Outpatient Pediatric Pulse Oximeters in Africa

The primary objective of this clinical trial is to evaluate the performance of three pulse oximeters during outpatient care within Cape Town, South Africa. This objective will be achieved through generating evidence on how, why, for whom, to what extent and at what cost can paediatric pulse oximetry devices improve the management of hypoxemic children. This will be done with two inter-linked studies: Aim 1: Determine the impact of two novel paediatric pulse oximeter devices on the correct management of hypoxaemia. If the investigators find these devices improve healthcare worker assessments and decision making, it could improve clinical outcomes for children in low-resource contexts. Aim 2: Describe the burden of hypoxaemia and risks for mortality amongst children presenting with acute respiratory infections in a low-resource setting in Cape Town. By establishing the burden and need, a clearer investment case for pulse oximetry can be made for this context.

Background: Lower respiratory tract infections (LRI) remain the leading infectious cause of death globally for children younger than five years.1 Alarmingly, >50% of LRI deaths occurred in low and middle-income countries (LMICs) in sub-Saharan Africa, with inequitable distribution both between and within countries.2 Key quality-of-care implementation gaps have hampered the effectiveness of the World Health Organization (WHO) Integrated Management of Childhood Illnesses (IMCI) guidelines used for pediatric LRI care in LMICs.3 Evaluations of Integrated Management of Childhood Illness (IMCI) guidelines have identified inadequate triaging and therefore results in a failure to identify children at higher risk of death. Interventions to improve the sensitivity and specificity of the IMCI approach in identifying severely ill children could improve outcomes. Routine use of pulse oximetry, to non-invasively measure peripheral oxyhemoglobin saturation (SpO2), is poorly implemented at the primary healthcare (PHC) levels in LMICs, and therefore provides one such opportunity to improve IMCI assessments. Hypoxemia - a low SpO2, is associated with increased mortality in children with LRI(1).4 Hypoxemia prevalence amongst children with pneumonia in African contexts has been estimated at 28%, and in outpatient settings as 23%.5 As hypoxemia is a key mortality risk factor, effectively identifying these children early in the care-seeking pathway is fundamental to reducing mortality in low-resource contexts.6 While SpO2 is recommended by IMCI for children with suspected pneumonia, pulse oximetry devices for measuring SpO2 are not widely implemented in PHCs in LMICs, where most children first access care. In Malawi 16% of nearly 700 outpatient encounters with suspected LRI had a SpO2 measured, and >40% of children eligible for hospitalization were not referred. Since few children have SpO2 collected during outpatient care, referral decisions are largely based on subjective clinical danger signs. This then has knock-on effects on receipt of oxygen treatment. While pulse oximetry implementation in PHCs has been slow to scale-up, there is evidence of utility and feasibility. In Malawi, healthcare workers successfully measured the SpO2 on 94% of >14,000 children and were >2 times more likely to correctly refer a child when the child's SpO2 was low. This work also demonstrated >60% of hypoxemic children would not have been referred in the absence of an SpO2 measurement.7 One explanation for slow adoption is the lack of appropriate devices, that have been designed specifically for spot-checks amongst children in outpatient LMIC settings - a population with specific oximetry needs. Important features of such a device are being low cost, robust, able to cope with poor perfusion and motion artefact, good battery life and reliable.8 Mobile phones are relatively inexpensive, widely available, and increasingly utilized for healthcare - 'mobile Health (mHealth)', while electronic Health (eHealth) is when electronic services - like the internet - support healthcare. In LMICs mobile phones offer the potential for expanded healthcare access and quality of care both as a medical device and as a platform for eHealth services, such as the digital health management information system (HMIS) used in sub-Saharan Africa - District Health Information Software 2 (DHIS2). Developing a mobile-based pulse oximeter, with interoperability to store and upload data directly into a patients DHIS2 record has the potential to improve the management of paediatric hypoxaemia. The Phefumela Project, meaning "breathe" in a local South African language, will evaluate the impact of two different novel paediatric pulse oximeters, both designed specifically for this population in a high burden setting in South Africa. Aim 1: Determine the impact of two novel paediatric pulse oximeter devices on the correct management of hypoxaemia. If the investigators find these devices improve healthcare worker assessments and decision making, it could improve clinical outcomes for children in low-resource contexts. Aim 2: Describe the burden of hypoxaemia and risks for mortality amongst children presenting with acute respiratory infections in a low-resource setting in Cape Town. By establishing the burden and need, a clearer investment case for pulse oximetry can be made for this context. Setting: The study site is the large Khayelitsha community, which includes 6 primary healthcare clinics (PHCs), 2 community health centers (CHCs), 1 community day centers (CDCs) and one government district hospital serving its catchment area, Khayelitsha Hospital. The nearest tertiary referral government hospital is Tygerberg, which serves 40% of the provincial paediatric population. Khayelitsha township has a population of approximately 450,000 and is 90.5% Black African. Khayelitsha has a very young population, with >40% of its residents under 19 years of age, residing in informal housing with high caregiver unemployment and limited running water access. HIV (human immunodeficiency virus) and tuberculosis prevalence is high; maternal HIV prevalence is 29.5% - the highest in the Western Cape Province - and the tuberculosis incidence of 1,389/100,000 population exceeded the national average of 834/100,000 in 2017.9 Khayelitsha Hospital has an average bed occupancy rate over 130% capacity, and its 47 bed emergency center cares for about 120 patients daily. A 2015 study characterized the pediatric case mix over six months at the emergency center, reporting >80% of pediatric patients were <5 years old, nearly 2/3 were triaged at an emergent level, and the most common diagnosis was LRIs (22.0%, n=70/317). Of 58 children with pneumonia, 5 (8.5%) died.9 Pulse oximeter devices: The investigators will be using and comparing three different pulse oximeter devices during this study, two (the Phefumla and Lifebox-01 (LB-01)) are not commercially available. The Contec device is widely available and currently used clinically in this setting in South Africa. Throughout this study, the Contec device will be considered the reference or control standard.

