Table 2.
CERQual summary of findings (SoFs)
| Review finding | Studies contributing to review finding | CERQual assessment | Explanation of confidence in the evidence assessment |
| Philosophy of birth | |||
| Beliefs about birth: across HIC and MICs health professionals reported varying beliefs about birth. These included a common approach to birth shared by obstetricians and midwives who valued the physiological process and worked effectively as a team to make it happen (recognising it as an empowering process for women and only intervening when medically necessary), to labour and vaginal birth as a fatally flawed physiological process with CS the preferable means to an end. This dichotomy of beliefs reflected competing ideologies of birth and shaped the importance individuals attached to CS rate reduction. In MIC, while some obstetricians who preferred CS made reference to perinatal mortality and morbidity gains, this was not the experience of the few female Chinese obstetricians who actually had CDMR, nor the preference of Iranian obstetricians who expressed concerns about having to deal with comorbidities caused by previous CSs. Beliefs were influenced by professional training, personal experience and practice setting. | 44–46, 54, 57–62, 64–66 | Moderate confidence | Thirteen studies with minor to significant methodological limitations. Rich data from 14 countries across four geographical regions, high-income and middle- income levels and high and low CS rates. Reasonable level of coherence with uncertain confidence in LICs. |
| Beliefs about what constitutes necessary and unnecessary CS: some health professionals reported CS rates as determined by factors beyond their control (ie, uncertain obstetric history, unfolding obstetric circumstance and clinical indications), but between health professionals, there was no clear consensus as to what they believed to be clinical indications across time (ie, breech), place (ie, availability and access) and parity (ie, women with a previous CS). Some senior doctors and midwives expressed concerns that less experienced staff are more likely to perform CS based on vague indications and spoke favourably about wanting junior staff to consult them more for a second opinion. Other senior staff suggested second opinion policies only work where both doctors are in attendance at the hospital. While some residents also reported wanting improved communication, they feared seeking a second opinion would negatively impact their clinical credibility and career. | 47, 54–57, 63 | Low confidence | Six studies with minor to moderate methodological limitations. Thin data, with major concerns about coherence across settings. |
| Beliefs about the evidence base surrounding caesarean section: health professionals’ views about research evidence varied. Most health professionals recognised that guidelines represent the national or international evidence base, which sensitised them to reflect on their practice, providing a potential mechanism for change. Most health professionals wanted more evidence of transferability to their own practice context, particularly in MIC and LIC contexts, where audit was not common. Not all health professionals believed available evidence to be valid, applicable to their practice or feasible to implement and spoke about keeping-up-to-date with the latest evidence as challenging. Across resource settings, obstetricians and midwives expressed concerns about evidence of risks associated with CS as incomplete. Some health professionals who valued guidelines were also very clear they took other factors into account in actual decision making (ie, interpersonal relationships and patient’s unique characteristics). | 54–55, 57–59, 61–64 | Moderate confidence | Ten studies with minor to moderate methodological limitations. Rich data from across three geographical regions but limited data from LICs. High coherence across HICs and MICs. Uncertain confidence in LICs. |
| Belief in need to reduce unnecessary CS and receptiveness to change: across resource settings, health professionals reported concerns about high CS rates and associated morbidity. In Iran and Tanzania, some health professionals spoke about colleagues who performed CS for non-medical reasons as contravening medicines underlying ethical principle to do no harm. In European settings, health professionals experienced interventions targeted to reduce unnecessary CS as most acceptable where this vision was shared within and between multidisciplinary groups. In the UK and Scandinavia, health professionals from organisations that achieved success in reducing rates had positive attitudes towards critical self-reflection (including audit, second opinion and continuing medical education) and felt supported by colleagues and opinion leaders. Across resource settings, health professionals acknowledged concerted action to reduce unnecessary CS as challenging but achievable and intrinsically rewarding where there was respect, accountability and shared responsibility to support women achieve a vaginal birth. | 54–55, 57–59, 61–64 | Moderate confidence | Nine studies with no to moderate methodological limitations. Thick data from Europe. Only one study from African region contributed to this finding. High coherence. |
| Social and cultural context | |||
| Fear of blame and recrimination (including medicolegal concerns): across HIC, MIC and LICs health professionals reported fear of litigation as an important influence on their low threshold for performing CS (although no one had actual experience of litigation in LIC). Predominantly in North America, health professionals described medicolegal concerns as an underlying factor in non-compliance to guideline recommendations. Across urban and rural settings with or without 24 hours obstetrical and anaesthesia coverage, obstetricians and midwives weighed up the balance of professional identity risk with not intervening, a poor outcome ensuing and a medicolegal case against them. Also in North America, some obstetricians were opposed to second-opinion policies because of the difficulties in medicolegal responsibilities that could ensue. In North America, some European countries and Africa, midwives and obstetricians expressed concerns about threats to their professional identity and career prospects posed by internal audit and feedback. A few health professionals welcomed guidelines as providing a defendable basis for their practice, while other midwives and obstetricians were undeterred in their commitment to intervene only when necessary. | 45, 54–55, 57–58, 61, 63–64 | Moderate confidence | Eight studies, with no to moderate methodological limitations. Rich data from five countries. Moderate coherence. |
