Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Achieving affordable critical care in low-income and middle-income countries

Achieving affordable critical care in low-income and middle-income countries Commentary Achieving affordable critical care in low-income and middle- income countries  1,2 3 1,2 1 Hugo C Turner, Nguyen Van Hao, Sophie Yacoub, Van Minh Tu Hoang, 4 1,2 2,5 1,2 David A Clifton, Guy E Thwaites, Arjen M Dondorp, C Louise Thwaites, Nguyen Van Vinh Chau To cite: Turner HC, Hao NV, InTroduCTIon Summary box Yacoub S, et al. Achieving In 2016, an estimated 8.6 million prema- affordable critical care in low- ture deaths occurred in low-income and ► Improving the quality and availability of critical care income and middle-income middle-income countries (LMICs) from countries. BMJ Global Health is essential for reducing the burden of preventable 2019;4:e001675. doi:10.1136/ causes that ‘should not occur in the presence deaths in low-income and middle-income countries. bmjgh-2019-001675 of timely and effective healthcare’. Improving ► The conventional high-income country model, based on resource-intensive intensive care units with ex- the quality and availability of critical illness Handling editor Seye Abimbola pensive monitoring and supportive equipment and care in LMICs is essential if this burden is to 1 2 large numbers of highly trained staff, is unlikely to be reduced, and even more important over Received 29 April 2019 be suitable for these settings. the coming years as populations age and the Revised 18 May 2019 ► Currently, costs severely restrict access to critical Accepted 25 May 2019 prevalence of comorbidities, such as cardio- care in low-income and middle-income countries, vascular disease and diabetes, increase. and there is an urgent need to develop an alternative Currently, capacity for critical illness care affordable critical care model for these settings. 3–5 in many LMICs is limited. In high-income ► Innovative technology and digital health may offer countries, there are generally between 5 and part of the solution and enable the development of 30 intensive care unit (ICU) beds per 100 an affordable, sustainable and scalable model of 2 3 000 people. The limited data available indi- critical care in resource-limited settings. cate that in LMICs, there are between 0.1 and 2.5 ICU beds per 100 000 people. Many higher than for high-income countries, due countries are also transitioning from low to to importation taxes and non-competitive lower–middle income status, receiving less © Author(s) (or their employer(s)) 2019. Re-use pricing structures. international healthcare aid which may limit permitted under CC BY. Maintaining operability of expensive ICU resources available for expanding capacity. Published by BMJ. equipment is a further challenge in LMICs While, the expansion of private health- Oxford University Clinical where there may be frequent power cuts and care systems in LMICs may partly meet the Research Unit, Wellcome Africa high ambient temperatures and humidity. increased demand, the quality of care deliv- Asia Programme, Ho Chi Minh Restricted availability of maintenance staff City, Vietnam ered by such providers is variable and will be 2 2 7 Nuffield Department of and replacement parts, means that equipment unaffordable for many. Medicine, Centre for Tropical is often non-functioning or cannot be used to Careful physiological monitoring is the Medicine and Global Health, 8 2 10 its full potential. Additionally, the paucity cornerstone of good critical illness care. In University of Oxford, Oxford, UK 3 of appropriately trained staff and limited the conventional high-income setting ICU Hospital for Tropical Diseases, infection control measures can result in more Ho Chi Minh City, Vietnam model, monitoring is achieved with expensive Department of Engineering frequent complications, which may worsen equipment, high-quality laboratory support 11 12 Science, Institute of Biomedical outcomes and further increase costs. and large numbers of highly trained staff. In Engineering, University of Costing studies conducted in high-income LMICs, this model is usually impractical as Oxford, Oxford, UK 5 countries have reported average costs of ICU the required resources are either unavailable Mahidol-Oxford Tropical 4 5 9 care between US$1700 and 4500 per day Medicine Research Unit, Faculty or too expensive. The figure 