Fulfilling the Promise of Artificial Intelligence in the Health Sector: Let's Get Real.

OBJECTIVES: This study aimed to showcase the potential and key concerns and risks of artificial intelligence (AI) in the health sector, illustrating its application with current examples, and to provide policy guidance for the development, assessment, and adoption of AI technologies to advance policy objectives. METHODS: Nonsystematic scan and analysis of peer-reviewed and gray literature on AI in the health sector, focusing on key insights for policy and governance. RESULTS: The application of AI in the health sector is currently in the early stages. Most applications have not been scaled beyond the research setting. The use in real-world clinical settings is especially nascent, with more evidence in public health, biomedical research, and "back office" administration. Deploying AI in the health sector carries risks and hazards that must be managed proactively by policy makers. For AI to produce positive health and policy outcomes, 5 key areas for policy are proposed, including health data governance, operationalizing AI principles, flexible regulation, skills among health workers and patients, and strategic public investment. CONCLUSIONS: AI is not a panacea, but a tool to address specific problems. Its successful development and adoption require data governance that ensures high-quality data are available and secure; relevant actors can access technical infrastructure and resources; regulatory frameworks promote trustworthy AI products; and health workers and patients have the information and skills to use AI products and services safely, effectively, and efficiently. All of this requires considerable investment and international collaboration.

District division administrative disaggregation data framework for monitoring leaving no one behind in the National Health Insurance Fund of Sudan: achieving sustainable development goals in 2030.

BACKGROUND: The aim of this study is to monitor the concept of 'leaving no one behind' in the Sustainable Development Goals (SDGs) to track the implications of the mobilization of health care resources by the National Health Insurance Fund (NHIF) of Sudan. METHODS: A cross-sectional study was used to monitor 'leaving no one behind' in NHIF by analyzing the secondary data of the information system for the year 2016. The study categorized the catchment areas of health care centers (HCCS) according to district administrative divisions, which are neighborhood, subdistrict, district, and zero. The District Division Administrative Disaggregation Data (DDADD) framework was developed and investigated with the use of descriptive statistics, maps of Sudan, the Mann-Whitney test, the Kruskal-Wallis test and health equity catchment indicators. SPSS ver. 18 and EndNote X8 were also used. RESULTS: The findings show that the NHIF has mobilized HCCs according to coverage of the insured population. This mobilization protected the insured poor in high-coverage insured population districts and left those living in very low-coverage districts behind. The Mann-Whitney test presented a significant median difference in the utilization rate between catchment areas (P value < 0.001). The results showed that the utilization rate of the insured poor who accessed health care centers by neighborhood was higher than that of the insured poor who accessed by more than neighborhood in each state. The Kruskal-Wallis test of the cost of health care services per capita in each catchment area showed a difference (P value < 0.001) in the median between neighborhoods. The cost of health care services in low-coverage insured population districts was higher than that in high-coverage insured population districts. CONCLUSION: The DDADD framework identified the inequitable distribution of health care services in low-density population districts leaves insured poor behind. Policymakers should restructure the equation of health insurance schemes based on equity and probability of illness, to distribute health care services according to needs and equity, and to remobilize resources towards districts left behind.

Physical rehabilitation in the context of a landslide that occurred in Brazil.

