Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app

Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Ausführliche Beschreibung

Gespeichert in:

Autor*in:	James O’Donovan [verfasserIn] Ken Kahn [verfasserIn] MacKenzie MacRae [verfasserIn] Allan Saul Namanda [verfasserIn] Rebecca Hamala [verfasserIn] Ken Kabali [verfasserIn] Anne Geniets [verfasserIn] Alice Lakati [verfasserIn] Simon M. Mbae [verfasserIn] Niall Winters [verfasserIn]

Format:	E-Artikel
Sprache:	Englisch

Erschienen:	2022

Schlagwörter:	Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training

Übergeordnetes Werk:	In: Human Resources for Health - BMC, 2003, 20(2022), 1, Seite 8
Übergeordnetes Werk:	volume:20 ; year:2022 ; number:1 ; pages:8

Links:	Link aufrufen Link aufrufen Link aufrufen Journal toc

DOI / URN:	10.1186/s12960-021-00699-5

Katalog-ID:	DOAJ046901515

Internformat


LEADER	01000caa a22002652 4500
001	DOAJ046901515
003	DE-627
005	20230308114851.0
007	cr uuu---uuuuu
008	230227s2022 xx \|\|\|\|\|o 00\| \|\|eng c
024	7		\|a 10.1186/s12960-021-00699-5 \|2 doi
035			\|a (DE-627)DOAJ046901515
035			\|a (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289
040			\|a DE-627 \|b ger \|c DE-627 \|e rakwb
041			\|a eng
050		0	\|a R5-920
050		0	\|a RA1-1270
100	0		\|a James O’Donovan \|e verfasserin \|4 aut
245	1	0	\|a Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
264		1	\|c 2022
336			\|a Text \|b txt \|2 rdacontent
337			\|a Computermedien \|b c \|2 rdamedia
338			\|a Online-Ressource \|b cr \|2 rdacarrier
520			\|a Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.
650		4	\|a Machine learning
650		4	\|a Artificial intelligence
650		4	\|a Supervision
650		4	\|a Community Health Worker
650		4	\|a Digital Health
650		4	\|a Training
653		0	\|a Medicine (General)
653		0	\|a Public aspects of medicine
700	0		\|a Ken Kahn \|e verfasserin \|4 aut
700	0		\|a MacKenzie MacRae \|e verfasserin \|4 aut
700	0		\|a Allan Saul Namanda \|e verfasserin \|4 aut
700	0		\|a Rebecca Hamala \|e verfasserin \|4 aut
700	0		\|a Ken Kabali \|e verfasserin \|4 aut
700	0		\|a Anne Geniets \|e verfasserin \|4 aut
700	0		\|a Alice Lakati \|e verfasserin \|4 aut
700	0		\|a Simon M. Mbae \|e verfasserin \|4 aut
700	0		\|a Niall Winters \|e verfasserin \|4 aut
773	0	8	\|i In \|t Human Resources for Health \|d BMC, 2003 \|g 20(2022), 1, Seite 8 \|w (DE-627)373756585 \|w (DE-600)2126923-3 \|x 14784491 \|7 nnns
773	1	8	\|g volume:20 \|g year:2022 \|g number:1 \|g pages:8
856	4	0	\|u https://doi.org/10.1186/s12960-021-00699-5 \|z kostenfrei
856	4	0	\|u https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 \|z kostenfrei
856	4	0	\|u https://doi.org/10.1186/s12960-021-00699-5 \|z kostenfrei
856	4	2	\|u https://doaj.org/toc/1478-4491 \|y Journal toc \|z kostenfrei
912			\|a GBV_USEFLAG_A
912			\|a SYSFLAG_A
912			\|a GBV_DOAJ
912			\|a GBV_ILN_11
912			\|a GBV_ILN_20
912			\|a GBV_ILN_22
912			\|a GBV_ILN_23
912			\|a GBV_ILN_24
912			\|a GBV_ILN_31
912			\|a GBV_ILN_39
912			\|a GBV_ILN_40
912			\|a GBV_ILN_60
912			\|a GBV_ILN_62
912			\|a GBV_ILN_63
912			\|a GBV_ILN_65
912			\|a GBV_ILN_69
912			\|a GBV_ILN_73
912			\|a GBV_ILN_74
912			\|a GBV_ILN_95
912			\|a GBV_ILN_110
912			\|a GBV_ILN_151
912			\|a GBV_ILN_161
912			\|a GBV_ILN_170
912			\|a GBV_ILN_206
912			\|a GBV_ILN_213
912			\|a GBV_ILN_230
912			\|a GBV_ILN_285
912			\|a GBV_ILN_293
912			\|a GBV_ILN_602
912			\|a GBV_ILN_2003
912			\|a GBV_ILN_2005
912			\|a GBV_ILN_2009
912			\|a GBV_ILN_2011
912			\|a GBV_ILN_2014
912			\|a GBV_ILN_2055
912			\|a GBV_ILN_2111
912			\|a GBV_ILN_4012
912			\|a GBV_ILN_4037
912			\|a GBV_ILN_4112
912			\|a GBV_ILN_4125
912			\|a GBV_ILN_4126
912			\|a GBV_ILN_4249
912			\|a GBV_ILN_4305
912			\|a GBV_ILN_4306
912			\|a GBV_ILN_4322
912			\|a GBV_ILN_4323
912			\|a GBV_ILN_4324
912			\|a GBV_ILN_4325
912			\|a GBV_ILN_4367
912			\|a GBV_ILN_4700
951			\|a AR
952			\|d 20 \|j 2022 \|e 1 \|h 8

