Identifying effective treatments is a critical issue for Alzheimer’s dementia (AD). The pathological amyloid deposits of AD result in disruption of the balance between long-term potentiation (LTP) and long-term depression (LTD) of neuronal cells and synaptic plasticity. Brain stimulation in dementia research, especially with relatively safe tDCS, has been taken seriously recently. In theory, tDCS affects long-term synaptic plasticity through LTP and LTD, thereby improving cognitive ability. Recently, an increasing number of studies have been conducted to evaluate the efficacy of tDCS in AD and concluded a positive therapeutic effect. Currently, there are no studies of tDCS for AD in Taiwan. In this study, we investigate the effects of tDCS in AD.

Using a double-blind, randomized and sham- controlled trial design, Sixteen AD aged 55-90 years (8 active, mean age 73.88 and 8 sham, mean age 74.75) were included in the study. AD diagnostics is according to DSM-5 criteria. The CDR ratings of AD participants ranged from 0.5 to 2. All subjects completed ten consecutive daily sessions in which they received either an active or a sham tDCS over the left dorsal lateral prefrontal cortex (anodal) and a cathodal electrode on the right supraorbital area. In each session, we applied a current intensity of 2 mA and an electrode size of 35 cm2 for 30 min in the active group. All subjects received a series of neuropsychological tests, which included CDR, MMSE, CASI and WCST, before and after these treatment sessions on the first day and 4 weeks later. Chi- square test, Wilcoxon signed ranks test and Mann-Whitney U test were used to assess the differences in participant demographic characteristics and to compare the differences among groups.

The active group showed significant improvement in total correct item, Conceptual level Responses (reflecting insight into the correct sorting principles), Categories Completed (reflecting overall success), and Trials to complete first categories (reflecting initial conceptual ability) of WCST 4 weeks later after the final stimulation. There were no statistically significant differences between before and after the 10-session course for the sham group.

tDCS stimulation improves cognitive operation and Conceptual Ability of AD.

Neurofibrillary tangles (NFT) formed by tau proteins, a pathological hallmark of Alzheimer’s disease, are a common co-pathology in people with Lewy body dementias, which include dementia with Lewy bodies (DLB) and Parkinson’s disease dementia (PDD).

To investigate the prevalence of tau in Lewy body dementia, and its association with clinical outcomes.

A systematic search was conducted on Medline, Embase and PubMed using the search term: (“dementia with Lewy bodies” OR “diffuse Lewy body disease”) AND (“tau protein” OR “tauopathy” OR “neurofibrillary tangle”). A total of 42 articles met the inclusion criteria for data extraction. Random-effect meta-analyses were performed to obtain pooled estimates for prevalence, and risk ratios (RR) or standardised mean difference (SMD) for clinical outcomes measures.

Braak NFT stage ≥III was observed in 67% (n=1399, 95%CI 59%-76%) of DLB and 52% (n=429, 95%CI 26%-78%) of PDD at autopsy. Abnormal CSF phosphorylated-tau levels were present in 27% (n=705, 95%CI 23%-30%) of DLB and 15% (n=172, 95%CI 5%-24%) of PDD cases. Higher tau burden in DLB was associated with reduced likelihood of manifesting visual hallucinations (RR 0.56; 95%CI 0.40-0.77) and motor parkinsonism (RR 0.62; 95%CI 0.40-0.98), lower diagnostic accuracy of DLB during life (RR 0.49; 95%CI 0.38-0.64) and worse cognition prior to death (SMD 0.60; 95%CI 0.44-0.76).

Tau is more common in DLB than PDD and may negatively impact clinical diagnostic accuracy in people with DLB. Prospective longitudinal studies are needed to understand the roles of co-morbid neuropathologies in Lewy body dementias.

Cerebral multi-morbidity is common in older people with dementia, including people with dementia with Lewy bodies (DLB). We describe the first Australian-based, longitudinal observational biomarker study of DLB.

To investigate the frequency and influence of Alzheimer’s disease (AD) pathology (amyloid-β and tau) and cerebrovascular disease on clinical symptoms and disease outcome in DLB.

