Tinkering with symptoms, causes and solutions: Tracing the enactments of multiple chronic illnesses in specialised outpatient check-ups.

People living with multiple chronic illnesses and an increasing need for acute care is a global health challenge, which questions the conventional ways of managing illness. A central issue is how medical practices can become more patient-centred and aligned with the everyday life of patients. Communicative strategies for eliciting the patient's goals and preferences are often proposed. In this article, we draw on ethnographic data from fieldwork conducted during 2019-2020 in health-care settings and among people living with multiple chronic illness(es) and repeated acute admissions in Denmark. Inspired by science and technology studies of chronic illness and care, we trace the enactments of illness and illness work in a patient trajectory marked by persistent symptoms and medical complexity. We analyse three medical encounters, and we show how 'tinkering' with clinical signs and utterances in each encounter constantly enacts new versions, shaping how the patient could and should live with his illness. We argue that specialised outpatient check-ups for these patients must provide space for continuous tinkering with the concrete effects of illness in everyday life.

Using Explainable Artificial Intelligence to Predict Potentially Preventable Hospitalizations: A Population-Based Cohort Study in Denmark.

BACKGROUND: The increasing aging population and limited health care resources have placed new demands on the healthcare sector. Reducing the number of hospitalizations has become a political priority in many countries, and special focus has been directed at potentially preventable hospitalizations. OBJECTIVES: We aimed to develop an artificial intelligence (AI) prediction model for potentially preventable hospitalizations in the coming year, and to apply explainable AI to identify predictors of hospitalization and their interaction. METHODS: We used the Danish CROSS-TRACKS cohort and included citizens in 2016-2017. We predicted potentially preventable hospitalizations within the following year using the citizens' sociodemographic characteristics, clinical characteristics, and health care utilization as predictors. Extreme gradient boosting was used to predict potentially preventable hospitalizations with Shapley additive explanations values serving to explain the impact of each predictor. We reported the area under the receiver operating characteristic curve, the area under the precision-recall curve, and 95% confidence intervals (CI) based on five-fold cross-validation. RESULTS: The best performing prediction model showed an area under the receiver operating characteristic curve of 0.789 (CI: 0.782-0.795) and an area under the precision-recall curve of 0.232 (CI: 0.219-0.246). The predictors with the highest impact on the prediction model were age, prescription drugs for obstructive airway diseases, antibiotics, and use of municipality services. We found an interaction between age and use of municipality services, suggesting that citizens aged 75+ years receiving municipality services had a lower risk of potentially preventable hospitalization. CONCLUSION: AI is suitable for predicting potentially preventable hospitalizations. The municipality-based health services seem to have a preventive effect on potentially preventable hospitalizations.

Discharge readiness as an infrastructure: Negotiating the transfer of care for elderly patients in medical wards.

The discharge from hospital is an essential care transition for elderly people with chronic illness, specifically because the responsibility for treatment and care is transferred between locations and healthcare staff. To optimise the use of healthcare resources in a time of progressively shorter hospital admissions and increasingly streamlined hospital care, discharges are highlighted as important moments to be handled with caution. Yet, discharges are expected to be "early" and "quick" procedures to maintain a flow of patients through the hospital. In this qualitative article, we use ethnographic methods to investigate how this apparent contradiction unfolds in everyday discharge situations through the work of establishing discharge readiness in three medical wards in a middle-sized Danish hospital. We use the lens of infrastructure to help us see how elements like patient screens, screen meetings, clinical (and embodied) signs and community health care criteria are interrelated in the work of establishing discharge readiness of patients. By looking closely into specific discharge situations, we analyse the way care needs are defined and how care work is transferred, and we identify the inherent uncertainties for health care professionals, patients and relatives. We show how clinical signs take precedence over embodied experience, and how complex situations are reduced to workable problems to enable discharge.

Interventions to Prevent Potentially Avoidable Hospitalizations: A Mixed Methods Systematic Review.

