Abstract
The central subject of this thesis is mental well-being in patients with eating disorders (EDs). We aim to provide a base of knowledge about this subject by exploring mental well-being from different perspectives, which we outlined in three overarching themes: 1) perspectives of individuals with an ED history, 2) cross-sectional associations between mental well-being, psychopathology and personality, and 3) longitudinal effects in mental health during outpatient treatment. We start in chapter one with a general introduction of the central subjects in this thesis.
Perspectives of individuals with an ED history
In the second chapter, we present the results of a systematic review and qualitative meta-analysis in which criteria for recovery are examined according to people who have recovered themselves. Eighteen relevant studies are included to identify criteria that are consistently described across these publications. In each study, we have extracted and counted the reported criteria based on a predefined mental health framework. We learn that recovery is primarily explained in terms of improved mental well-being, although symptom remission is also important.
Cross-sectional associations between mental well-being, psychopathology and personality
In the third chapter, we explore how patients with EDs experience mental well-being and compare these outcomes with the general population. Including a representative sample of 468 female ED patients, we find that patients experience lower levels of mental well-being, although a substantial proportion experiences adequate well-being. We find that well-being is correlated with several patient characteristics. Corresponding with the dual-continua model of mental health, we find overall low correlations between well-being and ED psychopathology and moderate to strong correlations between well-being and general psychopathology.
In chapter four, we use psychometric network analysis in a representative sample of 905 ED patients to arrive at a deeper understanding of the relationships between well-being and psychopathology. We estimate a network on a mental health domain level (i.e. emotional, psychological and social well-being, general and ED specific psychopathology) and a network with the underlying symptoms that make up these domains. Corresponding with the results of chapter three, we learn that ED psychopathology is only weakly connected with well-being domains. Psychological well-being is the most central node in the domain network. Central underlying mental health symptoms we find are feeling depressed, feeling worthless, purpose in life and self‐acceptance. We also learn that mental health symptoms such as self-acceptance, environmental mastery, interest in life and feeling depressed act as bridge symptoms linking well-being with psychopathology.
In chapter five, we explore associations between maladaptive personality trait facets and emotional, psychological and social well-being in 1187 female ED patients referred for specialised treatment. We find that personality traits lead to a statistically significant increase of the explained variance in emotional (38%), psychological (39%) and social well-being (26%) in addition to background and illness characteristics. The maladaptive personality trait facets anhedonia and depression are negatively associated with all three well-being domains. We argue that maladaptive personality functioning is related to the experience of well-being.
Longitudinal effects in mental health during outpatient treatment
The studies in chapters three to five are cross-sectional studies using a measurement at a single time-point. These studies do not depict changes over time within persons. In order to understand change in patients, we use routine outcome monitoring (ROM) data collected during outpatient treatment in chapter six. ROM data was collected every three months in 442 patients for a year, resulting in five measures. We apply latent growth mixture modelling (LGMM) to understand change in well-being and ED psychopathology. LGMM allows to model intraindividual change by estimating classes of patients with similar change trajectories. Both for well-being and ED psychopathology, we find three distinct classes. We also examine predictors for class membership and find that general psychopathology, early symptom change, hope for recovery, and the ED type are predictive for ED psychopathology class membership. General psychopathology and intrinsic motivation are predictive for well-being class membership.
In chapter seven, we present our last study, which expands on chapter six by using a larger dataset of 1250 patients with the same five measurements during a year of treatment. This dataset enabled us to perform panel data network analyses to understand associations between mental health domains during treatment. We used the same domains as included in the network analysis in chapter four. Panel data network analysis has the advantage that average within-person associations can be estimated separately from the between-person effects. The between-person effects in panel network analysis can be understood as effects also measured in cross-sectional studies, with the addition that in the panel study, they represent average stable means over time. Within-person effects are further modelled on two levels, depicted in a temporal and contemporaneous network. The temporal network depicts average within-person directed associations over time. Temporal networks allow understanding whether certain domains change before others change. These predictive effects can be estimated within domains (autoregressive effects) and between domains (cross-lagged effects). The contemporaneous network depicts average within-person effects within time points after controlling for temporal effects. These effects can be understood as associations occurring in a shorter period than measured (in our case, three months). From this study, we learn that changes in ED psychopathology are predictive of further changes in ED psychopathology over time and minor improvements in the other domains (temporal effects). Other domains were not predictive of changes in ED psychopathology. Also, we learn that psychological well-being plays a central role in the within-person contemporaneous network. Psychological well-being is considered an influential domain for experiencing mental health within time points with stable effects over time. We consider that ED psychopathology should be the primary focus of treatments because it changes relatively independent from other mental health domains. In addition, it may be warranted to address well-being in treatment as a secondary focus, in particular psychological well-being.
