"I Am Now Five Steps Ahead": How Co-design Platforms Sustain Kenyan Community Health Volunteer Engagement.

OBJECTIVES: The aim of this study is to examine how humancentered design (HCD) as a platform for co-production might function to explain community health volunteer (CHV) motivation in self-directed and self-funded community health activities. Sustaining engagement has been difficult for CHVs who lack monetary incentives, expense reimbursement, and are rarely given opportunity to give their own voice to local health priorities. DESIGN: Qualitative study utilized focus group discussions 12 months post intervention and included both an inductive and deductive level of analysis. SETTING: Three community health units (CHU) representing Kenya's diversity were selected with the local Ministry of Health including peri-urban slum, rural agrarian, and a unit where informal day labor and rented housing was the norm. PARTICIPANTS: The participants were selected according to Kenya's community health strategy norms and had previously had the standard basic community health training. INTERVENTION: A 3-day training rooted in HCD utilized multiple quality improvement tools (asset mapping, root cause analysis, key drivers) in order to help CHVs uncover unarticulated community needs and assumptions and encourage behavior change. Action plans with Plan-Do-Study-Act cycles were tracked longitudinally. RESULTS: Key themes were self-interest, common goal, gratitude/indebtedness. Additional thematic analysis identified altruism as supporting sustained engagement. CONCLUSIONS: This study supports HCD as a platform for sustained CHV engagement. It builds the evidence for self-interest, common goal, and gratitude/indebtedness as sustaining factors. These factors are also seen in process-based theories that operationalize and measure trust building reciprocity cycles that mirror the iterative P-D-S-A cycles seen in HCD.

Outpatient management of urinary tract infections by medical officers in Nairobi, Kenya: lack of benefit from audit and feedback on adherence to treatment guidelines.

INTRODUCTION: Acute uncomplicated urinary tract infections are common in outpatient settings but are not treated optimally. Few studies of the outpatient use of antibiotics for specific diagnoses have been done in sub-Saharan Africa, so little is known about the prescribing patterns of medical officers in the region. METHODS: Aga Khan University has 16 outpatient clinics throughout the Nairobi metro area with a medical officer specifically assigned to that clinic. A baseline assessment of evaluation and treatment of suspected UTI was performed from medical records in these clinics. Then the medical officer from each of the 16 clinics was recruited from each clinic was recruited with eight each randomized to control vs. feedback groups. Both groups were given a multimodal educational session including locally adapted UTI guidelines and emphasis on problems identified in the baseline assessment Each record was scored using a scoring system that was developed for the study according to adequacy of history, physical examination, clinical diagnosis matching recorded data, diagnostic workup and treatment. Three audits were done for both groups; baseline (audit 1), post-CME (audit 2), and a final audit, which was after feedback for the feedback group (audit 3). The primary analysis assessed overall guideline adherence in the feedback group versus the CME only group. RESULTS: The overall scores in both groups showed significant improvement after the CME in comparison to baseline and for each group, the scores in most domains also improved. However, audit 3 showed persistence of the gains attained after the CME but no additional benefit from the feedback. Some deficiencies that persisted throughout the study included lack of workup of possible STI and excess use of non-UTI laboratory tests such as CBC, stool culture and H. pylori Ag. After the CME, the use of nitrofurantoin rose from only 4% to 8% and cephalosporin use increased from 49 to 67%, accompanied by a drop in quinolone use. CONCLUSION: The CME led to modest improvements in patient care in the categories of history taking, treatment and investigations, but feedback had no additional effect. Future studies should consider an enforcement element or a more intensive feedback approach.

Moving beyond audit: driving system learning using a novel mortality classification system in a tertiary training hospital in Kenya.