Indexfelder

author_variant	j o jo k k kk m m mm a s n asn r h rh k k kk a g ag a l al s m m smm n w nw
matchkey_str	article:14784491:2022----::nlsn32dgtluevsritrcinbtenomnthatwresngnanknahdvlpetetnadaiain
hierarchy_sort_str	2022
callnumber-subject-code	R
publishDate	2022
allfields	10.1186/s12960-021-00699-5 doi (DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289 DE-627 ger DE-627 rakwb eng R5-920 RA1-1270 James O’Donovan verfasserin aut Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app 2022 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine Ken Kahn verfasserin aut MacKenzie MacRae verfasserin aut Allan Saul Namanda verfasserin aut Rebecca Hamala verfasserin aut Ken Kabali verfasserin aut Anne Geniets verfasserin aut Alice Lakati verfasserin aut Simon M. Mbae verfasserin aut Niall Winters verfasserin aut In Human Resources for Health BMC, 2003 20(2022), 1, Seite 8 (DE-627)373756585 (DE-600)2126923-3 14784491 nnns volume:20 year:2022 number:1 pages:8 https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 kostenfrei https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/toc/1478-4491 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700 AR 20 2022 1 8
spelling	10.1186/s12960-021-00699-5 doi (DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289 DE-627 ger DE-627 rakwb eng R5-920 RA1-1270 James O’Donovan verfasserin aut Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app 2022 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine Ken Kahn verfasserin aut MacKenzie MacRae verfasserin aut Allan Saul Namanda verfasserin aut Rebecca Hamala verfasserin aut Ken Kabali verfasserin aut Anne Geniets verfasserin aut Alice Lakati verfasserin aut Simon M. Mbae verfasserin aut Niall Winters verfasserin aut In Human Resources for Health BMC, 2003 20(2022), 1, Seite 8 (DE-627)373756585 (DE-600)2126923-3 14784491 nnns volume:20 year:2022 number:1 pages:8 https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 kostenfrei https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/toc/1478-4491 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700 AR 20 2022 1 8
allfields_unstemmed	10.1186/s12960-021-00699-5 doi (DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289 DE-627 ger DE-627 rakwb eng R5-920 RA1-1270 James O’Donovan verfasserin aut Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app 2022 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine Ken Kahn verfasserin aut MacKenzie MacRae verfasserin aut Allan Saul Namanda verfasserin aut Rebecca Hamala verfasserin aut Ken Kabali verfasserin aut Anne Geniets verfasserin aut Alice Lakati verfasserin aut Simon M. Mbae verfasserin aut Niall Winters verfasserin aut In Human Resources for Health BMC, 2003 20(2022), 1, Seite 8 (DE-627)373756585 (DE-600)2126923-3 14784491 nnns volume:20 year:2022 number:1 pages:8 https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 kostenfrei https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/toc/1478-4491 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700 AR 20 2022 1 8
allfieldsGer	10.1186/s12960-021-00699-5 doi (DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289 DE-627 ger DE-627 rakwb eng R5-920 RA1-1270 James O’Donovan verfasserin aut Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app 2022 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine Ken Kahn verfasserin aut MacKenzie MacRae verfasserin aut Allan Saul Namanda verfasserin aut Rebecca Hamala verfasserin aut Ken Kabali verfasserin aut Anne Geniets verfasserin aut Alice Lakati verfasserin aut Simon M. Mbae verfasserin aut Niall Winters verfasserin aut In Human Resources for Health BMC, 2003 20(2022), 1, Seite 8 (DE-627)373756585 (DE-600)2126923-3 14784491 nnns volume:20 year:2022 number:1 pages:8 https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 kostenfrei https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/toc/1478-4491 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700 AR 20 2022 1 8
allfieldsSound	10.1186/s12960-021-00699-5 doi (DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289 DE-627 ger DE-627 rakwb eng R5-920 RA1-1270 James O’Donovan verfasserin aut Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app 2022 Text txt rdacontent Computermedien c rdamedia Online-Ressource cr rdacarrier Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such. Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine Ken Kahn verfasserin aut MacKenzie MacRae verfasserin aut Allan Saul Namanda verfasserin aut Rebecca Hamala verfasserin aut Ken Kabali verfasserin aut Anne Geniets verfasserin aut Alice Lakati verfasserin aut Simon M. Mbae verfasserin aut Niall Winters verfasserin aut In Human Resources for Health BMC, 2003 20(2022), 1, Seite 8 (DE-627)373756585 (DE-600)2126923-3 14784491 nnns volume:20 year:2022 number:1 pages:8 https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 kostenfrei https://doi.org/10.1186/s12960-021-00699-5 kostenfrei https://doaj.org/toc/1478-4491 Journal toc kostenfrei GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700 AR 20 2022 1 8
language	English
source	In Human Resources for Health 20(2022), 1, Seite 8 volume:20 year:2022 number:1 pages:8
sourceStr	In Human Resources for Health 20(2022), 1, Seite 8 volume:20 year:2022 number:1 pages:8
format_phy_str_mv	Article
institution	findex.gbv.de
topic_facet	Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training Medicine (General) Public aspects of medicine
isfreeaccess_bool	true
container_title	Human Resources for Health
authorswithroles_txt_mv	James O’Donovan @@aut@@ Ken Kahn @@aut@@ MacKenzie MacRae @@aut@@ Allan Saul Namanda @@aut@@ Rebecca Hamala @@aut@@ Ken Kabali @@aut@@ Anne Geniets @@aut@@ Alice Lakati @@aut@@ Simon M. Mbae @@aut@@ Niall Winters @@aut@@
