- Amie Wilson, doctoral researcher1,
- David Lissauer, clinical research fellow1,
- Shakila Thangaratinam, senior lecturer in obstetrics and maternal medicine2,
- Khalid S Khan, professor of women’s health and clinical epidemiology2,
- Christine MacArthur, professor of maternal and child epidemiology3,
- Arri Coomarasamy, reader in reproductive medicine and gynaecology1
- 1School of Clinical and Experimental Medicine, College of Medical and Dental Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
- 2Centre for Health Sciences, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK
- 3School of Health and Population Sciences, College of Medical and Dental Sciences, University of Birmingham, UK
- Correspondence to: A Coomarasamy, Academic Unit (University of Birmingham), Birmingham Women’s Foundation Trust, Birmingham B15 2TG, UK
- Accepted 11 March 2011
Objective To review the effectiveness and safety of clinical officers (healthcare providers trained to perform tasks usually undertaken by doctors) carrying out caesarean section in developing countries compared with doctors.
Design Systematic review with meta-analysis.
Data sources Medline, Embase, Cochrane Central Register of Controlled Trials, CINAHL, BioMed Central, the Reproductive Health Library, and the Science Citation Index (inception-2010) without language restriction.
Study selection Controlled studies.
Data extraction Information was extracted from each selected article on study characteristics, quality, and outcome data. Two independent reviewers extracted data.
Results Six non-randomised controlled studies (16 018 women) evaluated the effectiveness of clinical officers carrying out caesarean section. Meta-analysis found no significant differences between the clinical officers and doctors for maternal death (odds ratio 1.46, 95% confidence interval 0.78 to 2.75; P=0.24) or for perinatal death (1.31, 0.87 to 1.95; P=0.19). The results were heterogeneous, with some studies reporting a higher incidence of both outcomes with clinical officers. Clinical officers were associated with a higher incidence of wound infection (1.58, 1.01 to 2.47; P=0.05) and wound dehiscence (1.89, 1.21 to 2.95; P=0.005). Two studies accounted for confounding factors.
Conclusion Clinical officers and doctors did not differ significantly in key outcomes for caesarean section, but the conclusions are tentative owing to the non-randomised nature of the studies. The increase in wound infection and dehiscence may highlight a particular training need for clinical officers.
Many developing countries have a shortage of trained doctors. Rural areas are particularly affected, as doctors predominantly congregate in urban areas.1 Various problems have been linked with the depletion in the workforce, including HIV (either because of death, sickness, or fear of exposure to the disease), the migration of trained staff, and the lack of resources and personal income.1 2 3 4
In some developing countries clinical officers were temporarily posted to alleviate the shortage of medical doctors.3 5 However, they have now become a more permanent strategy, being described as the “backbone” of healthcare in several settings.5 Clinical officers have a separate training programme to medical doctors, but their roles include many medical and surgical tasks usually carried out by doctors, such as anaesthesia, diagnosis and treatment of medical conditions, and prescribing. The perceived benefits of using clinical officers compared with doctors are reduced training and employment costs as well as enhanced retention within the local health systems.3 4 6
The scope of practice of a clinical officer within obstetrics is often determined by the country in which they work.2 In 19 out of 47 sub-Saharan African countries, clinical officers are authorised to provide obstetric care, yet in only five countries are they permitted to carry out caesarean sections and other emergency obstetric surgery.5 Given that caesarean section is the most common major surgical procedure in sub-Saharan Africa7 and must be delivered in a timely fashion to save a mother’s life,8 clinical officers could potentially play an important part in increasing accessibility and availability of emergency obstetric care, particularly caesarean section. However, uncertainty exists about their role,1 training, effectiveness, and safety. Given the central role that clinical officers increasingly have in the provision of obstetric care, we systematically reviewed and meta-analysed the effectiveness of clinical officers in caesarean section.
We searched databases for relevant literature on clinical officers within obstetrics in the developing world, with particular attention to maternal and perinatal mortality rates and adverse outcomes. We searched Medline, Embase, Cochrane Central Register of Controlled Trials, CINAHL, BioMed Central, the Reproductive Health Library, and the Science Citation Index (from inception to August 2010). Hand searches complemented electronic searches, and we checked reference lists. Search terms were “clinical officer”, “medical officer”, “assistant medical officer”, “medex”, and “non physician clinicians”. No language restrictions were applied to the search.
We selected controlled studies that compared clinical officers and medically trained doctors for caesarean section in the developing world setting and that reported on any clinically relevant maternal or perinatal outcomes. The electronic searches were firstly scrutinised and full manuscripts of relevant studies were obtained. A final decision on inclusion or exclusion of manuscripts was made after two reviewers (AW and DL) had examined these manuscripts. Information was extracted from each selected article on study characteristics, quality, and outcome data. Descriptive studies were also examined to explore further the role of the clinical officer.