Clinical classification systems have proliferated since the APGAR score was introduced in 1953. Numerical scores and classification systems enable qualitative clinical descriptors to be transformed into categorical data, with both clinical utility and ability to provide a common language for learning. The clarity of classification rubrics embedded in a mortality classification system provides the shared basis for discussion and comparison of results. Mortality audits have been long seen as learning tools, but have tended to be siloed within a department and driven by individual learner need. We suggest that the learning needs of the system are also important. Therefore, the ability to learn from small mistakes and problems, rather than just from serious adverse events, remains facilitated.We describe a mortality classification system developed for use in the low-resource context and how it is 'fit for purpose,' able to drive both individual trainee, departmental and system learning. The utility of this classification system is that it addresses the low-resource context, including relevant factors such as limited prehospital emergency care, delayed presentation, and resource constraints. We describe five categories: (1) anticipated death or complication following terminal illness; (2) expected death or complication given clinical situation, despite taking preventive measures; (3) unexpected death or complication, not reasonably preventable; (4) potentially preventable death or complication: quality or systems issues identified and (5) unexpected death or complication resulting from medical intervention. We document how this classification system has driven learning at the individual trainee level, the departmental level, supported cross learning between departments and is being integrated into a comprehensive system-wide learning tool.

Preparedness, resilience and unmet needs of informal caregivers of advanced cancer patients in a Regional Mission Hospital in Kenya: Qualitative Study.

BACKGROUND: Cancer is the third highest cause of death in Kenya. Eighty percent of cancer cases arrive at advanced stages, when there is nothing that can be done to cure them, and palliative care is the best alternative. Although the majority of end-of-life care in Kenya is provided at home, little is known about the caregivers' preparedness, resilience and continued unmet needs. The goal of this qualitative study was to explore caregivers' perceived preparedness, resilience and continued unmet needs in their caregiving role to patients with advanced stages of cancer. METHODS: A purposive sampling method was used to identify and recruit twelve informal, home-based caregivers of patients with advanced cancer from Kijabe Palliative Clinic data base. Interviews were conducted in patients' homes. The data was analyzed using interpretive phenomenological analysis approach. Ethical considerations were observed. Participants were kept anonymous and confidentiality. RESULTS: Competing tasks, lack of preparedness in handling end-of-life care for patients in advanced stages of cancer were the main concerns. Continued unmet needs and financial stresses, and vulnerability for female caregivers all contributed to increased caregiver burden in this study. Caregivers were however determined and resilient amidst challenges that faced them, they exhibited hope against hopelessness. Some caregivers were vulnerable and faced potential for abuse following anticipated loss of their family member exacerbated psychosocial stress and needs CONCLUSION: Informal caregivers had common unmet needs related to caring for their advanced cancer patients. Whilst family caregivers had huge caregiver burden, insurmountable practical challenges related to role overload and competing tasks, they remained resilient though unprepared in giving end of life care. RECOMMENDATIONS: Caregivers should also be examined, prepared, and supported during clinic reviews. More research is needed on the use of telephones for caregiver follow-up, the impact of introducing caregiver-targeted screening tools on caregiver quality of life and their impact on enhancing caregiver well-being in order to prepare & support them adequately for the caregiving role.

Cancer referrals at African Inland Church Kijabe Hospital, Kiambu County (2014-2020) and the impact of COVID-19.

In Kenya, cancer is the third leading cause of death. The African Inland Church Kijabe Hospital (AICKH) is a level 4 missionary hospital. The hospital serves the Kenyan population in many areas, including cancer care, and some of these services were affected during the coronavirus disease 2019 (COVID-19) pandemic. We aimed to leverage a recently established hospital-based cancer registry of patients treated at AICKH between 2014 and 2020 to describe the cancer cases and patient referral patterns seen at AICKH during the COVID-19 pandemic in 2020. A cross-sectional retrospective survey was conducted through medical records abstraction in the surgery, breast clinic, palliative care and pathology departments. A total of 3279 cases were included in the study, with females accounting for 58.1% of the cases. The top-three cancers overall were breast (23.0%), oesophagus (20.5%) and prostate (8.6%). There was a minimal increase in the number of cancer cases in 2020 (1.7%) compared with 2019, with an increase of 19.3% in 2019 compared with 2018. In conclusion, AICKH is one of the few hospitals in Kenya where a large number of cancer patients seek healthcare, and referral of cancer cases changed in 2020, which may be due to the COVID-19 pandemic. Future efforts can leverage this registry to determine the impacts of cancer diagnosis and treatment on survival outcomes.