BACKGROUNDS: The efforts to develop research and training on physical rehabilitation in regards to disasters is considered recent worldwide. In the late evening of the 11th up until the 12th of January of 2011, the most massive natural disaster occurred in Brazil with extremely heavy downpour, abrupt flood, as well as landslides on multiple areas of the Mountain Region of Rio de Janeiro. The objective of this research was to investigate the challenges in terms of physical rehabilitation provided by this event METHODS: The cross-sectional mixed method's study, which was conducted in the city of Nova Friburgo, used two different data sources: hospital records on traumatic injuries pre and post disaster, and interviews with key informants - victims who suffered injuries related to the disaster, professionals from rehabilitation services in the municipality, and also the city's health service management. Pearson's chi-squared test was performed to evaluate statistical significance between the week of a given incident and the type of injury. Interviews were transcribed and analysed through content analysis. RESULTS: A total of 2326 hospital records and 27 interviews were analysed. The proportion of traumatic injury in the municipal emergency service increased from 16% in the prior week, to 40% in the week post-disaster (p <  0.0001). Different injuries were identified: multiple fractures, crushing, amputation, perforation of soft tissues, inhalation of dust and establishment of chronic conditions through stress. Despite this scenario, out of the 16 health professionals interviewed, twelve did not observe an increase in the demand for outpatient rehabilitation services after the disaster. Interviews with the victims revealed that the pathways for care ran into different barriers. From 11 victims interviewed, only one received complete physiotherapy care through the public health services in the city, while all others hired additional assistance, received volunteer services, had assistance in other cities or remained without rehabilitation. CONCLUSIONS: The needs for rehabilitation increased after the disaster; however, the demand was repressed due to different barriers such as competing needs and possible lack of medical referral. Recommendations were made, including the action of performing a search of victims with rehabilitation needs.

Examination of the relationship between management and clinician perception of patient safety climate and patient satisfaction.

PURPOSE: The aim of this study was to explore the relationship between managers and clinicians' agreement on deeming the patient safety climate as high or low and the patients' satisfaction with those organizations. DATA SOURCES/STUDY SETTING: We used two secondary data sets: the Hospital Survey on Patient Safety Culture (2012) and the Hospital Consumer Assessment of Healthcare Providers and Systems (2012). METHODOLOGY/APPROACH: We used ordinary least squares regressions to analyze the relationship between the extent of agreement between managers and clinicians' perceptions of safety climate in relationship to patient satisfaction. The dependent variables were four Hospital Consumer Assessment of Healthcare Providers and Systems patient satisfaction scores: communication with nurses, communication with doctors, communication about medicines, and discharge information. The main independent variables were four groups that were formed based on the extent of managers and clinicians' agreement on four patient safety climate domains: communication openness, feedback and communication about errors, teamwork within units, and teamwork across units. FINDINGS: After controlling for hospital and market-level characteristics, we found that patient satisfaction was significantly higher if managers and clinicians reported that patient safety climate is high or if only clinicians perceived the climate as high. Specifically, manager and clinician agreement on high levels of communication openness (ß = 2.25, p = .01; ß = 2.46, p = .05), feedback and communication about errors (ß = 3.0, p = .001; ß = 2.89, p = .01), and teamwork across units (ß = 2.91, p = .001; ß = 3.34, p = .01) was positively and significantly associated with patient satisfaction with discharge information and communication about medication. In addition, more favorable perceptions about patient safety climate by clinicians only yielded similar findings. PRACTICE IMPLICATIONS: Organizations should measure and examine patient safety climate from multiple perspectives and be aware that individuals may have varying opinions about safety climate. Hospitals should encourage multidisciplinary collaboration given that staff perceptions about patient safety climate may be associated with patient satisfaction.

Informing Targeted Interventions to Optimize the Cascade of HIV Care Using Cluster Analyses of Health Resource Use Among People Living with HIV/AIDS.

Increased awareness of the secondary preventive benefits of antiretroviral treatment (ART) has strengthened the desire to optimize health care systems' response to HIV/AIDS. We identified clusters of health resource utilization (HRU) among people living with HIV (PLHIV) to inform targeted interventions aimed to optimize the cascade of HIV care. Using linked population-level health databases in British Columbia, Canada, we selected two analytic samples of PLHIV with 3 years of follow-up between 2006-2011 that were classified as intermittently retained in care or intermittently engaged in ART, and executed a probabilistic model-based clustering analysis for each sample with 5 and 9 quarterly HRU variables, respectively. We found clear HRU profile differences among both samples with similar HIV-related care: one featured active involvement in non-HIV care, the other little or no health care interaction following linkage to care. Differential reengagement intervention strategies capitalizing on missed opportunities in non-HIV care and further engaging physicians delivering HIV care are needed to optimize the response to the HIV epidemic.

What are seasonal and meteorological factors are associated with the number of attendees at a sexual health service? An observational study between 2002-2012.

