Healthcare workers’ behaviors and personal determinants associated with providing adequate sexual and reproductive healthcare services in sub-Saharan Africa: a systematic review

The following electronic databases were searched (from January 1990 until October 2015): PubMed, EMBASE (through OVID), Cochrane central (through Cochrane library), CINAHL (through EBSCOhost), and PsychINFO (through EBSCOhost) using a search strategy that was developed based on the key words identified from the study objective, which was adapted according to the technical requirements of each specific database. The full search strategy is provided as Additional file 1.

The search period included studies published since January 1990 until October 2015. This period was chosen because of the MDG 2015 (target 5) pertaining to maternal and child health; which includes amongst others, efforts to reduce maternal mortality, teenage pregnancy, and unmet needs of SRH (family planning) services. Therefore, this review sought to identify studies which were either addressing, or attempting to, or were related to the means and efforts aiming to attain the MDG 5 targets in SRH services. The studies were restricted to: peer-reviewed and academic journals using the specific database filters available and by manual assessment to limit studies to primary studies only (e.g., no editorials, or reviews), studies undertaken in the sub-Saharan African region only, and studies that were published in English language.

For studies to be included in this review, studies had to pass the eligibility criteria for inclusion: studies with participants as professional healthcare workers, including doctors, nurses, midwives, obstetricians and gyneacologists (Population); if the studies were focusing on behavioral and personal determinants of professional healthcare workers, such as education, knowledge and skills of professional healthcare workers in SRH services, studies involving contraceptives, abortion (TOP), pregnancy, childbirth; studies investigating factors associated with access to, utilization of, and quality of SRH services and related services (Intervention); if the primary outcomes were quality of healthcare in SRH services, access to SRH healthcare services, utilization of SRH services, and or the secondary outcomes were adolescent pregnancy, maternal health outcomes, and child health outcomes (Outcome); if study participants were working in the SRH services, including gyneacology and obstetric care (GOC) services, as well as maternal and obstetric units (MOU); and the studies involving services rendered under general public health facilities, such as primary healthcare clinics (PHC), community health centers (CHC), secondary and tertiary hospitals, as well as academic hospitals (Setting). Any types of study designs were eligible for inclusion provided that they reported quantitative data. Only studies with quantitative data focusing on either one or more of the primary and secondary outcomes were eligible. The rationale for the inclusion criteria of only quantitative studies was because this review sought to quantify the relationship between behaviours, determinants, and primary or secondary outcomes in order to inform future intervention development in terms of relevance of specific determinants. Because this review is focusing on determinants rather than comparing interventions or treatment, it does not have a comparator (No Control). The full PRISMA checklist is provided as Additional file 2. Two independent reviewers (KJ and RC) performed the selection process of the eligible studies. Studies which do not meet the inclusion criteria according to the independent reviewers were brought forward for discussion and reaching consensus regarding the study’s eligibility for inclusion in this review.

The search was conducted in a step-wise manner with three phases. The first phase was a broad search that was developed with the guidance of an experienced librarian. The second phase involved searching of specific electronic databases based on the results of the first phase using the provided search terms that were revised and adapted from the first phase. The third and last phase was conducted for additional sources of data. This involved a manual search of reference lists of included reports and articles, and other online resources. Search results were imported into the Endnote database and duplicates were screened for and deleted. Articles went through an extensive screening process in order to identify sources of relevant data.

Initially, titles were scanned for relevance and titles obviously not meeting the inclusion criteria were excluded. Then, the screening of abstracts of those titles that met the inclusion criteria began, and abstracts not meeting the inclusion criteria were excluded. Following the screening of abstracts was the reading of full-text articles of the included abstracts, and articles not meeting all the inclusion criteria were excluded, as well as the articles that did not have full text available. The reasons for exclusion of the full-text articles were either: the article did not report on the behaviours of healthcare workers, or the article reported behaviours based on qualitative findings, or the article reported behaviours of non-professional healthcare workers such as community healthcare workers. Finally, the rigorous and extensive assessment of the included full-text articles as they met the inclusion criteria was conducted. Again, the two independent reviewers (KJ and RC) performed the process described above and met regularly to discuss their assessment until consensus was reached.