Methodological quality assessment
We assessed the selected studies for methodological quality using the Newcastle-Ottawa scale.9 The controlled studies were evaluated for representativeness, selection, and comparability of the cohorts, ascertainment of the intervention and outcome, and the length and adequacy of follow-up. The risk of bias was regarded as low if a study obtained four stars for selection, two for comparability, and three for ascertainment of exposure.9 The risk of bias was considered to be medium in studies with two or three stars for selection, one for comparability, and two for exposure. Any study scoring one or zero stars for selection, comparability, or exposure was deemed to have a high risk of bias.
We used the random effects model to pool the odds ratios from individual studies. Heterogeneity of treatment effects was evaluated using forest plots, χ2 and I2 tests; the terms low, moderate, and high heterogeneity were assigned to I2 values of over 25%, 50%, and 75%, respectively. Where possible we present data for adjusted estimates on the forest plot to account for confounding factors. Analyses were done using Revman 5.0 statistical software.
Six non-randomised controlled cohort studies (16 018 women) were included in the review (table 1[t1] and fig 1⇓).1 3 8 10 11 12 When methodological quality was assessed on the Newcastle-Ottawa scale, most studies had a medium risk for selection bias and medium to high risk for comparability and outcome assessment (table 2[t2]).
All six studies compared maternal mortality. The meta-analysis showed no statistically significant difference between the clinical officers and doctors (odds ratio 1.46, 95% confidence interval 0.78 to 2.75; P=0.24, fig 2⇓). However, the analysis found significant heterogeneity (P=0.03), which was moderate (I2=60%). In one8 of the two studies3 8 that showed an increase in maternal mortality with clinical officers in the crude analysis, the increase was no longer statistically significant when the analysis was adjusted for rural setting, previous caesarean section, haemorrhage, other perioperative medical complications, and the level of training of the surgeon (adjusted odds ratio 1.4, 95% confidence interval 0.7 to 2.9). The second study3 that showed an increase in maternal mortality with the clinical officers also adjusted the analysis, but for reported diagnosis and referral status; the adjusted estimates were not, however, provided. The overall maternal mortality rate in the six studies was high, at 1.2%.
Five studies1 3 8 10 12 (15 665 women) compared perinatal mortality. The meta-analysis showed no significant difference between the groups (odds ratio 1.31, 95% confidence interval 0.87 to 1.95; P=0.19, fig 2). The analysis found significant heterogeneity (P<0.01), which was high (I2=88%). In one8 of the two studies3 8 that showed an increase in perinatal mortality with clinical officers in the crude analysis, the increase was no longer statistically significant when adjusted for confounding factors (adjusted odds ratio 1.1, 95% confidence interval 0.8 to 1.3). The overall perinatal mortality rate in the five studies was high, at 10.7%.
Wound infection and wound dehiscence
Two studies3 10 (4436 women) compared the rates of wound infection. The meta-analysis found a significant increase in wound infection with clinical officers (odds ratio 1.58, 95% confidence interval 1.01 to 2.47; P=0.05, fig 2). Heterogeneity was not significant (P=0.40, I2=0%).
Three studies1 3 10 (6507 women) compared the rates of wound dehiscence. The meta-analysis showed a significant increase in wound dehiscence when clinical officers carried out caesarean sections compared with doctors (odds ratio 1.89, 95% confidence interval 1.21 to 2.95; P=0.005, fig 2). Evidence of significant heterogeneity was lacking (P=0.38, I2=0%).
Training of clinical officers
All six papers gave training details of clinical officers; training length and specification varied between countries. In Zaire11 and Burkina Faso,3 nurses attend a two year training course to become clinical officers, with an additional 1-2 years of surgical training in Zaire. In Malawi8 10 and Mozambique,1 clinical officers require a three year health foundation course, with a year as an intern at a hospital or in surgical training. In Tanzania,12 clinical officers undergo three years’ medical training, with a further two years in clinical training plus three months in surgery and three months in obstetrics. In Burkina Faso,3 clinical officers are required to undergo a six month curriculum in emergency surgery to carry out operative obstetric care.
The meta-analysis did not show a statistically significant difference in maternal or perinatal mortality in caesarean sections carried out by clinical officers compared with doctors. However, when the outcomes of wound dehiscence and wound infection were assessed, both were significantly more frequent in caesarean sections carried out by clinical officers.