Background: The demand for healthcare is increasing due to an aging population, more people living with chronic diseases and medical comorbidities. To manage this demand, political institutions call for action to reduce the potentially avoidable hospitalizations. Quantitative and qualitative aspects should be considered to understand how and why interventions work, and for whom. The aim of this mixed methods systematic review was to identify and synthesize evidence on interventions targeting avoidable hospitalizations from the perspectives of the citizens and the healthcare professionals to improve the preventive healthcare services. Methods and Results: A mixed methods systematic review was conducted following the JBI methodology using a convergent integrated approach to synthesis. The review protocol was registered in PROSPERO, reg. no. CRD42020134652. A systematic search was undertaken in six databases. In total, 45 articles matched the eligibility criteria, and 25 of these (five qualitative studies and 20 quantitative studies) were found to be of acceptable methodological quality. From the 25 articles, 99 meaning units were extracted. The combined evidence revealed four categories, which were synthesized into two integrated findings: (1) Addressing individual needs through care continuity and coordination prevent avoidable hospitalizations and (2) Recognizing preventive care as an integrated part of the healthcare work to prevent avoidable hospitalizations. Conclusions: The syntheses highlight the importance of addressing individual needs through continuous and coordinated care practices to prevent avoidable hospitalizations. Engaging healthcare professionals in preventive care work and considering implications for patient safety may be given higher priority. Healthcare administers and policy-makers could support the delivery of preventive care through targeted educational material aimed at healthcare professionals and simple web-based IT platforms for information-sharing across healthcare settings. The findings are an important resource in the development and implementation of interventions to prevent avoidable hospitalizations, and may serve to improve patient safety and quality in preventive healthcare services.Systematic Review Registration: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=134652, identifier: CRD42020134652.

Person-centred care to prevent hospitalisations - a focus group study addressing the views of healthcare providers.

BACKGROUND: The primary healthcare sector comprises various health services, including disease prevention at local level. Research shows that targeted primary healthcare services can prevent the development of acute complications and ultimately reduce the risk of hospitalisations. While interdisciplinary collaboration has been suggested as a means to improve the quality and responsiveness of personal care needs in preventive services, effective implementation remains a challenge. To improve the quality and responsiveness of primary healthcare and to develop initiatives to support the interdisciplinary collaboration in preventive services, there is a need to investigate the views of primary healthcare providers. The aim of this study was to investigate perceptions of preventive care among primary healthcare providers by examining their views on what constitutes a need for hospitalisation, and which strategies are found useful to prevent hospitalisation. Further, to explain how interdisciplinary collaboration can be supported with a view to providing person-centred care. METHODS: Five focus group interviews were conducted with 27 healthcare providers, including general practitioners, social and healthcare assistants, occupational therapists, physiotherapists, home care nurses, specialist nurses and acute care nurses. Interviews were transcribed, and analysed with qualitative content analysis. RESULTS: Three categories emerged from the analysis: 1) Mental and social conditions influence physical functioning and hospitalisation need, 2) Well-established primary healthcare services are important to provide person-centred care through interdisciplinary collaboration and 3) Interdisciplinary collaboration in primary healthcare services is predominantly focussed on handling acute physical conditions. These describe that the healthcare providers are attentive towards the influence of mental, social and physical conditions on the risk of hospitalisation, entailing a focus on person-centred care. Nevertheless, in the preventive services, interdisciplinary collaboration focusses primarily on handling acute physical conditions, which constitutes a barrier for interdisciplinary collaboration. CONCLUSIONS: By focusing on the whole person, it could be possible to provide more person-centred care through interdisciplinary collaboration and ultimately to prevent some hospitalisations. Stakeholders at all levels should be informed about the relevance of considering mental, social and physical conditions to improve the quality and responsiveness of primary healthcare services and to develop initiatives to support interdisciplinary collaboration.

Healthcare Use in the Five Years Before a First Infertility Diagnosis: A Danish Register-Based Case-Control Study in the CROSS-TRACKS Cohort.

