Quality of Caesarean Sections in Rural Sub-Saharan Africa: A Prospective Study in Cameroon ()
1. Introduction
The world Health Organisation (WHO) has acknowledged that poor quality of care increases maternal and newborn morbidity and mortality and envisions a world where “every pregnant woman and newborn receives quality care throughout the pregnancy, childbirth and the postnatal period” [1] [2]. Worldwide, caesarean section (CS) is the most common obstetrical surgery and a major reproductive health intervention to improve maternal and foetal outcomes if appropriately performed [3]. In Sub-Saharan Africa (SSA), CS rates despite a slight increase over the past two decades remain below the WHO target (5% - 15%) and several studies have depicted several shortcomings in terms of quality: appropriateness, safety, access, timeliness, effectiveness and efficiency [3] [4] [5] [6]. As a result, maternal and perinatal mortality rates following CS in Africa are 50 times higher than in high-income countries [4] [7]. That tremendous reality must be addressed through adequate research-informed policies.
There is a significant research gap on the quality of CS in resource poor Sub-Saharan African settings with widespread task shifting in obstetrical care [3] [8] [9] [10]. In Cameroon, available literature on CS have reported low rates and poor maternal-foetal outcomes but so far no study has specifically assessed the quality of CS [11] [12] [13] [14] [15]. In view of filling that research gap and in line with the WHO’s commitment to improve quality of maternal and newborn care in health facilities, this study’s primary objective was to assess the quality of CS in a rural setting in a country where the overwhelming majority of literature on CS is based on urban health facilities. The secondary objective was to describe accessibility, pre-operative assessment, operative and post-operative course of CS in this setting.
2. Materials and Methods
2.1. Design and Setting
This quality assessment CS study was conducted from February 1, 2020 and April 30, 2020 in three hospitals in the Menoua Division in the West-Region of Cameroon (Dschang District Hospital, the Penka Michel District Hospital and Saint Vincent de Paul Catholic Hospital in Dschang) which is mainly rural and has over 300,000 inhabitants. Two of those health facilities are state-owned: the Dschang District Hospital (DDH) and the Penka Michel District Hospital (PMDH) and the third is faith-based: the Saint Vincent de Paul Hospital in Dschang (SVPH). Those three health facilities manage most of obstetrical emergencies in the Menoua Division.
The maternity of the DDH has a capacity of 27 beds and is led by an obstetrician-gynaecologist assisted by 5 general practitioners, 6 midwives and 5 nurses. The surgical theatre is available round the clock every day and anaesthesia is delivered by trained and/or certified nurses. The DDH’s paediatric department has a neonatal unit and is led by a paediatrician. The multifunctional biology laboratory is led by a senior biological engineer. The pharmacy is run by 3 pharmacists and the hospital doesn’t have a blood bank. The yearly average number of deliveries was 1412 from 2016 to 2019.
The maternity of the SVPH has a capacity of 16 beds and is headed by a general practitioner assisted by 2 midwifes and 4 nurses. The surgical theatre is available round the clock every day and anaesthesia is delivered by trained and/or certified nurses. The SVPH’s paediatric department is led by a general practitioner and doesn’t have a neonatal unit. The multifunctional biology laboratory is managed by a biological technician, and carries out only basic tests. The pharmacy is run by a community worker and there is no blood bank. The yearly average number of deliveries was 822 from 2016 to 2019.
The maternity of the PMDH has a capacity of 11beds and is headed by a general practitioner assisted by 1 midwife and 1 nurse. The surgical theatre is available round the clock every day and anaesthesia is delivered by trained and/or certified nurses. The PMDH’s paediatric department is led by a general practitioner and doesn’t have a neonatal unit. The biology laboratory is managed by a biological technician and carries out only basic tests. The pharmacy is run by a community worker and there is no blood bank. The yearly average of deliveries was 132 from 2016 to 2019.
