Background: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice.... Show moreBackground: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice. We implemented a CDSS to aid general practitioners (GPs) in treating patients with urinary tract infections (UTIs), which are a significant health burden worldwide. Objective: This study aims to prospectively assess the impact of this CDSS on treatment success and change in antibiotic prescription behavior of the physician. In doing so, we hope to identify drivers and obstacles that positively impact the quality of health care practice with ML. Methods: The CDSS was developed by Pacmed, Nivel, and Leiden University Medical Center (LUMC). The CDSS presents the expected outcomes of treatments, using interpretable decision trees as ML classifiers. Treatment success was defined as a subsequent period of 28 days during which no new antibiotic treatment for UTI was needed. In this prospective observational study, 36 primary care practices used the software for 4 months. Furthermore, 29 control practices were identified using propensity score-matching. All analyses were performed using electronic health records from the Nivel Primary Care Database. Patients for whom the software was used were identified in the Nivel database by sequential matching using CDSS use data. We compared the proportion of successful treatments before and during the study within the treatment arm. The same analysis was performed for the control practices and the patient subgroup the software was definitely used for. All analyses, including that of physicians' prescription behavior, were statistically tested using 2-sided z tests with an alpha level of .05. Results: In the treatment practices, 4998 observations were included before and 3422 observations (of 2423 unique patients) were included during the implementation period. In the control practices, 5044 observations were included before and 3360 observations were included during the implementation period. The proportion of successful treatments increased significantly from 75% to 80% in treatment practices (z=5.47, P<.001). No significant difference was detected in control practices (76% before and 76% during the pilot, z=0.02; P=.98). Of the 2423 patients, we identified 734 (30.29%) in the CDSS use database in the Nivel database. For these patients, the proportion of successful treatments during the study was 83%-a statistically significant difference, with 75% of successful treatments before the study in the treatment practices (z=4.95; P<.001). Conclusions: The introduction of the CDSS as an intervention in the 36 treatment practices was associated with a statistically significant improvement in treatment success. We excluded temporal effects and validated the results with the subgroup analysis in patients for whom we were certain that the software was used. This study shows important strengths and points of attention for the development and implementation of an ML-based CDSS in clinical practice. Trial Registration: ClinicalTrials.gov NCT04408976; https://clinicaltrials.gov/ct2/show/NCT04408976 Show less
Background: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice.... Show moreBackground: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice. We implemented a CDSS to aid general practitioners (GPs) in treating patients with urinary tract infections (UTIs), which are a significant health burden worldwide.Objective: This study aims to prospectively assess the impact of this CDSS on treatment success and change in antibiotic prescription behavior of the physician. In doing so, we hope to identify drivers and obstacles that positively impact the quality of health care practice with ML.Methods: The CDSS was developed by Pacmed, Nivel, and Leiden University Medical Center (LUMC). The CDSS presents the expected outcomes of treatments, using interpretable decision trees as ML classifiers. Treatment success was defined as a subsequent period of 28 days during which no new antibiotic treatment for UTI was needed. In this prospective observational study, 36 primary care practices used the software for 4 months. Furthermore, 29 control practices were identified using propensity score-matching. All analyses were performed using electronic health records from the Nivel Primary Care Database. Patients for whom the software was used were identified in the Nivel database by sequential matching using CDSS use data. We compared the proportion of successful treatments before and during the study within the treatment arm. The same analysis was performed for the control practices and the patient subgroup the software was definitely used for. All analyses, including that of physicians’ prescription behavior, were statistically tested using 2-sided z tests with an α level of .05.Results: In the treatment practices, 4998 observations were included before and 3422 observations (of 2423 unique patients) were included during the implementation period. In the control practices, 5044 observations were included before and 3360 observations were included during the implementation period. The proportion of successful