Hospital readiness for the provision of care to critically ill patients in Tanzania- an in-depth cross-sectional study

BACKGROUND

Critical illness is a state of ill health with vital organ dysfunction, a high risk of imminent death if care is not provided and potential for reversibility. The burden of critical illness is high, especially in low- and middle-income countries. Critical care can be provided as Essential Emergency and Critical Care (EECC)- the effective, low-cost, basic care that all critically ill patients should receive in all parts of all hospitals in the world- and advanced critical care- complex, resource-intensive care usually provided in an intensive care unit. The required resources may be available in the hospital and yet not be ready in the wards for immediate use for critically ill patients. The ward readiness of these resources, although harder to evaluate, is likely more important than their availability in the hospital. This study aimed to assess the ward readiness for EECC and the hospital availability of resources for EECC and for advanced critical care in hospitals in Tanzania.

METHODS

An in-depth, cross-sectional study was conducted in five purposively selected hospitals by visiting all wards to collect data on all the required 66 EECC and 161 advanced critical care resources. We defined hospital-availability as a resource present in the hospital and ward-readiness as a resource available, functioning, and present in the right place, time and amounts for critically ill patient care in the wards. Data were analyzed to calculate availability and readiness scores as proportions of the resources that were available at hospital level, and ready at ward level respectively.

RESULTS

Availability of EECC resources in hospitals was 84% and readiness in the wards was 56%. District hospitals had lower readiness scores (less than 50%) than regional and tertiary hospitals. Equipment readiness was highest (65%) while that of guidelines lowest (3%). Availability of advanced critical care resources was 31%.

CONCLUSION

Hospitals in Tanzania lack readiness for the provision of EECC- the low-cost, life-saving care for critically ill patients. The resources for EECC were available in hospitals, but were not ready for the immediate needs of critically ill patients in the wards. To provide effective EECC to all patients, improvements are needed around the essential, low-cost resources in hospital wards that are essential for decreasing preventable deaths.

Receive, Sustain, and Flow: A simple heuristic for facilitating the identification and treatment of critically ill patients during their hospital journeys

Background

Hospital patients can become critically ill anywhere in a hospital but their survival is affected by problems of identification and adequate, timely, treatment. This is issue of particular concern in lower middle-income countries' (LMICs) hospitals where specialised units are scarce and severely under-resourced. "Cross-sectional" approaches to improving narrow, specific aspects of care will not attend to issues that affect patients' care across the length of their experience. A simpler approach to understanding key issues across the "hospital journey" could help to deliver life-saving treatments to those patients who need it, wherever they are in the facility.

Methods

We carried out 31 narrative interviews with frontline health workers in five Kenyan and five Tanzanian hospitals from November 2020 to December 2021 during the COVID-19 pandemic and analysed using a thematic analysis approach. We also followed 12 patient hospital journeys, through the course of treatment of very sick patients admitted to the hospitals we studied.

Results

Our research explores gaps in hospital systems that result in lapses in effective, continuous care across the hospital journeys of patients in Tanzania and Kenya. We organise these factors according to the Systems Engineering Initiative for Patient Safety (SEIPS) approach to patient safety, which we extend to explore how these issues affect patients across the course of care. We discern three repeating, recursive phases we term Receive, Sustain, and Flow. We use this heuristic to show how gaps and weaknesses in service provision affect critically ill patients' hospital journeys.

Conclusion

Receive, Sustain, and Flow offers a heuristic for hospital management to identify and ameliorate limitations in human and technical resources for the care of the critically ill.