BACKGROUND: Open access to sexual health services may be inefficient if there are substantial unpredictable fluctuations in presentations. Our aim was to determine whether the number of presentations over the last 11 years was associated with certain factors. METHODS: This study involved all individuals presenting to Melbourne Sexual Health Centre (MSHC) from 2002 to 2012. The outcome measure was the number of presentations during a clinical session (half day). RESULTS: There were 270,070 presentations to the clinic among 86,717 individuals. The factors associated with the largest difference in mean presentations per session were morning or afternoon (60 vs 51 per session), days of the week (57-67 per session), months of the year (93-112 per day), year (77-131 per day), maximum temperatures of <15 °C vs. ≥30 °C (56-62 per morning session) and 5 working days after holiday periods (61 vs 54). A multiple linear regression model using these factors explained 64% of the variation in attendances per session. Peak attendance rates (>90th centile) were also strongly correlated with these same variables. Higher-risk heterosexuals (≤25 years of age) attended more commonly in the afternoons (37% of heterosexuals) than in the mornings (30%). No factor other than year of attendance substantially influenced the proportion of higher-risk men who have sex with men (MSM) (≥10 partners per year) who attended. CONCLUSIONS: A considerable proportion of the variability in presentations was explained by known factors that could predict client presentations to sexual health services and therefore allow optimal allocation of resources to match demand.

Retrospective studies of end-of-life resource utilization and costs in cancer care using health administrative data: a systematic review.

BACKGROUND: There has been an increase in observational studies using health administrative data to examine the nature, quality, and costs of care at life's end, particularly in cancer care. AIM: To synthesize retrospective observational studies on resource utilization and/or costs at the end of life in cancer patients. We also examine the methods and outcomes of studies assessing the quality of end-of-life care. DESIGN: A systematic review according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and AMSTAR (A Measurement Tool to Assess Systematic Reviews) methodology. DATA SOURCES: We searched MEDLINE, Embase, CINAHL, and York Centre for Research and Dissemination (1990-2011). Independent reviewers screened abstracts of 14,424 articles, and 835 full-text manuscripts were further reviewed. Inclusion criteria were English-language; at least one resource utilization or cost outcome in adult cancer decedents with solid tumors; outcomes derived from health administrative data; and an exclusive end-of-life focus. RESULTS: We reviewed 78 studies examining end-of-life care in over 3.7 million cancer decedents; 33 were published since 2008. We observed exponential increases in service use and costs as death approached; hospital services being the main cost driver. Palliative services were relatively underutilized and associated with lower expenditures than hospital-based care. The 15 studies using quality indicators demonstrated that up to 38% of patients receive chemotherapy or life-sustaining treatments in the last month of life and up to 66% do not receive hospice/palliative services. CONCLUSION: Observational studies using health administrative data have the potential to drive evidence-based palliative care practice and policy. Further development of quality care markers will enhance benchmarking activities across health care jurisdictions, providers, and patient populations.

Healthcare service problems reported in a national survey of South Africans.

OBJECTIVE: To identify common types of health service problems reported by South African adults during their most recent visit to a healthcare provider. DESIGN: Secondary analysis of South Africa's cross-sectional General Household Survey (GHS). SETTING: Nationally representative weighted sample of households in South Africa. PARTICIPANTS: 23,562 household representatives interviewed during the 2010 GHS. MAIN OUTCOME MEASURE: Problems experienced during the most recent visit to the usual healthcare provider. RESULTS: In total, 43.8% of participants reported experiencing at least one problem during their last visit; 19.1% reported multiple problems. The most common problems experienced were a long waiting time (34.8% of household representatives), needed drugs not being available (14.1%) and staff who were rude or uncaring or turned patients away (10.1%). Of the 73.6% of participants using public providers, 54.9% reported at least one problem; of the 26.4% of participants using private providers, only 18.0% reported a problem, usually cost. Similar differences in reported problems at public and private providers were reported for all racial/ethnic groups and income groups. Black Africans reported more problems than other population groups due in large part to being significantly more likely to use public providers. CONCLUSIONS: Addressing commonly reported problem areas-in particular, long waiting times, unavailable medications and staff who are perceived as being unfriendly-might help prevent delayed care seeking, increase the acceptability of healthcare services and reduce remaining health disparities in South Africa.