Purpose: Infertility may affect somatic and mental health later in life. Nevertheless, health status before diagnosed infertility is sparsely studied in women. We aimed to describe healthcare use in primary and secondary care before a first infertility diagnosis and compare use between cases and controls. Materials and Methods: The case-control study was based on register data and used incidence density sampling. From the CROSS-TRACKS Cohort, we included women residing in the Horsens area in Denmark in 2012-2018 (n = 54,175). Eligible women were aged 18-40 years, nulliparous, and living in heterosexual relationships. Cases were women with a first infertility diagnosis in the Danish National Patient Registry (index date). Five controls were matched on age, birth year, and calendar time. Through linkage to Danish national health registries, we identified general practitioner (GP) attendance, paraclinical examinations, hospital contacts, diagnoses, and redeemed prescriptions. Healthcare use from one year to five years before index date was compared with conditional logistic regression. Results: We identified 711 cases and 3555 controls. At one year before index date, cases consulted their GP (odds ratio (OR) = 5.2, 95% confidence interval (CI): 3.2, 8.3) and visited hospital (OR = 1.2, 95% CI: 1.0, 1.4) and redeemed prescriptions (OR = 2.3 95% CI: 1.9, 2.7) more often compared to controls. Cases more often had blood and hemoglobin tests performed, redeemed more drugs related to genitourinary and hormonal diseases, and were more often diagnosed with endocrine and genitourinary diseases in the year before a first infertility diagnosis compared to controls. Cases and controls had comparable healthcare use from five years to one year before a first infertility diagnosis. Conclusion: Cases and controls had similar healthcare use from five years to one year before a first infertility diagnosis. However, cases had a higher healthcare use in the year preceding a first infertility diagnosis compared to controls.

The Framing of machine learning risk prediction models illustrated by evaluation of sepsis in general wards.

Problem framing is critical to developing risk prediction models because all subsequent development work and evaluation takes place within the context of how a problem has been framed and explicit documentation of framing choices makes it easier to compare evaluation metrics between published studies. In this work, we introduce the basic concepts of framing, including prediction windows, observation windows, window shifts and event-triggers for a prediction that strongly affects the risk of clinician fatigue caused by false positives. Building on this, we apply four different framing structures to the same generic dataset, using a sepsis risk prediction model as an example, and evaluate how framing affects model performance and learning. Our results show that an apparently good model with strong evaluation results in both discrimination and calibration is not necessarily clinically usable. Therefore, it is important to assess the results of objective evaluations within the context of more subjective evaluations of how a model is framed.

Cohort profile: CROSS-TRACKS: a population-based open cohort across healthcare sectors in Denmark.

PURPOSE: This paper describes the open cohort CROSS-TRACKS, which comprises population-based data from primary care, secondary care and national registries to study patient pathways and transitions across sectors while adjusting for sociodemographic characteristics. PARTICIPANTS: A total of 221 283 individuals resided in the four Danish municipalities that constituted the catchment area of Horsens Regional Hospital in 2012-2018. A total of 96% of the population used primary care, 35% received at least one transfer payment and 66% was in contact with a hospital at least once in the period. Additional clinical information is available for hospital contacts (eg, alcohol intake, smoking status, body mass index and blood pressure). A total of 23% (n=8191) of individuals aged ≥65 years had at least one potentially preventable hospital admission, and 73% (n=5941) of these individuals had more than one. FINDINGS TO DATE: The cohort is currently used for research projects in epidemiology and artificial intelligence. These projects comprise a prediction model for potentially preventable hospital admissions, a clinical decision support system based on artificial intelligence, prevention of medication errors in the transition between sectors, health behaviour and sociodemographic characteristics of men and women prior to fertility treatment, and a recently published study applying machine learning methods for early detection of sepsis. FUTURE PLANS: The CROSS-TRACKS cohort will be expanded to comprise the entire Central Denmark Region consisting of 1.3 million residents. The cohort can provide new knowledge on how to best organise interventions across healthcare sectors and prevent potentially preventable hospital admissions. Such knowledge would benefit both the individual citizen and society as a whole.