Pragmatic 3-arm cluster randomised controlled trial (cRCT), conducted over a 24-month period, with a 3-month pilot and 3-month intervention embedding period. Clusters are defined as outpatient facilities (PHC, CDC, CHC), and the primary and secondary child-level outcomes will be assessed through a prospective cohort study. An embedded mixed-methods concurrent process evaluation will be conducted.

Researchers who are responsible for collecting data will not be informed of facility allocation. The key of which facilities are in which arm will be held by Stellenbosch and Johns Hopkins University, and the researcher at Karolinska Institute who will conduct the primary analysis will not have access to this information until after the primary analysis has been done and approved by the Trial Steering Committee. Process and economic evaluation data will also be analysed using a masked code for allocation arm, and only unblinded after the primary analysis is completed.

Controls will be managed routinely by the clinic staff, including the triage and clinical examination pathway through the facility, treatment and referral decisions, all of which can include the standard care device. After the child has been given a primary diagnosis and decision on onward care (i.e. referral or home-based management), the study data collector will conduct an exit interview. This interview will include the following topics: the clinical examination conducted by the healthcare worker; current intentions for onward care (e.g. are planning to go to hospital, how are the healthcare worker is travelling); extracting clinical information from the patient medical record - including oxygen saturation measurement. Finally, the study data collector will conduct oxygen saturation measurements using the control device and a reference device.

Phefumla arm participants will be managed by the clinic staff, including the triage and clinical examination pathway through the facility, treatment and referral decisions, all of which can include the Phefumla device. After the child has been given a primary diagnosis and decision on onward care (i.e. referral or home-based management), the study data collector will conduct an exit interview. This interview will include the following topics: the clinical examination conducted by the healthcare worker; current intentions for onward care (e.g. if planning to go to hospital, how is the healthcare worker is travelling); extracting clinical information from the patient medical record - including oxygen saturation measurement. Finally, the study data collector will conduct oxygen saturation measurements using the Phefumla device and a reference device.

LB-01 arm participants will be managed by the clinic staff, including the triage and clinical examination pathway through the facility, treatment and referral decisions, all of which can include the LB-01 device. After the child has been given a primary diagnosis and decision on onward care (i.e. referral or home-based management), the study data collector will conduct an exit interview. This interview will include the following topics: the clinical examination conducted by the healthcare worker; current intentions for onward care (e.g. are if planning to go to hospital, how is the healthcare worker travelling); extracting clinical information from the patient medical record - including oxygen saturation measurement. Finally, the study data collector will conduct oxygen saturation measurements using the LB-01 device and a reference device.

The Phefumla device uses the Motorola Moto G Power mobile phone. The device utilizes an Android 10 operating system and has 64 gigabyte memory with 4 gigabyte random access memory (RAM). The battery is a 5000 milliampere lithium polymer rechargeable battery, which should last at least 24 hours with minimal phone use. Data can be stored on the device and integration with information systems is planned. The reflectance sensor works on a variety of body parts including the finger, toe, and forehead.

The LB-01 probe uses transmissive oximetry with the light-emitting diode (LED) and photodetector (PD) positioned opposed to one another when placed on body tissues like fingers, and is used with the Acare pulse oximeter device. The LB-01 probe is an elongated clip sensor with an offset optics location near the hinge, permitting stable positioning on the child's big toe. By incorporating softer hollow silicone pads this design grasps the foot while placing the optics over the toe, to minimize movement artifact, an important issue for child measurements. The soft pads allow comfortable use across the smaller foot of neonates, and the design remains similar enough to a conventional finger sensor that it can be used on adult fingers as well.

The numerator is the number of children who have a biologically plausible oxygen saturation measurement achieved, the oxygen saturation measurement is documented by the healthcare worker, and an appropriate referral recommendation has been provided by the healthcare worker. All three of these conditions need to be met to be considered correct management. The denominator will be all the eligible recruited children, with suspected LRI.

The proportion of caregivers who permit the healthcare worker to measure the oxygen saturation on the child.

Amongst children with an oxygen saturation <94%, the proportion who present to the referral hospital within 48 hours.

Amongst children with an oxygen saturation <94%, the proportion who present to the referral hospital within 48 hours and are given oxygen treatment.