The study will recruit 100 people with mild to moderate probable DLB, who will undergo comprehensive clinical and cognitive assessments. Scales targeting DLB-specific clinical features (such as cognitive fluctuations and rapid eye movement sleep behaviour disorder) are administered. Biomarker protocols incorporate blood sampling (including ApoE genotyping and systemic inflammatory markers), molecular imaging (amyloid-β [18F-NAV 4694], tau [18F-MK6240], VMAT2 [18F-AV133] PET scans), 3-tesla magnetic resonance imaging and optional lumbar puncture. Clinical assessments are completed 6 - monthly and imaging 18-monthly. Participants are also invited to register for post-mortem brain tissue donation.

Thirty participants with probable DLB have been enrolled to date (mean age 75.4 years, range 64-82; 87% male). All participants have mild to moderate cognitive impairment (mean MMSE 25, range 17-30). Approximately 64% of the participants were amyloid-β positive. Study procedure tolerability has been excellent with no adverse events reported.

There is significant overlap of AD-related proteinopathies in people with DLB. Understanding the impact of multi-morbidity is essential in the development of effective treatment strategies. This study supports the feasibility of intensive, longitudinal biomarker studies in DLB in the Australian setting.

Early diagnosis of mild cognitive impairment is important in Alzheimer's disease management before brain damage is profoundly established and irreversible. Eye-tracking technology is a sensitive method to measure cognitive impairments in dementia and MCI. We examined the saccade movement deficits in amnestic MCI and compared them with the normal controls and Alzheimer to define early cognitive markers in MCI.

This study was a cross-sectional observational study. Twenty-one patients with AD, 40 patients with aMCI, and 59 normal participants were examined by eye tracking using anti-saccade task and pro-saccade task with ‘gap’ and ‘overlap’ procedures.

Patients with Alzheimer's made more errors, and corrected fewer errors than a-MCI and age-matched controls. Moreover, a-MCI had higher error rates in the prosaccade gap and overlap (38±1.5, p≤ 0.001; 21± 1.8, p≤ 0.001) and antisaccade gap and overlap (64± 1.4, p≤ 0.001; 45± 1.6, p≤ 0.001) than normal controls. Compared with the control group, a-MCI also showed more uncorrected responses in the prosaccade gap (6± 0.5, p≤ 0.001) and antisaccade gap and overlap (13± 0.4, p≤ 0.001; 10± 0.7, p≤ 0.001). Saccade Omission also revealed significant differences between normal controls and amnestic mild cognitive impairment in prosaccade (p≤ 0.001) and antisaccade (p≤ 0.001) tasks, in both gap and overlap paradigms.

Error proportion, target omission and uncorrected saccades impairments in a- a-MCI, support the concept of executive function deterioration, as an early marker of neurocognitive disorder. Our findings also confirm inhibitory and working memory impairments t in a-MCI.

Depression and cognitive impairments often co-occur in older adults and account for a high disease burden. Insight into how affect and cognitive function influence one another on a daily basis could be helpful in the diagnostic process and treatment decisions for individual patients. However, little is known about the daily associations between affect and cognitive function in older adults and it is unknown whether these associations differ per person. Therefore, the objective of this study was to get insight into the temporal associations between affect and cognitive function within individual older adults.

For this single-subject study eight older adults with depression and cognitive impairments filled in electronic diaries for 63 consecutive days in their home environment. The diaries included a questionnaire evaluating positive affect (PA), negative affect (NA) and a computerized cognitive test battery assessing working memory reaction time (WMRT) and visual learning accuracy (VLA). Time-series analyses using Vector Autoregressive (VAR) modelling were conducted for each individual separately. Granger causality tests were used to determine the temporal direction of the individual associations. Cumulative Orthogonalized Impulse Response Function (COIRF) analyses were performed in order to determine the cumulative effect size over a 10-day period. The contemporaneous associations were derived from the correlation between the residuals in the VAR model.

