Stillbirths and neonatal deaths remain a huge challenge in the care of pregnant women, especially in developing countries [
Being cognizant of the distribution of stillbirths (fresh and macerated) and deaths within the immediate postpartum period may help to detect shortcomings in the quality of antenatal and obstetric care given to the pregnant woman, hence prioritize appropriate intervention programmes [
A cross-sectional retrospective study was carried out at Bansang General Hospital and Armed Forces Provisional Ruling Council (AFPRC) Hospital. These hospitals are located in the North and South bank of The Gambia, respectively, in two different health regions. Comprehensive EOC is available most of the time, mainly provided by Cuban Medical doctors. There is no Gambian medical doctor or obstetrician in any of these hospitals. The first class of medical doctors educated in The Gambia completed their education in 2007. The two hospitals serve a population of nearly 600,000 and are referral points for nearly 30 peripheral health centres and or dispensaries. Basic EOC is not available at any of these peripheral health centres. Thus, women in either the North Bank or South Bank with obstetric complications are referred to Bansang or AFPRC hospitals. Most of these women are referred during labour. The Government of The Gambia has adopted the primary health care (PHC) strategy to make health care more accessible to the rural population. Villages with more than 400 inhabitants have resident traditional birth attendants (TBA) who have eight weeks formal training in antenatal, intrapartum and postpartum care of the mother and child. These TBAs are being supervised by a community health nurse (CHN) who is in charge of a cluster of villages. Antenatal care is provided by mobile reproductive and child health clinics from the health centres and the two hospitals.
We used data from hospital records on all women who gave birth at or after 28 completed weeks of gestation from 1st July 2008 to 31st December 2008. Data was abstracted from maternity case notes, admission, and delivery registers. Midwives or doctors attending a birth complete a standardised form to be filled in upon admission and immediately after delivery. The form contains important information about maternal health and complications during pregnancy and the intrapartum period. It also contains information about the newborn. A precoded case abstraction questionnaire was used. Data abstraction was done by the principal investigator and assisted by research assistants, mainly midwives.
The main outcome measure was stillbirth rate calculated as deaths per 1000 births. The eligibility criteria was based on the World Health Organization’s (WHOs) international comparison of viability; that is birth weight of
For each birth, demographic and obstetric explanatory factors were captured. The demographic variables included maternal age in years, categorized in three groups:
A total of 1,849 maternity admissions were recorded during the six months period. We excluded 224 (12.2%) who had not delivered. Twenty-one births were further excluded due to missing information on the vital status and birth weight, and 25 who weighed less than 1000 g. We also excluded 60 deliveries that occurred before reaching the hospital. The final data set for this analysis was 1,519. The Ethics Committee of Norway and the Joint Gambia Government and Medical Research Council Review board approved the study. Permission to carry out the study was achieved from the Ministry of Health of The Gambia and the chief executive officers of Bansang and AFPRC hospitals.
All 1,519 institutional births were included in the analysis. Frequency analysis and cross-tabulations were used to determine the frequency and percentage of stillbirth and early neonatal mortality. Overall, stillbirth was calculated as a proportion of all births while early neonatal mortality was presented as a proportion of live births. Fresh and macerated stillbirths were calculated as a proportion of stillbirths. Multiple births were initially excluded, but repeating the analysis including multiple births the stillbirth rate remained largely unchanged. Thus we decided to maintain multiple births in the analyses. Univariate association between covariates and stillbirths were assessed with chi-square test or Fisher’s exact test as appropriate. All
The total number of deliveries over the six months period was 1,519. Of these, 237 were stillbirths, representing a stillbirth rate of 156 (95% CI 138–174) per 1000 births. Of the 237 recorded stillbirths, 137 (57.8%) were fresh stillbirth. We recorded 6 early hospital neonatal deaths giving a hospital neonatal mortality rate of 5 (95% CI 2–10) per 1000.
More than half (54.9%) of the women were between 20–34 years old. Forty-six percent were primiparous. Nearly all (99.3%) attended antenatal care at least once. Most of the women (89%) had a spontaneous vaginal delivery, 8% delivered by caesarean section, while 3.4% of the births were assisted breech delivery. Twenty-two percent of the women were referred from peripheral health centres. Of the 1,519 recorded deliveries the partograph was not used in 958 (63%). Overall, 53 (3.5%) of the women had a breech presentation at delivery (Table
Demographic/reproductive and obstetric factors associated with stillbirth.