Inclusion Criteria: 0-59 months of age inclusive presenting to care for an acute condition the includes observed and/or caregiver history of either cough and/or difficult breathing residing in clinic catchment area caregiver agrees to provide contact details including phone number and/or residential address caregiver agrees to be contacted after two weeks by the study staff caregiver is able and willing to provide written informed consent Exclusion Criteria: 60 months of age or older presenting to care for a non-acute condition or an acute condition that does not include either observed or caregiver history of cough and/or difficult breathing does not reside in the clinic catchment area caregiver does not agree to provide contact details caregiver does not agree to be contact by study staff after two weeks caregiver unable to provide written informed consent

De-identified data will be made available after publication of the primary analysis. Other researchers who submit a written request to the principal investigator with analysis plan, data sharing agreement, and institutional review board approval will be afforded access to individual participant data.

Data will become available after publication of the primary analysis and will be available for at least 5 years.

Written request to the principal investigator, analysis plan, data sharing agreement, and institutional review board approval.

GBD 2019 Under-5 Mortality Collaborators. Global, regional, and national progress towards Sustainable Development Goal 3.2 for neonatal and child health: all-cause and cause-specific mortality findings from the Global Burden of Disease Study 2019. Lancet. 2021 Sep 4;398(10303):870-905. doi: 10.1016/S0140-6736(21)01207-1. Epub 2021 Aug 17.

Reiner RC, Welgan CA, Casey DC, Troeger CE, Baumann MM, Nguyen QP, Swartz SJ, Blacker BF, Deshpande A, Mosser JF, Osgood-Zimmerman AE, Earl L, Marczak LB, Munro SB, Miller-Petrie MK, Rodgers Kemp G, Frostad J, Wiens KE, Lindstedt PA, Pigott DM, Dwyer-Lindgren L, Ross JM, Burstein R, Graetz N, Rao PC, Khalil IA, Davis Weaver N, Ray SE, Davis I, Farag T, Brady OJ, Kraemer MUG, Smith DL, Bhatt S, Weiss DJ, Gething PW, Kassebaum NJ, Mokdad AH, Murray CJL, Hay SI. Identifying residual hotspots and mapping lower respiratory infection morbidity and mortality in African children from 2000 to 2017. Nat Microbiol. 2019 Dec;4(12):2310-2318. doi: 10.1038/s41564-019-0562-y. Epub 2019 Sep 30.

WHO. World Health Organization Integrated Management of Childhood Illness ( IMCI ) Chart Booklet-Standard. Geneva (Switzerland): World Health Organization 2014; : 1-80.

Lazzerini M, Sonego M, Pellegrin MC. Hypoxaemia as a Mortality Risk Factor in Acute Lower Respiratory Infections in Children in Low and Middle-Income Countries: Systematic Review and Meta-Analysis. PLoS One. 2015 Sep 15;10(9):e0136166. doi: 10.1371/journal.pone.0136166. eCollection 2015.

Rahman AE, Hossain AT, Nair H, Chisti MJ, Dockrell D, Arifeen SE, Campbell H. Prevalence of hypoxaemia in children with pneumonia in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Glob Health. 2022 Mar;10(3):e348-e359. doi: 10.1016/S2214-109X(21)00586-6.

McCollum ED, King C, Hammitt LL, Ginsburg AS, Colbourn T, Baqui AH, O'Brien KL. Reduction of childhood pneumonia mortality in the Sustainable Development era. Lancet Respir Med. 2016 Dec;4(12):932-933. doi: 10.1016/S2213-2600(16)30371-X. Epub 2016 Nov 12. No abstract available.

McCollum ED, King C, Deula R, Zadutsa B, Mankhambo L, Nambiar B, Makwenda C, Masache G, Lufesi N, Mwansambo C, Costello A, Colbourn T. Pulse oximetry for children with pneumonia treated as outpatients in rural Malawi. Bull World Health Organ. 2016 Dec 1;94(12):893-902. doi: 10.2471/BLT.16.173401. Epub 2016 Oct 11. Erratum In: Bull World Health Organ. 2017 Jan 1;95(1):81.

King C, Boyd N, Walker I, Zadutsa B, Baqui AH, Ahmed S, Islam M, Kainja E, Nambiar B, Wilson I, McCollum ED. Opportunities and barriers in paediatric pulse oximetry for pneumonia in low-resource clinical settings: a qualitative evaluation from Malawi and Bangladesh. BMJ Open. 2018 Jan 30;8(1):e019177. doi: 10.1136/bmjopen-2017-019177.

Richards D, Hunter L, Forey K, et al. Demographics and predictors of mortality in children undergoing resuscitation at Khayelitsha Hospital, Western Cape, South Africa. SAJCH South African Journal of Child Health 2018; 12: 127-31

Glasgow RE, Harden SM, Gaglio B, Rabin B, Smith ML, Porter GC, Ory MG, Estabrooks PA. RE-AIM Planning and Evaluation Framework: Adapting to New Science and Practice With a 20-Year Review. Front Public Health. 2019 Mar 29;7:64. doi: 10.3389/fpubh.2019.00064. eCollection 2019.