[Citation analysis of research articles from Norwegian health enterprises, 2005-2011]. / Siteringsanalyse av vitenskapelige artikler fra norske helseforetak 2005-11.

BACKGROUND: The citation frequency of a publication is often interpreted as an expression of its scientific impact. Previous citation analyses of Norwegian medical research have either focused on universities and university hospitals or on subject areas at the national level. Such analyses have paid little attention to other health enterprises, despite a strong increase in their research activity during the last decade. MATERIAL AND METHOD: For all health enterprises with more than 25 publications in the Web of Science during the period 2005-2011 we have calculated field normalized citation indexes at the institution and subject levels. RESULTS: On the whole, research undertaken by the health enterprises is frequently cited, and some medium-sized health enterprises stand out in terms of their high average citation indexes: Helse Stavanger Health Enterprise, Diakonhjemmet Hospital and Helse Nord-Trøndelag Health Enterprise, although Oslo University Hospital Health Enterprise and Helse Bergen Health Enterprise account for more than half of the most cited articles. INTERPRETATION: In citation analyses at the aggregated level, highly and infrequently cited research groups/departments may balance each other. This appears to be the case in the largest health enterprises. Some medium-sized health enterprises that have a more concentrated research portfolio will thus accumulate higher average citation indexes than the largest university hospitals.

[Eye care management in Russian Federation].

The article presents an analysis of Russian eye care performance indicators based on federal and sector statistics over the recent years provided by the Ministry of Health of the Russian Federation, that is the incidence of eye diseases, eye care equipment provision, inpatient and outpatient volumes. Legal acts of the Russian Federation on health system in general and eye care in particular were taken into consideration when preparing the section on organizational matters. Problems of human resources, efficiency of specialists' time management, hospital beds use, and administrative issues in particular regions and Russia as a whole are discussed.

Assessing the role of prevention partnerships in STD prevention: a review of comprehensive STD prevention systems progress reports.

Systematic analysis of STD programme data contributes to a national portrait of sexually transmitted disease (STD) prevention activities, including research and evaluation specifically designed to optimise programme efficiency and impact. We analysed the narrative of the 2009 annual progress reports of the US Comprehensive STD Prevention Systems cooperative agreement for 58 STD programmes, concentrating on programme characteristics and partnerships. Programmes described 516 unique partnerships with a median of seven organisations cited per STD programme. Non-profit organisations (including service providers) were most frequently cited. Higher gonorrhoea morbidity was associated with reporting more partnerships; budget problems were associated with reporting fewer. Challenges to engaging in partnerships included budget constraints, staff turnover and low interest. Data provide a source of information for judging progress in programme collaboration and for informing a sustained programme-focused research and evaluation agenda.

Embedding health policy and systems research into decision-making processes in low- and middle-income countries.

Attention is increasingly directed to bridging the gap between the production of knowledge and its use for health decision-making in low- and middle-income countries (LMICs). An important and underdeveloped area of health policy and systems research (HPSR) is the organization of this process. Drawing from an interdisciplinary conception of embeddedness, a literature review was conducted to identify examples of embedded HPSR used to inform decision-making in LMICs. The results of the literature review were organized according to the World Health Organization's Building Blocks Framework. Next, a conceptual model was created to illustrate the arrangement of organizations that produce embedded HPSR and the characteristics that facilitate its uptake into the arena of decision-making. We found that multiple forces converge to create context-specific pathways through which evidence enters into decision-making. Depending on the decision under consideration, the literature indicates that decision-makers may call upon an intricate combination of actors for sourcing HPSR. While proximity to decision-making does have advantages, it is not the position of the organization within the network, but rather the qualities the organization possesses, that enable it to be embedded. Our findings suggest that four qualities influence embeddedness: reputation, capacity, quality of connections to decision-makers, and quantity of connections to decision-makers and others. In addition to this, the policy environment (e.g. the presence of legislation governing the use of HPSR, presence of strong civil society, etc.) strongly influences uptake. Through this conceptual model, we can understand which conditions are likely to enhance uptake of HPSR in LMIC health systems. This raises several important considerations for decision-makers and researchers about the arrangement and interaction of evidence-generating organizations in health systems.

