Predicting student performance using sequence classification with time-based windows.

A growing number of universities worldwide use various forms of online and blended learning as part of their academic curricula. Furthermore, the recent changes caused by the COVID-19 pandemic have led to a drastic increase in importance and ubiquity of online education. Among the major advantages of e-learning is not only improving students' learning experience and widening their educational prospects, but also an opportunity to gain insights into students' learning processes with learning analytics. This study contributes to the topic of improving and understanding e-learning processes in the following ways. First, we demonstrate that accurate predictive models can be built based on sequential patterns derived from students' behavioral data, which are able to identify underperforming students early in the course. Second, we investigate the specificity-generalizability trade-off in building such predictive models by investigating whether predictive models should be built for every course individually based on course-specific sequential patterns, or across several courses based on more general behavioral patterns. Finally, we present a methodology for capturing temporal aspects in behavioral data and analyze its influence on the predictive performance of the models. The results of our improved sequence classification technique are capable to predict student performance with high levels of accuracy, reaching 90% for course-specific models.

Analysis of the relationship between treatment networks and the evolution of patients with Type 2 Diabetes Mellitus.

Type 2 Diabetes Mellitus (T2DM) is a chronic disease that has been increasing in prevalence in recent years and that can cause severe complications. To ensure patient care is administered correctly, it is necessary for medical treatment teams to be both multidisciplinary and cohesive. The analysis of health processes is a constant challenge due to their high variability and complexity. This paper proposes a method based on the analysis of social networks to detect treatment networks, and to identify a relationship between these networks and patient evolution, as measured by glycated hemoglobin (HbA1c) levels. The networks were segmented based on patient adherence to their medical appointments and their mean time of delay. We applied this method on a sample of 1574 patients diagnosed with T2DM. Results show that participatory treatment -in which a patient sees a particular group of professionals on a recurrent basis - together with high levels of adherence are associated to those patients who improve their HbA1c levels in the case of high levels of adherence, while those who continually experience referrals to different professionals, remain unstable and, in some cases, get worse. On the other hand, in order to maintain a patient as stable, continuous control of the patient is enough, regardless of the recurrence to the same professionals.

Team Collaboration Networks and Multidisciplinarity in Diabetes Care: Implications for Patient Outcomes.

Prevalence of type 2 diabetes mellitus (T2DM) has almost doubled in recent decades and commonly presents comorbidities and complications. T2DM is a multisystemic disease, requiring multidisciplinary treatment provided by teams working in a coordinated and collaborative manner. The application of social network analysis techniques in the healthcare domain has allowed researchers to analyze interaction between professionals and their roles inside care teams. We studied whether the structure of care teams, modeled as complex social networks, is associated with patient progression. For this, we illustrate a data-driven methodology and use existing social network analysis metrics and metrics proposed for this research. We analyzed appointment and HbA1c blood test result data from patients treated at three primary health care centers, representing six different practices. Patients with good metabolic control during the analyzed period were treated by teams that were more interactive, collaborative and multidisciplinary, whereas patients with worsening or unstable metabolic control were treated by teams with less collaboration and more continuity breakdowns. Results from the proposed metrics were consistent with the previous literature and reveal relevant aspects of collaboration and multidisciplinarity.

Analysis of the Relationship between the Referral and Evolution of Patients with Type 2 Diabetes Mellitus.

Type 2 Diabetes Mellitus (T2DM) is a chronic disease that has risen in prominence in recent years and can cause serious complications. Several studies show that the level of adherence to different types of treatment has a direct correlation with the positive evolution of chronic diseases. While such studies relate to patient adherence to medication, those that concern adherence to medical appointments do not distinguish between the different disciplines that attend to or refer patients. This study analyses the relationship between adherence to referrals made by three distinct disciplines (doctors, nurses, and nutritionists) and the results of HbA1c tests from a sample of 2290 patients with T2DM. The aim is to determine whether a relationship exists between patient improvement and the frequency with which they attend scheduled appointments in a timely manner, having been previously referred from or to a particular discipline. Results showed that patients tended to be more adherent when their next appointment is with a doctor, and less adherent when it is with a nurse or nutritionist. Furthermore, patients that remained stable had higher rates of adherence, whereas those with lower adherence tended to be more decompensated. The results can enable healthcare professionals to monitor patients and place particular emphasis on those who do not attend their scheduled appointments in a timely manner.

Multidisciplinary Collaboration in the Treatment of Patients With Type 2 Diabetes in Primary Care: Analysis Using Process Mining.

BACKGROUND: Public health in several countries is characterized by a shortage of professionals and a lack of economic resources. Monitoring and redesigning processes can foster the success of health care institutions, enabling them to provide a quality service while simultaneously reducing costs. Process mining, a discipline that extracts knowledge from information system data to analyze operational processes, affords an opportunity to understand health care processes. OBJECTIVE: Health care processes are highly flexible and multidisciplinary, and health care professionals are able to coordinate in a variety of different ways to treat a diagnosis. The aim of this work was to understand whether the ways in which professionals coordinate their work affect the clinical outcome of patients. METHODS: This paper proposes a method based on the use of process mining to identify patterns of collaboration between physician, nurse, and dietitian in the treatment of patients with type 2 diabetes mellitus and to compare these patterns with the clinical evolution of the patients within the context of primary care. Clustering is used as part of the preprocessing of data to manage the variability, and then process mining is used to identify patterns that may arise. RESULTS: The method is applied in three primary health care centers in Santiago, Chile. A total of seven collaboration patterns were identified, which differed primarily in terms of the number of disciplines present, the participation intensity of each discipline, and the referrals between disciplines. The pattern in which the three disciplines participated in the most equitable and comprehensive manner had a lower proportion of highly decompensated patients compared with those patterns in which the three disciplines participated in an unbalanced manner. CONCLUSIONS: By discovering which collaboration patterns lead to improved outcomes, health care centers can promote the most successful patterns among their professionals so as to improve the treatment of patients. Process mining techniques are useful for discovering those collaborations patterns in flexible and unstructured health care processes.

Multidisciplinary collaboration in primary care: a systematic review.

Background: Several studies have discussed the benefits of multidisciplinary collaboration in primary care. However, what remains unclear is how collaboration is undertaken in a multidisciplinary manner in concrete terms. Objective: To identify how multidisciplinary teams in primary care collaborate, in regards to the professionals involved in the teams and the collaborative activities that take place, and determine whether these characteristics and practices are present across disciplines and whether collaboration affects clinical outcomes. Methods: A systematic literature review of past research, using the MEDLINE, ScienceDirect and Web of Science databases. Results: Four types of team composition were identified: specialized teams, highly multidisciplinary teams, doctor-nurse-pharmacist triad and physician-nurse centred teams. Four types of collaboration within teams were identified: co-located collaboration, non-hierarchical collaboration, collaboration through shared consultations and collaboration via referral and counter-referral. Two combinations were commonly repeated: non-hierarchical collaboration in highly multidisciplinary teams and co-located collaboration in specialist teams. Fifty-two per cent of articles reported positive results when comparing collaboration against the non-collaborative alternative, whereas 16% showed no difference and 32% did not present a comparison. Conclusion: Overall, collaboration was found to be positive or neutral in every study that compared collaboration with a non-collaborative alternative. A collaboration typology based on objective measures was devised, in contrast to typologies that involve interviews, perception-based questionnaires and other subjective instruments.