For one out of eight participants higher NA was associated with better WMRT the next day (cumulative effect size=0.345; p=0.005). For another participant higher NA and lower PA were associated with worse WMRT at the same time (r=-0.369; p=0.003 and r=0.352; p=0.005, respectively). For a third participant better VLA was associated with lower NA (cumulative effect size=-0.569; p<0.001) and higher PA the next day (cumulative effect size=0.223; p=0.001). The other five participants showed no contemporaneous or lagged association between affect and cognitive function.

For the majority of individuals we found no contemporaneous or temporal association between affect and cognitive functioning. For the others, the associations differed in direction, sign and size. This highlights heterogeneity even in a small and relatively homogeneous sample. Future studies should evaluate how individual data can be used in personalizing diagnoses and treatments.

Anxiety and depression are highly prevalent among people with dementia (PWD), and has many negative outcomes, such as increased behavioral problems, decreased ability to live independently, and increased risk of nursing home placement, which may affect the person’s quality of life (QoL). Many cross-sectional studies have investigated factors associated with QoL in PWD in nursing homes, but few longitudinal studies have investigated how anxiety and depression affect the course of QoL of PWD in nursing homes. This study aimed to explore the association between QoL and anxiety, depression, and comorbid anxiety and depression in PWD in nursing homes at a 12-month follow-up.

In all, 298 PWD ≥ 65 years old from 17 Norwegian nursing homes were assessed with Norwegian version of the Rating Anxiety in Dementia scale (RAID-N). Anxiety was defined as RAID-N score ≥ 12. QoL was assessed by Quality of Life in Late-Stage Dementia (QUALID) scale. Depression was assessed by the Cornell scale of depression in dementia (CSDD), defined as CSDD score ≥ 10. The assessments were made at baseline and after a mean follow-up period of 350 days (SD 12.3). A growth mixture model revealed two distinct trajectories of QUALID scores. Participants’ characteristics and related risk factors associated to QUALID score trajectories were analyzed by logistic regression models.

In trajectory group 1 (206 participants), a lower proportion of participants had anxiety, and a more stable and better QoL, compared to trajectory group 2 (92 participants) with a higher proportion of participants with anxiety and poorer QoL at a 12-month follow-up. Participants with comorbid anxiety and depression had the worst QoL than the participants with only anxiety or only depression. Adjusted multivariate logistic regression analysis revealed that more severe impairment in activities of daily living, presence of anxiety, depression, agitation, and use of antipsychotics at baseline, were associated with belonging to the trajectory group with poor QoL.

It is important among nursing home personnel to increase awareness of and skills in identifying anxiety and depression, and to initiate adequate management to enhance Qol of PWD in nursing homes.

Note: The material was published in September 11, 2018 in Plos one.

Our aim is to use the growth mixture model (GMM) to distinguish different trajectories of cognitive change in Chinese geriatric population and identify risk factors for cognitive decline in each subpopulation.

We obtained data from the Chinese Longitudinal Health Longevity Survey, using the Chinese Mini-Mental State Examination (C-MMSE) as a proxy for cognitive function. We applied the GMM to identify heterogeneous subpopulations and potential risk factors.

Our sample included 2850 older adults, 1387 (48.7%) male and 1463 (51.3%) female with age range of 62 to 108 (average of 72.3). Using GMM and best fit statistics, we identified two distinct subgroups in respect to their longitudinal cognitive function: cognitively stable (91.4%) group with 0.42 C-MMSE points decline per 3 years, and cognitively declining (8.6%) group with 4.76 C-MMSE points decline per 3 years. Of note, vision impairment and hearing impairment had the highest associations with cognitive decline, with stronger association found in the cognitively declining group than the cognitively stable group. Cognitive activities were protective in both groups. Diabetes was associated with cognitive decline in cognitive declining group. Physical activities, social activities and intake of fresh vegetables, fruits, and fish products were protective in cognitive stable group.

Using GMM, we identified heterogeneity in trajectories of cognitive change in Chinese elders. Moreover, we found risk factors specific to each subgroup, which should be considered in future studies.

Henry Brodaty, Megan Heffernan, Gavin Andrews, Kaarin Anstey, Maria Fiatarone Singh, Louisa Jorm, Nicola Lautenschlager, Anthony Maeder, John McNeill, Perminder Sachdev, Michael Valenzuela, Maintain Your Brain Collaborative Team.