Strengths and limitations of the review
All of the six studies examined were comparative cohort studies. As they were not randomised trials, there is the potential for bias. When methodological quality was assessed on the Newcastle-Ottawa Scale there was a medium risk of selection bias and a medium to high risk of bias in comparability and outcome assessment for most studies. For example, in one study,1 elective caesarean sections were exclusively carried out by doctors, whereas emergencies were carried out by both doctors and clinical officers. As elective caesarean section is associated with better outcomes than emergency caesarean section,13 this arrangement would have conferred an advantage to doctors. Furthermore, clinical officers tend to be located in rural areas,3 where access to lifesaving facilities such as blood transfusion and high dependency care may not be available. Another study8 tackled such issues by adjusting for rural setting, previous caesarean section, haemorrhage, other perioperative medical complications, and the level of training of the surgeon. Their initial analysis showed an excess in maternal and perinatal mortality associated with clinical officers. However, when adjustments were made for the relevant factors, the difference in these outcomes was no longer statistically significant. This suggests the possibility of more high risk cases in the clinical officer group in this study. It is also plausible that the bias could be in the other direction. For instance, the perceived severity of the situation may have resulted in a doctor rather than a clinical officer carrying out the caesarean section. This may cause bias in favour of clinical officers. Although most studies reported no differences in patient characteristics1 3 8 10 or indication for caesarean section,1 3 8 10 11 12 and some studies adjusted for various factors,3 8 residual confounding can still exist.
Maternal and perinatal outcomes were statistically significantly heterogeneous, which may reflect the diversity of the setting and the population, indications for surgery, surgical approach and training, and role of the clinical officers in these studies. Given such clinical heterogeneity, it is unsurprising that statistical heterogeneity was identified in the analyses. Formal exploration of the reasons for statistical heterogeneity by study features was limited owing to the small number of studies identified in our review. However, when confounding factors were adjusted for, the observed heterogeneity decreased.
Although we acknowledge caution when interpreting the findings of this meta-analysis owing to the non-randomised nature of the included studies, the present study remains the best current evidence on these outcomes.
Clinical officers were associated with an increase in wound infection and dehiscence. This was consistent in the two studies that examined these outcomes. We speculate that these outcomes may be associated with surgical technique and a need for enhanced training. One study1 highlighted that 97% of caesarean sections were through a vertical abdominal incision, which is known to be associated with increased wound dehiscence and other adverse outcomes when compared with horizontal incisions.14 Thus there may be substantial scope for improvement in surgical technique. Evidence shows that specialist training of clinical officers can improve outcomes. One study8 measured the incidence of maternal death from anaesthesia, when administered by clinical officers who had or had not received formal training. The maternal mortality rate was much higher in those who had not received training compared with those who had (2.4% v 0.9%).
Our review assesses the important and specific role of clinical officers in carrying out caesarean section, which is an immediate determinant of outcome. However, this must be placed within the wider context of the many distant and intermediate determinants of maternal health and mortality15 (see web extra on bmj.com). Although little work has been done to assess the role of clinical officers in tackling these wider determinants, they can have an important impact on these factors through, for example, increasing access to services5 16 and a role in family planning2and broader preventive health programmes5 17 to reduce maternal mortality. Furthermore, part of the value of the clinical officer role is that their job can be adapted to suit local needs and conditions. Yet as there are no internationally agreed curriculums or scope of practice guidelines,2 the importance of evaluating clinical officers in their specific setting needs to be recognised.
Our meta-analysis suggests that the provision of caesarean section by clinical officers does not result in a significant increase in maternal or perinatal mortality. Enhanced access to emergency obstetric surgery through greater deployment of clinical officers, in countries with poor coverage by doctors, can form part of the solution to meet Millennium Development Goals 4 (reducing child mortality) and 5 (improving maternal health).
What is already known on this topic
When compared with doctors, clinical officers cost less to train and employ, and are retained better within health systems in developing countries
Clinical officers are the backbone of obstetric care in many developing countries, performing as much as 4/5th of caesarean sections in some countries
There is uncertainty about the effectiveness and safety of clinical officers performing caesarean section surgery
What this study adds
Meta-analysis of six controlled studies found no differences between the clinical officers and doctors for maternal and perinatal death after caesarean section
Wound dehiscence and wound infection were found to be significantly more frequent in caesarean sections performed by clinical officers
Cite this as: BMJ 2011;342:d2600
We thank Paul Fenton for constructive feedback on the manuscript.
Contributors: AW and AC conceived the review. AW and DL carried out the search, study selection, and data extraction. AW, ST, and AC analysed the results. AW, AC, and DL drafted the manuscript. CM and KSK provided critical revision of the manuscript. All authors approved the final version of the manuscript. AC is the guarantor.
Funding: This study was funded by the registered charity Ammalife (www.ammalife.org) and Birmingham Women’s NHS foundation Trust R&D department.
Competing interests: All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that no support from any institution for the submitted work; no relationships with any institution that might have an interest in the submitted work in the previous 3 years; their spouses, partners, or children have no financial relationships that may be relevant to the submitted work; and have no non-financial interests that may be relevant to the submitted work.
Ethical approval: Not required.
Data sharing: No additional data available.
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits use, distribution, and reproduction in any medium, provided the original work is properly cited, the use is non commercial and is otherwise in compliance with the license. See: http://creativecommons.org/licenses/by-nc/2.0/ and http://creativecommons.org/licenses/by-nc/2.0/legalcode.