1 shows the 13 14 of Tropical Medicine, Mahidol (adjusted to 2014 prices). The delivery of predicted costs of providing a high-income University, Bangkok, Thailand critical care is less expensive in LMICs largely country model ICU bed in Vietnam. Although because of much lower labour costs; for this is just one case study, it highlights the Correspondence to example, a study based in an Indian hospital magnitude of those costs. Counterintui- Dr Hugo C Turner; hturner@ oucru. org estimated the average daily cost of ICU care tively, equipment costs can be substantially Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 1 BMJ Global Health Figure 1 The costs for the monitoring and supportive equipment associated with an intensive care unit bed in Vietnam. Costs are based on quotes from commercial medical equipment distributors in Vietnam (2018 prices). Costs are not annualised. was US$109 (2014 prices). Although this amount may learning algorithms can be used to analyse ICU patients’ appear low, the average annual healthcare expendi- physiological data, learn from them and create comput- ture per capita across LMICs is only around 5% that of er-assisted decision-support systems. With simple modi- high-income countries. fications, low-cost wearables can feed data into AI systems Furthermore, in LMICs, critical care costs are often which can then guide treatment decisions and diagnos- not fully covered by the health/insurance systems and tics. A key advantage of this approach is that AI systems patients’ and their families can incur high out-of-pocket can compensate for the noisy, artefactual signals that 17–19 expenses. Currently costs severely restrict access to typically arise from wearables. In high-income settings, AI ICU care in LMICs, particularly for the socioeconom- algorithms have been shown to improve the management 24 25 ically disadvantaged and uninsured, and there is an of sepsis and lower mortality. The ability of AI systems urgent need to develop an alternative affordable critical to continuously learn and adapt means that computer-as- care model for LMICs. sisted clinical decision-support systems can be tailored to the needs of a specific context or setting. Thus algo- rithms could be created to help in the management of WHaT Can be done? diseases such as malaria, dengue and tetanus, which are The emergence of new technologies, means there are uncommon in high-income ICU settings but are signifi- huge opportunities to expand capacity and improve the 10 21 22 cant problems in LMICs. Importantly, existing libraries care of critically ill patients in LMICs. of analogous data sets acquired from western clinical A substantial proportion of critical care costs in LMICs settings can be used, along with smaller quantities of are to cover staffing and fixed asset equipment costs as LMICs physiological data to permit ‘transfer learning’, in opposed to actual medications and laboratory tests. which complex predictive models can be retrained and Methods impacting these may be a way of reducing costs, recalibrated for use with low-cost sensors. allowing expansion of capacity as well as improving the As well as physiological monitoring, point-of-care diag- care quality. nostic and imaging devices are also increasingly available Low-cost wearable devices offer a potentially afford- and affordable. Devices such as hand-held ultrasound able approach to physiological monitoring in LMICs, probes connecting to a mobile phone could make equip- reducing the need for expensive commercial equipment ping LMIC ICUs more feasible and cheaper, aiding and, combined with artificial intelligence (AI), may also diagnosis and management of patients. While a high improve care quality. Wearable devices, such as fitness degree of training is currently required to acquire and trackers, have been used in ICU populations in high-in- come countries and have shown good correlation with interpret images, future AI systems may provide operator conventionally-derived ECG data. AI and machine guidance for inexperienced users and perform image 2 Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 BMJ Global Health Contributors HCT, CLT and NVVC conceived the manuscript. All authors interpretation, reducing requirements for highly trained contributed to the writing. staff and making the provision of critical care services Funding HCT, SY, VMTH and GET (089276/B/09/7). AMD is