Hospital care for critical illness in low-resource settings: lessons learned during the COVID-19 pandemic

Care for the critically ill patients is often considered synonymous with a hospital having an intensive care unit. However, a focus on Essential Emergency and Critical Care (EECC) may obviate the need for much intensive care. Severe COVID-19 presented a specific critical care challenge while also being an exemplar of critical illness in general. Our multidisciplinary team conducted research in Kenya and Tanzania on hospitals' ability to provide EECC as the COVID-19 pandemic unfolded. Important basic inputs were often lacking, especially sufficient numbers of skilled health workers. However, we learnt that higher scores on resource readiness scales were often misleading, as resources were often insufficient or not functional in all the clinical areas they are needed. By following patient journeys, through interviews and group discussions, we revealed gaps in timeliness, continuity and delivery of care. Generic challenges in transitions between departments were identified in the receipt of critically ill patients, the ability to sustain monitoring and treatment and preparation for any subsequent transition. While the global response to COVID-19 focused initially on providing technologies and training, first ventilators and later oxygen, organisational and procedural challenges seemed largely ignored. Yet, they may even be exacerbated by new technologies. Efforts to improve care for the critically ill patients, which is a complex process, must include a whole system and whole facility view spanning all areas of patients' care and their transitions and not be focused on a single location providing 'critical care'. We propose a five-part strategy to support the system changes needed.

The hidden emotional labour behind ensuring the social value of research: Experiences of frontline health policy and systems researchers based in Kenya during COVID-19

Health policy and systems research (HPSR) is a multi-disciplinary, largely applied field of research aimed at understanding and strengthening the performance of health systems, often with an emphasis on power, policy and equity. The value of embedded and participatory HPSR specifically in facilitating the collection of rich data that is relevant to addressing real-world challenges is increasingly recognised. However, the potential contributions and challenges of HPSR in the context of shocks and crises are not well documented, with a particular gap in the literature being the experiences and coping strategies of the HPSR researchers who are embedded in health systems in resource constrained settings. In this paper, we draw on two sets of group discussions held among a group of approximately 15 HPSR researchers based in Nairobi, Kenya, who were conducting a range of embedded HPSR studies throughout the COVID-19 pandemic. The researchers, including many of the authors, were employed by the KEMRI-Wellcome Trust Research Programme (KWTRP), which is a long-standing multi-disciplinary partnership between the Kenya Medical Research Institute and the Wellcome Trust with a central goal of contributing to national and international health policy and practice. We share our findings in relation to three inter-related themes: 1) Ensuring the continued social value of our HPSR work in the face of changing priorities; 2) Responding to shifting ethical procedures and processes at institutional and national levels; and 3) Protecting our own and front-line colleagues' well-being, including clinical colleagues. Our experiences highlight that in navigating research work and responsibilities to colleagues, patients and participants through the pandemic, many embedded HPSR staff faced difficult emotional and ethical challenges, including heightened forms of moral distress, which may have been better prevented and supported. We draw on our findings and the wider literature to discuss considerations for funders and research leads with an eye to strengthening support for embedded HPSR staff, not only in crises such as the on-going COVID-19 pandemic, but also more generally.

COVID-19 and unintended steps towards further equity in global health research

There was, and possibly still is, potential for COVID-19 to disrupt power inequities and contribute to positive transformation in global health research that increases equity. While there is consensus about the need to decolonise by transforming global health, and a roadmap outlining how we could approach it, there are few examples of steps that could be taken to transform the mechanics of global health research. This paper contributes lessons learnt from experiences and reflections of our diverse multinational team of researchers involved in a multicountry research project. We demonstrate the positive impact on our research project of making further steps towards improving equity within our research practices. Some of the approaches adopted include redistributing power to researchers from the countries of interest at various stages in their career, by involving the whole team in decisions about the research; meaningfully involving the whole team in research data analysis; and providing opportunities for all researchers from the countries of interest to voice their perspectives as first authors in publications. Although this approach is consistent with how research guidance suggests research should be run, in reality it does not often happen in this way. The authors of this paper hope that by sharing our experience, we can contribute towards discussions about the processes required to continue developing a global health sector that is equitable and inclusive.

Same label, different patients: Health-workers' understanding of the label 'critical illness'

Background

During the course of patients' sickness, some become critically ill, and identifying them is the first important step to be able to manage the illness. During the course of care provision, health workers sometimes use the term 'critical illness' as a label when referring to their patient's condition, and the label is then used as a basis for communication and care provision. Their understanding of this label will therefore have a profound impact on the identification and management of patients. This study aimed to determine how Kenyan and Tanzanian health workers understand the label 'critical illness'.

