1
|
McLaughlin M, Pellé KG, Scarpino SV, Giwa A, Mount-Finette E, Haidar N, Adamu F, Ravi N, Thompson A, Heath B, Dittrich S, Finette B. Development and Validation of Manually Modified and Supervised Machine Learning Clinical Assessment Algorithms for Malaria in Nigerian Children. Front Artif Intell 2022; 4:554017. [PMID: 35187469 PMCID: PMC8851346 DOI: 10.3389/frai.2021.554017] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/20/2020] [Accepted: 08/11/2021] [Indexed: 11/13/2022] Open
Abstract
It is currently estimated that 67% of malaria deaths occur in children under-five years (WHO, 2020). To improve the identification of children at clinical risk for malaria, the WHO developed community (iCCM) and clinic-based (IMCI) protocols for frontline health workers using paper-based forms or digital mobile health (mHealth) platforms. To investigate improving the accuracy of these point-of-care clinical risk assessment protocols for malaria in febrile children, we embedded a malaria rapid diagnostic test (mRDT) workflow into THINKMD’s (IMCI) mHealth clinical risk assessment platform. This allowed us to perform a comparative analysis of THINKMD-generated malaria risk assessments with mRDT truth data to guide modification of THINKMD algorithms, as well as develop new supervised machine learning (ML) malaria risk algorithms. We utilized paired clinical data and malaria risk assessments acquired from over 555 children presenting to five health clinics in Kano, Nigeria to train ML algorithms to identify malaria cases using symptom and location data, as well as confirmatory mRDT results. Supervised ML random forest algorithms were generated using 80% of our field-based data as the ML training set and 20% to test our new ML logic. New ML-based malaria algorithms showed an increased sensitivity and specificity of 60 and 79%, and PPV and NPV of 76 and 65%, respectively over THINKD initial IMCI-based algorithms. These results demonstrate that combining mRDT “truth” data with digital mHealth platform clinical assessments and clinical data can improve identification of children with malaria/non-malaria attributable febrile illnesses.
Collapse
Affiliation(s)
- Megan McLaughlin
- THINKMD, Burlington, VT, United States
- *Correspondence: Megan McLaughlin, ; Karell G. Pellé, ; Samuel V. Scarpino,
| | - Karell G. Pellé
- FIND, Geneva, Switzerland
- *Correspondence: Megan McLaughlin, ; Karell G. Pellé, ; Samuel V. Scarpino,
| | - Samuel V. Scarpino
- Network Science Institute, Northeastern University, Boston, MA, United States
- Santa Fe Institute, Santa Fe, NM, United States
- Vermont Complex Systems Center, University of Vermont, Burlington, VT, United States
- *Correspondence: Megan McLaughlin, ; Karell G. Pellé, ; Samuel V. Scarpino,
| | | | | | | | | | | | | | - Barry Heath
- THINKMD, Burlington, VT, United States
- Department of Pediatrics, University of Vermont, Burlington, VT, United States
| | | | - Barry Finette
- THINKMD, Burlington, VT, United States
- Department of Pediatrics, University of Vermont, Burlington, VT, United States
| |
Collapse
|
2
|
Baker K, Petzold M, Mucunguzi A, Wharton-Smith A, Dantzer E, Habte T, Matata L, Nanyumba D, Okwir M, Posada M, Sebsibe A, Nicholson J, Marasciulo M, Izadnegahdar R, Alfvén T, Källander K. Performance of five pulse oximeters to detect hypoxaemia as an indicator of severe illness in children under five by frontline health workers in low resource settings - A prospective, multicentre, single-blinded, trial in Cambodia, Ethiopia, South Sudan, and Uganda. EClinicalMedicine 2021; 38:101040. [PMID: 34368660 PMCID: PMC8326731 DOI: 10.1016/j.eclinm.2021.101040] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/01/2020] [Revised: 06/30/2021] [Accepted: 07/06/2021] [Indexed: 11/29/2022] Open
Abstract
BACKGROUND Low blood oxygen saturation (SpO2), or hypoxaemia, is an indicator of severe illness in children. Pulse oximetry is a globally accepted, non-invasive method to identify hypoxaemia, but rarely available outside higher-level facilities in resource-constrained countries. This study aims to evaluate the performance of different types of pulse oximeters amongst frontline health workers in Cambodia, Ethiopia, South Sudan, and Uganda. METHODS Five pulse oximeters (POx) which passed laboratory testing, out of an initial 32 potential pulse oximeters, were evaluated by frontline health workers for performance, defined as agreement between the SpO2 measurements of the test device and the reference standard. The study protocol is registered with the Australia New Zealand Clinical Trials Registry (Ref: ACTRrn12615000348550). FINDINGS Two finger-tip pulse oximeters (Contec and Devon), two handheld pulse oximeters (Lifebox and Utech), and one phone pulse oximeter (Masimo) passed the laboratory testing. They were evaluated for performance on 1,313 children under five years old by 207 frontline health workers between February and May 2015. Phone and handheld pulse oximeters had greater overall agreement with the reference standard (56%; 95% CI 0.52 - 0.60 to 68%; 95% CI 0.65 - 0.71) than the finger-tip POx (31%; 95% CI 0.26 to 0.36 and 47%; 95% CI 0.42 to 0.52). Fingertip POx performance was substantially lower in the 0-2 month olds; having just 17% and 25% agreement. The finger-tip devices more often underreported SpO2 readings (mean difference -7.9%; 95%CI -8.6,-7.2 and -3.9%; 95%CI -4.4,-3.4), and therefore over diagnosed hypoxaemia in the children assessed. INTERPRETATION While the Masimo phone pulse oximeter performed best, all handheld POx with age-specific probes performed well in the hands of frontline health workers, further highlighting their suitability as a screening tool of severe illness. The poor performance of the fingertip POx suggests they should not be used in children under five by frontline health workers. It is essential that POx are performance tested on children in routine settings (in vivo), not only in laboratories or controlled settings (in vitro), before being introduced at scale. FUNDING Bill & Melinda Gates Foundation [OPP1054367].