supported by outside of major urban hospitals significantly more the Wellcome Trust (106698/B14/Z) CLT is supported by the Wellcome Trust feasible. (107367/Z/15/Z). Competing interests None declared. patient consent for publication Not required. WHaT SHould Happen nexT? provenance and peer review Not commissioned; externally peer reviewed. The potential for new technology to transform health- care in LMICs is now widely accepted, but most innova- data availability statement All data are contained within the main body of the text. tions remain at the proof-of-concept stage or have only 22 26 open access This is an open access article distributed in accordance with the been tested in small pilot studies. The current chal- Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits lenge is bridging the gap between proof-of-concept and others to copy, redistribute, remix, transform and build upon this work for any 22 27 actual large-scale implementation. purpose, provided the original work is properly cited, a link to the licence is given, Moving forward, there is an urgent need to conduct and indication of whether changes were made. See: https:// creativecommons. org/ licenses/ by/ 4. 0/. implementation trials, to assess the actual effectiveness and feasibility of using these new digital technologies for critical care in LMICs. Successful innovation can only take place in close collaboration with end-user commu- nities and a real understanding of the contextual need. ReFeRenCes 1. Adhikari NKJ, Fowler RA, Bhagwanjee S, et al. Critical care In addition, the wide variety of critical care capacity and and the global burden of critical illness in adults. The Lancet facilities within and between many LMICs means that 2010;376:1339–46. 2. Dondorp AM, Iyer SS, Schultz MJ. Critical care in Resource- new technologies should be designed to fit within the Restricted settings. JAMA 2016;315:753–4. existing infrastructure. This will therefore require more 3. Schultz MJ, Dunser MW, Dondorp AM, et al. Current challenges in than a simple design process but an active two-way part- the management of sepsis in ICUs in resource-poor settings and suggestions for the future. Intensive Care Med 2017;43:612–24. nership between all stakeholders and with considerations 4. Murthy S, Leligdowicz A, Adhikari NKJ. Intensive care unit regarding scale-up taken into account from the start of capacity in low-income countries: a systematic review. Plos One 2015;10:e0116949. the process. 5. Baelani I, Jochberger S, Laimer T, et al. Availability of critical care The use of these new technologies also needs to be resources to treat patients with severe sepsis or septic shock part of broader strategies to improve ICU performance. in Africa: a self-reported, continent-wide survey of anaesthesia providers. Crit Care 2011;15. Other potential strategies for improving the delivery of 6. Dieleman J, Campbell M, Chapin A, et al. Evolution and patterns critical care in LMICs include improving organisational of global health financing 1995–2014: development assistance for health, and government, prepaid private, and out-of-pocket health structures, empowerment of nurses and locally gener- spending in 184 countries. The Lancet 2017;389:1981–2004. 2 28 ated clinical guidelines. In addition, in most LMICs, 7. Sheikh K, Saligram PS, Hort K. What explains regulatory failure? analysing the architecture of health care regulation in two Indian critical care is not currently a well-developed specialty. states. Health Policy Plan 2015;30:39–55. Consequently, the development of training and capacity 8. Baker T, Khalid K, Acicbe O, et al. Critical care of tropical disease building programmes is particularly important—not only in low income countries: report from the task Force on tropical diseases by the World Federation of societies of intensive and for ICU physicians but also for nurses and other clinical critical care medicine. J Crit Care 2017;42:351–4. personnel. It is vital that these training programmes, as 9. Dünser MW, Baelani I, Ganbold L. A review and analysis of intensive care medicine in the least developed countries. Crit Care Med well as covering specific ICU clinical skills also include 2006;34:1234–42. basic management and organisational aspects of crit- 10. Mekontso Dessap A. Frugal innovation for critical care. Intensive ical care. Successful initiatives such as Train-the-Trainer Care Med 2019;45:252–4. 