Methods

A total of 10 hospitals-five in Kenya and five in Tanzania-were visited. In-depth interviews were conducted with 30 nurses and physicians from different departments in the hospitals who had experience in providing care for sick patients. We conducted a thematic analysis of the translated and transcribed interviews, synthesized findings and developed an overarching set of themes which captured healthcare workers' understandings of the label 'critical illness'.

Results

Overall, there does not appear to be a unified understanding of the label 'critical illness' among health workers. Health workers understand the label to refer to patients in four thematic ways: (1) those in a life-threatening state; (2) those with certain diagnoses; (3) those receiving care in certain locations; and (4) those in need of a certain level of care.

Conclusion

There is a lack of a unified understanding about the label 'critical illness' among health workers in Tanzania and Kenya. This potentially hampers communication and the selection of patients for urgent life-saving care. A recently proposed definition, "a state of ill health with vital organ dysfunction, a high risk of imminent death if care is not provided and the potential for reversibility", could be useful for improving communication and care.

A systematic review of clinical, epidemiological and demographic predictors of tuberculosis in children with pneumonia

Background

Tuberculosis (TB) can present as acute, severe pneumonia in children, but features which distinguish TB from other causes of pneumonia are not well understood. We conducted a systematic review to determine the prevalence and to explore clinical and demographic predictors of TB in children presenting with pneumonia over three decades.

Methods

We searched for peer-reviewed, English language studies published between 1990 and 2020 that included children aged between 1 month and 17 years with pneumonia and prospectively evaluated for TB. There were 895 abstracts and titles screened, and 72 full text articles assessed for eligibility.

Results

Thirteen clinical studies, two autopsy studies and one systematic review were included in analyses. Majority of studies were from Africa (12/15) and included children less than five years age. Prevalence of bacteriologically confirmed TB in children with pneumonia ranged from 0.2% to 14.8% (median = 3.7%, interquartile range (IQR) = 5.95) and remained stable over the three decades. TB may be more likely in children with pneumonia if they have a history of close TB contact, HIV infection, malnutrition, age less than one year or failure to respond to empirical antibiotics. However, these features have limited discriminatory value as TB commonly presents as acute severe pneumonia – with a short duration of cough, and clinical and radiographic features indistinguishable from other causes of pneumonia. Approximately half of patients with TB respond to initial empirical antibiotics, presumably due to bacterial co-infection, and follow-up may be critical to detect and treat TB.

Conclusion

TB should be considered as a potential cause or comorbidity in all children presenting with pneumonia in high burden settings. Clinicians should be alert to the presence of known risk factors for TB and bacteriological confirmation sought whenever possible. Quality data regarding clinical predictors of TB in childhood pneumonia are lacking. Further, prospective research is needed to better understand predictors and prevalence of TB in childhood pneumonia, particularly in TB endemic settings outside of Africa and in older children. Children of all ages with pneumonia should be included in research on improved, point-of-care TB diagnostics to support early case detection and treatment.

Improving facility-based care: eliciting tacit knowledge to advance intervention design

Attention has turned to improving the quality and safety of healthcare within health facilities to reduce avoidable mortality and morbidity. Interventions should be tested in health system environments that can support their adoption if successful. To be successful, interventions often require changes in multiple behaviours making their consequences unpredictable. Here, we focus on this challenge of change at the mesolevel or microlevel. Drawing on multiple insights from theory and our own empirical work, we highlight the importance of engaging managers, senior and frontline staff and potentially patients to explore foundational questions examining three core resource areas. These span the physical or material resources available, workforce capacity and capability and team and organisational relationships. Deficits in all these resource areas may need to be addressed to achieve success. We also argue that as inertia is built into the complex social and human systems characterising healthcare facilities that thought on how to mobilise five motive forces is needed to help achieve change. These span goal alignment and ownership, leadership for change, empowering key actors, promoting responsive planning and procurement and learning for transformation. Our aim is to bridge the theory-practice gap and offer an entry point for practical discussions to elicit the critical tacit and contextual knowledge needed to design interventions. We hope that this may improve the chances that interventions are successful and so contribute to better facility-based care and outcomes while contributing to the development of learning health systems.