Antepartum SARS-CoV-2 infection and adverse birth outcomes in South African women.

Background: SARS-CoV-2 infection in pregnant women has been associated with severe illness in the women and higher rates of premature delivery. There is, however, paucity of data on the impact of the timing of SARS-CoV-2 infection and on symptomatic or asymptomatic infections on birth outcomes. Data from low-middle income settings is also lacking. Methods: We conducted a longitudinal study from April 2020 to March 2021, in South Africa, where symptomatic or asymptomatic pregnant women were investigated for SARS-CoV-2 infection during the antepartum period. We aimed to evaluate if there was an association between antepartum SARS-CoV-2 infection on birth outcomes. SARS-CoV-2 infection was investigated by nucleic acid amplification test (NAAT), histological examination was performed in a sub-set of placentas. Results: Overall, 793 women were tested for SARS-CoV-2 antenatally, including 275 (35%) who were symptomatic. SARS-CoV-2 infection was identified in 138 (17%) women, of whom 119 had symptoms (COVID-19 group) and 19 were asymptomatic. The 493 women who were asymptomatic and had a negative SARS-CoV-2 NAAT were used as the referent comparator group for outcomes evaluation. Women with COVID-19 compared with the referent group were 1.66-times (95% confidence interval (CI) = 1.02-2.71) more likely to have a low-birthweight newborn (30% vs 21%) and 3.25-times more likely to deliver a very low-birthweight newborn (5% vs 2%). Similar results for low-birthweight were obtained comparing women with SARS-CoV-2 confirmed infection (30%) with those who had a negative NAAT result (22%) independent of symptoms presentation. The placentas from women with antenatal SARS-CoV-2 infection had higher percentage of chorangiosis (odds ratio (OR) = 3.40, 95% CI = 1.18-.84), while maternal vascular malperfusion was more frequently identified in women who tested negative for SARS-CoV-2 (aOR = 0.28, 95% CI = 0.09-0.89). Conclusions: Our study demonstrates that in a setting with high HIV infection prevalence and other comorbidities antenatal SARS-CoV-2 infection was associated with low-birthweight delivery.

Active Intrapartum SARS-CoV-2 Infection and Pregnancy Outcomes.

OBJECTIVE: Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) infection during pregnancy has been associated with poor pregnancy outcomes. There is, however, not much information on the impact of the timing of SARS-CoV-2 infection on pregnancy outcomes, and studies from low-middle income settings are also scarce. STUDY DESIGN: We conducted a cross-sectional study from April to December 2020, in South Africa, to assess the association of SARS-CoV-2 infection on a nasal swab at the time of labor with fetal death, preterm birth, low birth weight, or pregnancy-induced complications. When possible, maternal blood, cord blood, and placenta were collected. SARS-CoV-2 infection was investigated by a nucleic acid amplification test (NAAT). RESULTS: Overall, 3,117 women were tested for SARS-CoV-2 on a nasal swab, including 1,562 (50%) healthy women with uncomplicated term delivery. A positive NAAT was detected among 132 (4%) women. Adverse birth outcomes or pregnancy-related complications were not associated with SARS-CoV-2 infection at the time of labor. Among SARS-CoV-2-infected women, an NAAT-positive result was also obtained from 6 out of 98 (6%) maternal blood samples, 8 out of 93 (9%) cord-blood samples, 14 out of 54 (26%) placentas, and 3 out of 22 (14%) nasopharyngeal swabs from newborns collected within 72 hours of birth. Histological assessment of placental tissue revealed that women with SARS-CoV-2 nasal infection had a higher odds (3.82, 95% confidence interval: 1.20, 12.19) of chronic chorioamnionitis compared with those without SARS-CoV-2 infection. CONCLUSION: Our study demonstrates that intrapartum, SARS-CoV-2 infection was not associated with evaluated poor outcomes. In utero fetal and placental infections and possible vertical and/or horizontal viral transfer to the newborn were detected among women with nasal SARS-CoV-2 infection. KEY POINTS: · Intrapartum SARS-CoV-2 infection was not associated with evaluated poor outcomes.. · In utero fetal and placental infections were detected among women with nasal SARS-CoV-2 infection.. · Women with SARS-CoV-2 nasal infection had a higher odds of chronic chorioamnionitis..