In the absence of disease modifying interventions for Alzheimer’s disease (AD) and other dementias there is an increased interest in dementia prevention. An issue for population-based lifestyle preventative approaches is scalability. An internet-based multicomponent Maintain Your Brain (MYB) randomised trial is currently underway.

Invitations to participate in Maintain Your Brain were emailed or mailed to people aged 55-77 years from the 45 and Up study, a population-based cohort study of one in ten people aged 45 years and older (n = 267,000). To be enrolled into the study participants were required to have risk factors that made them eligible for at least one of the available modules. Modules were designed to address physical inactivity and health risks associated with inactivity (Physical Activity), adherence to a Mediterranean-type diet and health risks associated with poor nutrition (Nutrition), cognitive activity (Brain Training) and mental wellbeing (Peace of Mind).

During recruitment 96,418 invitations were sent and 14,064 (14%) provided consent. Of those who completed baseline over 90% were eligible for at least two modules of the intervention. Overall, 6,236 (44%) were enrolled resulting in an overall recruitment rate of 6%, or 50% of those eligible at screening. The final sample was 64% female with a mean age of 64.9yrs (SD = 5.8, range 55-77 yrs) and mean years of education of 12.9yrs (SD = 3.0, range 2-22 yrs).

Recruitment rates in MYB were comparable to other clinical trials targeting older people and who included online recruitment strategies. This is promising for the willingness of participants to engage with trials that use online rather than traditional in-person methods. Although the baseline withdrawal number (3%) remained stable after randomisation (4%), a further 25% of participants did not attempt any activities. These rates continued at the annual assessments when 4% withdrew, 31% completed no follow-up activities, 14% partially completed and 55% completed all.

Online lifestyle interventions have capacity to reach broad segments of the 55-77 year old population, the majority of whom may benefit from support activities shown to be effective in reducing risk of cognitive decline and ultimately delay onset of dementia. The challenge with this new approach is encouraging continued engagement with the program over time.

Relationship quality (RQ) in dyads of persons with dementia and their family carers is important both as a clinical outcome and as a determinant of health and quality of life. In previous work we studied RQ using baseline data of a large-scale European longitudinal study on timely access to and use of community formal services in dementia (EU-JPND Acticare). We concluded that neuropsychiatric symptoms and carer stress contributed to discrepancies in RQ ratings within the dyad, which were less favourable when reported by family carers. This and other associations (e.g. between carer-rated RQ and sense of coherence) were cautiously interpreted, in the context of a cross-sectional analysis.

To analyse how carer-reported RQ varies over time and to examine its most important influencing factors.

We present preliminary longitudinal analyses from the Actifcare cohort study of 451 community-dwelling persons with dementia and their primary carers in eight European countries (12-month follow-up). Comprehensive assessments included the Positive Affect Index (PAI) to assess RQ, persons with dementia’s neuropsychiatric symptoms, persons with dementia and carers’ unmet needs, carers’ anxiety and depression, social support, sense of coherence and stress.

Carers’ mean PAI scores decreased over the 12 -month period. The person with dementia neuropsychiatric symptoms and unmet needs, and carers’ perceived social support were significant predictors of carers’ RQ change.

We analysed carer-reported RQ variation over time and predictors in a large European sample of persons with dementia and their family carers. As expected, RQ decreased over the oneyear follow-up period as the disease progressed. Its main predictors in this sample (neuropsychiatric symptoms and the person’s unmet needs, together with carers’ social support) can all influence the impact that caregiving has on the carer and on how time and energy-consuming caregiving is. The role of increased clinical symptoms (also affecting communication difficulties), together with carers’ exhaustion, must be equated. Overall, these results may help us to tailor interventions addressing RQ and potentially improve dementia outcomes.

Research into the experiences of those with a diagnosis of young onset dementia is extremely limited and the research that does exist tends to be loss orientated.

This paper will focus on a doctoral study which aimed to explore the full spectrum of lived experience of those with young onset dementia, whilst considering the unique impact of diagnosis at a younger age and the possibility of personal growth.