Diagnostic Challenges in Childhood Pulmonary Tuberculosis-Optimizing the Clinical Approach

The management of childhood tuberculosis (TB) is hampered by the low sensitivity and limited accessibility of microbiological testing. Optimizing clinical approaches is therefore critical to close the persistent gaps in TB case detection and prevention necessary to realize the child mortality targets of the End TB Strategy. In this review, we provide practical guidance summarizing the evidence and guidelines describing the use of symptoms and signs in decision making for children being evaluated for either TB preventive treatment (TPT) or TB disease treatment in high-TB incidence settings. Among at-risk children being evaluated for TPT, a symptom screen may be used to differentiate children who require further investigation for TB disease before receiving TPT. For symptomatic children being investigated for TB disease, an algorithmic approach can inform which children should receive TB treatment, even in the absence of imaging or microbiological confirmation. Though clinical approaches have limitations in accuracy, they are readily available and can provide valuable guidance for decision making in resource-limited settings to increase treatment access. We discuss the trade-offs in using them to make TB treatment decisions.

Essential Emergency and Critical Care: a consensus among global clinical experts

BACKGROUND

Globally, critical illness results in millions of deaths every year. Although many of these deaths are potentially preventable, the basic, life-saving care of critically ill patients are often overlooked in health systems. Essential Emergency and Critical Care (EECC) has been devised as the care that should be provided to all critically ill patients in all hospitals in the world. EECC includes the effective care of low cost and low complexity for the identification and treatment of critically ill patients across all medical specialties. This study aimed to specify the content of EECC and additionally, given the surge of critical illness in the ongoing pandemic, the essential diagnosis-specific care for critically ill patients with COVID-19.

METHODS

In a Delphi process, consensus (>90% agreement) was sought from a diverse panel of global clinical experts. The panel iteratively rated proposed treatments and actions based on previous guidelines and the WHO/ICRC's Basic Emergency Care. The output from the Delphi was adapted iteratively with specialist reviewers into a coherent and feasible package of clinical processes plus a list of hospital readiness requirements.

RESULTS

The 269 experts in the Delphi panel had clinical experience in different acute medical specialties from 59 countries and from all resource settings. The agreed EECC package contains 40 clinical processes and 67 requirements, plus additions specific for COVID-19.

CONCLUSION

The study has specified the content of care that should be provided to all critically ill patients. Implementing EECC could be an effective strategy for policy makers to reduce preventable deaths worldwide.

Employing learning health system principles to advance research on severe neonatal and paediatric illness in Kenya

We have worked to develop a Clinical Information Network (CIN) in Kenya as an early form of learning health systems (LHS) focused on paediatric and neonatal care that now spans 22 hospitals. CIN's aim was to examine important outcomes of hospitalisation at scale, identify and ultimately solve practical problems of service delivery, drive improvements in quality and test interventions. By including multiple routine settings in research, we aimed to promote generalisability of findings and demonstrate potential efficiencies derived from LHS. We illustrate the nature and range of research CIN has supported over the past 7 years as a form of LHS. Clinically, this has largely focused on common, serious paediatric illnesses such as pneumonia, malaria and diarrhoea with dehydration with recent extensions to neonatal illnesses. CIN also enables examination of the quality of care, for example that provided to children with severe malnutrition and the challenges encountered in routine settings in adopting simple technologies (pulse oximetry) and more advanced diagnostics (eg, Xpert MTB/RIF). Although regular feedback to hospitals has been associated with some improvements in quality data continue to highlight system challenges that undermine provision of basic, quality care (eg, poor access to blood glucose testing and routine microbiology). These challenges include those associated with increased mortality risk (eg, delays in blood transfusion). Using the same data the CIN platform has enabled conduct of randomised trials and supports malaria vaccine and most recently COVID-19 surveillance. Employing LHS principles has meant engaging front-line workers, clinical managers and national stakeholders throughout. Our experience suggests LHS can be developed in low and middle-income countries that efficiently enable contextually appropriate research and contribute to strengthening of health services and research systems.