Living on the Edge of Possibility: Ethical Issues in the Care of Critically Ill Patients in Resource-Limited Settings.

The birth, expansion, and sustenance of critical care medicine as a specialty have often presented ethical challenges and dilemmas to health care workers in diverse settings. In addition to critical services being provided at the extreme end of a disease process, they are often in limited supply. The authors present patterns of inception and development of this crucial service as they have witnessed in rural Africa. Furthermore, they present the ethical challenges, both typical and unique, as they have experienced them in a tertiary referral center in Kenya.

Maternal-foetal transfer of severe acute respiratory syndrome coronavirus 2 antibodies among women with and those without HIV infection.

In pregnant women, antibodies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein cross the placenta and can be detected in cord-blood at the time of delivery. We measured SARS-CoV-2 full-length antispike IgG in blood samples collected from women living with HIV (WLWHIV) and without HIV when presenting for labour, and from paired cord-blood samples. Antispike IgG was measured in maternal blood at delivery on the Luminex platform. Cord-blood samples from newborns of women in with detectable antispike IgG were analysed. The IgG geometric mean concentrations (GMCs) and the percentage of cord-blood samples with detectable antispike IgG were compared between WLWHIV and without HIV. A total of 184 maternal and cord-blood pairs were analysed, including 47 WLWHIV and 137 without HIV. There was no difference in antispike GMCs between WLWHIV and without HIV [157 binding antibody units (BAU)/ml vs. 187 BAU/ml; P  = 0.17)]. Cord-blood samples from newborns of WLWHIV had lower GMCs compared with those without HIV (143 vs. 205 BAU/ml; P  = 0.033). Cord-to-maternal blood antibody ratio was 1.0 and similar between the two HIV groups. In WLWHIV, those who were 30 years old or less had lower cord-to-maternal blood antibody ratio (0.75 vs. 1.10; P  = 0.037) and their newborns had lower cord-blood GMCs (94 vs. 194 BAU/ml; P  = 0.04) compared with the older women. Independently of maternal HIV infection status, there was efficient transplacental transfer of antispike antibodies. The GMCs in cord-blood from newborns of WLWHIV were lower than those in HIV-unexposed newborns.

Understanding intra- and interprofessional team and teamwork processes by exploring facility-based neonatal care in kenyan hospitals.