11. Arabi YM, Phua J, Koh Y, et al. Structure, organization, and delivery and peer-to-peer programmes have been shown to be of critical care in Asian ICUs. Crit Care Med 2016;44:e940–8. 29 30 successful in LMICs and could be further expanded. 12. Dat VQ, Long NT, Giang KB, et al. Healthcare infrastructure capacity to respond to severe acute respiratory infection (SARI) and sepsis in Vietnam: a low-middle income country. J Crit Care 2017;42:109–15. 13. Halpern NA, Pastores SM. Critical care medicine in the United States ConCluSIon 2000-2005: an analysis of bed numbers, occupancy rates, payer mix, and costs. Crit Care Med 2010;38:65–71. Improving the quality and availability of critical care is 14. Tan SS, Bakker J, Hoogendoorn ME, et al. Direct cost analysis essential for reducing the burden of preventable deaths of intensive care unit stay in four European countries: applying a standardized costing methodology. Value in Health 2012;15:81–6. in LMICs. There will be no one size fits all solution to this 15. Agrawal A, Gandhe M, Gandhe S, et al. Study of length of stay and problem and a multifaceted approach is required. Never- average cost of treatment in medicine intensive care unit at tertiary care center. J Health Res Rev 2017;4:24–9. theless, greater utilisation of new technologies could be 16. World Bank. World development indicators. Secondary world an important part of the solution. Through such inno- development indicators. Available: https:// data. worldbank. org/ vation, critical care capacity could not only be increased indicator 17. Jayaram R, Ramakrishnan N. Cost of intensive care in India. Indian J but also be improved in quality and at a reduced cost. Crit Care Med 2008;12:55–61. However, designing and implementing sustainable and 18. Woldeamanuel YW, Andemeskel AT, Kyei K, et al. Case fatality of adult tetanus in Africa: systematic review and meta-analysis. J scalable solutions is a significant challenge, requiring Neurol Sci 2016;368:292–9. strong collaborations and real understanding between 19. Shukla VV, Nimbalkar SM, Ganjiwale JD, et al. Direct cost of critical all stakeholders. illness associated healthcare expenditures among children admitted Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 3 BMJ Global Health in pediatric intensive care unit in rural India. Indian J Pediatr 25. Komorowski M, Celi LA, Badawi O, et al. The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive 2016;83:1065–70. care. Nat Med 2018;24:1716–20. 20. Divatia JV, Iyer S. Ten major priorities for intensive care in India. 26. Lundin J, Dumont G. Medical mobile technologies – what is needed Intensive Care Med 2015;41:1468–71. for a sustainable and scalable implementation on a global scale? 21. Rodriguez-Manzano J, Chia PY, Yeo TW, et al. Improving dengue Global Health Action 2017;10. diagnostics and management through innovative technology. Curr 27. Norrie J. The challenge of implementing AI models in the ICU. Infect Dis Rep 2018;20. Lancet Respir Med 2018;6:886–8. 22. Wahl B, Cossy-Gantner A, Germann S, et al. Artificial intelligence (AI) 28. Diaz JV, Riviello ED, Papali A, et al. Global critical care: moving and global health: how can AI contribute to health in resource-poor forward in resource-limited settings. Annals of Global Health settings? BMJ Global Health 2018;3:e000798. 2019;85. 23. Kroll RR, McKenzie ED, Boyd JG, et al. Use of wearable devices 29. Beane A, Padeniya A, De Silva AP, et al. Closing the theory to for post-discharge monitoring of ICU patients: a feasibility study. J practice gap for newly qualified doctors: evaluation of a peer- Intensive Care 2017;5. delivered practical skills training course for newly qualified doctors in 24. Shimabukuro DW, Barton CW, Feldman MD, et al. Effect of a preparation for clinical practice. Postgrad Med J 2017;93:592–6. machine learning-based severe sepsis prediction algorithm on 30. Tunpattu S, Newey V, Sigera C, et al. A short, structured skills patient survival and hospital length of stay: a randomised clinical training course for critical care physiotherapists in a lower-middle trial. BMJ Open Respir Res 2017;4. income country. Physiother Theory Pract 2018;34:714–22. 4 Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMJ Global Health British Medical Journal

Achieving affordable critical care in low-income and middle-income countries

Download PDF
4 pages

Loading next page...
