In the previous sections, a general framework for the development of AD was outlined and stages of AD from preclinical AD to AD dementia were described. SCD was then defined for the present work and a comprehensive overview of the rather heterogeneous literature on SCD was given. Finally, a working model, describing how SCD might evolve and interact with objective cognitive decline throughout the course of AD, was presented. Implications from this working model for SCD as a predictor of either future AD dementia or higher likelihood of AD pathology were briefly discussed.
11This is the interpretation of the author and not explicitly mentioned in the original paper of Jessen et al. (2014a).
Based on the present literature it can be cautiously concluded that SCD is associated with future cognitive decline, higher risk of AD dementia, and higher likelihood of AD pathology. Thus, it is potentially useful as an indicator of incipient AD with implications both for clinical practice and research. However, although there is a growing body of evidence, there are also studies that have not found associations with longitudinal outcomes or biomarker profiles. Furthermore, there has been a wide heterogeneity both in the samples studied (preclinical AD, MCI, mixed samples) and in the definition of SCD (see section 2.4.2). This two-fold heterogeneity (samples and assessment) might explain some of the equivocal findings in the research literature (section 2.4.3), especially as the validity of SCD may vary depending on the stage of AD, as outlined above (section 2.4.4.). The aspects just mentioned pose a serious difficulty to compare results across studies and there are many questions not fully answered yet. These questions refer to the proposed working model in section 2.4.4 and to the measurement aspect of SCD outlined in section 2.4.2. Such questions may be:
Is SCD a valid feature specifically in the early clinical phase of AD?
How strong is the predictive validity of SCD in subjects with MCI? Does it vary across different levels of objective cognitive impairment?
What factors influence SCD and/or moderate its association to incident AD dementia or biomarkers of AD pathology?
Are there certain aspects of SCD which are more closely related to AD and thus of a higher predictive value than others? If yes, are these aspects the same for different stages of AD?
Section 3 of this thesis contains empirical research that addresses some of these questions. In the following subsections, the study aims and hypotheses of the three empirical studies presented herein are described. Study 1 is based on a population based sample of the German “Study on Ageing, Cognition and Dementia (AgeCoDe)” while study 2 and 3 are based on a memory-clinic MCI sample of the German Competence Network for dementia (DCN).
2.5.1 Study 1 (longitudinal study, AgeCoDe sample): Memory-related SCD (with vs. without concerns) as a predictor of AD dementia in individuals with normal cognition, early and late MCI.
Building on previous research that showed associations of SCD with higher risk of future AD dementia, study 1 (previously published; Jessen et al. 2014b) addresses two aspects. First, it compares the AD dementia risk of subjects with pure memory- related SCD (but no cognitive impairment; i.e. pre-MCI SCD) to that of patients with either early or late MCI. Early MCI comprises patients with evidence of very mild objective memory impairment on neuropsychological tests (memory performance between 1.0 to 1.5 SD below the norm) while late MCI patients are more advanced in their objective memory impairment (more than 1.5 SD below the norm). Both MCI groups have SCD as defined in the MCI criteria. Second, the study evaluates whether SCD with concerns/worries (SCD+C) is associated with a higher risk of developing incident AD dementia compared to SCD without concerns (SCD-C). It is hypothesized that the qualitative appraisal of a self-experienced cognitive decline as worrisome (i.e.
SCD+C) is important and might further elevate the risk of AD compared to SCD-C.
2.5.2 Study 2: (longitudinal study, DCN sample): Significance of memory-related SCD in a clinical sample of MCI patients: Interaction with objective memory impairment.
Study 2 (previously published; Wolfsgruber et al. 2014b) deals with a key aspect of the working model presented in the previous section. The model stated that, during the early course of symptomatic AD, SCD might evolve as a result of self-perceived, intra-individual decline that is difficult to detect on cross-sectional neuropsychological tests. However, with the evolvement of more severe cognitive impairment, SCD might wane as a result of reduced insight into symptoms. If these assumptions from the working model were true, then the predictive value of SCD should be most predictive in very mildly impaired MCI patients and then decrease with increasing levels of memory impairment. Study 2 tests this hypothesis derived from the working model by examining the risk of incident AD dementia by SCD+C, objective memory performance and the proposed interaction of both factors in a large sample of memory clinic MCI patients with different levels of cognitive impairment. Some of these patients could be considered as early MCI patients (see definition in 2.5.1.) while others have more advanced, multi-domain cognitive impairment in the range of late MCI.
2.5.3 Study 3 (cross-sectional biomarker study, DCN sample): Biomarker correlates of memory-related SCD in MCI patients.
Study 3 (Wolfsgruber et al. 2015) finally addresses an under-researched topic, namely the relationship between measures of SCD and biomarkers of AD in the stage of MCI. While most research on SCD and biomarkers has focused on the pre-MCI stage, the MCI stage has been somewhat overlooked presumably due to the equivocal findings regarding the presence or absence of SCD in this patient group and the importance of objective cognitive testing in MCI. SCD might be influenced by several confounding factors in MCI: depressive symptoms, objective impairment level (interaction hypothesis proposed in section 2.5.2 applies here too) and education (as a cognitive reserve proxy). This study examines whether measures of SCD may predict abnormal CSF biomarkers of AD in a sample of MCI patients while taking into account the aforementioned factors as covariates.
3 Empirical Studies