Explainable artificial intelligence model to predict acute critical illness from electronic health records.

Acute critical illness is often preceded by deterioration of routinely measured clinical parameters, e.g., blood pressure and heart rate. Early clinical prediction is typically based on manually calculated screening metrics that simply weigh these parameters, such as early warning scores (EWS). The predictive performance of EWSs yields a tradeoff between sensitivity and specificity that can lead to negative outcomes for the patient. Previous work on electronic health records (EHR) trained artificial intelligence (AI) systems offers promising results with high levels of predictive performance in relation to the early, real-time prediction of acute critical illness. However, without insight into the complex decisions by such system, clinical translation is hindered. Here, we present an explainable AI early warning score (xAI-EWS) system for early detection of acute critical illness. xAI-EWS potentiates clinical translation by accompanying a prediction with information on the EHR data explaining it.

Early detection of sepsis utilizing deep learning on electronic health record event sequences.

BACKGROUND: The timeliness of detection of a sepsis incidence in progress is a crucial factor in the outcome for the patient. Machine learning models built from data in electronic health records can be used as an effective tool for improving this timeliness, but so far, the potential for clinical implementations has been largely limited to studies in intensive care units. This study will employ a richer data set that will expand the applicability of these models beyond intensive care units. Furthermore, we will circumvent several important limitations that have been found in the literature: (1) Model evaluations neglect the clinical consequences of a decision to start, or not start, an intervention for sepsis. (2) Models are evaluated shortly before sepsis onset without considering interventions already initiated. (3) Machine learning models are built on a restricted set of clinical parameters, which are not necessarily measured in all departments. (4) Model performance is limited by current knowledge of sepsis, as feature interactions and time dependencies are hard-coded into the model. METHODS: In this study, we present a model to overcome these shortcomings using a deep learning approach on a diverse multicenter data set. We used retrospective data from multiple Danish hospitals over a seven-year period. Our sepsis detection system is constructed as a combination of a convolutional neural network and a long short-term memory network. We assess model quality by standard concepts of accuracy as well as clinical usefulness, and we suggest a retrospective assessment of interventions by looking at intravenous antibiotics and blood cultures preceding the prediction time. RESULTS: Results show performance ranging from AUROC 0.856 (3 h before sepsis onset) to AUROC 0.756 (24 h before sepsis onset). Evaluating the clinical utility of the model, we find that a large proportion of septic patients did not receive antibiotic treatment or blood culture at the time of the sepsis prediction, and the model could, therefore, facilitate such interventions at an earlier point in time. CONCLUSION: We present a deep learning system for early detection of sepsis that can learn characteristics of the key factors and interactions from the raw event sequence data itself, without relying on a labor-intensive feature extraction work. Our system outperforms baseline models, such as gradient boosting, which rely on specific data elements and therefore suffer from many missing values in our dataset.

Implementation of patient involvement methods in the clinical setting: A qualitative study exploring the health professional perspective.

RATIONALE, AIMS AND OBJECTIVES: Patient involvement is increasingly recognized as a key component on the international health care agenda. This attention has brought a need for developing generic and standardized open-source methods, tools, and guidelines on how to systematically implement patient involvement initiatives in the clinical setting. The large-scale project the User-involving Hospital was initiated to implement two systematic methods for patient involvement at a Danish university hospital, but the required methods can only be implemented if embraced by the health professionals. This evaluation study aimed to explore the health professional perspective on the development and implementation of shared decision making (SDM) and user-led health care. Specifically, the objectives were to identify the most crucial preconditions for success and to translate the findings into practice recommendations. METHOD: The study was based on a simple questionnaire survey and a qualitative descriptive analysis of semistructured focus group interviews with representatives of 21 multidisciplinary clinical teams (nine interviews) and 18 health professional department managers (six interviews). RESULTS: Two years after the initiation of the User-involving Hospital, 13 out of 21 developed patient involvement initiatives were fully incorporated into clinical practice. Five domains were found significant for successful development and implementation of the patient involvement methods: the patients' perspectives, composition of multidisciplinary teams, bottom-up and skill building, support from management, and information sharing with colleagues. CONCLUSIONS: The findings draw attention to several significant factors for successful implementation of large-scale patient involvement initiatives in hospitals, including the importance of having both a top-down and bottom-up approach and of active listening to the patients' perspectives. On the basis of these findings, the study outlines four recommendations incorporating the five identified key domains, which may inspire future projects on systematic development and implementation of patient-involvement initiatives based on either shared decision making or user-led health care in the clinical setting.