 
/lp/british-medical-journal/achieving-affordable-critical-care-in-low-income-and-middle-income-kca0b2voHb

References (62)

Publisher
British Medical Journal
Copyright
© Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY. Published by BMJ.
eISSN
2059-7908
DOI
10.1136/bmjgh-2019-001675
Publisher site
See Article on Publisher Site

Abstract

Commentary Achieving affordable critical care in low-income and middle- income countries  1,2 3 1,2 1 Hugo C Turner, Nguyen Van Hao, Sophie Yacoub, Van Minh Tu Hoang, 4 1,2 2,5 1,2 David A Clifton, Guy E Thwaites, Arjen M Dondorp, C Louise Thwaites, Nguyen Van Vinh Chau To cite: Turner HC, Hao NV, InTroduCTIon Summary box Yacoub S, et al. Achieving In 2016, an estimated 8.6 million prema- affordable critical care in low- ture deaths occurred in low-income and ► Improving the quality and availability of critical care income and middle-income middle-income countries (LMICs) from countries. BMJ Global Health is essential for reducing the burden of preventable 2019;4:e001675. doi:10.1136/ causes that ‘should not occur in the presence deaths in low-income and middle-income countries. bmjgh-2019-001675 of timely and effective healthcare’. Improving ► The conventional high-income country model, based on resource-intensive intensive care units with ex- the quality and availability of critical illness Handling editor Seye Abimbola pensive monitoring and supportive equipment and care in LMICs is essential if this burden is to 1 2 large numbers of highly trained staff, is unlikely to be reduced, and even more important over Received 29 April 2019 be suitable for these settings. the coming years as populations age and the Revised 18 May 2019 ► Currently, costs severely restrict access to critical Accepted 25 May 2019 prevalence of comorbidities, such as cardio- care in low-income and middle-income countries, vascular disease and diabetes, increase. and there is an urgent need to develop an alternative Currently, capacity for critical illness care affordable critical care model for these settings. 3–5 in many LMICs is limited. In high-income ► Innovative technology and digital health may offer countries, there are generally between 5 and part of the solution and enable the development of 30 intensive care unit (ICU) beds per 100 an affordable, sustainable and scalable model of 2 3 000 people. The limited data available indi- critical care in resource-limited settings. cate that in LMICs, there are between 0.1 and 2.5 ICU beds per 100 000 people. Many higher than for high-income countries, due countries are also transitioning from low to to importation taxes and non-competitive lower–middle income status, receiving less © Author(s) (or their employer(s)) 2019. Re-use pricing structures. international healthcare aid which may limit permitted under CC BY. Maintaining operability of expensive ICU resources available for expanding capacity. Published by BMJ. equipment is a further challenge in LMICs While, the expansion of private health- Oxford University Clinical where there may be frequent power cuts and care systems in LMICs may partly meet the Research Unit, Wellcome Africa high ambient temperatures and humidity. increased demand, the quality of care deliv- Asia Programme, Ho Chi Minh Restricted availability of maintenance staff City, Vietnam ered by such providers is variable and will be 2 2 7 Nuffield Department of and replacement parts, means that equipment unaffordable for many. Medicine, Centre for Tropical is often non-functioning or cannot be used to Careful physiological monitoring is the Medicine and Global Health, 8 2 10 its full potential. Additionally, the paucity cornerstone of good critical illness care. In University of Oxford, Oxford, UK 3 of appropriately trained staff and limited the conventional high-income setting ICU Hospital for Tropical Diseases, infection control measures can result in more Ho Chi Minh City, Vietnam model, monitoring is achieved with expensive Department of Engineering frequent complications, which may worsen equipment, high-quality laboratory support 11 12 Science, Institute of Biomedical outcomes and further increase costs. and large numbers of highly trained staff. In Engineering, University of Costing studies conducted in high-income LMICs, this model is usually impractical as Oxford, Oxford, UK 5 countries have reported average costs of ICU the required resources are either unavailable Mahidol-Oxford Tropical 4 5 9 care