publishDateDaySort_date	2022-01-01T00:00:00Z
hierarchy_top_id	373756585
id	DOAJ046901515
language_de	englisch
fullrecord	<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">DOAJ046901515</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230308114851.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">230227s2022 xx \|\|\|\|\|o 00\| \|\|eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1186/s12960-021-00699-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)DOAJ046901515</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="050" ind1=" " ind2="0"><subfield code="a">R5-920</subfield></datafield><datafield tag="050" ind1=" " ind2="0"><subfield code="a">RA1-1270</subfield></datafield><datafield tag="100" ind1="0" ind2=" "><subfield code="a">James O’Donovan</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Machine learning</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Artificial intelligence</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Supervision</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Community Health Worker</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Digital Health</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Training</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Medicine (General)</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Public aspects of medicine</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Ken Kahn</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">MacKenzie MacRae</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Allan Saul Namanda</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Rebecca Hamala</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Ken Kabali</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Anne Geniets</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Alice Lakati</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Simon M. Mbae</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Niall Winters</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Human Resources for Health</subfield><subfield code="d">BMC, 2003</subfield><subfield code="g">20(2022), 1, Seite 8</subfield><subfield code="w">(DE-627)373756585</subfield><subfield code="w">(DE-600)2126923-3</subfield><subfield code="x">14784491</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:20</subfield><subfield code="g">year:2022</subfield><subfield code="g">number:1</subfield><subfield code="g">pages:8</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1186/s12960-021-00699-5</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1186/s12960-021-00699-5</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">https://doaj.org/toc/1478-4491</subfield><subfield code="y">Journal toc</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_DOAJ</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_11</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_31</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_39</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_63</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_74</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_161</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_170</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_206</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_285</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_293</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2003</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4012</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4126</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4367</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">20</subfield><subfield code="j">2022</subfield><subfield code="e">1</subfield><subfield code="h">8</subfield></datafield></record></collection>
callnumber-first	R - Medicine
author	James O’Donovan
spellingShingle	James O’Donovan misc R5-920 misc RA1-1270 misc Machine learning misc Artificial intelligence misc Supervision misc Community Health Worker misc Digital Health misc Training misc Medicine (General) misc Public aspects of medicine Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
authorStr	James O’Donovan
ppnlink_with_tag_str_mv	@@773@@(DE-627)373756585
format	electronic Article
delete_txt_mv	keep
author_role	aut aut aut aut aut aut aut aut aut aut
collection	DOAJ
remote_str	true
callnumber-label	R5-920
illustrated	Not Illustrated
issn	14784491
topic_title	R5-920 RA1-1270 Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app Machine learning Artificial intelligence Supervision Community Health Worker Digital Health Training
topic	misc R5-920 misc RA1-1270 misc Machine learning misc Artificial intelligence misc Supervision misc Community Health Worker misc Digital Health misc Training misc Medicine (General) misc Public aspects of medicine
topic_unstemmed	misc R5-920 misc RA1-1270 misc Machine learning misc Artificial intelligence misc Supervision misc Community Health Worker misc Digital Health misc Training misc Medicine (General) misc Public aspects of medicine
topic_browse	misc R5-920 misc RA1-1270 misc Machine learning misc Artificial intelligence misc Supervision misc Community Health Worker misc Digital Health misc Training misc Medicine (General) misc Public aspects of medicine
format_facet	Elektronische Aufsätze Aufsätze Elektronische Ressource
format_main_str_mv	Text Zeitschrift/Artikel
carriertype_str_mv	cr
hierarchy_parent_title	Human Resources for Health
hierarchy_parent_id	373756585
hierarchy_top_title	Human Resources for Health
isfreeaccess_txt	true
familylinks_str_mv	(DE-627)373756585 (DE-600)2126923-3