In chapter eight, we present a general discussion and reflect on the findings of the studies. We conclude that we arrived at a better understanding of the role of mental well-being in ED patients. The strengths and limitations of the research and the implications and directions for further research are also outlined.
Perspectives of individuals with an ED history
In the second chapter, we present the results of a systematic review and qualitative meta-analysis in which criteria for recovery are examined according to people who have recovered themselves. Eighteen relevant studies are included to identify criteria that are consistently described across these publications. In each study, we have extracted and counted the reported criteria based on a predefined mental health framework. We learn that recovery is primarily explained in terms of improved mental well-being, although symptom remission is also important.
Cross-sectional associations between mental well-being, psychopathology and personality
In the third chapter, we explore how patients with EDs experience mental well-being and compare these outcomes with the general population. Including a representative sample of 468 female ED patients, we find that patients experience lower levels of mental well-being, although a substantial proportion experiences adequate well-being. We find that well-being is correlated with several patient characteristics. Corresponding with the dual-continua model of mental health, we find overall low correlations between well-being and ED psychopathology and moderate to strong correlations between well-being and general psychopathology.
In chapter four, we use psychometric network analysis in a representative sample of 905 ED patients to arrive at a deeper understanding of the relationships between well-being and psychopathology. We estimate a network on a mental health domain level (i.e. emotional, psychological and social well-being, general and ED specific psychopathology) and a network with the underlying symptoms that make up these domains. Corresponding with the results of chapter three, we learn that ED psychopathology is only weakly connected with well-being domains. Psychological well-being is the most central node in the domain network. Central underlying mental health symptoms we find are feeling depressed, feeling worthless, purpose in life and self‐acceptance. We also learn that mental health symptoms such as self-acceptance, environmental mastery, interest in life and feeling depressed act as bridge symptoms linking well-being with psychopathology.
In chapter five, we explore associations between maladaptive personality trait facets and emotional, psychological and social well-being in 1187 female ED patients referred for specialised treatment. We find that personality traits lead to a statistically significant increase of the explained variance in emotional (38%), psychological (39%) and social well-being (26%) in addition to background and illness characteristics. The maladaptive personality trait facets anhedonia and depression are negatively associated with all three well-being domains. We argue that maladaptive personality functioning is related to the experience of well-being.
Longitudinal effects in mental health during outpatient treatment
The studies in chapters three to five are cross-sectional studies using a measurement at a single time-point. These studies do not depict changes over time within persons. In order to understand change in patients, we use routine outcome monitoring (ROM) data collected during outpatient treatment in chapter six. ROM data was collected every three months in 442 patients for a year, resulting in five measures. We apply latent growth mixture modelling (LGMM) to understand change in well-being and ED psychopathology. LGMM allows to model intraindividual change by estimating classes of patients with similar change trajectories. Both for well-being and ED psychopathology, we find three distinct classes. We also examine predictors for class membership and find that general psychopathology, early symptom change, hope for recovery, and the ED type are predictive for ED psychopathology class membership. General psychopathology and intrinsic motivation are predictive for well-being class membership.
In chapter seven, we present our last study, which expands on chapter six by using a larger dataset of 1250 patients with the same five measurements during a year of treatment. This dataset enabled us to perform panel data network analyses to understand associations between mental health domains during treatment. We used the same domains as included in the network analysis in chapter four. Panel data network analysis has the advantage that average within-person associations can be estimated separately from the between-person effects. The between-person effects in panel network analysis can be understood as effects also measured in cross-sectional studies, with the addition that in the panel study, they represent average stable means over time. Within-person effects are further modelled on two levels, depicted in a temporal and contemporaneous network. The temporal network depicts average within-person directed associations over time. Temporal networks allow understanding whether certain domains change before others change. These predictive effects can be estimated within domains (autoregressive effects) and between domains (cross-lagged effects). The contemporaneous network depicts average within-person effects within time points after controlling for temporal effects. These effects can be understood as associations occurring in a shorter period than measured (in our case, three months). From this study, we learn that changes in ED psychopathology are predictive of further changes in ED psychopathology over time and minor improvements in the other domains (temporal effects). Other domains were not predictive of changes in ED psychopathology. Also, we learn that psychological well-being plays a central role in the within-person contemporaneous network. Psychological well-being is considered an influential domain for experiencing mental health within time points with stable effects over time. We consider that ED psychopathology should be the primary focus of treatments because it changes relatively independent from other mental health domains. In addition, it may be warranted to address well-being in treatment as a secondary focus, in particular psychological well-being.
In chapter eight, we present a general discussion and reflect on the findings of the studies. We conclude that we arrived at a better understanding of the role of mental well-being in ED patients. The strengths and limitations of the research and the implications and directions for further research are also outlined.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 28 Oct 2021 |
Place of Publication | Enschede |
Publisher | |
Print ISBNs | 978-90-365-5250-9 |
DOIs | |
Publication status | Published - 28 Oct 2021 |