Experts in the field of maternal and child healthcare, obstetric and gynaecology were also contacted and asked to recommend any new data sources not yet identified through the means described above, including unpublished data. Of the twelve experts contacted via email, only two responded with sources of information consisting of articles not available through the databases searched. The flow diagram below shows the review process from the initial searches of bibliographic databases to the final included studies (see Fig. 1).

Publication year, authors, location of study, research design, study participants and outcome measure, as well as data regarding determinants were key factors for data extraction. Data abstraction was carried out by the two independent reviewers (KJ and RC) and they met regularly to discuss their assessment and address any discrepancies until consensus was reached. Where consensus was not reached for a study, the particular study was taken to the next step for further screening and assessment until both reviewers were in consensus. This was also done in order to strengthen the quality of the analysis of the studies included, as well as not immediately excluding studies based on one reviewer’s assessment while studies could potentially be eligible when thoroughly assessed by more than one reviewer. Data from these studies included was extracted and tabulated in Microsoft Excel. Quality assessment of the studies included was assessed using the STROBE checklist for observational studies tool which covers data collection, methods, results, discussion and other information relevant for this review such as the study population. Quantitative data collection methods (e.g., surveys), type of study design (e.g., cross-sectional designs), and description of study participants (e.g., nurses) were among the key criteria in the quality assessment of the studies.

A narrative analysis approach was used to synthesize the data. For all studies included in this review, data was extracted into a table eliciting information on: (1) first author and year of publication; (2) study sample and population; (3) study design; (4) study outcome; and (5) the determinant(s) with summary measures of the study findings (e.g., attitudes towards SRH services, knowledge of specific SRH components, use of specific SRH resources, and obstetric skills). Furthermore, studies were assessed and data was synthesized for potential to be associated with or impact on the primary or secondary outcomes.

A majority of the studies were from Nigeria (n = 9) and Ethiopia (n = 7), followed by Kenya (n = 2), Uganda (n = 2), Zambia (n = 2), Senegal (n = 2), Ghana (n = 2), Zimbabwe (n = 2), and South Africa (n = 2). Malawi, Swaziland, Cameroon, and Rwanda were represented by one study each, and one study represented Mali, Tanzania and Mozambique in one. Of the 35 studies included (see Fig. 1), 21 of them investigated association or influence on the primary outcomes, which affect quality of SRH services and possibly access to and utilization of the services, as described above (see Table 2). Fourteen studies did not investigate this association and only reported on healthcare workers’ behaviours and related factors in SRH services in a descriptive sense.

Of the 35 studies included, 28% (n = 10) of them investigated basic obstetric care skills among professional healthcare workers, 23% (n = 8) investigated knowledge of SRH service components, 20% (n = 7) investigated behaviours of professional healthcare workers with regards to certain SRH services. The remaining 29% (n = 10) studies investigated availability of resources and other essential components of SRH services, as factors associated with adequate quality service provision in SRH services. A majority of the studies, 71% (n = 25) were cross-sectional studies, some were population and hospital surveys 9% (n = 3), exploratory studies 6% (n = 2), observational studies 6% (n = 2), intervention studies 6% (n = 2), and only one was a retrospective study 3% (n = 1).

The included studies investigated determinants associated with adequate, quality service provision in sexual and reproductive healthcare (SRH) services and other related factors. The determinants included basic knowledge and skills regarding SRH among professional healthcare workers working in SRH services, as well as their behaviours and attitudes towards providing SRH services. Other factors associated with the behaviours and personal determinants of healthcare workers in SRH, such as the availability of essential SRH resources are also reported. The results are presented under the research questions which this review sought to answer.