Five individuals took part in semi-structured interviews within which they were invited to bring along artefacts that helped them to share their experiences. Transcribed interviews were analysed using Interpretative Phenomenological Analysis and superordinate and subordinate themes were identified.

Four superordinate themes emerged: ‘Fear’, ‘Anger’, ‘Sadness’, ‘Contentment’ with thirteen subordinate themes. The findings indicated that participants experienced feelings of fear and vulnerability in response to their diagnosis. Participants felt angry that they did not have a voice, not enough was being done for those with dementia and they were being stereotyped. Participants spoke of a depressive state of mind in which they grieved for their past self, experienced isolation and loneliness, and feelings of hopelessness and despair. Finally, participants spoke of a sense of contentment in relation to a preserved self, a sense of living alongside their dementia and a desire to live in the present; making the most of the here and now.

The findings highlight the transitory nature of participants’ experiences as a result of multiple and repeated challenges to psychological equanimity. It suggested that these experiences could best be understood using lenses of lifespan development, bereavement and grief. The findings suggest that approaches to dementia care that focus on developing present-mindedness, acceptance and compassion may be particularly helpful.

People with young-onset dementia (YOD) living in nursing homes may experience poor quality of life (QoL) due to advanced dementia, high prevalence of neuropsychiatric symptoms and psychotropic drug use. However, the course of QoL in institutionalized people with YOD and factors that predict this course are unclear. This knowledge could help health professionals identify appropriate interventions to improve QoL in YOD.

To explore the course of QoL in institutionalized people with YOD and resident-related predictors of that course.

Secondary analyses were conducted with longitudinal data from the Behavior and Evolution in Young-ONset Dementia (BEYOND)-II study. A total of 278 people with YOD were recruited from 13 YOD special care units in the Netherlands. QoL was measured by the proxy assessment of Quality of Life in Dementia (QUALIDEM) questionnaire at four assessments over 18 months. Independent variables included age, gender, dementia subtype, length of stay, dementia severity, neuropsychiatric symptoms and psychotropic drug use at baseline. Multilevel modeling adjusted for correlation within nursing homes and residents was used to determine the course and predictors of QoL.

The total QUALIDEM score (range: 0–111) decreased over 18 months with a statistically significant decline of 0.73 points per six months. A significant increase of QoL over time was seen in the subscales “Care relationship”, “Positive self-image”, and “Feeling at home”. However, a significant decline was observed in the subscales “Positive affect”, “Social relations”, and “Something to do”. Residents’ course of QoL was positively associated with the baseline scores of the QoL, age and longer duration of stay; however, being male, having advanced dementia, Alzheimer’s disease and high rates of neuropsychiatric symptoms at baseline were negatively associated with the course of QoL

Longitudinal changes in QoL in residents with YOD were small over 18 months and QUALIDEM subscales showed multidirectional changes. The largest QoL decline in the subscale “Positive affect” suggests that interventions should be targeted to improve positive emotions, in particular for male residents with neuropsychiatric symptoms and advanced dementia.

• Prof.dr. Richard C. Oude Voshaar, University of Groningen, The Netherlands (r.c.oude.voshaar@umcg.nl)

• Prof.dr. Ivan Aprahamian, Faculty of Medicine of Jundiaí, Brazil (ivan.aprahamian@gmail.com)

Depressive disorder has been conceptualised as a disorder of accelerated ageing. Furthermore, meta - analyses have shown that depression associated with excess mortality, although most studies can be criticised for insufficient confounder control. These characteristics of depression resemble the characteristics of biomedical frailty. Biomedical frailty can be assessed with the frailty index (FI) based on the deficit accumulation model. This model postulates that the proportion of at least 30 ageing-related health deficits reflects biological age on top of chronological age. The characteristics of the FI are independent of the specific health deficits included, as long as health deficits from different physiological systems are included. The FI is the most accurate frailty model for predicting mortality, the ultimate outcome of increasing frailty states. Because of its continuous nature, the FI is sensitive to change which enables us to study trajectories of frailty.