2.2. Participants
Sampling was exhaustive and consecutive. We prospectively included all women who delivered by caesarean section (after 28 completed weeks of pregnancy) and received postoperative care in the study hospitals between February 1, 2020 and April 30, 2020. The quality of caesarean sections was assessed according to the analytical framework of Dujardin and Delvaux (1998) based on its four pillars (access, diagnosis, procedure, postoperative care of CS) [8]. The quality of caesarean section was classified as acceptable (very good score: 18-20/20 and good score: 17-16/20) or unacceptable (poor score: 14-15/20 and very poor score ≤13). Maternal complications (present or absent) and foetal outcome (Apgar score ≥7 over 10 and Apgar score < 7 over 10) were noted.
Data were collected from hospital records, patients’ files and face-to-face interviews. Five categories of variables were retrieved: 1) sociodemographic and obstetrical characteristics (age, level of formal education, habitat, marital status, gravidity, parity, gestational age, total number and timing of antenatal consultations, number and content of obstetrical ultrasounds and number of uterine scars), 2) operative and peri-operative variables (pre-operative assessment: blood pressure, pulse rate, temperature, symphyseal-fundal height, foetal presentation, foetal heart rate, cervical dilatation, state of the foetal membranes, maternal weight and height, estimated foetal weight, clinical pelvimetry, use of partogram, Robson’s class, qualification of personnel who indicated the surgery; operative variables: total duration of the caesarean section, incision-extraction duration, availability of the surgical kit, availability and composition of the operative team, type of anaesthesia, qualification of the surgeon, qualification of the anaesthetist, supplementary procedures; postoperative parameters; effectiveness and report of nursing care, delays in caregiving, missed care, duration of postoperative hospital stay, duration of total hospital stay, availability of blood donors, blood transfusion), 3) geographic and financial accessibility to caesarean section (means of transport to the hospital, distance to hospital, mode of admission into the hospital, availability of users’ fees for CS, the overall cost of the hospital stay) and 4) maternofoetal outcomes (Apgar’s score, birthweight, early neonatal complications, early maternal complications, maternal and neonatal death). We used the Dujardin’s model and selected criteria from context-relevant literature to set the quality score (Table 1) [8] [9] [10].
2.3. Statistics
Data were entered into a predefined excel sheet (Microsoft Excel 2010 software) analysed using SPSS® (Statistical Package for Social Sciences) 25 software to generate tables of frequencies or 2 × 2 tables. Khi2 was used when necessary to compare proportions or Student-test (t-test) when necessary to compare means (replaced by Wilcoxon-test for abnormal distribution). Association between variables was measured using bivariate analyses followed by a multivariate analysis including independent variables having a P-value < 0.05 on bivariate analysis.
2.4. Ethical considerations
The study was approved by the institutional ethics board of the University of “Montagnes” at Bangangté in the West region of Cameroon. Investigators strictly complied with the Helsinki guidelines on research involving human subjects (written informed consent, anonymity, administrative clearance and confidentiality were compulsory).
![]()
Table 1. Quality score for cesarean section.
aN: number of indicators; b8 clinical parameters (blood pressure, pulse rate, temperature, symphyseal-fundal height, fœtal presentation, fœtal heart tone, cervical dilation, state of the foetal membranes); c6 staffs (surgeon, assistant-surgeon, anaesthetist, assistant-anaesthetist, midwife, scrub technician); dAppropriateness of the CS indication was determined by and independent Gynaecologist-Obstetrician working in the region where the study was conducted.
3. Results
During the study period, there were 538 deliveries and 120 (22.3%) CS. Figure 1 describes the participants’ inclusion process. Seventy-eight of the 120 (65.0%) CS were emergency procedures and 83 (69.2%) of CS were recorded in state-owned hospitals (Dschang District Hospital and Penka-Michel District Hospital) and 37 (30.8%) were registered in the faith-based hospital (Saint Vincent de Paul Hospital of Dschang).