Effect of enhancing audit and feedback on uptake of childhood pneumonia treatment policy in hospitals that are part of a clinical network: a cluster randomized trial

BACKGROUND

The World Health Organization (WHO) revised its clinical guidelines for management of childhood pneumonia in 2013. Significant delays have occurred during previous introductions of new guidelines into routine clinical practice in low- and middle-income countries (LMIC). We therefore examined whether providing enhanced audit and feedback as opposed to routine standard feedback might accelerate adoption of the new pneumonia guidelines by clinical teams within hospitals in a low-income setting.

METHODS

In this parallel group cluster randomized controlled trial, 12 hospitals were assigned to either enhanced feedback (n = 6 hospitals) or standard feedback (n = 6 hospitals) using restricted randomization. The standard (network) intervention delivered in both trial arms included support to improve collection and quality of patient data, provision of mentorship and team management training for pediatricians, peer-to-peer networking (meetings and social media), and multimodal (print, electronic) bimonthly hospital specific feedback reports on multiple indicators of evidence guideline adherence. In addition to this network intervention, the enhanced feedback group received a monthly hospital-specific feedback sheet targeting pneumonia indicators presented in multiple formats (graphical and text) linked to explicit performance goals and action plans and specific email follow up from a network coordinator. At the start of the trial, all hospitals received a standardized training on the new guidelines and printed booklets containing pneumonia treatment protocols. The primary outcome was the proportion of children admitted with indrawing and/or fast-breathing pneumonia who were correctly classified using new guidelines and received correct antibiotic treatment (oral amoxicillin) in the first 24 h. The secondary outcome was the proportion of correctly classified and treated children for whom clinicians changed treatment from oral amoxicillin to injectable antibiotics.

RESULTS

The trial included 2299 childhood pneumonia admissions, 1087 within the hospitals randomized to enhanced feedback intervention, and 1212 to standard feedback. The proportion of children who were correctly classified and treated in the first 24 h during the entire 9-month period was 38.2% (393 out of 1030) and 38.4% (410 out of 1068) in the enhanced feedback and standard feedback groups, respectively (odds ratio 1.11; 95% confidence interval [CI] 0.37-3.34; P = 0.855). However, in exploratory analyses, there was evidence of an interaction between type of feedback and duration (in months) since commencement of intervention, suggesting a difference in adoption of pneumonia policy over time in the enhanced compared to standard feedback arm (OR = 1.25, 95% CI 1.14 to 1.36, P < 0.001).

CONCLUSIONS

Enhanced feedback comprising increased frequency, clear messaging aligned with goal setting, and outreach from a coordinator did not lead to a significant overall effect on correct pneumonia classification and treatment during the 9-month trial. There appeared to be a significant effect of time (representing cumulative effect of feedback cycles) on adoption of the new policy in the enhanced feedback compared to standard feedback group. Future studies should plan for longer follow-up periods to confirm these findings.

TRIAL REGISTRATION

US National Institutes of Health-ClinicalTrials.gov identifier (NCT number) NCT02817971 . Registered September 28, 2016-retrospectively registered.

An observational study of monitoring of vital signs in children admitted to Kenyan hospitals: an insight into the quality of nursing care?

Background

Measurement and correct interpretation of vital signs is part of routine clinical care. Repeated measurement enhances early recognition of deterioration, may help prevent morbidity and mortality and is a standard of care in most countries.

Objective

To examine documentation of vital signs by clinicians for admissions to paediatric wards in Kenyan hospitals, to describe monitoring frequency by nurses and explore factors influencing frequency.

Methods

Vital signs information (temperature, respiratory and pulse rate) for the first 48 hours of admission was collected from case records of children admitted with non-surgical conditions to 13 Kenyan county hospitals between September 2013 and April 2016. A mixed effect negative binomial regression model was used to explore whether the severity of illness (indicated by danger signs or severe diagnostic episodes) is associated with increased vital signs observation frequency.