In this paper, a description of healthcare workers’ behaviours and personal determinants associated with these behaviours are reported. These behaviours and personal determinants have potential to negatively affect SRH access and utilization of services, as well as quality of care.

Among the studies investigating knowledge, eight of them focused on knowledge of partograph (n = 8), a component of basic emergency obstetric care (EmOC), and its association with its utilization in SRH services. These studies were undertaken in Nigeria (n = 5), Ethiopia (n = 2), and Ghana (n = 1) [22‐29]. All these studies found that knowledge of partograph had a strong association with its utilization in SRH services. Knowledge of partograph was found to be higher in obstetric care workers working in hospitals than in those working in health centers in Ethiopia [25]. While the knowledge was higher among those working in hospitals, the utilization was more frequent in the health centers than in hospitals [25]. The knowledge of partograph components was found to be poor among the auxillary nurses and midwives in Nigeria [24]. More years of experience of a healthcare worker with the healthcare system in Nigeria was associated with better partograph utilization in terms of frequency and skills of partograph use [27].

A majority of the studies (n = 10) investigated knowledge of other components of basic EmOC and found: basic EmOC knowledge among the obstetric care workers was poor, EmOC knowledge and skills and confidence in providing the services was improved by training, higher knowledge scores of some life-saving techniques, such as neonatal resuscitation were associated with having performed the technique recently and with geographical location [30‐39]. However, type of healthcare worker such as being a nurse or a midwife did not have any associations with higher knowledge scores among healthcare workers in Ethiopia [32].

Knowledge of active management of the third stage of labour (AMTSL) was fairly adequate among the healthcare workers in Zambia, although a few of the workers still practiced cord clamping and cutting, a procedure no longer recommended as part of AMTSL [23]. Only a third of healthcare workers were aware of the fundal massage as a component of AMTSL [33]. The knowledge and use of maternal ultrasound among midwives in Zambia was improved after training on maternal ultrasound [34]. Knowledge of basic EmOC among the different professions in SRH services in Nigeria was better among doctors compared to other professions, but no difference was found between age groups [38]. Knowledge of symphysiotomy, a procedure performed to widen the pelvis in order to ease childbirth when there is a mechanical problem, was moderate among all participants in the Zimbabwean study and all could describe how to perform it [39]. In particular, rural midwives in Zimbabwe regarded the technique as a lifesaving one appropriate for remote areas, and more than half of them thought that it was necessary to be taught [39].

With regards to PMTCT, knowledge among nurses and midwives was moderate Nigeria, with knowledge and behaviour being positively correlated [40, 41]. Knowledge of PMTCT among nurses and midwives who manages pregnant women living with HIV and AIDS in Nigeria was significantly better from those who did not manage pregnant women living with HIV and AIDS [40]. Knowledge of emergency contraceptive, on the one hand, was fair among the different cadre of healthcare workers in Uganda [42]. On the other hand, 1 in 4 healthcare workers did not know the time limit within which EC was effective. Knowledge of postpartum infections and management of hypertensive complications was fairly poor among the healthcare workers in Mali [43]. However, high knowledge scores on immunization issues were found among healthcare workers in Nigeria [44].

With regards to continuous education, a model often used in the healthcare setting to continuously improve knowledge and skills of healthcare workers, nearly half of the participants in a study in Swaziland had no continuous education training and felt that they needed guidance or supervision on adolescent sexual and reproductive health care, although some of them wanted to be trained on post-abortion care while some wanted to be trained on how to perform abortions [45]. A small proportion of the healthcare workers wanted the government to train healthcare workers in adolescent sexual and reproductive health services [45]. In Nigeria however, a large proportion of healthcare workers (68%) had adequate training in immunization, according to Ehiri et al. [44].