treatments increased significantly from 75% to 80% in treatment practices (z=5.47, P<.001). No significant difference was detected in control practices (76% before and 76% during the pilot, z=0.02; P=.98). Of the 2423 patients, we identified 734 (30.29%) in the CDSS use database in the Nivel database. For these patients, the proportion of successful treatments during the study was 83%—a statistically significant difference, with 75% of successful treatments before the study in the treatment practices (z=4.95; P<.001).Conclusions: The introduction of the CDSS as an intervention in the 36 treatment practices was associated with a statistically significant improvement in treatment success. We excluded temporal effects and validated the results with the subgroup analysis in patients for whom we were certain that the software was used. This study shows important strengths and points of attention for the development and implementation of an ML-based CDSS in clinical practice. Show less
Background: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice.... Show moreBackground: There is increasing attention on machine learning (ML)-based clinical decision support systems (CDSS), but their added value and pitfalls are very rarely evaluated in clinical practice. We implemented a CDSS to aid general practitioners (GPs) in treating patients with urinary tract infections (UTIs), which are a significant health burden worldwide.Objective: This study aims to prospectively assess the impact of this CDSS on treatment success and change in antibiotic prescription behavior of the physician. In doing so, we hope to identify drivers and obstacles that positively impact the quality of health care practice with ML.Methods: The CDSS was developed by Pacmed, Nivel, and Leiden University Medical Center (LUMC). The CDSS presents the expected outcomes of treatments, using interpretable decision trees as ML classifiers. Treatment success was defined as a subsequent period of 28 days during which no new antibiotic treatment for UTI was needed. In this prospective observational study, 36 primary care practices used the software for 4 months. Furthermore, 29 control practices were identified using propensity score-matching. All analyses were performed using electronic health records from the Nivel Primary Care Database. Patients for whom the software was used were identified in the Nivel database by sequential matching using CDSS use data. We compared the proportion of successful treatments before and during the study within the treatment arm. The same analysis was performed for the control practices and the patient subgroup the software was definitely used for. All analyses, including that of physicians’ prescription behavior, were statistically tested using 2-sided z tests with an α level of .05.Results: In the treatment practices, 4998 observations were included before and 3422 observations (of 2423 unique patients) were included during the implementation period. In the control practices, 5044 observations were included before and 3360 observations were included during the implementation period. The proportion of successful treatments increased significantly from 75% to 80% in treatment practices (z=5.47, P<.001). No significant difference was detected in control practices (76% before and 76% during the pilot, z=0.02; P=.98). Of the 2423 patients, we identified 734 (30.29%) in the CDSS use database in the Nivel database. For these patients, the proportion of successful treatments during the study was 83%—a statistically significant difference, with 75% of successful treatments before the study in the treatment practices (z=4.95; P<.001).Conclusions: The introduction of the CDSS as an intervention in the 36 treatment practices was associated with a statistically significant improvement in treatment success. We excluded temporal effects and validated the results with the subgroup analysis in patients for whom we were certain that the software was used. This study shows important strengths and points of attention for the development and implementation of an ML-based CDSS in clinical practice. Show less
Internet connectivity is widely considered to be a game changer for knowledge economies of developing countries. The arrival of submarine fibre-optic underwater cables in East Africa in 2009 and... Show moreInternet connectivity is widely considered to be a game changer for knowledge economies of developing countries. The arrival of submarine fibre-optic underwater cables in East Africa in 2009 and 2010 is seen by many as a strong case in point. The fast evolution of the information and communication technology (ICT) landscape of Kenya and Rwanda that ensued has attracted the attention of actors from private investors, development agencies, NGOs, policymakers and many other groups. Kenya became a role model for its widespread adoption of mobile money services and a burgeoning ICT application development sector; Rwanda's government became known for its explicitly ICT-oriented development agenda as well as large-scale ICT projects in government, health and education that aimed to latch onto