Results

We examined 54 800 admission episodes with an overall mortality 6.1%. Nurse to bed ratios were very low (1:10 to 1:41 across hospitals). Admitting clinicians documented all or no vital signs in 57.0% and 8.4% cases respectively. For respiratory and pulse rates there was pronounced even end-digit preference (an indicator of incorrect information) and high frequency recording of specific values (P < 0.001) suggesting approximation. Monitoring frequency was explored in 41 738 children. Those with inpatient stays ≥48 hours were expected to have a vital signs count of 18, hospitals varied but most did not achieve this benchmark (median 9, range 2-30). There were clinically small but significant associations between vital signs count and presence of multiple severe illnesses or presence of severe pallor (adjusted relative risk ratio = 1.04, P < 0.01, 95% confidence interval CI = 1.02-1.06 and 1.05, P = 0.02, 95% CI = 1.01-1.09, respectively).

Conclusions

Data suggest accurate admission measures are sometimes missing especially for pulse and respiratory rates, possibly linked to manual measurement. Monitoring frequency is often low in the high risk population studied probably indicating how quality of nursing care is undermined by considerable human resource shortages.

Comparable outcomes among trial and nontrial participants in a clinical trial of antibiotics for childhood pneumonia: a retrospective cohort study

OBJECTIVES

We compared characteristics and outcomes of children enrolled in a randomized controlled trial (RCT) comparing oral amoxicillin and benzyl penicillin for the treatment of chest indrawing pneumonia vs. children who received routine care to determine the external validity of the trial results.

STUDY DESIGN AND SETTING

A retrospective cohort study was conducted among children aged 2-59 months admitted in six Kenyan hospitals. Data for nontrial participants were extracted from inpatient records upon conclusion of the RCT. Mortality among trial vs. nontrial participants was compared in multivariate models.

RESULTS

A total of 1,709 children were included, of whom 527 were enrolled in the RCT and 1,182 received routine care. History of a wheeze was more common among trial participants (35.4% vs. 11.2%; P < 0.01), while dehydration was more common among nontrial participants (8.6% vs. 5.9%; P = 0.05). Other patient characteristics were balanced between the two groups. Among those with available outcome data, 14/1,140 (1.2%) nontrial participants died compared to 4/527 (0.8%) enrolled in the trial (adjusted odds ratio, 0.7; 95% confidence interval: 0.2-2.1).

CONCLUSION

Patient characteristics were similar, and mortality was low among trial and nontrial participants. These findings support the revised World Health Organization treatment recommendations for chest indrawing pneumonia.

Does audit and feedback improve the adoption of recommended practices? Evidence from a longitudinal observational study of an emerging clinical network in Kenya

Background

Audit and feedback (A&F) is widely used in healthcare but there are few examples of how to deploy it at scale in low-income countries. Establishing the Clinical Information Network (CIN) in Kenya provided an opportunity to examine the effect of A&F delivered as part of a wider set of activities to promote paediatric guideline adherence.

Methods

We analysed data collected from medical records on discharge for children aged 2–59 months from 14 Kenyan hospitals in the CIN. Hospitals joined CIN in phases and for each we analysed their initial 25 months of participation that occurred between December 2013 and March 2016. A total of 34 indicators of adherence to recommendations were selected for evaluation each classified by form of feedback (passive, active and none) and type of task (simple or difficult documentation and those requiring cognitive work). Performance change was explored graphically and using generalised linear mixed models with attention given to the effects of time and use of a standardised paediatric admission record (PAR) form.

Results

Data from 60 214 admissions were eligible for analysis. Adherence to recommendations across hospitals significantly improved for 24/34 indicators. Improvements were not obviously related to nature of feedback, may be related to task type and were related to PAR use in the case of documentation indicators. There was, however, marked variability in adoption and adherence to recommended practices across sites and indicators. Hospital-specific factors, low baseline performance and specific contextual changes appeared to influence the magnitude of change in specific cases.

