Deriving prevalence estimates of depressive symptoms throughout middle and old age in those living in the community

Practical Strategies for Extreme Missing Data Imputation in Dementia Diagnosis

Accurate computational models for clinical decision support systems require clean and reliable data but, in clinical practice, data are often incomplete. Hence, missing data could arise not only from training datasets but also test datasets which could consist of a single undiagnosed case, an individual. This work addresses the problem of extreme missingness in both training and test data by evaluating multiple imputation and classification workflows based on both diagnostic classification accuracy and computational cost. Extreme missingness is defined as having ~50% of the total data missing in more than half the data features. In particular, we focus on dementia diagnosis due to long time delays, high variability, high attrition rates and lack of practical data imputation strategies in its diagnostic pathway. We identified and replicated the extreme missingness structure of data from a real-world memory clinic on a larger open dataset, with the original complete data acting as ground truth. Overall, we found that computational cost, but not accuracy, varies widely for various imputation and classification approaches. Particularly, we found that iterative imputation on the training dataset combined with a reduced-feature classification model provides the best approach, in terms of speed and accuracy. Taken together, this work has elucidated important factors to be considered when developing a predictive model for a dementia diagnostic support system.

Age, sex and period estimates of Australia's mental health over the last 17 years

OBJECTIVE

Describing the long-term mental health of Australians is limited as many reports rely on cross-sectional studies which fail to account for within-person changes and age-related developmental processes which may bias estimates which ignore these phenomena. We examined the 17-year trajectories of mental health in 27,519 Australian adults.

METHODS

Household panel data of 27,519 participants aged 18 years and over from the Household, Income and Labour Dynamics in Australia Survey provided at least one observation of mental health over a 17-year period from 2001. On average, participants reported 7.6 observations. Mental Health was assessed annually using the Short-Form Health Survey-36 mental health scale.

RESULTS

Over time, there were only very small changes in mental health and only for the youngest and oldest adults. Over time, there was consistent evidence for better metal health with increasing age, although for the very old, there appear to be substantial declines. These patterns were consistent between sex. In line with an existing literature, males reported better mental health over life span, although the declines of mental health in very-late-life are particularly pronounced for males.

CONCLUSION

Decline in mental health was only reported by the youngest and oldest respondents, and was notable only in the last 4-5 years. However, the magnitude of the decline was small and further follow-up will be needed to determine whether this is a trend of substantive declining mental health for these specific age cohorts. In contrast, the more consistent finding is that there has been no substantive change in the level of mental health in Australia over the last 17 years. Analysis of the mental health trajectories of baseline age-cohorts confirmed that age differences are consistent over time.

Affective disorders in the elderly in different European countries: Results from the MentDis_ICF65+ study

OBJECTIVES

Affective disorders are among the most prevalent disorders in the elderly. The present study aims to examine the sociodemographic and clinical correlates of major depressive disorder (MDD) and dysthymia in different European and Associated countries using standardized interview techniques. Furthermore, service utilization for the elderly with depression is assessed.

METHODS

The MentDis_ICF65+ study is a cross-sectional survey (N = 3,142) that was conducted in six different European and Associated countries (Germany, Italy, Spain, Switzerland, England and Israel) with a subsample of n = 463 elderly with any depressive disorder.

RESULTS

Sociodemographic and clinical correlates, such as gender, age and symptom severity, were significantly associated with MDD and dysthymia in the elderly. Only 50% of elderly with any depressive disorder were treated with psycho- or pharmacotherapy.

CONCLUSION

Our findings identified sociodemographic and clinical characteristics for depression risk in the elderly and highlight the need to improve service delivery to older adults who suffer from depression.

The impact of childhood parental quality on mental health outcomes in older adults

OBJECTIVE

Parental bonding is cited as a determinant of mental health outcomes in childhood, adolescence and early-mid adulthood. Examination of the long-term impact for older adults is limited. We therefore examine the long-term risk of perceived poor parental bonding on mental health across the lifespan and into early-old age.