Negative attitudes towards provision of SRH services to adolescents in Ethiopia was significantly associated with healthcare workers’ being married, lower levels of education, being a health extension worker, lack of training on RH services, and non-use of family planning by healthcare workers themselves [17]. A majority of healthcare workers in both Kenya (82%) and Zambia (86%) reported that they would advise adolescents to abstain from sex when they seek contraceptives [46]. With regards to provision of contraceptives to young women, healthcare workers had a minimum age restriction, as well as restrictions according to clients’ marital status in Senegal [47]. However, healthcare workers with more education and those who had received continuing education on adolescent sexuality and reproduction showed a tendency towards more youth- friendly attitudes and were in support of contraceptive use among adolescents [46]. Furthermore, these two countries concurred with the notion of adolescent boys being taught the dangers of masturbation [46]. On the positive attitudes of healthcare workers, a majority of the obstetric care workers in Zimbabwe agreed that adequate knowledge about prenatal services, accessible and acceptable services, affordable services, needs-focused prenatal services, prompt services and perceived benefits would influence women and adolescents’ decisions to utilize prenatal services [13].

Religion was found to be a factor among healthcare workers in Swaziland as some healthcare workers reported that contraceptive use was against their religion and that youth should not indulge in sex [48]. Furthermore, masturbation is regarded as a sinful act and those who perform it are believed to experience mental disorders or fertility problems by a majority of healthcare workers in both Kenya (n = 264) and Zambia (n = 331) [46]. In general, masturbation is considered a taboo in the southern region of SSA. Nurses and midwives disagreed that abortion should be provided to adolescent girls with unwanted pregnancies, on both cultural and religious grounds in Kenya (80%) and Zambia (94%) [46]. Nearly a third (26%) of the nurses and midwives in both countries reported that they would feel annoyed if an adolescent girl would report with abortion-induced related symptoms [46].

In Ethiopia, availability of essential drugs, equipment and other supplies was unsatisfactory, and discrepancies were found between providers and clients with regards to services provided and services received [45]. A majority of the facilities in Ethiopia did not function fully for EmOC as per their level, as well as important ANC components were incomplete and unsatisfactory [45]. Ndikom & Onibokun (2007) [40] found a significant relationship between behaviours and availability of resources. While facilities were adequately equipped with immunization supplies in Nigeria, essential drugs supply was inadequate in all centers; as well as emergency care facilities [40]. Various healthcare system constraints, such as human resources, availability of essential medical supplies including drugs, and challenges with the referral system were reported as key factors influencing decision-making for clinical care in Ghana [49].

Printed protocols and guidelines were the most commonly selected aids used in decision-making for clinical care and the availability of guidelines was associated with higher competency of physicians, health technicians, and obstetric nurses [49]. However, study participants also reported healthcare systems’ facilitators in decision-making for clinical care in Ghana. The facilitators included availability and consultation of workshop materials, expert advice and telephone calls to other colleagues, and access to various locally (institutionally) modified or developed guidelines and protocols e.g., guidelines on the management of postpartum [49].

Contrary to Oduro-Mensah, et al., [49], Implementation of clinical guidelines was less than optimal, with certain sections of the guidelines better implemented than others by healthcare workers in Rwanda [50]. Furthermore, higher competency in EmOC skills was linked to the availability of guidelines which further enhanced competency of other healthcare workers with fewer than 10 years of experience in Mali [43].

The major factors found to influence behaviour towards PMTCT services were mainly fear of getting infected, irregular supply of resources like gloves, goggles, and sharps boxes, and an irregular supply of water [40]. A majority of healthcare workers in Kenya were in favour of equal treatment for post abortion care (PAC) regardless of the age or marital status of clients [51]. All healthcare workers in Kenya deemed PAC services important. Sexual intercourse after marriage was supported by less than half of the healthcare workers in the Kenyan study [51]. Private doctors in South Africa considered caesarean section (CS) safer than a “normal” delivery, a convenience and financially enticing to the doctor [52]. Public doctors believed that CS is performed by obstetricians due to a lack of proper (training) management of complications or difficult delivery in South Africa [52].