Socio-demographical and obstetrical characteristics of participants
Table 2 describes the characteristics of participants. Mean (standard deviation (SD)) age was 27.6 (6.3) years with extremes of 17 and 42 years. Participants between age 15 - 24 years accounted for 41.7% (50/120) of the sample size. The level of education of participants was secondary for 51.7% and tertiary for 47.5% of participants. The mean pre-operative parity (SD) was 1.56 (1.48) and primiparous accounted for 21.7% (26/120) of the sample size. The mean (SD) gestational age was 38.7 (2.6) with extremes of 31 and 43 weeks. One hundred and fourteen (95.0%) participants did antenatal consultation during the third trimester and 38.4% (46/120) had a previous caesarean section. The following Robson’s group were predominant: group 5 (35.0%), group 1 (17.5%) and group 3 (17.5%).
Access to caesarean section and pre-operative assessment
Table 3 illustrates assessments for access to caesarean section and pre-operative clinical evaluation of participants, 11.5% (14/120) of participants lived more than 15 kilometers away from the hospital. Motorbikes were the mode of transport to the maternity for 72.5% (87/120) of participants. None of the women was carried to the hospital with an ambulance. Patients and families who could totally afford the fees for caesarean section constituted 36.7% (44/120). After the
![]()
Figure 1. Selection of participants in the three hospitals.
![]()
Table 2. Characteristics of participants
PMDH: Penka Michel district Hospital, SVPH: Saint Vincent de Paul Hospital, DDH: Dschang District Hospital, ANC: Antenatal Consultation.
![]()
Table 3. Accessibility to caesarean section and pre-operative diagnosis.
PMDH: Penka Michel district Hospital, SVPH: Saint Vincent de Paul Hospital, DDH: Dschang District Hospital.
indication of CS, none was delayed for because of lack of user’s fees for CS. The quarter of participants 31 (25.8%) were referred from lower tier health facilities. Pre-operative clinical assessment was properly documented for 26 (21.6%) cases and was done by a general practitioner in 80.9% (97/120) of cases. All the indications of CS were appropriate.
Surgical characteristics of caesarean sections
Table 4 describes operative CS parameters. The average (SD) duration of CS was 40.1 (9.2) minutes ranging from 20 to 60 minutes. Mean (SD) foetal extraction time (skin incision – foetal extraction) was 7.57 (1.99) minutes. The caesarean section set (material for surgery and anaesthesia) and the operating room were immediately available in 100% and 99% of cases respectively. The surgical team was not complete (<6 staffs) in 56 (43.3%) cases. Anaesthesia was done by nurses trained on daily anaesthesia practices without a diploma in 17 (14.2%) cases and by certified nurse anaesthetist in 103 (85.8%) of cases. CS was done by a general practitioner and by nurses in 86 (71.7%) and 14 (11.7%) cases respectively. None of the study hospitals had a blood bank and blood donors were
![]()
Table 4. Assessment of surgical parameters of caesarean sections.
alocal anaesthesia performed to dissect the abdominal wall till the peritoneum then general anaesthesia as from the hysterotomy (in view of shortening the duration of exposure of the foetus to ketamine). PMDH: Penka Michel district Hospital, SVPH: Saint Vincent de Paul Hospital, DDH: Dschang District Hospital.
available during the surgery for 1.6% (2/120) cases. Anaesthesia was spinal in 80.0% (96/120) of cases. Laparotomy was done according to the Pfannenstiel’s technique in 97.5% (117/120 of cases and the time from skin incision to foetal extraction was below 10 minutes in 95.0% (114/120) of cases.
Post-operative care and materno-foetal outcomes
Table 5 summarizes the post-operative care and the materno-neonatal outcomes. Postoperative nursing care was compliant with medical prescription in 95.8% (115/120) cases. Delays in caregiving were observed in 1.7% (2/120) of cases. Intensive immediate postoperative care was not effective in 45.0% (54/120) of cases. Apgar score was below 7 over 10 in 14.2% (17) of cases of which 6 freshes still births. No maternal death was recorded and maternal complications were recorded in 11.7% (14) of cases.