between US$1700 and 4500 per day Medicine Research Unit, Faculty or too expensive. The figure 1 shows the 13 14 of Tropical Medicine, Mahidol (adjusted to 2014 prices). The delivery of predicted costs of providing a high-income University, Bangkok, Thailand critical care is less expensive in LMICs largely country model ICU bed in Vietnam. Although because of much lower labour costs; for this is just one case study, it highlights the Correspondence to example, a study based in an Indian hospital magnitude of those costs. Counterintui- Dr Hugo C Turner; hturner@ oucru. org estimated the average daily cost of ICU care tively, equipment costs can be substantially Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 1 BMJ Global Health Figure 1 The costs for the monitoring and supportive equipment associated with an intensive care unit bed in Vietnam. Costs are based on quotes from commercial medical equipment distributors in Vietnam (2018 prices). Costs are not annualised. was US$109 (2014 prices). Although this amount may learning algorithms can be used to analyse ICU patients’ appear low, the average annual healthcare expendi- physiological data, learn from them and create comput- ture per capita across LMICs is only around 5% that of er-assisted decision-support systems. With simple modi- high-income countries. fications, low-cost wearables can feed data into AI systems Furthermore, in LMICs, critical care costs are often which can then guide treatment decisions and diagnos- not fully covered by the health/insurance systems and tics. A key advantage of this approach is that AI systems patients’ and their families can incur high out-of-pocket can compensate for the noisy, artefactual signals that 17–19 expenses. Currently costs severely restrict access to typically arise from wearables. In high-income settings, AI ICU care in LMICs, particularly for the socioeconom- algorithms have been shown to improve the management 24 25 ically disadvantaged and uninsured, and there is an of sepsis and lower mortality. The ability of AI systems urgent need to develop an alternative affordable critical to continuously learn and adapt means that computer-as- care model for LMICs. sisted clinical decision-support systems can be tailored to the needs of a specific context or setting. Thus algo- rithms could be created to help in the management of WHaT Can be done? diseases such as malaria, dengue and tetanus, which are The emergence of new technologies, means there are uncommon in high-income ICU settings but are signifi- huge opportunities to expand capacity and improve the 10 21 22 cant problems in LMICs. Importantly, existing libraries care of critically ill patients in LMICs. of analogous data sets acquired from western clinical A substantial proportion of critical care costs in LMICs settings can be used, along with smaller quantities of are to cover staffing and fixed asset equipment costs as LMICs physiological data to permit ‘transfer learning’, in opposed to actual medications and laboratory tests. which complex predictive models can be retrained and Methods impacting these may be a way of reducing costs, recalibrated for use with low-cost sensors. allowing expansion of capacity as well as improving the As well as physiological monitoring, point-of-care diag- care quality. nostic and imaging devices are also increasingly available Low-cost wearable devices offer a potentially afford- and affordable. Devices such as hand-held ultrasound able approach to physiological monitoring in LMICs, probes connecting to a mobile phone could make equip- reducing the need for expensive commercial equipment ping LMIC ICUs more feasible and cheaper, aiding and, combined with artificial intelligence (AI), may also diagnosis and management of patients. While a high improve care quality. Wearable devices, such as fitness degree of training is currently required to acquire and trackers, have been used in ICU populations in high-in- come countries and have shown good correlation with interpret images, future AI systems may provide operator conventionally-derived ECG data. AI and machine guidance for inexperienced users and perform image 2 Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 BMJ Global Health Contributors HCT, CLT and NVVC conceived the manuscript. All authors interpretation, reducing requirements for highly trained contributed to the writing. staff and making the provision of critical care services Funding