Attitudes and behaviours of maternal health care providers in interactions with clients: A systematic review

Background: High maternal mortality and morbidity persist, in large part due to inadequate access to timely and quality health care. Attitudes and behaviours of maternal health care providers (MHCPs) influence health care seeking and quality of care. Methods: Five electronic databases were searched for studies from January 1990 to December 2014. Included studies report on types or impacts of MHCP attitudes and behaviours towards their clients, or the factors influencing these attitudes and behaviours. Attitudes and behaviours mentioned in relation to HIV infection, and studies of health providers outside the formal health system, such as traditional birth attendants, were excluded. Findings: Of 967 titles and 412 abstracts screened, 125 full-text papers were reviewed and 81 included. Around two-thirds used qualitative methods and over half studied public-sector facilities. Most studies were in Africa (n = 55), followed by Asia and the Pacific (n = 17). Fifty-eight studies covered only negative attitudes or behaviours, with a minority describing positive provider behaviours, such as being caring, respectful, sympathetic and helpful. Negative attitudes and behaviours commonly entailed verbal abuse (n = 45), rudeness such as ignoring or ridiculing patients (n = 35), or neglect (n = 32). Studies also documented physical abuse towards women, absenteeism or unavailability of providers, corruption, lack of regard for privacy, poor communication, unwillingness to accommodate traditional practices, and authoritarian or frightening attitudes. These behaviours were influenced by provider workload, patients’ attitudes and behaviours, provider beliefs and prejudices, and feelings of superiority among MHCPs. Overall, negative attitudes and behaviours undermined health care seeking and affected patient well-being. Conclusions: The review documented a broad range of negative MHCP attitudes and behaviours affecting patient well-being, satisfaction with care and care seeking. Reported negative patient interactions far outweigh positive ones. The nature of the factors which influence health worker attitudes and behaviours suggests that strengthening health systems, and workforce development, including in communication and counselling skills, are important. Greater attention is required to the attitudes and behaviours of MHCPs within efforts to improve maternal health, for the sake of both women and health care providers. Five electronic databases were searched: the Cochrane Library, CINAHL Complete, Medline (PubMed), Popline and PsychInfo. Search strings were developed based on identifying key words and medical subject headings related to the population (MHCPs in LMICs), the “intervention” (attitudes and behaviours), and potential outcomes (satisfaction, acceptability, access, utilization, and health-seeking behaviours). The full search strategy is included as Additional file 1. Reference lists of included studies and reviews located on the topic were examined to identify additional literature. Retrieved records were imported into the reference management software EndNote X4 and assessed against inclusion and exclusion criteria in three stages – screening of titles, abstracts, and finally full texts. This study was limited to literature published in English from January 1990 to 1 December 2014. As the aim was to explore the breadth of the research undertaken on MHCP attitudes and behaviours in LMICs, all types of study design were included. MHCPs were defined as trained providers (such as medical doctors, nurses, midwives and paramedics) delivering antenatal, abortion, childbirth or postnatal services (including family planning) up to one year after childbirth. Studies on experiences of HIV-positive women within maternal health services were not included here as HIV itself incurs marked stigma and discrimination, with corresponding implications for service utilization and health outcomes [27–33]. Given that provider attitudes and behaviours towards HIV likely differ considerably from other conditions, this was considered a separate review and beyond the scope of this study. The LMICs included were drawn from the World Bank’s classification of countries’ income status in July 2012. Studies were included if they reported on the types of attitudes and behaviours, the factors influencing these, and/or the impacts resulting from certain attitudes and behaviours. Reports which simply stated that the attitude or behaviour was ‘positive’ or ‘negative’ without providing additional details on the type of attitude or behaviour, or the influences or impacts of the positive or negative attitudes and behaviours were excluded. We also excluded studies related to health care for children; case studies of the experience of one patient or one MHCP only; and studies describing factors which influence quality of care without specifying the impact of MHCP behaviours and attitudes. A thematic analysis approach was used to synthesize the evidence located. Text relevant to attitudes and behaviours, and their influences and impacts, was extracted from full-text documents and those that were similar or conceptually-related were grouped together. Thus, for example, insulting and humiliating speech, shouting and scolding were classified as ‘verbal abuse’; whilst ignoring patients or being uncaring, dismissive or hostile were classified as ‘rudeness’. Selected quotations from participants as reported in the studies were copied verbatim to further illustrate dominant themes or notable exceptions to these. For each paper included in the review, information was extracted into a standardized data tool on: (1) study characteristics (first author and year of publication, study design and setting); (2) study population; (3) type of facility (public or private) and health worker cadre; (4) type of attitude or behaviour, grouped as positive and negative; (5) factors influencing attitudes and behaviours; and (6) impact of attitudes and behaviours.