Caesarean section quality scores
Figure 2 and Table 6 show classes of quality scores and mean quality scores respectively. Quality scores were unacceptable for 55.3% of caesarean sections (scores ≤ 15/20). The mean (SD) quality score for the whole sample was 16.33 (1.60). The mean (SD) global quality score in state-owned hospitals was significantly lower than in the faith-based hospital: (15.59 (1.24) versus 18.00 (0.91); p < 0.001). The Faith-based hospital had 2.7% unacceptable quality CS against
![]()
Table 5. Assessment of postoperative care and materno-foetal outcome of caesarean sections.
PMDH: Penka Michel district Hospital, SVPH: Saint Vincent de Paul Hospital, DDH: Dschang District Hospital.
78.3% in state-owned hospitals.
Partial quality scores were also significantly higher in faith-based hospitals than in state-owned hospitals: pre-operative assessment, surgical parameters and post-operative care (Table 6). Factors associated with poor quality scores are summarized in Table 7. On bivariate analysis the following factors were associated with unacceptable quality of caesarean section: state-owned hospital (cOR (95%CI): 130 (16.66 - 1014.31); total costs of CS < 180 USD (cOR (95%CI)): 28.97 (9.81 - 85.56), surgery done by a nurse: (cOR (95%CI)): 15.78 (1.23 - 27.08); incompleteness of the surgical team (cOR (95%CI)): 2.85 (1.33 - 6.09); general anaesthesia (cOR (95%CI)): 3.62 (1.37 - 11.47) and Group 5 of the Robson’s classification (cOR (95%CI)): 1.54(0.72 - 3.32). After adjustment, only state-owned hospitals (aOR (95%CI): 92.30 (6.33 - 1345.15)) remained associated with unacceptable quality scores.
![]()
Figure 2. Distribution of patients by clusters of quality scores of CS and health centers.
![]()
Table 6. Distribution of caesareans following mean quality scores.
![]()
Table 7. Factors associated with unacceptable quality of caesarean section.
USD: United States Dollar; ANC: Antenatal consultation.
4. Discussion
Overall, more than half of caesarean sections had unacceptable quality in rural setting at west Cameroon. The CS rate (22.3%) in this study is 10-fold higher than the national average, 20-fold higher than the rural average and similar to CS rates noted in hospitals in the urban Cameroonian areas [3] [6] [11] [13] [14] [15]. This high rate of CS in this milieu could be explained by the referral status of the study hospitals. Like in others studies in Cameroonian hospitals, a high proportion (25.8%) of women who underwent CS in our study were referred from other health facilities [11] [13] [14] [15]. Similar CS and referral rates were reported in a rural referral hospital in Tanzania [16]. The high proportion of unacceptable quality of CS we found is similar to those reported in several sub-Saharan African (SSA) countries where it is (inter alia) responsible of the high CS associated maternal and neonatal morbidity [1] [17] [18]. This could be explained by the fact that in SSA countries, there is poor accessibility of CS, poor medical equipment and lack of qualified health professionals.
Sub-group quality scores were significantly different for pre-operative diagnosis, surgical procedure and resources and post-operative care. Though preoperative clinical assessment was completely documented for only one fifth of women in our study, all decisions of CS were appropriate. This is quite different from a large-scale study in Burkina-Faso that reported 24% of non-medically (inappropriately) indicated CS [19]. Determinants of inappropriate CS in this Burkina-Faso study were (inter alia) urban residence and decision to CS by general practitioners (GPs). This difference from our results could be explained by the rural residence of all our participants. Besides, considering that four fifths and one fifth of CSs in our study were indicated by GPs and obstetricians respectively, we can hypothesise that long-term lone practice in rural resource-poor setting has improved diagnostic skills of our rural GPs above those of urban GPs in the Kabore’s series. We can also hypothesize that the very low acceptability of CS in our rural settings led to a selection of parturients such that only or mostly those with absolute indications of CS underwent surgery.
