Sinds het begin van de hiv-epidemie zijn er bepaalde groepen mensen die een verhoogd risico hebben om hiv (het virus dat aids kan veroorzaken) te krijgen. Deze zogenaamde ‘sleutelgroepen’ – zoals transgender personen, sekswerkers en mannen die seks hebben met mannen (MSM) – hebben niet alleen een verhoogd risico door hun seksuele gedrag, maar krijgen ook te maken met uitsluiting, stigma en discriminatie. Daardoor is hun toegang tot seksuele gezondheidszorg vaak beperkt, terwijl juist deze groepen gebaat zijn bij toegankelijke zorg die aansluit op hun specifieke behoeften. In veel delen van de wereld, waaronder Sub-Sahara Afrika, ontbreekt het nog steeds aan betrouwbare informatie over de hiv-ziektelast onder deze groepen. Ook in Nederland bestaan er kennishiaten, waardoor er bijvoorbeeld maar weinig bekend was over de impact van de coronapandemie op de gezondheid, veiligheid en het welzijn van sekswerkers. Het overkoepelende doel van dit proefschrift is te onderzoeken hoe de hiv- en seksuele gezondheidszorg voor sekswerkers, transgender personen en MSM in Sub-Sahara Afrika en Nederland verbeterd kan worden.
This thesis evaluates the quality of complex care in the Netherlands and explores predictive models to enhance care efficiency, focusing on acute myeloid leukemia (AML), metastatic non-small cell lung carcinoma (NSCLC), acute oncological care, and high-need, high-cost (HNHC) patients. Using data from the Netherlands Cancer Registry, Chapters 2–4 reveal persistent regional disparities in NSCLC diagnostics and AML treatment with intensive remission induction chemotherapy (ICT), despite increased adoption over time.
Structural aspects of complex care were examined through quantitative and qualitative methods. Chapter 5 highlights the association between higher ICT patient hospital volume and reduced mortality risk. However, Chapter 6 emphasizes that implementing structural changes requires a careful, structured approach that builds upon the existing system.
Finally, Chapters 7–9 present predictive models designed to improve care efficiency for complex care patients. These models demonstrate the potential to serve as case identifiers, ensuring that patients receive care tailored to their needs.
In conclusion, this thesis finds that the quality of complex care in the Netherlands can be significantly enhanced by reducing practice variation, implementing structural changes with consideration, and integrating predictive models into routine clinical practice. These advancements could be realized through the adoption of a learning healthcare system—an approach characterized by continuous, transparent, and adaptive improvement, wherein information technology is used to measure and enhance healthcare quality.
This thesis examines cancer prognosis (Part 1) and treatment effects (Part 2) using statistical models. In Part 1, nationwide data from Dutch cancer patients (1989–2019) showed improved life expectancy due to better detection and treatments, though some cancers have poor outcomes. For melanoma patients undergoing sentinel lymph node biopsy, a model was developed to predict recurrence and mortality risks. Part 2 highlights that treatment decisions should consider personalized effects rather than averages. For patients with diffuse large B-cell lymphoma, high-risk patients benefit most from intensified treatment. Methodological advances were made in quantifying individualized treatment effects. This thesis makes five recommendations: 1) use life expectancy as a measure for prognosis and treatment effects, 2) use simple models, 3) update existing models before developing a new one but validate new models if you do, 4) strive to make personalized predictions, and 5) critically evaluate data analysis to ensure accurate information influences decisions.
Patiënten met Barrett slokdarm hebben een verhoogd risico op het ontwikkelen van een adenocarcinoom van de slokdarm. Endoscopische surveillance met het nemen van biopten in dit weefsel wordt uitgevoerd om vroege neoplastische veranderingen tijdig te detecteren en indien nodig te behandelen. In de praktijk is er door een hoge interobserver variabiliteit tussen pathologen en sampling error bij het nemen van biopten een grote kans op misclassificatie van de histologische diagnose. Dat veroorzaakt wisselende diagnoses in mate van ernst van neoplasie met als gevolg oversurveillance en overbehandeling, resulterende in een hoge druk op de gezondheidszorg.
In dit proefschrift wordt er een predictiemodel opgesteld, waarbij per individuele Barrett patient meerdere metingen van de graad van dysplasie worden meegenomen, evenals veelbelovende immuunhistochemische biomarkers. Door de multipele metingen per patient wordt het risico van een enkele misclassificatie in histologische diagnose verkleind. Tevens faciliteert dit model een gepersonaliseerde risico-inschatting die per meetmoment geüpdate kan worden.
De voorspellende waarde van dit model wordt hoger geschat dan wanneer surveillance volgens de huidige richtlijn wordt uitgevoerd. Tevens laten simulatiestudies zien dat wanneer dit predictiemodel in de praktijk gebruikt zou worden, er een geschatte 65% van de patiënten geen surveillance meer zou hoeven ondergaan door een dermate geschat laag risico op het ontwikkelen van slokdarmkanker en een totale afname van de kosten met 25%, met gelijkblijvende effectiviteit. In afwachting van externe validatie in een bestaand (niet gesimuleerd) cohort, zou deze alternatieve strategie belangrijk kunnen zijn om veel onnodige scopieën in de toekomst te kunnen voorkomen met als gevolg het vrijkomen van capaciteit om de kwaliteit van surveillance in hoogrisico patiënten te verbeteren.