Collapse
Affiliation(s)
- Kevin Baker
- Malaria Consortium, London, United Kingdom
- Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden
- Corresponding author at: Kevin Baker, Malaria Consortium, The Green House, 244-254 Cambridge Heath Road, London, E2 9DA
| | - Max Petzold
- School of Public Health and Community Medicine, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden
- University of the Witwatersrand, Johannesburg, South Africa
| | | | | | | | | | | | | | | | | | | | | | | | | | - Tobias Alfvén
- Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden
- Sachs’ Children and Youth Hospital, Stockholm, Sweden
| | - Karin Källander
- Malaria Consortium, London, United Kingdom
- Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden
- Programme Division, Health Section, UNICEF, New York, United States
| |
Collapse
|
3
|
Winders WT, Garbern SC, Bills CB, Relan P, Schultz ML, Trehan I, Kivlehan SM, Becker TK, McQuillan R. The effects of mobile health on emergency care in low- and middle-income countries: A systematic review and narrative synthesis. J Glob Health 2021; 11:04023. [PMID: 33828846 PMCID: PMC8021077 DOI: 10.7189/jogh.11.04023] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022] Open
Abstract
BACKGROUND In resource-constrained settings, mobile health (mHealth) has varied applications. While there is strong evidence for its use in chronic disease management, the applications of mHealth for management of acute illness in low- and middle-income countries (LMICs) are not as well described. This review systematically explores current available evidence on the effectiveness of mHealth interventions at improving health outcomes in emergency care settings in LMICs. METHODS A systematic search of the literature was performed in accordance with PRISMA guidelines, utilizing seven electronic databases and manual searches to identify peer-reviewed literature containing each of three search elements: mHealth, emergency care (EC), and LMICs. Articles quality was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria. RESULTS After removing duplicates, 6498 studies met initial search criteria; 108 were eligible for full text review and 46 met criteria for inclusion. Thirty-six pertained to routine emergency care, and 10 involved complex humanitarian emergencies. Based on the GRADE criteria, 15 studies were rated as "Very Low" quality, 24 as "Low" quality, 6 as "Moderate" quality, and 1 as "High" quality. Eight studied data collection, 9 studied decision support, 15 studied direct patient care, and 14 studied health training. All 46 studies reported positive impacts of mHealth on EC in LMICs. CONCLUSIONS Mobile health interventions can be effective in improving provider-focused and patient-centered outcomes in both routine and complex EC settings. Future investigations focusing on patient-centered outcomes are needed to further validate these findings.