HCT, SY, VMTH and GET (089276/B/09/7). AMD is supported by outside of major urban hospitals significantly more the Wellcome Trust (106698/B14/Z) CLT is supported by the Wellcome Trust feasible. (107367/Z/15/Z). Competing interests None declared. patient consent for publication Not required. WHaT SHould Happen nexT? provenance and peer review Not commissioned; externally peer reviewed. The potential for new technology to transform health- care in LMICs is now widely accepted, but most innova- data availability statement All data are contained within the main body of the text. tions remain at the proof-of-concept stage or have only 22 26 open access This is an open access article distributed in accordance with the been tested in small pilot studies. The current chal- Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits lenge is bridging the gap between proof-of-concept and others to copy, redistribute, remix, transform and build upon this work for any 22 27 actual large-scale implementation. purpose, provided the original work is properly cited, a link to the licence is given, Moving forward, there is an urgent need to conduct and indication of whether changes were made. See: https:// creativecommons. org/ licenses/ by/ 4. 0/. implementation trials, to assess the actual effectiveness and feasibility of using these new digital technologies for critical care in LMICs. Successful innovation can only take place in close collaboration with end-user commu- nities and a real understanding of the contextual need. ReFeRenCes 1. Adhikari NKJ, Fowler RA, Bhagwanjee S, et al. Critical care In addition, the wide variety of critical care capacity and and the global burden of critical illness in adults. The Lancet facilities within and between many LMICs means that 2010;376:1339–46. 2. Dondorp AM, Iyer SS, Schultz MJ. Critical care in Resource- new technologies should be designed to fit within the Restricted settings. JAMA 2016;315:753–4. existing infrastructure. This will therefore require more 3. Schultz MJ, Dunser MW, Dondorp AM, et al. Current challenges in than a simple design process but an active two-way part- the management of sepsis in ICUs in resource-poor settings and suggestions for the future. Intensive Care Med 2017;43:612–24. nership between all stakeholders and with considerations 4. Murthy S, Leligdowicz A, Adhikari NKJ. Intensive care unit regarding scale-up taken into account from the start of capacity in low-income countries: a systematic review. Plos One 2015;10:e0116949. the process. 5. Baelani I, Jochberger S, Laimer T, et al. Availability of critical care The use of these new technologies also needs to be resources to treat patients with severe sepsis or septic shock part of broader strategies to improve ICU performance. in Africa: a self-reported, continent-wide survey of anaesthesia providers. Crit Care 2011;15. Other potential strategies for improving the delivery of 6. Dieleman J, Campbell M, Chapin A, et al. Evolution and patterns critical care in LMICs include improving organisational of global health financing 1995–2014: development assistance for health, and government, prepaid private, and out-of-pocket health structures, empowerment of nurses and locally gener- spending in 184 countries. The Lancet 2017;389:1981–2004. 2 28 ated clinical guidelines. In addition, in most LMICs, 7. Sheikh K, Saligram PS, Hort K. What explains regulatory failure? analysing the architecture of health care regulation in two Indian critical care is not currently a well-developed specialty. states. Health Policy Plan 2015;30:39–55. Consequently, the development of training and capacity 8. Baker T, Khalid K, Acicbe O, et al. Critical care of tropical disease building programmes is particularly important—not only in low income countries: report from the task Force on tropical diseases by the World Federation of societies of intensive and for ICU physicians but also for nurses and other clinical critical care medicine. J Crit Care 2017;42:351–4. personnel. It is vital that these training programmes, as 9. Dünser MW, Baelani I, Ganbold L. A review and analysis of intensive care medicine in the least developed countries. Crit Care Med well as covering specific ICU clinical skills also include 2006;34:1234–42. basic management and organisational aspects of crit- 10. Mekontso Dessap A. Frugal innovation for critical care. Intensive ical care. Successful initiatives such as Train-the-Trainer Care Med 2019;45:252–4. 