Characteristics | Total birth | Stillbirth | Crude OR (95% CI) | Adjustedaa OR (95% CI) |
---|---|---|---|---|
Maternal age (yrs) | ||||
509(33.5) | 61(12.0) | 1 | 1. | |
20–34 | 827(54.4) | 140(16.9) | 1.20(0.80–1.81) | 0.97(0.57–1.65) |
183(12.0) | 36(19.7) | 1.80(1.14–2.83)* | 1.39(0.64–2.93) | |
Parity | ||||
0 | 703(46.3) | 77(11.0) | 1 | 1. |
1–3 | 380(25.0) | 66(17.4) | 1.31(0.92–1.86) | 0.48(0.28–0.78) |
436(28.7) | 94(21.6) | 2.24(1.61–3.10)*** | 0.35(0.20–0.63) | |
Residence | ||||
PHC Village | 473(31.1) | 108(22.8) | 2.13(1.59–2.80)*** | 1.14(0.78–1.66) |
Non-PHC Village | 1046(68.9) | 129(12.3) | 1 | . |
Admission status | ||||
Referred | 303(19.9) | 153(50.5) | 13.75(10.02–18.84)*** | 3.82(2.24–6.51)*** |
Booked | 1216(80.1) | 84(6.9) | 1 | 1. |
Antenatal Care | ||||
No | 10(0.7) | 5(50.0) | 5.50(1.58–19.16)** | 4.45(0.84–23.43) |
Yes | 1509(99.3) | 232(15.4) | 1 | 1. |
Mode of delivery | ||||
Spontaneous Vaginal | 1340(88.2) | 191(14.3) | 1 | 1 |
Assisted breech | 53(3.5) | 17(32.1) | 1.80(1.16–2.80)** | 1.64(0.74–3.60) |
Caesarean Section | 126(8.3) | 29(23.0) | 0.63(0.31–1.29) | 4.35(2.46–7.70)*** |
Used of Patograph | ||||
No | 958(63.1) | 187(19.5) | 2.48(1.78–3.45)*** | 1.70(1.13–2.56)** |
Yes | 561(36.9) | 50(8.9) | 1 | 1. |
Obstetric complication | ||||
Yes | 370 | 164(44.3) | 11.74(8.58–16.06)*** | 6.68(3.84–11.62)*** |
No | 1149 | 73(6.4) | 1 | 1. |
Type of birth | ||||
Multiple | 111(7.3) | 20(18.0) | 1.20(0.73–1.99) | 2.01(1.05–3.86)** |
Singleton | 1408(92.7) | 217(15.4) | 1 | 1. |
Birth weight | ||||
278(18.3) | 114(41.0) | 6.32(4.67–8.55)*** | 4.48(3.04–6.59)*** | |
1241(81.7) | 123(9.9) | 1 |
aaAdjusted for all variables listed in the table.
Table
The association between maternal demographic/obstetric factors and fresh stillbirths are presented in Table
Maternal demographic and obstetric factors associated with fresh stillbirth (FSB) (%).
Profile | FSB (%) | Crude OR (95% CI) | Adjustedaa OR (95% CI) |
---|---|---|---|
Age (years) | |||
54.1 | 0.72(0.39–1.32) | 0.76(0.27–2.15) | |
20–34 | 62.1 | 1.32(0.58–3.01) | 1.91(0.42–8.41) |
47.2 | 1 | 1 | |
Parity | |||
0 | 59.7 | 1.24(0.64–2.40) | 1.08(0.37–3.12) |
1–3 | 54.5 | 1 | 1 |
≥4 | 58.5 | 1.05(0.57–1.94) | 0.48(0.15–1.48) |
Recruiting hospital | |||
Bansang | 49.6 | 1 | 1 |
AFPRC | 67.9 | 2.15(1.26–3.66)** | 1.42(0.66–3.01) |
Residence | |||
PHC Village | 56.5 | 1 | 1 |
Non-PHC Village | 58.9 | 1.11(0.66–1.85) | 0.78(0.38–1.58) |
Antepartum admission | |||
Yes | 52.8 | 0.65(0.39–1.10) | 0.80(0.30–2.10) |
No | 63.2 | 1 | 1 |
Timing of complication | |||
Antepartum | 50.4 | 1 | 1 |
Intrapartum | 78.4 | 3.57(1.52–8.40)** | 3.14(1.01–9.76)** |
Partograph used | |||
No | 58.3 | 1.10(0.59–2.06) | 1.73(0.70–4.30) |
Yes | 56.0 | 1 | 1 |
Mode of delivery | |||
Spontaneous vaginalBreech/others | 55.5 | 1 | 1 |
Breech/others | 64.7 | 1.43(0.40–5.18) | 1.23(0.42–3.57 |
Caesarean Section | 72.4 | 2.15(0.91–5.11) | 1.32(0.32–5.45) |
Birth weight (grms) | |||
48.2 | 1 | ||
67.7 | 2.15(1.27–3.63)** | 1.67(0.81–3.44) |
aaAdjusted for all variables listed in the table.