Collapse
Affiliation(s)
- W Tyler Winders
- School of Public Health, University of Edinburgh, Edinburgh, UK
| | - Stephanie C Garbern
- Department of Emergency Medicine, Alpert Medical School of Brown University, Providence, Rhode Island, USA
| | - Corey B Bills
- Department of Emergency Medicine, University of Colorado School of Medicine, Denver, Colorado, USA
| | - Pryanka Relan
- Department of Emergency Medicine, Emory Healthcare Network, Atlanta, Georgia, USA
| | - Megan L Schultz
- Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin, USA
| | - Indi Trehan
- Departments of Pediatrics and Global Health, University of Washington, Seattle, Washington, USA
| | - Sean M Kivlehan
- Department of Emergency Medicine, Brigham and Women's Hospital, Boston, Massachusetts, and Harvard Humanitarian Initiative, Cambridge, Massachusetts, USA
| | - Torben K Becker
- Department of Emergency Medicine, University of Florida, Gainesville, Florida, USA
| | - Ruth McQuillan
- Usher Institute of Population Health Sciences and Informatics, University of Edinburgh, Edinburgh, UK
| |
Collapse
|
4
|
Mahmood H, Mckinstry B, Luz S, Fairhurst K, Nasim S, Hazir T. Community health worker-based mobile health (mHealth) approaches for improving management and caregiver knowledge of common childhood infections: A systematic review. J Glob Health 2020; 10:020438. [PMID: 33437462 PMCID: PMC7774026 DOI: 10.7189/jogh.10.020438] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022] Open
Abstract
Background Children in lower middle-income countries (LMICs) are more at risk of dying, than those in High Income Countries (HICs), due to highly prevalent deadly yet preventable childhood infections. Alongside concerns about the incidence of these infections, there has been a renewed interest in involving community health workers (CHWs) in various public health programs. However, as CHWs are increasingly asked to take on different tasks there is a risk that their workload may become unmanageable. One solution to help reduce this burden is the use of mobile health (mHealth) technology in the community through behaviour change. Considering there are various CHWs based mHealth approaches on illness management and education, therefore, we aimed to appraise the available literature on effectiveness of these mHealth approaches for caregivers to improve knowledge and management about common under-five childhood infections with respect to behaviour change. Methods We searched six databases between October to December 2019 using subject heading (Mesh) and free text terms in title or abstract in US English. We included multiple study types of children under-five or their caregivers who have been counselled, educated, or provided any health care service by CHWs for any common paediatric infectious diseases using mHealth. We excluded articles published prior to 1990 and those including mHealth technology not coming under the WHO definition. A data extraction sheet was developed and titles, abstracts, and selected full text were reviewed by two reviewers. Quality assessment was done using JBI tools. Results We included 23 articles involving around 300 000 individuals with eight types of study designs. 20 studies were conducted in Africa, two in Asia, and one in Latin America mainly on pneumonia or respiratory tract infections followed by malaria and diarrhoea in children. The most common types of Health approaches were mobile applications for decision support, text message reminders and use of electronic health record systems. None of the studies employed the use of any behaviour change model or any theoretical framework for selection of models in their studies. Conclusions Coupling mhealth with CHWs has the potential to benefit communities in improving management of illnesses in children under-five. High quality evidence on impact of such interventions on behaviour is relatively sparse and further studies should be conducted using theoretically informed behaviour change frameworks/models. Registration PROPSERO Registration number: CRD42018117679
Collapse
Affiliation(s)
- Hana Mahmood
- Maternal, Neonatal and Child Health Research Network (MNCHRN), Pakistan.,NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | - Brian Mckinstry
- NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | - Saturnino Luz
- NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | - Karen Fairhurst
- NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | - Sumaira Nasim
- NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | - Tabish Hazir
- NIHR Global Health Research Unit on Respiratory Health (RESPIRE), Usher Institute, the University of Edinburgh, Edinburgh, UK
| | | |
Collapse
|
5
|
Ansermino JM, Wiens MO, Kissoon N. Evidence and Transparency are Needed to Develop a Frontline Health Worker mHealth Assessment Platform. Am J Trop Med Hyg 2019; 101:948. [PMID: 32519659 PMCID: PMC6779222 DOI: 10.4269/ajtmh.19-0411a] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022] Open
|
6
|
Finette BA, McLaughlin M, Scarpino SV, Canning J, Grunauer M, Teran E, Bahamonde M, Quizhpe E, Shah R, Swedberg E, Rahman KA, Khondker H, Chakma I, Muhoza D, Seck A, Kabore A, Nibitanga S, Heath B. Authors' Response. Am J Trop Med Hyg 2019; 101:949-950. [PMID: 32519660 PMCID: PMC6779201 DOI: 10.4269/ajtmh.19-0411b] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022] Open
Affiliation(s)
- Barry A Finette
- University of Vermont Robert Larner College of Medicine and University of Vermont Children's HospitalBurlington, VermontTHINKMD, Inc.Burlington, Vermont
| | | | | | | | | | | | | | - Edy Quizhpe
- University of San Francisco de Quito-Ecuador Ministry of Health-AffiliateQuito, Ecuador
| | | | | | | | | | - Ituki Chakma
- Save the Children-International BangladeshDhaka, Bangladesh
| | | | - Awa Seck
- UNICEF-Burkina FasoOuagadougou, Burkina Faso
| | | | | | - Barry Heath
- University of Vermont Robert Larner College of Medicine and University of Vermont Children's HospitalBurlington, VermontTHINKMD, Inc.Burlington, Vermont
| |
Collapse
|