Psychosocial health in early childhood plays a crucial role in long-term well-being, yet disparities persist due to socioeconomic and environmental factors, especial for children in the early life. This thesis investigates the relative factors associated with psychosocial health in young children, focusing on three key areas. Part I explores the influence of socioeconomic status and parental migration background on psychosocial health in two-year-olds and its development over time. Part II examines the impact of stressful life events on psychosocial and general health from ages two to four. Part III addresses parental help-seeking for psychosocial concerns and the factors related to parenting stress. By identifying early risk factors and the role of parental behaviors, this research provides insights for policymakers and healthcare professionals to support mental health interventions for young children and their families.
Dementia is a brain condition affecting memory, thinking, learning, language, judgment, mood, and personality. Key features of dementia are that symptoms get worse over time and they become severe enough to disrupt daily life. Rather than being one specific disease, dementia includes a range of disorders that cause permanent brain damage. The most common type is Alzheimer’s disease. Dementia is common among older people and affects over 55 million people worldwide. This thesis introduces a new microsimulation model for dementia and demonstrates its use in addressing epidemiological and health-economic questions. The thesis is divided into three parts. In Part I – Dementia Disease Prognosis, two reviews of the literature are presented that investigate how long people with dementia live after their diagnosis, how long they live independently and how much time they spend in different severity stages. In Part II – Projections of Dementia Outcomes, the development of the new microsimulation model MISCAN-Dementia is presented and its use is shown in two applications that predict the future burden of dementia in terms of dementia prevalence, and costs and quality of life. In Part III – Dementia Prevention and Intervention, the newly developed microsimulation model is used to evaluate prevention strategies for dementia. Two studies reporting on the participation in comparative modeling exercises for the evaluation of interventions are also included. In summary, this thesis introduces a new microsimulation model for dementia, providing valuable insights into the future burden of the disease.
This thesis examines the interrelations and contributing factors of frailty, polypharmacy, medication-related problems, and loneliness in older European community-dwelling adults, using data from the Urban Health Centres Europe (UHCE) Project. It is structured into three main parts:
Part I focuses on frailty, divided into physical, psychological, and social domains. Findings indicate various demographic, lifestyle, and health factors associated with these frailty domains. Notably, physical frailty was shown to have a reciprocal relationship with psychological frailty and a unidirectional association with social frailty, highlighting the complex interplay between these domains.
Part II investigates the factors linked with polypharmacy and a high risk of medication-related problems. Significant associations were found with sociodemographic characteristics, health status, and healthcare usage. Additionally, a cycle was observed where higher levels of frailty increased the risk of medication-related problems, which in turn could lead to increased frailty, suggesting a bidirectional relationship.
Part III explores the relationship between loneliness and frailty. It was found that loneliness, particularly social loneliness, significantly contributes to all frailty domains. A bidirectional association between overall loneliness and frailty was identified, with emotional loneliness specifically influencing psychological frailty.
The thesis underscores the need for a holistic approach to address these interconnected issues, recommending the integration of medical, psychological, and social elements into health strategies to enhance aging well-being and resilience.
Value-based healthcare in inherited bleeding disorders
Value-based healthcare (VBHC) was first introduced in 2006 as a strategy to combat the rising healthcare cost and unwanted variation in outcomes and quality of care. According to VBHC framework, healthcare organizations should aim to maximize patient value, where value is defined as the health outcomes that matter to patients relative to the cost of achieving those outcomes. From 2011 onwards, hospitals in the Netherlands have been implementing VBHC principles to improve the care for patients with various medical conditions. The Dutch interpretation of VBHC, however, deviates slightly from the original concept. While the objective of maximizing patient value through the routine collection of outcomes that matter to patients prevails, Dutch healthcare organizations often bypass the immediate need to measure cost. In addition, in the Dutch context, emphasis is placed on the use of outcome information to improve shared decision making and to create a culture of continuous learning and quality improvement.
This PhD thesis aimed to assess the added value of the implementation of VBHC, as defined within the Dutch healthcare context, in routine clinical care for patients with an inherited bleeding disorder. To achieve this aim, this thesis includes research on all the steps necessary to facilitate the implementation and evaluation of VBHC in this patient population. This thesis provides a recommendation on which patient-reported outcomes should be measured in the care for patients with inherited bleeding disorders, and which patient-reported outcomes should be used to collect these outcomes. In addition, this thesis provides insight into the possible variation in the care provision for patients with hemophilia which can be used as both a starting point for continuous learning and quality improvement initiatives, as well as guide the collection and use of outcome information. Moreover, this thesis provides insight into patient experiences with the collection of outcome information in routine care provision.