11. Arabi YM, Phua J, Koh Y, et al. Structure, organization, and delivery and peer-to-peer programmes have been shown to be of critical care in Asian ICUs. Crit Care Med 2016;44:e940–8. 29 30 successful in LMICs and could be further expanded. 12. Dat VQ, Long NT, Giang KB, et al. Healthcare infrastructure capacity to respond to severe acute respiratory infection (SARI) and sepsis in Vietnam: a low-middle income country. J Crit Care 2017;42:109–15. 13. Halpern NA, Pastores SM. Critical care medicine in the United States ConCluSIon 2000-2005: an analysis of bed numbers, occupancy rates, payer mix, and costs. Crit Care Med 2010;38:65–71. Improving the quality and availability of critical care is 14. Tan SS, Bakker J, Hoogendoorn ME, et al. Direct cost analysis essential for reducing the burden of preventable deaths of intensive care unit stay in four European countries: applying a standardized costing methodology. Value in Health 2012;15:81–6. in LMICs. There will be no one size fits all solution to this 15. Agrawal A, Gandhe M, Gandhe S, et al. Study of length of stay and problem and a multifaceted approach is required. Never- average cost of treatment in medicine intensive care unit at tertiary care center. J Health Res Rev 2017;4:24–9. theless, greater utilisation of new technologies could be 16. World Bank. World development indicators. Secondary world an important part of the solution. Through such inno- development indicators. Available: https:// data. worldbank. org/ vation, critical care capacity could not only be increased indicator 17. Jayaram R, Ramakrishnan N. Cost of intensive care in India. Indian J but also be improved in quality and at a reduced cost. Crit Care Med 2008;12:55–61. However, designing and implementing sustainable and 18. Woldeamanuel YW, Andemeskel AT, Kyei K, et al. Case fatality of adult tetanus in Africa: systematic review and meta-analysis. J scalable solutions is a significant challenge, requiring Neurol Sci 2016;368:292–9. strong collaborations and real understanding between 19. Shukla VV, Nimbalkar SM, Ganjiwale JD, et al. Direct cost of critical all stakeholders. illness associated healthcare expenditures among children admitted Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675 3 BMJ Global Health in pediatric intensive care unit in rural India. Indian J Pediatr 25. Komorowski M, Celi LA, Badawi O, et al. The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive 2016;83:1065–70. care. Nat Med 2018;24:1716–20. 20. Divatia JV, Iyer S. Ten major priorities for intensive care in India. 26. Lundin J, Dumont G. Medical mobile technologies – what is needed Intensive Care Med 2015;41:1468–71. for a sustainable and scalable implementation on a global scale? 21. Rodriguez-Manzano J, Chia PY, Yeo TW, et al. Improving dengue Global Health Action 2017;10. diagnostics and management through innovative technology. Curr 27. Norrie J. The challenge of implementing AI models in the ICU. Infect Dis Rep 2018;20. Lancet Respir Med 2018;6:886–8. 22. Wahl B, Cossy-Gantner A, Germann S, et al. Artificial intelligence (AI) 28. Diaz JV, Riviello ED, Papali A, et al. Global critical care: moving and global health: how can AI contribute to health in resource-poor forward in resource-limited settings. Annals of Global Health settings? BMJ Global Health 2018;3:e000798. 2019;85. 23. Kroll RR, McKenzie ED, Boyd JG, et al. Use of wearable devices 29. Beane A, Padeniya A, De Silva AP, et al. Closing the theory to for post-discharge monitoring of ICU patients: a feasibility study. J practice gap for newly qualified doctors: evaluation of a peer- Intensive Care 2017;5. delivered practical skills training course for newly qualified doctors in 24. Shimabukuro DW, Barton CW, Feldman MD, et al. Effect of a preparation for clinical practice. Postgrad Med J 2017;93:592–6. machine learning-based severe sepsis prediction algorithm on 30. Tunpattu S, Newey V, Sigera C, et al. A short, structured skills patient survival and hospital length of stay: a randomised clinical training course for critical care physiotherapists in a lower-middle trial. BMJ Open Respir Res 2017;4. income country. Physiother Theory Pract 2018;34:714–22. 4 Turner HC, et al. BMJ Global Health 2019;4:e001675. doi:10.1136/bmjgh-2019-001675

Journal

BMJ Global Health – British Medical Journal

Published: Jun 19, 2019

Keywords: health economics; health systems; public health

There are no references for this article.