BACKGROUND: Within intensive care settings such as neonatal intensive care units, effective intra- and interprofessional teamwork has been linked to a significant reduction of errors and overall improvement in the quality of care. In Kenya, previous studies suggest that coordination of care among healthcare teams providing newborn care is poor. Initiatives aimed at improving intra- and interprofessional teamwork in healthcare settings largely draw on studies conducted in high-income countries, with those from resource-constrained low and middle countries, particularly in the context of newborn care lacking. In this study, we explored the nature of intra- and interprofessional teamwork among health care providers in newborn units (NBUs) of three hospitals in Kenya, and the professional and contextual dynamics that shaped their interactions. METHODS: This exploratory qualitative study was conducted in three hospitals in Nairobi and Muranga Counties in Kenya. We adopted an ethnographic approach, utilizing both in-depth interviews (17) and non-participant observation of routine care provision in NBUs (250 observation hours). The study participants included: nurses, nursing students, doctors, and trainee doctors. All the data were thematically coded in NVIVO 12. RESULTS: The nature of intra- and interprofessional teamwork among healthcare providers in the study newborn units is primarily shaped by broader contextual factors and varying institutional contexts. As a result, several team types emerged, loosely categorized as the 'core' team which involves providers physically present in the unit most times during the work shift; the emergency team and the temporary ad-hoc teams which involved the 'core' team, support staff students and mothers. The emergence of these team types influenced relationships among providers. Overall, institutionalized routines and rituals shaped team relations and overall functioning. CONCLUSIONS: Poor coordination and the sub-optimal nature of intra-and interprofessional teamwork in NBUs are attributed to broader contextual challenges that include low staff to patient ratios and institutionalized routines and rituals that influenced team norming, relationships, and team leadership. Therefore, mechanisms to improve coordination and collaboration among healthcare teams in these settings need to consider contextual dynamics including institutional cultures while also targeting improvement of team-level processes including leadership development and widening spaces for more interaction and better communication.

Germs, Guns, and Fear in Disaster Response: A Rapid Qualitative Assessment to Understand Fear-Based Responses in the Population at Large: Lessons From Sierra Leone 2014-2015.

We set out to assess the feasibility of community-focused randomized qualitative assessment at the start of an emergency to identify the root causes of fear-based responses driving the pandemic. We used key informant interviews, focus group discussions, reviewing of government and non-government organization documents, combined with direct field observation. Data were recorded and analyzed for key-themes: (1) lack of evidence-based information about Ebola; (2) lack of support to quarantined families; (3) culturally imbedded practices of caring for ill family members; (4) strong feeling that the government would not help them, and the communities needed to help themselves: (5) distrust of nongovernmental organizations and Ebola treatment centers that the communities viewed as opportunistic. On-the-ground real-time engagement with stakeholders provided deep insight into fear-based-responses during the Ebola epidemic, formed a coherent understanding of how they drove the epidemic, presenting an alternative to the standard disaster-response United Nations-strategy, producing community-driven solutions with local ownership.

Exploring healthcare workers' perceptions on the use of morbidity and mortality audits as an avenue for learning and care improvement in Kenyan hospitals' newborn units.

BACKGROUND: In many sub-Saharan African countries, including Kenya, the use of mortality and morbidity audits in maternal and perinatal/neonatal care as an avenue for learning and improving care delivery is sub-optimal due to structural, organizational, and human barriers. While attempts to address these barriers have been reported, lots of emphasis has been paid to addressing the role of tangible inputs (e.g., availing guidelines and training staff in the success of mortality and morbidity audits), while process-related factors (i.e., the role of the people, their experiences, relationships, and motivations) remain inadequately explored. We examined the processes of neonatal audits, their potential in promoting learning from gaps in care and improving care delivery, with a deliberate focus on process-related factors that generally influence mortality and morbidity (M&M) audits. METHODS: This was an exploratory qualitative study, conducted in three hospitals, in Nairobi and Muranga counties. We employed a mix of in-depth interviews (17) and observation of 12 mortality and morbidity audit meetings. Our study participants included: nurses, doctors, trainee clinicians (i.e., junior doctors on internships), and nursing students involved in providing newborn care. These data were coded using NVivo12 employing a thematic content analysis approach. RESULTS: Perceived shortcomings in the conduct of M&M audits such as unclear structure was reported to have contributed to its sub-optimal nature in promoting learning. These shortcomings, in addition to hierarchy and power dynamics, poor implementation of audit recommendations, and negative experiences, (e.g., blame) also demotivated health workers from attendance and participation in audits. Despite these, positive outcomes linked to audit recommendations, such as revision of care protocols, were reported. Overall, leadership and a blame-free culture enabled positive changes and promoted learning from audit-identified modifiable factors. CONCLUSION: Our findings indicate that M&M audits provide a space for meaningful discussions, which may lead to learning and improvement in care delivery processes. However, a lack of participation, lack of observed positive outcomes, and negative experiences may reduce their usefulness. An enabling environment characterized by minimized effects of hierarchy and positive use of power and a blame-free culture may promote active participation, enhancing positive relationships and interactions thus promoting team learning.