title	Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
ctrlnum	(DE-627)DOAJ046901515 (DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289
title_full	Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
author_sort	James O’Donovan
journal	Human Resources for Health
journalStr	Human Resources for Health
callnumber-first-code	R
lang_code	eng
isOA_bool	true
recordtype	marc
publishDateSort	2022
contenttype_str_mv	txt
container_start_page	8
author_browse	James O’Donovan Ken Kahn MacKenzie MacRae Allan Saul Namanda Rebecca Hamala Ken Kabali Anne Geniets Alice Lakati Simon M. Mbae Niall Winters
container_volume	20
class	R5-920 RA1-1270
format_se	Elektronische Aufsätze
author-letter	James O’Donovan
doi_str_mv	10.1186/s12960-021-00699-5
author2-role	verfasserin
title_sort	analysing 3429 digital supervisory interactions between community health workers in uganda and kenya: the development, testing and validation of an open access predictive machine learning web app
callnumber	R5-920
title_auth	Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
abstract	Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.
abstractGer	Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.
abstract_unstemmed	Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.
collection_details	GBV_USEFLAG_A SYSFLAG_A GBV_DOAJ GBV_ILN_11 GBV_ILN_20 GBV_ILN_22 GBV_ILN_23 GBV_ILN_24 GBV_ILN_31 GBV_ILN_39 GBV_ILN_40 GBV_ILN_60 GBV_ILN_62 GBV_ILN_63 GBV_ILN_65 GBV_ILN_69 GBV_ILN_73 GBV_ILN_74 GBV_ILN_95 GBV_ILN_110 GBV_ILN_151 GBV_ILN_161 GBV_ILN_170 GBV_ILN_206 GBV_ILN_213 GBV_ILN_230 GBV_ILN_285 GBV_ILN_293 GBV_ILN_602 GBV_ILN_2003 GBV_ILN_2005 GBV_ILN_2009 GBV_ILN_2011 GBV_ILN_2014 GBV_ILN_2055 GBV_ILN_2111 GBV_ILN_4012 GBV_ILN_4037 GBV_ILN_4112 GBV_ILN_4125 GBV_ILN_4126 GBV_ILN_4249 GBV_ILN_4305 GBV_ILN_4306 GBV_ILN_4322 GBV_ILN_4323 GBV_ILN_4324 GBV_ILN_4325 GBV_ILN_4367 GBV_ILN_4700
container_issue	1
title_short	Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app
url	https://doi.org/10.1186/s12960-021-00699-5 https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289 https://doaj.org/toc/1478-4491
remote_bool	true
author2	Ken Kahn MacKenzie MacRae Allan Saul Namanda Rebecca Hamala Ken Kabali Anne Geniets Alice Lakati Simon M. Mbae Niall Winters
author2Str	Ken Kahn MacKenzie MacRae Allan Saul Namanda Rebecca Hamala Ken Kabali Anne Geniets Alice Lakati Simon M. Mbae Niall Winters
ppnlink	373756585
callnumber-subject	R - General Medicine
mediatype_str_mv	c
isOA_txt	true
hochschulschrift_bool	false
doi_str	10.1186/s12960-021-00699-5
callnumber-a	R5-920
up_date	2024-07-03T23:08:35.513Z
_version_	1803601172779499520
fullrecord_marcxml	<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">DOAJ046901515</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230308114851.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">230227s2022 xx \|\|\|\|\|o 00\| \|\|eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1186/s12960-021-00699-5</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)DOAJ046901515</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)DOAJb1cdcf9372674a1fb255ae91a27d4289</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="050" ind1=" " ind2="0"><subfield code="a">R5-920</subfield></datafield><datafield tag="050" ind1=" " ind2="0"><subfield code="a">RA1-1270</subfield></datafield><datafield tag="100" ind1="0" ind2=" "><subfield code="a">James O’Donovan</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">Computermedien</subfield><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Online-Ressource</subfield><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Background Despite the growth in mobile technologies (mHealth) to support Community Health Worker (CHW) supervision, the nature of mHealth-facilitated supervision remains underexplored. One strategy to support supervision at scale could be artificial intelligence (AI) modalities, including machine learning. We developed an open access, machine learning web application (CHWsupervisor) to predictively code instant messages exchanged between CHWs based on supervisory interaction codes. We document the development and validation of the web app and report its predictive accuracy. Methods CHWsupervisor was developed using 2187 instant messages exchanged between CHWs and their supervisors in Uganda. The app was then validated on 1242 instant messages from a separate digital CHW supervisory network in Kenya. All messages from the training and validation data sets were manually coded by two independent human coders. The predictive performance of CHWsupervisor was determined by comparing the primary supervisory codes assigned by the web app, against those assigned by the human coders and calculating observed percentage agreement and Cohen’s kappa coefficients. Results Human inter-coder reliability for the primary supervisory category of messages across the training and validation datasets was ‘substantial’ to ‘almost perfect’, as suggested by observed percentage agreements of 88–95% and Cohen’s kappa values of 0.7–0.91. In comparison to the human coders, the predictive accuracy of the CHWsupervisor web app was ‘moderate’, suggested by observed percentage agreements of 73–78% and Cohen’s kappa values of 0.51–0.56. Conclusions Augmenting human coding is challenging because of the complexity of supervisory exchanges, which often require nuanced interpretation. A realistic understanding of the potential of machine learning approaches should be kept in mind by practitioners, as although they hold promise, supportive supervision still requires a level of human expertise. Scaling-up digital CHW supervision may therefore prove challenging. Trial registration: This was not a clinical trial and was therefore not registered as such.