| Value attached to financial rewards associated with CS: some health professionals were outspoken about the economic incentives for CSs, particularly in private healthcare facilities. This included doctors in Tanzania, Iran, China and Nicaragua, as well as midwives in Iran and the USA. Some doctors considered CS to involve more work, which justified the payment; others blamed the system, while others still reported personally valuing this extra income. Some doctors, and midwives, were critical of insufficient monetary reward to staff labour and vaginal birth by comparison. | 45, 47, 55, 57–58, 60–61, 63 | Moderate confidence | Eight studies with minor to moderate methodological limitations. Rich data predominantly from middle-income countries. High coherence. |
| Preferences for CS as convenient: health professionals valued both the scheduling CS offers and the lesser time commitment it entails compared with labour and vaginal birth. Some health professionals described how CS was convenient for women too (for the same reasons), although others recognised while CS might be more convenient for them, it is not what every woman wants. | 46, 57–61, 63 | Moderate confidence | Seven studies with minor to moderate methodological limitations. Fairly rich data from two studies and convenience a theme in a third. High coherence. |
| Beliefs about women: across the world, health professionals reported women’s demand for a particular birth method as an important factor influencing rates of CS, NVD and VBAC. Some health professionals believed women now value CS as a consumer choice (available in public and private healthcare settings), others attributed increasing rates to women’s lower threshold for CS during labour. In HIC, MICs and one LIC (Tanzania), a few health professionals spoke about women’s innate ability to labour and birth as being diminished by rising BMIs, advanced maternal age, sedentary lifestyles and ‘western diseases’. Health professionals also perceived women as lacking in antenatal education, being influenced by their families and the plethora of information about birth available in the media and online. | 45–47, 54–61, 63–66 | High confidence | Fifteen studies with no to moderate methodological limitations. Thick data from 15 countries, across five world regions, high-income, middle-income and low-income settings with high CSRs. High coherence. |
| Dysfunctional teamwork, within the medical profession and including the marginalisation of midwives: health professionals reported dysfunctional teamwork within and between professionals as an important barrier to reducing unnecessary CS rates. Medicine’s entrenched hierarchies, lack of communication between maternity and theatre staff and difficult relationships between obstetricians, midwives and family doctors were all spoken about. Some midwives and obstetricians spoke passionately about the marginalisation of midwives and their exclusion from birth as counterproductive. | 47, 55–63, 65 | Moderate confidence | Eleven studies with minor to moderate methodological limitations. Thick data from across resource settings. High coherence. |
| Negotiation within the system | |||
| Organisation of care: across the world, health professionals perceived the maternity care system as insufficiently resourced (human and material). Midwives and obstetricians reported where CS was an important source of revenue operating facilities were a priority, and facilities for labouring women were poor and inadequately staffed. | 47,55-59,61-63,65 | Moderate confidence | Ten studies with no to moderate methodological limitations. Thin data from 13 countries, and thick data from Iran. High coherence. |
| Beliefs about need for high-level infrastructures: health professionals in HICs who were supportive of VBAC were flexible in their interpretation of guidelines and used them and available technologies in a facilitative way. Other health professionals, predominantly from MICs and LICs, but some from HICs, expressed concerns that a lack of human and technological resource made guideline recommendations unworkable in practice. In HICs where 24 hours obstetrical and anaesthesia cover was available, some health professionals reported women were still refused a trial of labour. | 47, 54–66 | Moderate confidence | Fourteen studies with no to moderate methodological limitations. Thick data from HICs and MICs. The finding may have higher confidence in settings where the level of resource is sufficient to sustain necessary CS. |
| Reluctance to change based on lack of training, skills or experience: some health professionals spoke about how preregistration and postregistration training has ill-equipped the next generation for a reduction in CS rates as they have little experience, competency or confidence in normal labour and vaginal birth. Others reported wanting specific training on recommendations to make them more acceptable in practice. Reasons for many health professionals lack of buy-in was multifactorial; see also organisation of care; beliefs about need for complex infrastructure; and beliefs about the clinical encounter and autonomous decision making. | 45, 47, 55–57, 59, 61, 65–66 | Low confidence | Nine studies with minor to significant methodological limitations. Thick data from one study. Extent of coherence unclear. |
| Views about the format, content and delivery of interventions: a few health professionals spoke about the importance of the tone of guidance as facilitative of reflection, not dictatorial, judgemental or threatening, at the same time as being clear about the need for change by avoiding the use of words such as ‘should’, ‘developmental’ or ‘pilot’. Some health professionals described how important it was for local opinion leaders to endorse projects, and where external facilitators were involved they are ‘credible’ and ‘grounded’, exercised cultural humility and understand the challenges within specific practice settings. In some HICs, health professionals talked about multidisciplinary/interprofessional team involvement meaning representatives from medicine (obstetrics, anaesthesia and paediatrics), nursing and midwifery, allied health professionals, quality, health records and scheduling in secondary care. | 55, 57, 59, 61–63 | Low confidence | Six studies with minor to significant methodological limitations. Thick data from one study. Extent of coherence unclear. |
| Beliefs about the clinical encounter and autonomous decision making: obstetricians and midwives’ views varied as to who they thought should have the final say in the decision to perform a CS. Some health professionals accepted a woman’s right to choose CS, many thought the decision should be shared, while others believed the decision could only be made by health professionals qualified to do so. Some health professionals expressed concern time constraints in practice limited their opportunities to facilitate informed decision making. Where teams had a shared approach, they reported informed decision making did happen and irrespective of who made the final decision everyone involved was reassured by the process. | 44–47, 54–55, 57–59, 61–64, 66 | Moderate confidence | Fourteen studies with no to significant methodological limitations. Thick data from HICs, MICs and one LIC. High coherence. |
BMI, body mass index; CS, caesarean section; HIC, high-income country; LICs, low-income countries; MICs, middle-income countries; VBAC, vaginal birth after caesarean.