METHODS

Participants (N = 1255) were aged 60-64 years of age and drawn from the Australian Life Histories and Health study. Quality of parental bonding was assessed with the Parental Bonding Instrument (PBI). Self-reported history of doctors' mental health diagnoses and current treatment for each participant was recorded. Current depression was assessed with the Centre for Epidemiologic Studies Depression-8 (CESD-8). Due to known gender differences in mental health rates across the lifespan, analyses were stratified by sex.

RESULTS

A bi-factor analysis of the PBI in a structural equation framework indicated perceived Poor Parental Quality as a risk for both ever and current depression for both sexes. For males, Over-Protective Fathers were a risk for ever and current depression, whilst overall Poor Parental Quality was a risk for reporting current depression treatment. Whilst a number of the risks associated with current depression and treatment were attenuated when controlling for current mood, parental quality remained a significant risk for having reported a lifetime diagnosis for depression and anxiety for men.

CONCLUSION

Our results extend the existing literature base and demonstrate that mental health risk attributed to poor perceived parental quality continues across the life-course and into early-old age.

Depressive symptoms in older adults awaiting cataract surgery

BACKGROUND

To assess the prevalence and predictors of depressive symptoms in a cohort of older adults awaiting cataract surgery and establish threshold vision at which depressive symptoms may emerge.

DESIGN

Analysis of cross-sectional baseline data from a longitudinal cohort study of patients aged ≥65 years on Australian public hospital cataract surgery waiting lists.

PARTICIPANTS

We included 329 participants enrolled October 2013-August 2015.

METHODS

Participants completed assessment of depressive symptoms, visual disability, quality of life, social participation and exercise frequency at least one month prior to cataract surgery. High and low contrast habitual vision was examined and systemic comorbidities noted.

MAIN OUTCOME MEASURE

Depressive symptoms prior to first eye cataract surgery.

RESULTS

The prevalence of depressive symptoms was 28.6% (94/329). Univariate analysis identified that participants with poorer high contrast vision, reduced quality of life, greater patient-reported visual disability, higher comorbidity score and who were taking more medications were more likely to exhibit signs of depression. Greater patient-reported visual disability (P = 0.02), reduced quality of life (P = 0.003) and a higher comorbidity score (P = 0.02) remained significantly associated with depressive symptoms in the multivariable model. Depressive symptoms emerged at a visual acuity of 6/12.

CONCLUSIONS

These findings demonstrate a high prevalence of depressive symptoms in older persons with cataract, emerging at modest levels of vision loss. Efficient referral processes, timely surgical management, and improved screening and coordinated treatment of depressive symptoms during the surgical wait may minimize the negative psychological effects of cataract in this already vulnerable population.

Prevalence of depressive symptoms and suicidal thoughts among elderly persons in rural Bangladesh

BACKGROUND

Depression, if broadly defined, is the commonest late-life mental disorder. We examined the distribution of depressive symptoms and suicidal thoughts, across age, sex, literacy, and marital status, among elderly individuals residing in rural Bangladesh and participating in a population-based study on health and aging.

METHODS

Prevalence figures of depressive symptoms were assessed with SRQ20 (n = 625), and possible social network and economic associations were examined. Morbidity accounts of depressive symptoms and suicidal thoughts were examined for a subsample that also underwent complete medical examination (n = 471).

RESULTS

We selected for analyses the items that corresponded to DSM-IV criteria and constructed a dichotomous variable. The prevalence was 45%, and most pronounced among the oldest women (70%). The overall prevalence of suicidal thoughts was 23%. Being a woman, illiterate or single were all risk factors for depressive symptoms and suicidal thoughts. These associations remained unaccounted for by the social network and economic variables. Co-residing with a child and having a high quality of contact were protective of both depressive symptoms and suicidal thoughts. The main findings were replicated in the subsample, where it was found that morbidities were also associated with the outcomes, independently of the four main predictors.