Of the 1339 live births registered during the study period, 11 maternal deaths were recorded representing a hospital maternal mortality rate of 822/100,000 live births (LB). Of the 11 recorded maternal deaths, 7 (1,169/100,000 LB) and 4 (541/100,000 LB) were in Bansang and AFPRC hospitals, respectively.
The stillbirth rate found in the two rural hospitals in The Gambia was unacceptably high, pegging at 156 per 1000 total births. The reported early neonatal deaths rate was 5/1000 live births. Presence of severe obstetric complication showed a close association with stillbirth, followed by low birth weight, caesarean section, and referral from a peripheral health facility. Stillbirth was also associated with nonused of the partograph, multiple pregnancy, and lack of antenatal care. In addition, obstetric complications during the intrapartum period were independently associated with fresh stillbirths.
Even though hospital-based data has a limitation in the correct appraisal of the magnitude of a problem in the general population, lack of nationwide vital registration system in many developing countries including The Gambia, has made population-based studies unfeasible. Some field reports on stillbirths from Zimbabwe [
The reported stillbirth rate in the current study is higher than in a recent hospital-based study by Cham et al., 116 per 1000 births [
Unavailability and high cost of transportation, poor road conditions, and time to arrange for transport from remote villages may increase the time to reach a health facility [
Traditionally, advanced maternal age is viewed as risk factor for pregnancy complications and adverse perinatal outcomes including stillbirths [
Almost all the women in our study (99%) attended antenatal care at least once. However, the percentage of stillbirth was much higher among mothers who did not attend antenatal care compared with those who did. This is consistent with results of other studies carried out elsewhere [
Noncompliance in completing the partograph is common in Gambian hospitals. A higher percentage of stillbirths were observed in situations where the partograph was not used. In a recent paper by Cham et al. on foetal outcome in severe maternal morbidity, the partograph was not used in any of the 725 identified hospital deliveries [
Complications during pregnancy and childbirth have been long known to increase the risk of perinatal death. In our study a higher percentage of women admitted with obstetric complications lost their babies either during pregnancy, labour or shortly after delivery. This is consistent with findings by Cham et al. 2009 [
Most of the stillbirths in our study were fresh, with a fresh to macerated stillbirths ratio of 1.3 : 1. This indicates that most of the deaths probably occurred during labour, Fresh stillbirths are often used as proxy for stillbirths due to acute intrapartum insults [
The unmet need for obstetric care is high in developing countries where most of the intrapartum stillbirths take place [
Our findings suggest that the stillbirth rate is unacceptably high in rural hospitals in The Gambia. The findings also reaffirmed the important contribution of severe obstetric complications on the birth outcome in these settings. We also demonstrated an association between stillbirths and nonuse of the partograph, as well as with assisted breech and caesarean section. Most of the stillbirths were fresh, suggesting that these deaths have occurred during labour or shortly before delivery, thus potentially avertable. Improved intrapartum care through safe, comprehensive essential, and emergency obstetric supported by emergency transport services and skilled personnel is warranted for improved foetal out-comes in low resource settings such as The Gambia.
No competing interest. The authors are responsible for the content and writing of this paper.
Millennium development goals
Emergency obstetric care
Armed forces provisional ruling council
Primary health care
Traditional birth attendant
Early neonatal neonatal mortality
Comprehensive emergency obstetric care.
This project was funded by the Institute of General Practice and Community Medicine, University of Oslo, Norway. The authors’ sincere thanks goes to the Chief Executive Officers (CEOs)/Principal Nursing officers of Bansang and Armed Forces Provisional Ruling Council (ARPRC) Hospital, and the Regional Director, and Regional Health Team-NBD/W for providing office space and support during the data collection period. The authors also deeply appreciate the cooperation received from the data managers, nurses, and midwives in the two hospitals. The authors are deeply indebted to our research assistants and driver for making the data abstraction possible.