Menu
Menu
Menu
Menu
Background: Computerized decision support systems (CDSS) and performance-based incentives (PBIs) can improve health-worker performance. However, there is minimal evidence on the combined effects of these interventions or perceived effects among maternal and child healthcare providers in low-resource settings. We thus aimed to explore the perceptions of maternal and child healthcare providers of CDSS support in the context of a combined CDSS-PBI intervention on performance in twelve primary care facilities in Ghana’s Upper East Region. Methods: We conducted a qualitative study drawing on semi-structured key informant interviews with 24 nurses and midwives, 12 health facility managers, and 6 district-level staff familiar with the intervention. We analysed data thematically using deductive and inductive coding in NVivo 10 software. Results: Interviewees suggested the combined CDSS-PBI intervention improved their performance, through enhancing knowledge of maternal health issues, facilitating diagnoses and prescribing, prompting actions for complications, and improving management. Some interviewees reported improved morbidity and mortality. However, challenges described in patient care included CDSS software inflexibility (e.g. requiring administration of only one intermittent preventive malaria treatment to pregnant women), faulty electronic partograph leading to unnecessary referrals, increased workload for nurses and midwives who still had to complete facility forms, and power fluctuations affecting software. Conclusion: Combining CDSS and PBI interventions has potential to improve maternal and child healthcare provision in low-income settings. However, user perspectives and context must be considered, along with allowance for revisions, when designing and implementing CDSS and PBIs interventions.
This qualitative study reanalyses 66 semi-structured interviews conducted with frontline health-workers, facility managers, and district supervisors at the end of the European Union-funded Quality of maternal and neonatal health (QUALMAT) intervention trial [15, 26, 28]. This cluster-randomized un-blinded controlled trial, conducted in Ghana’s Upper East region, aimed to assess the effect of a CDSS for maternal and neonatal health care services. The trial was conducted in twelve health facilities in Kassena-Nankana (KND) and Builsa districts and is described in Aninanya et al. [16]. KND (intervention district) has an estimated population of 152,000 served by one hospital in the district capital Navrongo, six health centres, one private clinic, and twenty-seven Community-based Health Planning and Services (CHPS) compounds. Builsa (comparison district) has an estimated population of 95,800 served by the district hospital in Sandema, six health centres, and thirteen CHPS compounds. MNH services were considered poor in Upper East Region, due to insufficient and demotivated staff; inadequate access to and use of computers, reproductive health guidance, and protocols; and inadequate diagnoses and referrals [29–31]. A cross-sectional study of 400 new mothers indicated 93% had delivered in health facilities and 97% had received antenatal care with 75% having four or more antenatal visits [32]. However, institutional MMR in the Upper East Region prior to the trial was 352 per 100,000 live births and estimated at 367 and 259 per 100,000 live births In KND and Builsa district respectively [33]. Nurses and midwives are responsible for provision of facility-based MNH services in these districts [34], with low productivity reported as some providers rarely used treatment guidelines or partograph, and missed health education opportunities [8]. CDSS and PBI interventions were implemented in six KND intervention facilities for two years (2012–2014). CDSS included computerized clinical support in implementing WHO ‘Pregnancy, Childbirth, Postpartum and Newborn Care: a guide for essential practice’ guidelines for 35 purposively selected and trained midwives, community health nurses, and facility managers. Each intervention facility received a laptop, with CDSS software and IT support, shared between three trained midwives and nurses, while facility managers were tasked with supervision. Performance was rewarded based on improvements within facilities rather than in comparison to other facilities. Individuals were identified for PBIs based on achieving key indicators determined by facility leadership, regional and district health directorates, and the PBI committee. PBI consisted of rewarding best-performing midwives and facilities with domestic appliances (e.g. blenders, fans, stoves, freezers, fridges, televisions, saucepans, cloths, kettles, microwaves) and recognition certificates at annual awards ceremonies, along with regular supervision, verbal appreciation, furniture, and small monthly allowances [15, 16]. Key performance indicators included proportion of ANC visits recorded in CDSS; proportions of pregnant women receiving tetanus vaccination, iron supplementation, safe sex and HIV counselling; proportion of births attended by skilled personnel; proportion of partographs completed in CDSS; partograph usage rate; proportion of women referred based on CDSS recommendation; and proportion of newborns vaccinated [16]. We included all 24 participating CDSS users, 6 facility managers, and 6 district-level staff in intervention facilities and used purposive non-probability sampling to select equal numbers of equivalent participants in comparison facilities. This enabled us to include non-intervention participants with similar workloads who could provide relevant insights. GAA obtained written informed consent from all participants before interview. We developed an interview guide from the literature and expert opinion. Topics included basic demographics, experiences of CDSS and PBI interventions if relevant, work challenges, and suggested solutions. GAA and two research assistants conducted in-person audio-recorded interviews in English, lasting 35–45 min, in locations such as homes and health facilities as chosen by participants. We used interviews with facility managers and district-level staff, because of their varied involvement supervising the intervention, to triangulate and crystalize findings from frontline providers. To ensure confidentiality, we assigned identification codes and did not include personal identifiers in study tools or outputs. Audio files were transcribed verbatim by two professional transcriptionists. GAA imported transcripts into Nvivo 10 (QSR International Pty Ltd, Victoria, Australia) data management software [35] and analysed them thematically using Braun and Clarke’s six-stage approach [36]. In summary, GAA read and became familiarised with the data, identified deductive codes based on interview guide topics and inductively coded new or unexpected topics arising in transcripts. She developed a coding structure iteratively, collating codes related to intervention effects, challenges, and suggestions into themes. GAA examined relationships between codes, compiled and summarised contents of each theme with support from co-authors, and conducted thematic mapping. GAA and NH refined and defined themes and sub-themes through discussion, further integration, and the reporting process. Triangulation of participant perspectives (i.e. midwives, nurses, medical assistants, district public health nurses, directors) at different health system levels and in both study arms helped improve validity.
N/A