Estimating the burden of mucormycosis infections in France (2005-2007) through a capture-recapture method on laboratory and administrative data.

BACKGROUND: Mucormycoses are rare but severe fungal infections whose incidence is increasing, particularly in immunosuppressed and diabetic patients. Following a retrospective study on the characteristics and outcomes of cases who were identified through two sources of information, we carried out a capture-recapture method to estimate the actual burden of the disease in France, 2005-2007. METHODS: An administrative dataset from the national hospital discharge system and a laboratory dataset from the National Reference Centre for Mycoses and Antifungals were combined to identify patients from 2005 to 2007. We applied capture-recapture equations to estimate the number of cases missed by both sources and to assess the advantages of each dataset, especially in terms of sensitivity. RESULTS: There were 94 mucormycosis cases included in the study: 30 and 31 in each respective source and 33 common to both. Capture-recapture showed that 28 cases were missed (expected total: 122 cases, CI95: 102-142). Each dataset had a sensitivity value below 53%. The merged set yielded a 77% sensitivity (66%-92%). CONCLUSION: This study highlights the importance of combining available sources when analysing rare infectious diseases. The proportion of 23% missed cases might seem acceptable given the scarcity of the disease, for which further knowledge is needed. However this proportion could decrease in the future, through the sensitization of clinicians, pathologists and mycologists together with the improving quality of discharge datasets.

Adherence to therapeutic and diagnostic recommendations in patients with femur fracture and at risk of re-fracture or death: results of an analysis of administrative databases.

The aim of the present study was to evaluate the application into clinical practice of therapeutic and diagnostic recommendations for the prevention of bone re-fracture in postmenopausal women after an hospitalization for hip fracture in clinical practice and to assess the relationship between the application of diagnostic recommendations and re-fracture or death risk. A retrospective cohort analysis was conducted. All female patients, at least 65 years old, and with an hospitalization with main or secondary diagnosis of hip fracture during the period 1 January 2006 - 31 December 2008, were included. Besides demographic characteristics and comorbidities, drug treatment prescriptions related to bone fracture or supplementary with calcium or vitamin D and prescriptions of recommended laboratory and instrumental diagnostic tests (e.g. spine radiography), were analysed. A total of 5,636 patients were included in the study. The prescription of a drug treatment aimed to reduce the risk of re-fracture was found in 16.3% of patients, among them 76.3% (699 patients) used bisphosphonates only, 17.1% (157 patients) strontium ranelate only and 4.9% (45 patients) used more than one treatment during the observation period. Among the patients who did not receive drug treatment, 17.5% made use of only supplemental calcium and vitamin D. The remaining part of patients (69.1%) received no treatment. The prescription of at least one laboratory test of first and second level was performed, respectively, on 53.7% and 43.1% of included patients, whereas the prescription of at least one instrumental test of first and second level was performed, respectively, on 5.9% and 0.8%. Although it is established that the prescription of the recommended tests and appropriate drug treatment are significantly associated with reduced risk of re-fracture and death, today the application of these recommendations is reduced.

Why health care process performance measures can have different relationships to outcomes for patients and hospitals: understanding the ecological fallacy.

Relationships between health care process performance measures (PPMs) and outcomes can differ in magnitude and even direction for patients versus higher level units (e.g., health care facilities). Such discrepancies can arise because facility-level relationships ignore PPM-outcome relationships for patients within facilities, may have different confounders than patient-level PPM-outcome relationships, and may reflect facility effect modification of patient PPM-outcome relationships. If a patient-level PPM is related to better patient outcomes, that care process should be encouraged. However, the finding in a multilevel analysis that the proportion of patients receiving PPM care across facilities nevertheless is linked to poor hospital outcomes would suggest that interventions targeting the health care facility also are needed.