Many of the studies included came from northern regions compared to the southern regions of SSA, although it is well known that inadequate SRH services are predominantly vast in the southern region. Methodologically, the majority of the studies included in this review were descriptive cross-sectional studies with a strong heterogeneity. This makes it hard to pool results from different studies. Moreover, it is possible that some interesting findings are in the grey literature, which this review did not access. The other limitation is the lack of information on adolescent specific obstetric care. However, this is not surprising, as obstetric care in general has been tailored around the general population of women and not adolescents.

The results of this review are not necessarily generalizable to each and every country in the SSA region due to the limited number of studies supporting this review’s findings. Hence, caution is warranted to draw general conclusions. However, the limited number of studies in SRH services in SSA emphasizes the need for more research in this field in order to draw conclusive and generalizable findings. Furthermore, findings of this review can be used to some extent to highlight specific areas needing intervention, but additional needs assessment maybe needed. It is clear that more research still needs to be done to uncover healthcare workers’ behaviours and associated factors which have a negative effect on access to and utilization of SRH, and quality of care in this field.

Healthcare workers behaviours, personal values, and morale significantly hinder the provision of adequate quality SRH services to young people, especially to adolescent girls. Adolescents’ SRH services need to be youth-friendly, and served by welcoming healthcare workers who have a positive attitude towards adolescents seeking SRH services. This kind of environment will further equip adolescents seeking SRH services with the knowledge and information they require to make informed choices regarding their SRH.

It has been shown that interactive forms of continuous education sessions on sexual and reproductive healthcare that enhance participant activity and provide the opportunity to practice skills can effect change in professional practice and, improve health care outcomes [59, 60]. The findings in this review of doctors having better knowledge of specific SRH components compared to nurses and midwives, highlights the importance of further education and training among nurses and midwives in providing SRH services.

Basic EmOC signal functions, as described by the WHO, cannot be fully performed if a health facility is inadequately equipped with necessary resources to carry all the signal functions. The availability, continuous and frequent utilization of SRH resources, such as the partograph are very important and cannot be over emphasized as they play a crucial role in maternal and child health outcomes. Therefore, increasing essential supplies and equipment in health facilities is very important to strengthening the healthcare system and improving health outcomes.

Continuous SRH education and training is necessary to adequately provide quality services, and therefore needs to be incorporated and enforced into the healthcare systems policies for professional development. The importance of comprehensive knowledge and skills to adequately care for women, newborn, children, and adolescents should be prioritized as it is critical in SRH services. Reducing maternal morbidity and mortality rates, including deaths of newborn, and children under five years of age, are a priority in the SDG 2016-2030. To ascertain that this specific target is achieved in SSA, serious efforts and investments towards adequate quality healthcare services are needed.

Religion, a socio-demographic factor associated with many behaviours, is commonly acknowledged in the SSA region and requires attention to assist healthcare workers who find themselves in conflict of their religion when rendering some SRH services. Healthcare workers in SRH services may benefit from policy guidelines and instructions pertaining to religious beliefs and clarifying the role of healthcare providers in this regard. Healthcare workers should be able to provide SRH services to women and adolescents when they require them, irrespective of their own personal views and religious beliefs.

There is a need for an in-depth understanding of the negative behaviours and attitudes of healthcare workers in SRH services, and their associated factors in order to develop interventions that will address them and consequently stimulate access to and utilization of SRH services by women and young people in SSA. Interventions to promote positive behaviours and attitudes of healthcare workers in SRH services are needed. Furthermore, future studies should be intervention-based focusing on changing the known and reported negative behaviours and attitudes of healthcare workers in SRH services into positive ones. Such intervention studies may include changing the negative behaviours and attitudes of healthcare workers with regards to young women’s use of contraceptives in some parts of SSA, taking into account the contextual differences in SSA regions. An SRH specific evaluation tool to regularly monitor and evaluate knowledge and skills of healthcare workers in SRH services would be useful to ensure continuous education and training, and improved SRH knowledge and skills.