Understanding Women's Choices: How Women's Perceptions of Quality of Care Influences Place of Delivery in a Rural Sub-County in Kenya. A Qualitative Study.

BACKGROUND: Maternal mortality is still unacceptably high in Kenya. The Kenyan Government introduced a free maternity service to overcome financial barriers to access. This policy led to a substantial increase in women's delivery options. This increase in coverage might have led to a reduction in quality of care. This study explores women's perceptions of quality of delivery services in the context of the free policy and how the perceptions lead to the choice of a place for delivery. METHODS: Our study site was Naivasha sub-County in Kenya, a rural context, whose geography encompasses pastoralists, rural agrarian, and high population density informal settlements near flower farms. Women from this area are from the lowest wealth quintile in Kenya. We conducted a qualitative study to explore the women's perceptions of quality of care based on their experiences during maternity care. The participants were women of reproductive age (18-49 years) attending antenatal care clinics at six health facilities in the sub-county. Six focus group discussions with 55 respondents were used. For inclusion, the women needed to have delivered a baby within the six months preceding the study. Interviews were recorded with consent, translated and transcribed. The interviews were analyzed using a thematic content approach. RESULTS: Four broad themes that determined the choice of health facility for delivery were identified: women's perceptions of clinical quality of care; the cost of delivery; distance to the health facility and management of primary health facilities. An unexpected theme was the presence of home deliveries amongst pastoralist women. These findings suggest that in this setting both process and structural dimensions of quality of care and financial and physical accessibility influence women's choices for place of delivery. CONCLUSION: This study expands our understanding of how women make choices regarding place of delivery. Understanding women's perceptions can provide useful insights to policy makers and facility managers on providing high quality patient centered maternity care necessary to sustain the increased utilization of maternity services at health facilities under the free maternity policy and further reductions in maternal mortality.

Coping with COVID: Developing a Rapid-cycle Frontline Quality-improvement Process to Support Employee Well-being and Drive Institutional Responsiveness in a Tertiary Care Faith-based Hospital in Rural Kenya.

The coronavirus disease 2019 (COVID-19) pandemic has demanded rapid institutional responses to meet the needs of patients and employees in the face of a serious new disease. To support the well-being of frontline staff, a series of debriefing sessions was used to drive a rapid-cycle quality-improvement process. The goals were to confidentially determine personal coping strategies used by staff, provide an opportunity for staff cross-learning, identify what staff needed most, and provide a real-time feedback loop for decision-makers to create rapid changes to support staff safety and coping. Data were collected via sticky notes on flip charts to protect confidentiality. Management reviewed the data daily. Institutional responses to problems identified during debrief sessions were tracked, visualized, addressed, and shared with staff. More than 10% of staff participated over a 2-week period. Feedback influenced institutional decisions to improve staff schedules, transportation, and COVID-19 training.

Eliciting women's preferences for place of child birth at a peri-urban setting in Nairobi, Kenya: A discrete choice experiment.

