study by [25], that among women who had ever given birth, the proportion of birth assisted by nurses/midwives had increased to about 79 percent. Also, the study by [34] conducted at Kenya con rms this nding. On the other hand, [24] concluded that the proportion of birth assisted by skilled birth attendants such as nurses/midwives has been low. This they attribute to the attitude of nurses/midwives towards patients at the facility.
It was also realized that older women were two times more likely to utilize the services of nurses/midwives compared with younger women during birth delivery. This is in line with the study by conducted by [7]. He concluded that younger women do not utilize the services of nurses/midwives because they do not have power in decision making in their homes even though it affects their own health. Also, a bivariate analysis suggested that older women were more likely to utilize maternal healthcare services than younger women [29]. Contrary to this, [35] concluded that younger women utilized the services of skilled attendants such as nurses/midwives because they are just starting child bearing and they are in a high risk so they tend to fear delivering without assistance.
Compared with women in the Western Region, women from the Greater Accra Region were more likely to utilize the services of nurses/midwives during birth delivery. In a study conducted at Nigeria con rms that Region was statistically signi cant factor that affected the use of skilled delivery [36]. A study by [37] posited different nding when they concluded that Region was not found to predict skilled attendance during delivery in their study in Ghana Using the 2008 Ghana Demographic and Health Survey. With reference to the occupation of the respondents, it was clear that women working in the clerical sector were two times more likely to utilize the services of nurses/midwives compared with those not working.
This con rms the study by [17] conducted at Ga East Municipality in Ghana that maternal occupation showed statistical association with the assisted delivery. Also, [38] in their study at Nigeria concluded that women who were working in a modern-sector occupation had 3.66 odds of receiving assistance from skilled attendants such as nurses/midwives during birth delivery.
The ndings show that women with higher education were twice more likely to utilize the services of nurses/midwives compared with those with no education. A study conducted at Southern Tanzania concluded that years spent in school showed a signi cant association with seeking skilled care such as doctors during delivery [32]. Women with more schooling years had the highest proportion of deliveries attended by doctors. [31] concluded in multivariate regression that the odds of being assisted by nurses/midwives during delivery increased with mother's education. Also, a study in Kenya [34], Sierra Leon, Niger and Mali [12] and Ghana [13] established that women with higher education were more likely to use skilled attendant such as nurses/midwives.
Finally, this section sought to assess birth delivery assisted by traditional birth attendants from 1988 to 2014. According to [39], unskilled birth attendants are also called traditional birth attendants which is a community-based provider of care during pregnancy and childbirth. In general, there has been a decline in birth assisted by traditional birth attendants in Ghana. This is contrary to the report from the Nepal Demographic Health Survey which reported that about 81 percent of deliveries are assisted by unskilled birth attendants. Also, a report by [40] also concluded that most women prefer to have a Traditional birth attendant at delivery because they perceived pregnancy and child birth as a natural phenomenon and does not require formal health services. Again, a study conducted in Talensi District of Ghana by [41] among women, TBAs and key informants stated majority of women were persisting in utilizing unskilled birth attendants largely because they believe services provided by unskilled birth attendants are good for them. The high usage of Traditional birth attendants during birth delivery will be as a result of ignorance, availability of TBAs and poor access to skilled services [41,42].
It is evident that women from the Northern region were twice more likely to utilize the services of TBAs during birth delivery compared with those in the Western region. This is in line with the study by [25] conducted at Northern part of Ghana. The study concluded that respondents from these places were more likely to utilize the services of Traditional birth attendants during birth delivery. In a study conducted to identify demographic, maternal and community predictors of skilled attendance at delivery among women, region was found to be associated with delivery assisted by traditional birth attendants [37].
Women who work in the agricultural sector are more likely to utilize the services of traditional birth attendants during delivery compared with those not working. This contradicts the study by [17] conducted at Ga East Municipality in Ghana that maternal occupation showed statistical association with the utilization of assisted delivery. Also, in a study at Nigeria concluded that women who were working in a modern-sector occupation had 3.66 odds of receiving assistance from skilled attendants such as nurses/midwives during birth delivery [38].
Some studies seem to suggest that rural women utilize the services of traditional birth attendants during birth delivery [24,33]. Consistent with these previous studies, the result shows that women from the rural areas were thrice more likely to utilize the services of traditional birth attendants during birth delivery compared with women from the urban areas. Women with no education were more likely to utilize the services of traditional birth attendants during delivery compared with those with higher education. This con rms the work of [31], that women with little or no education were more likely to utilize the services of TBAs during birth delivery. Also, a study at Kenya concluded that women with no education were more likely to use traditional birth attendants during birth delivery [30].