Redesigning end-of-life care for patients with congenital heart disease by scrutinizing trajectories toward death and evaluating curative and palliative care in the last year of life

Healthcare spending has never been higher and is expected to continue to rise in all parts of the world over the next few years. One solution to reducing healthcare costs is to reduce avoidable healthcare use, which is healthcare use that could have been avoided by proper prevention, precautions or timely referrals. Avoidable healthcare use is common, unfortunately. To gain a good understanding of this concept, a detailed review of the factors that contribute to differences in healthcare use and outcomes is necessary. The most widely used theoretical model for mapping healthcare use is the Andersen Behavioral Model of Health Services Use, developed in 1968. This model has been modified and expanded several times over the past 50 years as a result of research about the concepts and relationships of the model.

The model includes a lot of factors. However, not all relationships between these factors have been well investigated yet. Indeed, some important gaps in the literature can be identified. First, it remains unknown to what extent a person's identity influences healthcare use. The concept of "illness identity," the degree to which the illness is integrated into one's identity, has not yet been investigated in association with healthcare use. However, since this variable was found to be clinically relevant in patients with chronic conditions, illness identity could play a key role in explaining individual variation in healthcare use. Second, the Andersen model indicates that healthcare system characteristics are an enabling factor influencing healthcare use and patient-reported outcomes, such as quality of life or self-reported health. However, international comparisons of healthcare systems have not yet paid much attention to these outcome measures. Therefore, the exact relationship between healthcare systems characteristics and patient-reported outcomes is still unclear. Third, for some chronic patient populations, almost no research has been done about the end-of-life care use or use of palliative care.

Thus, there is a need for a comprehensive examination of the associations between illness identity, healthcare system characteristics, palliative care, healthcare use (at the end of life) and patient-reported outcomes. As a sample population to examine these associations, we selected adults with congenital heart diseases. This is a heterogeneous population with high healthcare use.

Thus, the goal of this doctoral project was to examine (predictors of) healthcare use and patient-reported outcomes in adults with congenital heart disease, guided by the Andersen Behavioral Model of Health Services Use.

In Part 1 of this PhD project, we examined the predictive value of illness identity for healthcare use and patient-reported outcomes. We found that illness identity was associated with healthcare use and patient-reported outcomes. In more detail, we found that patients who did not integrate their illness well into their identity (i.e., engulfment or rejection) reported more hospitalizations, visits to the primary care physician and medical specialist, and poorer health status, quality of life and psychological distress. In contrast, patients who integrated their illness well into their identity (i.e., acceptance) reported less healthcare use and better outcomes.

Part 2 examined the relationship between healthcare system characteristics and patient-reported outcomes in adults with congenital heart disease. We found that healthcare system characteristics were associated with these outcomes. More specifically, unhealthy behaviours and physical health were associated with elements of the healthcare system performance. We also showed that more physicians could ensure better physical and mental health and quality of life. In addition, more nurses were found to be related to better physical and psychological health and fewer unhealthy behaviours in adults with congenital heart disease.

Part 3 of this doctoral project describes the development of a large international research project on patient-reported outcomes in adults with congenital heart disease. The data of this project were used in Part 4.

In Part 4, we studied palliative and end-of-life care provision to adults with congenital heart disease. Worldwide, 45% of participants indicated that they had not discussed with their physician their health status in the future, and only 9% had written down what care they would like to receive at the end of life. Moreover, we found that care provision at the end of life is intensive and potentially avoidable, with only limited attention for palliative care. In more detail, in the last year of life, 87% of patients were hospitalised, 78% had emergency department contact, and 19% had an intensive care admission. Specialised palliative care was provided to 17% of patients, and to only 4% of patients who died due to cardiovascular causes.

In conclusion, this doctoral project contributed to knowledge about (predictors of) healthcare use and patient-reported outcomes of adults with congenital heart disease. Illness identity was found to be related to healthcare use and outcomes. In addition, associations were found between healthcare system characteristics and outcomes. Furthermore, there is reasonable evidence that end-of-life and palliative care for adults with congenital heart disease can be optimized. More research is needed to clarify the direction of the relationships and the extent to which the results described in this thesis apply to other chronic patient populations.