OBJECTIVE: Maternal and newborn mortality rates are high in peri-urban areas in cities in Kenya, yet little is known about what drives women's decisions on where to deliver. This study aimed at understanding women's preferences on place of childbirth and how sociodemographic factors shape these preferences. METHODS: This study used a Discrete Choice Experiment (DCE) to quantify the relative importance of attributes on women's choice of place of childbirth within a peri-urban setting in Nairobi, Kenya. Participants were women aged 18-49 years, who had delivered at six health facilities. The DCE consisted of six attributes: cleanliness, availability of medical equipment and drug supplies, attitude of healthcare worker, cost of delivery services, the quality of clinical services, distance and an opt-out alternative. Each woman received eight questions. A conditional logit model established the relative strength of preferences. A mixed logit model was used to assess how women's preferences for selected attributes changed based on their sociodemographic characteristics. RESULTS: 411 women participated in the Discrete Choice Experiment, a response rate of 97.6% and completed 20,080 choice tasks. Health facility cleanliness was found to have the strongest association with choice of health facility (ß = 1.488 p<0.001) followed respectively by medical equipment and supplies availability (ß = 1.435 p<0.001). The opt-out alternative (ß = 1.424 p<0.001) came third. The attitude of the health care workers (ß = 1.347, p<0.001), quality of clinical services (ß = 0.385, p<0.001), distance (ß = 0.339, p<0.001) and cost (ß = 0.0002 p<0.001) were ranked 4th to 7th respectively. Women who were younger and were the main income earners having a stronger preference for clean health facilities. Older married women had stronger preference for availability of medical equipment and kind healthcare workers. CONCLUSIONS: Women preferred both technical and process indicators of quality of care. DCE's can lead to the development of person-centered strategies that take into account the preferences of women to improve maternal and newborn health outcomes.

Understanding what women want: eliciting preference for delivery health facility in a rural subcounty in Kenya, a discrete choice experiment.

OBJECTIVE: To identify what women want in a delivery health facility and how they rank the attributes that influence the choice of a place of delivery. DESIGN: A discrete choice experiment (DCE) was conducted to elicit rural women's preferences for choice of delivery health facility. Data were analysed using a conditional logit model to evaluate the relative importance of the selected attributes. A mixed multinomial model evaluated how interactions with sociodemographic variables influence the choice of the selected attributes. SETTING: Six health facilities in a rural subcounty. PARTICIPANTS: Women aged 18-49 years who had delivered within 6 weeks. PRIMARY OUTCOME: The DCE required women to select from hypothetical health facility A or B or opt-out alternative. RESULTS: A total of 474 participants were sampled, 466 participants completed the survey (response rate 98%). The attribute with the strongest association with health facility preference was having a kind and supportive healthcare worker (ß=1.184, p<0.001), second availability of medical equipment and drug supplies (ß=1.073, p<0.001) and third quality of clinical services (ß=0.826, p<0.001). Distance, availability of referral services and costs were ranked fourth, fifth and sixth, respectively (ß=0.457, p<0.001; ß=0.266, p<0.001; and ß=0.000018, p<0.001). The opt-out alternative ranked last suggesting a disutility for home delivery (ß=-0.849, p<0.001). CONCLUSION: The most highly valued attribute was a process indicator of quality of care followed by technical indicators. Policymakers need to consider women's preferences to inform strategies that are person centred and lead to improvements in quality of care during delivery.

Implementation research and human-centred design: how theory driven human-centred design can sustain trust in complex health systems, support measurement and drive sustained community health volunteer engagement.

Human-centred design (HCD) can support complex health system interventions by navigating thorny implementation problems that often derail population health efforts. HCD is a pragmatic, 'practice framework', not an intervention protocol. It can build empathy by bringing patient voice, user perspective and innovation to construct and repair pieces of the intervention or health system. However, its emphasis on product development and process change with fixed end points has left it as an approach lacking explanatory power and reproducible measurement. Yet when informed by theory, the tremendous innovation potential of HCD can be harnessed to drive sustainability, mediate implementation problems, frame measurement constructs and ultimately improve population-level health outcomes. In attempting to mine, the potential of HCD we move beyond the pragmatic 'how it works', to the theoretical question, 'why it works'. In doing so, we explore a more fundamental human question, 'How can participation and engagement be sustained for impact in close to the community health systems?' In this exploration, we illustrate the power of HCD by linking it to our theory of trust building. The research method we utilize is that of a longitudinal process evaluation. We leverage the heterogeneity of five community health units from the diverse setting (rural, peri-urban slum) to better understand what works for whom and in what context by tracking 21 groups of community health volunteers (CHVs) over 12 months. We report results with a focus on the outlier case failure to illustrate the contrast with common features of sustained CHV engagement, where recurrent reciprocal cycles of trust building are demonstrated in the successful implementation of action plans in plan-do-study-act cycles for improvement. All was accomplished by CHVs with no external funding. We conclude by discussing how HCD could be unleashed if linked to theoretical frameworks, increasing ability to address implementation challenges in complex health systems.