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Machine learning</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Artificial intelligence</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Supervision</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Community Health Worker</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Digital Health</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Training</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Medicine (General)</subfield></datafield><datafield tag="653" ind1=" " ind2="0"><subfield code="a">Public aspects of medicine</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Ken Kahn</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">MacKenzie MacRae</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Allan Saul Namanda</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Rebecca Hamala</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Ken Kabali</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Anne Geniets</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Alice Lakati</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Simon M. Mbae</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="0" ind2=" "><subfield code="a">Niall Winters</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">In</subfield><subfield code="t">Human Resources for Health</subfield><subfield code="d">BMC, 2003</subfield><subfield code="g">20(2022), 1, Seite 8</subfield><subfield code="w">(DE-627)373756585</subfield><subfield code="w">(DE-600)2126923-3</subfield><subfield code="x">14784491</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:20</subfield><subfield code="g">year:2022</subfield><subfield code="g">number:1</subfield><subfield code="g">pages:8</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1186/s12960-021-00699-5</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doaj.org/article/b1cdcf9372674a1fb255ae91a27d4289</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1186/s12960-021-00699-5</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="856" ind1="4" ind2="2"><subfield code="u">https://doaj.org/toc/1478-4491</subfield><subfield code="y">Journal toc</subfield><subfield code="z">kostenfrei</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_DOAJ</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_11</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_20</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_22</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_23</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_24</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_31</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_39</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_40</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_60</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_62</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_63</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_65</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_69</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_73</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_74</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_95</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_110</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_151</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_161</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_170</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_206</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_213</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_230</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_285</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_293</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_602</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2003</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2005</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2009</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2011</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2014</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2055</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_2111</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4012</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4037</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4112</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4125</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4126</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4249</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4305</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4306</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4322</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4323</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4324</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4325</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4367</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ILN_4700</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">20</subfield><subfield code="j">2022</subfield><subfield code="e">1</subfield><subfield code="h">8</subfield></datafield></record></collection>
score	7.399455

Nicht das Richtige dabei?

Schreiben Sie uns!

Analysing 3429 digital supervisory interactions between Community Health Workers in Uganda and Kenya: the development, testing and validation of an open access predictive machine learning web app

Nicht das Richtige dabei?

Zugang & Verfügbarkeit

Vorhandene Bände

Nicht das Richtige dabei?