CONCLUSIONS

Prevalence figures for depressive symptoms among elderly in rural Bangladesh are high. Demographic, social network, and morbidity factors are independently associated with both depressive symptoms and suicidal thoughts. This is the first study to report prevalence figures for depressive symptoms in this population.

Soluble amyloid-β levels and late-life depression

Late-Life Major Depression (LLMD) is a complex heterogeneous disorder that has multiple pathophysiological mechanisms such as medical comorbidity, vascular-related factors and Alzheimer's disease (AD). There is an association between LLMD and AD, with LLMD possibly being a risk factor for, or early symptom of AD and vascular dementia. Whether depression is an etiologic risk factor for dementia, or part of the dementia prodrome remains controversial. AD has a long prodromal period with the neuropathologic features of the disease preceding the onset of clinical symptoms by as much as 15-20 years. Clinicopathological studies have provided robust support for the importance of Aβ42 in the pathogenesis of AD, but several other risk factors have also been identified. Given the relationship between Aβ42 and AD, a potential relationship between Aβ42 and LLMD would improve the understanding of the association between LLMD and AD. We reviewed 15 studies that analyzed the relationship between soluble Aβ42 and LLMD. For studies looking at plasma and/or cerebrospinal fluid (CSF) levels of Aβ42, the relationship between LLMD and soluble Aβ42 was equivocal, with some studies finding elevated Aβ42 levels associated with LLMD and others finding the opposite, decreased levels of Aβ42 associated with LLMD. It may be that there is poor reliability in the diagnosis of depression in late life, or variability in the criteria and the scales used, or subtypes of depression in late life such as early vs. late onset depression, vascular-related depression, and preclinical/comorbid depression in AD. The different correlations associated with each of these factors would be causing the inconsistent results for soluble Aβ42 levels in LLMD, but it is also possible that these patterns derive from disease stage-dependent differences in the trajectory of CSF Aβ42 during older age, or changes in neuronal activity or the sleep/wake cycle produced by LLMD that influence Aβ42 dynamics.

Gender differences in the trajectories of late-life depressive symptomology and probable depression in the years prior to death

BACKGROUND

Gender differences in depression are well established. Whether these differences persist into late life and in the years preceding death is less clear. There is a suggestion that there is no increased likelihood of depression in late life, but that there is an increase in depressive symptomology, particularly with proximity to death. We compared trajectories of probable depression and depressive symptomology between men and women over age and distance-to-death metrics to determine whether reports of depressive symptoms are more strongly related to age or mortality.

METHODS

Participants (N = 2,852) from the Dynamic Analyses to Optimise Ageing (DYNOPTA) project had a mean age of 75 years (SD = 5.68 years) at baseline and were observed for up to 16 years prior to death. Multi-level regression models estimated change in depressive symptomology and probable depression over two time metrics, increasing age, and distance-to-death.

RESULTS

Increases in depressive symptomology were reported over increasing age and in the years approaching death. Only male participants reported increased probable depression in the years preceding death. Models that utilized distance-to-death metrics better represented changes in late-life depression, although any changes in depression appear to be accounted for by co-varying physical health status.

CONCLUSIONS

As death approaches, there are increases in the levels of depressive symptomology even after controlling for socio-demographic and health covariates. In line with increases in suicide rates in late life, male participants were at greater risk of reporting increases in depressive symptomology.

What characterizes late-life depression?

This article analyzes late-life depression, looking carefully at what defines a person as elderly, the incidence of late-life depression, complications and differences in symptoms between young and old patients with depression, subsyndromal depression, bipolar depression in the elderly, the relationship between grief and depression, along with sleep disturbances and suicidal ideation.