'We just look at the well-being of the baby and not the money required': a qualitative study exploring experiences of quality of maternity care among women in Nairobi's informal settlements in Kenya.

OBJECTIVE: To examine how women living in an informal settlement in Nairobi perceive the quality of maternity care and how it influences their choice of a delivery health facility. DESIGN: Qualitative study. SETTINGS: Dandora, an informal settlement, Nairobi City in Kenya. PARTICIPANTS: Six focus group discussions with 40 purposively selected women aged 18-49 years at six health facilities. RESULTS: Four broad themes were identified: (1) perceived quality of the delivery services, (2) financial access to delivery service, (3) physical amenities at the health facility, and (4) the 2017 health workers' strike.The four facilitators that influenced women to choose a private health facility were: (1) interpersonal treatment at health facilities, (2) perceived quality of clinical services, (3) financial access to health services at the facility, and (4) the physical amenities at the health facility. The three barriers to choosing a private facility were: (1) poor quality clinical services at low-cost health facilities, (2) shortage of specialist doctors, and (3) referral to public health facilities during emergencies.The facilitators that influenced women to choose a public facility were: (1) physical amenities for dealing with obstetric emergencies and (2) early referral to public maternity during antenatal care services. Barriers to choosing a public facility were: (1) perception of poor quality clinical services, (2) concerns over security for newborns at tertiary health facilities, (3) fear of mistreatment during delivery, (4) use of unsupervised trainee doctors for deliveries, (5) poor quality of physical amenities, and (6) inadequate staffing. CONCLUSION: The study provides insights into decision-making processes for women when choosing a delivery facility by identifying critical attributes that they value and how perceptions of quality influence their choices.

The impact of the nurses', doctors' and clinical officer strikes on mortality in four health facilities in Kenya.

BACKGROUND: Health worker strikes are a significant threat to universal access to care globally and especially in sub Saharan Africa. Kenya's health sector has seen an increase in such industrial action. Globally there is limited data that has examined mortality related to such strikes in countries where emergency services were preserved. We sought to assess the mortality impact of an 100 day physician strike which was followed by 151 day nurses' strike and 20 day clinical officer strike in Kenya. METHODS: Monthly mortality data was abstracted from four public hospitals, Kenyatta National Referral Hospital, AIC Kijabe Hospital, Mbagathi Hospital and Siaya Hospital between December 2016 and March 2018. Differences in mortality were assessed using t-tests and multiple linear regression adjusting for facility, numbers of patients utilizing the hospital and department. RESULTS: There was a significant decline in the numbers of patients seen, comparing the non-strike and strike periods; beta (ß) coefficient - 649 (95% CI -950, - 347) p < 0.0001. The physicians' strike saw a significant decline in mortality (ß) coefficient - 19.0 (95%CI -29.2, - 8.87) p < 0.0001. Nurses and Clinical Officer strikes' did not significantly impact mortality. There was no mortality increase in the post-strike period beta (ß) coefficient 7.42 (95%CI -16.7, 1.85) p = 0.12. CONCLUSION: Declines in facility-based mortality during strike months was noted when compared to a non-striking facility, where mortality increased. The decline is possibly associated with the reduced patient volumes, and a possible change in quality of care. Public health facilities are congested and over-utilized by the local population majority of whom cannot afford even low cost private care. Health worker strikes in Kenya where the public health system is the only financially accessible option for 80% of the population pose a significant threat to universal access to care. Judicious investment in the health infrastructure and staffing may decrease congestion and improve quality of care with attendant mortality decline.