We found that surgical theatre, CS sets and staff were always readily available although blood and blood donors were generally not available. Therefore, challenging the comprehensiveness of emergency of obstetrical care in the study hospitals. This finding is common in SSA with poor medical equipment and lack of qualified health professionals. A striking finding is the difference in staffing of surgical theatre between state-owned and faith-based hospitals. Indeed, the minimum number of surgical staff for CS was not met in more than half of cases in state-owned hospitals versus 10% of cases in the faith-based hospital. Moreover, general practitioners performed more than two thirds of CSs overall; nurses did 11.7% of CS. Similarly, anaesthesia was conducted by certified nurses in more than four fifths of cases overall and non-certified nurses conducted anaesthesia in 14.2% of CS. Similarly, Kruk et al. reported that in districts hospitals of 3 SSA countries, mid-level staff provided more than a third of CS and anaesthesia thus confirming that planned task shifting in surgical practice is widespread in SSA [5] [20]. If task shifting is ineluctable to address the huge gap in skilled staff for comprehensive obstetrical care, appropriate training and monitoring are paramount to ensure safe practice and good outcomes of CS inter alia. The higher qualification and availability of staff during surgery in faith-based hospital may explain the lower complication rates we observed (Apgar score < 7: 13.5% versus 27.7%; maternal complications: 2.7% versus 15.7%). Overall maternal and neonatal outcomes of CS in our rural series were similar to those reported in urban Cameroonian hospitals and in other SSA settings [3] [4] [5] [6] [7] [9] [10] [12] [14] [15]. The way to curb those high rates is certainly through quality improvement. Our study is most likely the first to prospectively assess the quality of CS in Cameroon. Besides revealing the very high proportion of poor quality CS the study provides details on critical components to be targeted in view of addressing that crucial issue. The data used in this study are highly accurate because they were collected within 7 days of the CS from several sources (women, care providers and hospital records). This study has a main limitation. It has been conducted in a single administrative constituency and therefore may not be generalizable to all Cameroonian rural settings.
5. Conclusion
Quality of caesarean sections is generally poor in rural settings in West-Cameroon. The quantity and quality of staffs required for Caesarean sections in the hospitals are sometimes insufficient. State-owned hospital is the main factor associated to unacceptable quality of caesarean section, while faith-based hospital seems to have better quality scores. The poor quality of CS in this region could be addressed using the St Vincent de Paul’s Hospital as a model.
Ethics Approval and Informed Consent
The study was approved by the institutional ethics board of the “Université des Montagnes at Bangangte” in the West region of Cameroon (reference: 2020/133/ UdM/PR/CIE).
Data Availability (Where Applicable)
Data supporting this study are available from authors under reasonable request.
Acknowledgements
Authors are grateful to the staffs of the Dschang District Hospital, the Penka Michel District Hospital and the Saint Vincent de Paul’s District Hospital.
Authors’ Contributions
Conception JDNK, JTF, BK,
Study design JDNK, JTF, ACG,
Execution JDNK, JTF, ACG, ANN, LMT,
Acquisition of data ACG, ANN, LMT,
Analysis and interpretation of data JDNK, JTF, ACG, HDM,
Writing JDNK, JTF, ACG, JHF,
Critical revision BK, LMT, CN, PLMFZ
Critical review PMT, FYF, JHF
All the authors have approved the final version and have agreed on the journal.
Abbreviations
aOR: Adjusted Odds Ratio
cOR: Crude Odds Ratio
CS: Cesarean Section
DDH: Dschang District Hospital
PMDH: Penka Michel District Hospital
SD: Standard Deviation
SPSS: Statistical Package for Social Sciences
SSA: Sub-Saharan Africa
SVPH: Saint Vincent de Paul Hospital
UHC: Universal Health Coverage
USD: United States Dollars
WHO: World Health Organisation