Part II | Vascular risk factors for depression and apathy 105 6 population. This implies that studies in broad patient groups or the general population including those with minimally cognitively impaired patients were included in the review. Studies were excluded i. when the language was not English and ii. when the studies concerned specific populations, such as post-stroke patients, patients with dementia (including vascular dementia), with Parkinson’s disease or major depression. Study quality The quality of the case control, cross-sectional and longitudinal studies selected for review was judged against specific criteria for design and methodology. We used an adapted version of the evaluation scale for cross-sectional (not case-control) studies originally developed by Kuijpers et al.33 (online supplementary file 1). For case-control and longitudinal studies we used scales based on the Newcastle-Ottawa scale 34 (online supplementary file 2). Overall quality of a study was considered high when it attained at least 60% of the maximum score 35. Evaluation of the quality of apathy scales The apathy evaluation scale (AES) and the apathy subscale of the neuropsychiatric inventory (NPI) were considered of high quality 36 37. The 3 apathy items of the geriatric depression scale (GDS-3A) are validated by comparison with the apathy scale (sensitivity 69% and specificity 85% 29 38). The apathy scale (an abbreviated version of the AES) and therefore also the apathy items of the GDS were not granted the highest quality status in our evaluation based on the review by Clarke et al., 2011 37. Clinician- or informant-based information was considered of higher quality than self-reported in the older population where individuals may have been suffering from MCI 39. Evaluation of the quality of SSVD assessment SSVD on neuroimaging was operationalized as WMH, silent lacunar infarcts, cerebral microbleeds, or decreased cortical thickness onMRI scans 27. Diffusion tensor imaging (DTI) studies the diffusivity of water molecules in white matter as a model of the connectivity of this tissue and its markers (fractional anisotropy and diffusivity) are associated with SVD 40. Peripheral measures of atherosclerosis were operationalized as the ABI, IMT, and/or CFPWV. Although the ABI and CFPWV are measures of large artery atherosclerosis 41, we considered both measures proxies for SSVD as large artery and small vessel disease are closely related 42. Cardiovascular disease was included as an SSVD proxy, since it can lead to haemodynamic changes affecting the small vessels 42. Finally, being the strongest risk factor for SSVD, hypertension was also taken as an SSVD proxy 27 43. Studies were awarded an extra point was if SSVD proxies were measured rather than mentioned in an interview or derived from information provided by general practitioners. Self-reported SSVD was categorized as “low quality”.
RkJQdWJsaXNoZXIy MjY0ODMw