The objective of the present lecture is 1) to show that depressive disorder associated with accelerated biological ageing as indexed by the frailty index, 2) that depressive disorder is a risk factor for excess mortality, and 3) to show that this latter explanation is largely explain by biomedical frailty.

The Netherlands Study of Depression in Older persons (NESDO) is a clinical cohort of 378 depressed patients according to DSM-IV criteria and 132 non-depressed controls who have been followed up for six-year. The data of this cohort study enabled us to construct a frailty index based on 41 non-depression related health deficits. Linear mixed models showed that the frailty index had a significantly steeper increase over six years among depressed patients compared to their non-depressed counterparts.

Subsequently, we constructed a frailty index based on 64 health deficits in the LifeLines cohort study, a three-generation population-based study including 167,000 persons. In this population-based cohort, we replicated our finding that the presence of a depressive disorder according to DSM-IV criteria, but also anxiety disorders, were associated with an accelerated increase of the frailty index over a 5-year follow-up. Furthermore, Cox-regression showed that the presence of a depressive disorder or any anxiety disorder was also associated with excess mortality over a 10-year follow-up. These effects remained significant when adjusted for socio-demographic characteristics, lifestyle variables and multimorbidity, but disappeared when adjusted for the frailty index.

We conclude that depressive disorder can indeed be conceptualised as a disorder of accelerated biological ageing, as indexed by the frailty index. Moreover, biomedical frailty may be an explanatory factor for excess mortality found to be associated with affective disorders. Therefore, biomedical frailty seems a promising marker for risk stratification in geriatric psychiatry.

Depression is common in nursing homes, with newly admitted residents at a particularly high risk. Current prevention and treatment approaches have failed to impact on the high rates of depression and new approaches are required. We have developed a novel intervention that addresses individual psychological needs in newly admitted residents, which is implemented in collaboration with residents and facility staff. The Program to Enhance Adjustment to Residential Living (PEARL) is a 5-session intervention based on Self-Determination Theory that aims to tailor care to enhance residents’ autonomy, competence and relations.

A cluster randomised controlled trial was conducted to determine the effectiveness of PEARL in reducing depression in newly admitted nursing home residents, compared to a treatment as usual control condition. A total of 216 residents with normal cognition or mild-moderate cognitive impairment, living in 42 nursing homes in Melbourne, Australia, participated in the study. The sample included 76 men and 140 women, aged 62-99 years (M = 85.5, SD = 7.3). The primary outcome was level of depressive symptoms, assessed using the Cornell Scale for Depression in Dementia at baseline and at 8, 16 (primary endpoint), and 31 weeks follow-up. Multi-level modelling, accounting for within-facility clustering and repeated assessments, was used to determine the effectiveness of the intervention, employing an intention-to-treat approach.

At 16-weeks, fewer symptoms of depression were observed among those in the intervention group (M = 7.0, SD = 5.6) than in the control group (M = 8.6, SD = 6.5). Multi-level modelling of unadjusted data showed a significant condition (intervention, control) by time (T1, T2, T3) interaction (p = .021), indicating that the PEARL intervention led to a reduction in the occurrence of depressive symptoms. The treatment effect was maintained at 31 weeks follow-up (p = .004).

This study provides evidence that an intervention designed to address newly admitted residents’ individual psychological needs reduced symptoms of depression in subsequent months. This brief, simple intervention may be suitable for broad implementation across long-term care settings. However, future research to determine if the intervention can be implemented by trained facility staff in situ is warranted.

Depression and anxiety are two very common psychiatric disorders in late-life. They are markers of poor quality of life and are strongly associated with death among older people. Yet, few studies on these comorbidities have been conducted in the African population. This study aims to present the epidemiology of depression and anxiety among older people in Central Africa.

A cross-sectional population-based study was carried out in Republic of Congo (ROC) and Central African Republic (CAR) between 2011 - 2012 among older people aged ? 65 years (EPIDEMCA study). Data were collected using a standardized questionnaire and participants underwent a brief physical examination. Depression and anxiety symptoms were ascertained using a community version of the Geriatric Mental State (GMS-B3) and the Automated Geriatric Examination for Computer Assisted Taxonomy diagnostic system (AGECAT), probable cases were defined as having a GMS-AGECAT level of 3 or more. Logistic regression models were used to investigate the association between potential risk factors collected and each symptom.