fast-growing mobile subscription rates and 3G network roll-outs. For this report, we set out to examine the role that changing connectivity has played for a particular component of the ICT sector in Kenya and Rwanda: ICT-enabled business process outsourcing (BPO).1 BPO has been a priority in the national ICT strategies of both countries, so we anticipated this sector to provide a fertile ground for comparing expectations and realities of the role that changing connectivity has played following the deployment of fibre-optic cable infrastructure. The study outlined how policy, popular discourse and media got somewhat carried away by the promise of internet connectivity as the fuel for the growth of Kenya's and Rwanda's BPO sectors. The development of ICT sectors fell short of many original hopes. Internet connectivity proved to only function as a catalyst for economic growth in combination with other enablers, even for the examined sectors of connectivity-based enterprises. Competitive advantage is always relative, and, in the case of Kenya's and Rwanda's BPO sectors, India and other Asian BPO destinations have maintained the edge in international markets. Despite the overall positive evolution of ICT-based subsectors in Kenya and Rwanda, the role of internet connectivity for growth in knowledge economies continues to be a complicated one, including for connectivity-based enterprises. Future opportunities might actually lie in 'close' (local and regional) markets, and policymakers and indeed all economic actors will need to continue to learn and adjust to other unexpected developments brought about by internet connectivity. Show less
L'histoire récente de l'Afrique se caractérise par la 'révolution' des technologies de l'information et de la communication (TIC), plus spécifiquement de la téléphonie mobile. La présente thèse... Show moreL'histoire récente de l'Afrique se caractérise par la 'révolution' des technologies de l'information et de la communication (TIC), plus spécifiquement de la téléphonie mobile. La présente thèse explore la manifestation de la dynamique des TIC dans la société hadjeray de la région du Guéra, au Tchad, qui a connu violences politiques, mobilité et rupture au sein des familles et aujourd'hui 'retrouvailles' grâce à la téléphonie mobile et aux réseaux sociaux sur Internet. L'auteur, lui-même issu de cette société, a axé son travail sur le rôle de la communication pendant les crises, dans la mobilité et aujourd'hui dans la dynamique relationnelle née de la facilité qu'offrent les TIC. Il montre les appropriations réciproques entre les TIC et la population hadjeray, et notamment le rôle central des TIC dans la dynamique identitaire hadjeray. [Résumé ASC Leiden]. Show less
This is a paper about expectations surrounding a potentially highly transformative moment in East Africa's history: the arrival of underwater fibre-optic broadband communications cables into the... Show moreThis is a paper about expectations surrounding a potentially highly transformative moment in East Africa's history: the arrival of underwater fibre-optic broadband communications cables into the Indian Ocean port of Mombasa. It combines a media content analysis with findings from interviews with business owners in Kenya's nascent business process outsourcing (BPO) and software development sectors in order to explore how such moments of technological 'connectivity' are imagined, marketed and enacted within economic development. It argues that connectivity is not just a matter of boosting physical/material capacity but also about redressing conceptual connectivity; bringing places 'closer together' involves rehabilitating the images of places in peoples' minds and removing imagined senses of distance. As such, technologies of connectivity are marketed not just as tools of altered communications affordances, but more importantly, as momentary opportunities for revisiting the image of places from afar. Additionally, the cables reveal the importance of fostering internal linkages in order to better build international recognition and connections. 'Moments of expectation' that surround new ICT technologies reveal how discourse and representation play a strong role in enabling markets to form and change. The very idea of 'connectivity' itself is driving plans and policies throughout the region. Show less
This introductory chapter sketches globalization and Africa in broad theoretical terms, examining the meaning of the term globalization; the impact of globalization on daily life in Africa in... Show moreThis introductory chapter sketches globalization and Africa in broad theoretical terms, examining the meaning of the term globalization; the impact of globalization on daily life in Africa in economic as well as sociocultural terms; globalization as a historical phenomenon; the political aspects of globalization; its spatial dynamics: migration and transmigration, and the relationship of transnationalism with the emergence of new religious forms; and the contribution of anthropology to the field of globalization studies. Bibliogr., notes, ref. [ASC Leiden abstract] Show less