Chlamydia trachomatis infection in young adults - association with concurrent partnerships and short gap length between partners.

BACKGROUND: Sexually transmitted infections (STIs) continue to be highly prevalent in young people. New understanding of sexual risk behaviour is essential for future preventive initiatives. Studies based on self-reported STI history indicate that gap length between sexual partnerships is an important determinant in STI transmission, but little is known about the impact of concurrent partnerships and short gap length. This study aimed to examine the significance of concurrent partnerships and short gap length between serially monogamous partnerships in Chlamydia trachomatis-infected individuals compared to the general population. METHODS: A Danish cross-sectional study was conducted among individuals aged 15-29 years with a verified C. trachomatis infection and a sample of the background population. Participants answered a web-based questionnaire on sexual behaviour. Associations were identified in multivariate analyses. RESULTS: In total, 36% of the included young adults reported that they had two or more partners within the last year. Concurrent partnerships were frequent (46%), and the gap length between serially monogamous partnerships tended to be short (median gap length, 64 days, interquartile interval (IQI) = 31, 122). A strong association was found between concurrent partnerships (odds ratio (OR) = 12.5, 95% confidence interval (CI) = 7.7-20.4), short gap length between serially monogamous partnerships (OR = 10.0, 95% CI = 5.7-17.4) and having a verified C. trachomatis infection. CONCLUSIONS: C. trachomatis infection was strongly associated with concurrent partnerships or short gap length between serially monogamous partnerships. These findings have considerable implications for public health policy. Both types of risk factors should be considered in future preventive interventions aiming to reduce the spread of C. trachomatis infections.

Sexual behaviour among young Danes aged 15-29 years: a cross-sectional study of core indicators.

OBJECTIVES: Sexually transmitted infections and unwanted pregnancies occur at high rates among youth. Understanding sexual behaviour is essential for planning and implementing future effective preventive interventions. The present study examines the sexual behaviour in the general Danish population aged 15-29 years using the core indicators recommended by the European Centre for Disease Prevention and Control. METHODS: A nationwide cross-sectional study was conducted in Denmark among a random sample of 20 000 men and women in 2012. Respondents completed a web-based sexual behaviour questionnaire and data were linked to a nationally held demographic database. Core indicators for sexual behaviour frequency stratified by gender are presented as unweighted and weighted data after consideration of sociodemographic differences between respondents and non-respondents. RESULTS: Response rate was 20.4%. Condoms were used at sexual debut by 69.9% of women and 62.3% of men, while 14.3% of women and 15.1% of men used no contraceptives at sexual debut. Half of the respondents used condom alone at the latest sexual encounter with a steady partner (women 51.8%, men 55.2%), while 10% used no contraceptives. Having a sexual encounter with a casual partner decreased the likelihood of using condoms (women 43.7%, men 49.5%) and increased the likelihood of using no contraceptives (women 14.8%, men 20.9%). Data on sexual behaviour characteristics showed only minor changes when weighted for non-response. CONCLUSIONS: The findings call for interventions addressing the use of appropriate contraception at sexual debut and at last sexual encounter; this seems particularly important when the sexual partner is a casual partner.