Alcohol, hospital admissions, and falls in older adults: a longitudinal evaluation

BACKGROUND

There are limited data on the impacts of alcohol use in older adults. We aimed to evaluate self-reported hospital admissions and falls against current Australian alcohol consumption guidelines.

METHODS

We conducted a longitudinal analysis of data from five Australian cohort studies. The study comprised 16,785 people aged 65 years or older at baseline. Alcohol consumption was categorized using Australian guidelines in standard (10 g) drinks per day as "abstinent," "low-risk" (>0 ≤2), "long-term risk" (>2 ≤4), or "short-term risk" (>4). Separate generalized estimating equations for men and women, controlling for key demographic, and health variables (depression, diabetes, circulatory and musculoskeletal conditions) were used to examine the relationship of alcohol consumption with hospitalization and falls against a reference category of low-risk consumption.

RESULTS

Most participants were in the low (10,369, 62%) or abstinent (5,488, 33%) categories. Among women, all alcohol groups had greater odds of admission than low-risk users; among men, only the abstinent group had increased odds. For both genders, depression, diabetes, circulatory and musculoskeletal conditions all increased the odds of admission. For both genders, the unadjusted model showed that abstainers had increased odds of falling, with depression, diabetes, and for women, musculoskeletal conditions also associated with falls in the adjusted model.

CONCLUSION

These outcomes suggest that older women in particular could benefit from targeted alcohol consumption messages or interventions. In relation to falls, other health conditions appear better targets for intervention than alcohol use.

Alcohol and smoking consumption behaviours in older Australian adults: prevalence, period and socio-demographic differentials in the DYNOPTA sample

PURPOSE

Alcohol consumption and tobacco use are key risk factors for chronic disease and health burden across the adult lifespan. We estimate the prevalence of alcohol consumption and smoking by age and time period in adults from mid to old age.

METHODS

Participants (n = 50,652) were drawn from the Dynamic Analyses to Optimise Ageing (DYNOPTA) project and were compared with Australian National Health Survey data. Alcohol and smoking consumption DYNOPTA data were weighted to the estimated resident population of the sampling frame for each contributing study according to age and sex distributions within major statistical regions.

RESULTS

Comparisons in the rates of smoking and alcohol consumption between DYNOPTA and other national surveys were comparable. Males were more likely to be (RRR = 2.12) or have been smokers (RRR = 2.97), whilst females were more likely to be non-drinkers (RRR = 2.52). Period effects were also identified; higher prevalence rates in consumption of alcohol (RRR = 3.21) and smoking (RRR = 1.67) for those contributing studies from the early 1990's, in comparison with those studies from the latter half of the decade, were reported.

CONCLUSIONS

Over a decade, prevalence rates for high-risk consumption of alcohol and current smoking behaviour declined and suggest the possible impact of government health policy, with targeted-health policies, that included bans on public smoking, and a toughening of legislation against alcohol-related crime.

Stability and change in level of probable depression and depressive symptoms in a sample of middle and older-aged adults

BACKGROUND

Findings from studies investigating depression in adults in late life are mixed due to a lack of large longitudinal studies with the power necessary to yield reliable estimates of stability or change. We examined the long-term stability of probable depression and depressive symptomology over a 13-year period in the Dynamic Analyses to Optimize Ageing (DYNOPTA) project.

METHODS

Community-living participants (N = 35,200) were aged 45-103 at baseline, predominantly female (79%), partnered (73%), and educated to secondary school only (61%) and followed for up to 13 years.

RESULTS

At baseline, increased age was associated with lower prevalence of probable depression and depressive symptomology. Over time, prevalence of probable depression was stable while levels of depressive symptomology reported a small decline. However, this finding was not consistent for all age groups; there was evidence for increasing levels of depressive symptomology, but not probable depression, as individuals aged. This effect was particularly notable among males aged 70 plus years.

CONCLUSIONS

These results answer important questions relating to the longitudinal prevalence of probable depression and depressive symptomology in a sample of older Australians. These findings have policy implications for mental health service provision for older adults.