[The health system of Bolivia]. / Sistema de salud de Bolivia.

This paper describes the Bolivian health system, including its structure and organization, its financing sources, its health expenditure, its physical, material and humans resources, its stewardship activities and the its health research institutions. It also discusses the most recent policy innovations developed in Bolivia: the Maternal and Child Universal Insurance, the Program for the Extension of Coverage to Rural Areas, the Family, Community and Inter-Cultural Health Model and the cash-transfer program Juana Azurduy intended to strengthen maternal and child care.

[The health system of Brazil]. / Sistema de salud de Brasil.

This paper describes the Brazilian health system, which includes a public sector covering almost 75% of the population and an expanding private sector offering health services to the rest of the population. The public sector is organized around the Sistema Único de Saúde (SUS) and it is financed with general taxes and social contributions collected by the three levels of government (federal, state and municipal). SUS provides health care through a decentralized network of clinics, hospitals and other establishments, as well as through contracts with private providers. SUS is also responsible for the coordination of the public sector. The private sector includes a system of insurance schemes known as Supplementary Health which is financed by employers and/or households: group medicine (companies and households), medical cooperatives, the so called Self-Administered Plans (companies) and individual insurance plans.The private sector also includes clinics, hospitals and laboratories offering services on out-of-pocket basis mostly used by the high-income population. This paper also describes the resources of the system, the stewardship activities developed by the Ministry of Health and other actors, and the most recent policy innovations implemented in Brazil, including the programs saúde da Familia and Mais Saúde.

[The health system of Colombia]. / Sistema de salud de Colombia.

This document briefly describes the health conditions of the Colombian population and, in more detail, the characteristics of the Colombian health system. The description of the system includes its structure and coverage; financing sources; expenditure in health; physical material and human resources available; monitoring and evaluation procedures; and mechanisms through which the population participates in the evaluation of the system. Salient among the most recent innovations implemented in the Colombian health system are the modification of the Compulsory Health Plan and the capitation payment unit, the vertical integration of the health promotion enterprises and the institutions in charge of the provision of services and the mobilization of additional resources to meet the objectives of universal coverage and the homologation of health benefits among health regimes.

[The health system of Chile]. / Sistema de salud de Chile.

This paper describes the Chilean health system, including its structure, financing, beneficiaries, and its physical, material and human resources. This system has two sectors, public and private. The public sector comprises all the organisms that constitute the National System of Health Services, which covers 70% of the population, including the rural and urban poor, the low middle-class, the retired, and the self-employed professionals and technicians.The private sector covers 17.5% of the population, mostly the upper middle-class and the high-income population. A small proportion of the population uses private health services and pays for them out-of-pocket. Around l0% of the population is covered by other public agencies, basically the Health Services for the Armed Forces. The system was recently reformed with the establishment of a Universal System of Explicit Entitlements, which operates through a Universal Plan of Explicit Entitlements (AUGE), which guarantees timely access to treatment for 56 health problems, including cancer in children, breast cancer, ischaemic heart disease, HIV/AIDS and diabetes.

[The health system of Costa Rica]. / Sistema de salud de Costa Rica.

This paper describes the Costa Rican health system which provides health, water and sanitation services. The health component of the system includes a public and a private sector. The public sector is dominated by the Caja Costarricense de Seguro Social (CCSS), an autonomous institution in charge of financing, purchasing and delivering most of the personal health services in Costa Rica. CCSS is financed with contributions of the affiliates, employers and the state, and manages three regimes: maternity and illness insurance, disability, old age and death insurance, and a non-contributive regime. CCSS provides services in its own facilities but also contracts with private providers. The private sector includes a broad set of services offering ambulatory and hospital care. These services are financed mostly out-of-pocket, but also with private insurance premiums. The Ministry of Health is the steward of the system, in charge of strategic planning, sanitary regulation, and research and technology development. Among the recent policy innovations we can mention the establishment of the basic teams for comprehensive health care (EBAIS), the de-concentration of hospitals and public clinics, the introduction of management agreements and the creation of the Health Boards.