Overall 2002 participants were included in the EPIDEMCA study (500 in Brazzaville and 529 in Gamboma in ROC, 500 in Bangui and 473 in Nola (473) in CAR). Median age of the participants was 72 years [interquartile range: 68 – 78 years] and females were mostly represented (61.8%). Prevalence was 38.1% (95% Confidence Interval: 35.9% - 40.2%) for depression, 7.7% (95% CI: 6.5% - 8.9%) for anxiety and 5.7% (95% CI: 4.6% - 6.7%) for the co-occurrence of both disorders. For all three outcomes, prevalence was significantly higher among females and in rural areas. Only depression increased with age. Preliminary analyses showed that female sex, living in a rural area, and living without a partner were associated with the three outcomes (Odds Ratios from 1.59 to 3.27; p<0.01). In-depth results regarding correlates of depression, anxiety and the co-occurrence of both will be presented.

The prevalence of depression and anxiety was high among Central African older people. Evidence on the epidemiology of these common psychiatric symptoms are of importance for care management and also emphasize the need to maintain and/or strengthen social support around older people in the region.

role of global cognition in the association between Anemia and depression.

We examined the longitudinal relationships between hemoglobin concentrations or Anemia and depression and whether baseline cognitive function modifies these longitudinal relationships over 4 years of follow-up.

A total of 1608 community-dwelling older adults from the International Mobility in Aging Study (IMIAS) aged 65 to 74 years were recruited in Natal (Brazil), Manizales (Colombia), Kingston (Ontario, Canada), and Saint-Hyacinthe (Quebec, Canada). The study outcome was depression, defined by a score of 16 or over in the Center for Epidemiologic Studies Depression Scale (CES-D). Longitudinal associations over four years follow-up were examined using generalized estimating equations. Models reported were either unadjusted and adjusted for research sites, alcohol drinking status, body mass index, chronic conditions, activities of daily life disabilities, and polypharmacy.

Longuitinal relationships suggested an evidence of multiplicative interaction by baseline global cognition in which 1g/dL increase in hemoglobin concentrations there was a significant reduction in the risk of depression with a stronger effect among participants with good cognitive function (Odds Ratio (OR)=0.85, 95% CI: 0.78-0.92) compared to those with poor cognition (OR=0.89, 95% CI: 0.80-0.97). Anemia and poor cognition at baseline were associated with an increased risk of depression over four years of follow-up (OR=5.80, 95% CI: 1.84-18.23). Global cognition was also an effect modifier of the longitudinal association between the severity of Anemia and depression.

In international samples of older adults, hemoglobin concentrations, as well as the severity of Anemia, were independent risk factors for depression, and these associations differed by global cognitive function.

Kerstin Johansson, Karolina Thömkvist, Ingmar Skoog and Sacuiu SF* (*presenter)

To determine the effects of electroconvulsive therapy (ECT) in major depression in relation to the development of dementia during long-term follow-up.

In an observational clinical prospective study of consecutive patients 70 years and older diagnosed with major depression at baseline 2000-2004 (n=1090), who were free of dementia and received antidepressant treatment, with or without ECT, we sought to determine if cognitive decline (mild cognitive impairment and dementia) during 15 -year follow-up was associated with receiving ECT at baseline. The control group was selected among the participants in the Gothenburg H70 Birth Cohort Studies matched by age group and sex 1:1.

Among patients with affective syndromes 7% received ECT. During follow-up, 157 patients were diagnosed with dementia, equal proportions among those who received ECT (14.5%) and those who did not receive ECT (14.5%). The relation between ECT and cognitive decline remained non-significant irrespective antidepressive medication or presence of mild cognitive impairment at baseline.

Preliminary results indicate that ECT was not associated with the development of cognitive decline in the long-term in a hospital-based cohort of 70+ year-olds. The results remain to verify against controls from a representative community sample.