Correlates of subjective health across the aging lifespan: understanding self-rated health in the oldest old

OBJECTIVES

To examine the determinants of self-rated health (SRH) in different age groups of older adults, including the oldest old.

METHODS

Variables assessing physical health, difficulty with self-care, depressive symptoms, and cognitive impairment were pooled and harmonized from three Australian longitudinal studies of ageing (N = 5,222). The association of these with SRH was examined in older adults aged 60 to 64 years, 65 to 74 years, 75 to 84 years, and 85 years and older.

RESULTS

SRH was not associated with cognitive impairment or difficulty with self-care in the oldest old, and its association with physical health was diminished compared with younger groups. Depression showed a significant relationship in all age groups, conferring an approximately fourfold increase in the likelihood of poorer SRH.

DISCUSSION

As old age progresses, self-reports of poor health become most closely related to psychological symptoms. This explains some of the paradoxes of past literature and offers important insights for health professionals working with the oldest old.

Alcohol use and depression from middle age to the oldest old: gender is more important than age

BACKGROUND

Alcohol use disorders are associated with other mental health disorders in young adults, but there are few data on alcohol use and mental health outcomes in older adults, particularly the oldest old. This study examines the relationship between alcohol consumption and depressive symptoms.

METHODS

Data were collected from the Dynamic Analyses to Optimise Ageing (DYNOPTA) project, which has pooled nine Australian longitudinal studies. Alcohol consumption was classified using standard drinks (10 g alcohol)/day as: abstinent, low risk (<0-≤ 2 standard drinks), long-term risk (>2-≤ 4) and short-term risk (>4). Probable depression was classified from harmonized scores on various standard instruments (e.g. Centre for Epidemiological Studies Depression scale).

RESULTS

Overall, 39,104 (86%) participants contributed data. Alcohol classification at baseline showed 7,526 abstinent, 28,112 low risk, 2,271 long-term risk, and 1,195 short-term risk participants. Age ranged from 45 to 103 year (median 60). Using generalized estimating equations (GEE), there were significant gender by alcohol and gender by age interactions, so the analysis was split by gender. Among males, the abstinent and short-term risk groups had increased likelihood of depression: in females the abstinent, long- and short-term risk groups had increased odds of depression. Increased odds of depression was also associated with former and current smoking, younger age-group, not being partnered, leaving school before age 15 and increasing levels of health-impaired walking, dressing, or bathing.

CONCLUSION

The impact of alcohol use differs by gender, nevertheless those using higher levels of alcohol or who smoke should be screened for depression and may benefit from interventions.

Depressão em Idosos de uma Comunidade assistida pela Estratégia de Saúde da Família em Passo Fundo, RS

Objetivo: Identificar a prevalência de depressão em idosos atendidos pela Estratégia de Saúde da Família (ESF) e os fatores associados. Métodos: Estudo transversal com 151 idosos residentes em área adscrita à ESF de Passo Fundo/RS. Coletaram-se os dados utilizando questionário estruturado e Escala de Depressão Geriátrica GDS–15. Resultados: 33,1% são do sexo masculino, com faixa etária predominante entre 60-69 anos (50,3%), escolaridade média de 4,14 anos de estudo (DP: 2,63), 8,2% são analfabetos ou semianalfabetos, 21,9% declararam algum episódio depressivo e 2,6% outra doença psiquiátrica. Houve associação entre depressão e percepção de saúde, perda familiar, asma, ocorrência de fratura, insuficiência cardíaca e artrite (p < 0,01). Conclusões: os resultados mostraram que a prevalência de depressão em idosos na ESF estudada foi de 21,2%, desses, 17,9% com diagnóstico de depressão leve a moderada e 3,3% com depressão severa, necessitando de intervenções multidisciplinares e integrais visando reduzir os fatores associados e qualificar a vida dessas pessoas.