Conclusion

Our observational data suggest some change in multiple indicators of adherence to recommendations (aspects of quality of care) can be achieved in low-resource hospitals using A&F and simple job aides in the context of a wider network approach.

Blood Transfusion Delay and Outcome in County Hospitals in Kenya

Severe anemia is a leading indication for blood transfusion and a major cause of hospital admission and mortality in African children. Failure to initiate blood transfusion rapidly enough contributes to anemia deaths in sub-Saharan Africa. This article examines delays in accessing blood and outcomes in transfused children in Kenyan hospitals. Children admitted with nonsurgical conditions in 10 Kenyan county hospitals participating in the Clinical Information Network who had blood transfusion ordered from September 2013 to March 2016 were studied. The delay in blood transfusion was calculated from the date when blood transfusion was prescribed to date of actual transfusion. Five percent (2,875/53,174) of admissions had blood transfusion ordered. Approximately half (45%, 1,295/2,875) of children who had blood transfusion ordered at admission had a documented hemoglobin < 5 g/dl and 36% (2,232/6,198) of all children admitted with a diagnosis of anemia were reported to have hemoglobin < 5 g/dL. Of all the ordered transfusions, 82% were administered and documented in clinical records, and three-quarters of these (75%, 1,760/2,352) were given on the same day as ordered but these proportions varied from 71% to 100% across the 10 hospitals. Children who had a transfusion ordered but did not receive the prescribed transfusion had a mortality of 20%, compared with 12% among those transfused. Malaria-associated anemia remains the leading indication for blood transfusion in acute childhood illness admissions. Delays in transfusion are common and associated with poor outcomes. Variance in delay across hospitals may be a useful indicator of health system performance.

Building Learning Health Systems to Accelerate Research and Improve Outcomes of Clinical Care in Low- and Middle-Income Countries

Mike English and colleagues argue that as efforts are made towards achieving universal health coverage it is also important to build capacity to develop regionally relevant evidence to improve healthcare.

Oral amoxicillin versus benzyl penicillin for severe pneumonia among kenyan children: a pragmatic randomized controlled noninferiority trial

Evidence demonstrating noninferiority of oral amoxicillin vs benzyl penicillin for severe childhood pneumonia is largely drawn from Asian populations where mortality is low. This study confirms noninferiority and is expected to inform policy on treatment of pneumonia in sub-Saharan Africa.

Background. There are concerns that the evidence from studies showing noninferiority of oral amoxicillin to benzyl penicillin for severe pneumonia may not be generalizable to high-mortality settings.

Methods. An open-label, multicenter, randomized controlled noninferiority trial was conducted at 6 Kenyan hospitals. Eligible children aged 2–59 months were randomized to receive amoxicillin or benzyl penicillin and followed up for the primary outcome of treatment failure at 48 hours. A noninferiority margin of risk difference between amoxicillin and benzyl penicillin groups was prespecified at 7%.

Results. We recruited 527 children, including 302 (57.3%) with comorbidity. Treatment failure was observed in 20 of 260 (7.7%) and 21 of 261 (8.0%) of patients in the amoxicillin and benzyl penicillin arms, respectively (risk difference, −0.3% [95% confidence interval, −5.0% to 4.3%]) in per-protocol analyses. These findings were supported by the results of intention-to-treat analyses. Treatment failure by day 5 postenrollment was 11.4% and 11.0% and rising to 13.5% and 16.8% by day 14 in the amoxicillin vs benzyl penicillin groups, respectively. The most frequent cause of cumulative treatment failure at day 14 was clinical deterioration within 48 hours of enrollment (33/59 [55.9%]). Four patients died (overall mortality 0.8%) during the study, 3 of whom were allocated to the benzyl penicillin group. The presence of wheeze was independently associated with less frequent treatment failure.

Conclusions. Our findings confirm noninferiority of amoxicillin to benzyl penicillin, provide estimates of risk of treatment failure in Kenya, and offer important additional evidence for policy making in sub-Saharan Africa.

Clinical Trial Registration. NCT01399723.