Screening major depressive disorder using vocal acoustic features in the elderly by sex.

BACKGROUND: Vocal acoustic features are potential biomarkers of elderly depression. Previous automated diagnostic tests for depression have employed unstandardized speech samples, and few studies have considered differences in voice reactivity. We aimed to develop a voice-based screening test for depression measuring vocal acoustic features of elderly Koreans while they read a series of mood-inducing sentences (MIS). METHODS: In this case-control study, we recruited 61 individuals with major depressive disorder and 143 healthy controls (mean age [SD]: 72 [6]; female, 70%) from the community-dwelling elderly population. Participants were asked to read MIS and their variation pattern of acoustic features represented by the correlation distance between two MIS were analyzed as input features using the univariate feature selection technique and subsequently classified by AdaBoost. RESULTS: Acoustic features showing significant discriminatory performances were spectral and energy-related features for males (sensitivity 0.95, specificity 0.88, and accuracy 0.86) and prosody-related features for females (sensitivity 0.73, specificity 0.86, and accuracy 0.77). The correlation distance between negative and positive MIS was significantly shorter in the depressed group than in the healthy control (F = 18.574, P < 0.001). LIMITATIONS: Small sample size and relatively homogenous clinical profile of depression could limit the generalizability. CONCLUSIONS: While reading MIS, spectral and energy-related acoustic features for males and prosody-related features for females are good discriminators for major depressive disorder. These features may be used as biomarkers of depression in the elderly.

Sleep and cognitive decline: A prospective nondemented elderly cohort study.

OBJECTIVE: To investigate sleep disturbances that induce cognitive changes over 4 years in nondemented elderlies. METHODS: Data were acquired from a nationwide, population-based, prospective cohort of Korean elderlies (2,238 normal cognition [NC] and 655 mild cognitive impairment [MCI]). At baseline and 4-year follow-up assessments, sleep-related parameters (midsleep time, sleep duration, sleep latency, subjective sleep quality, sleep efficiency, and daytime dysfunction) and cognitive status were measured using the Pittsburgh Sleep Quality Index and Consortium to Establish a Registry for Alzheimer's Disease Assessment, respectively. We used logistic regression models adjusted for covariates including age, sex, education, apolipoprotein E genotype, Geriatric Depression Scale, Cumulative Illness Rating Scale, and physical activity. RESULTS: In participants with NC, long sleep latency (>30 minutes), long sleep duration (≥7.95 hours), and late midsleep time (after 3:00 am) at baseline were related to the risk of cognitive decline at 4-year follow-up assessment; odds ratio (OR) was 1.40 for long sleep latency, 1.67 for long sleep duration, and 0.61 for late midsleep time. These relationships remained significant when these variables maintained their status throughout the follow-up period. Newly developed long sleep latency also doubled the risk of cognitive decline. In those with MCI, however, only long sleep latency reduced the chance of reversion to NC (OR = 0.69). INTERPRETATION: As early markers of cognitive decline, long sleep latency can be used for elderlies with NC or MCI, whereas long sleep duration and relatively early sleep time might be used for cognitively normal elderlies only. Ann Neurol 2018;83:472-482.

Impacts of Illiteracy on the Risk of Dementia: A Global Health Perspective.

Despite its significance as a contributing factor for late-life dementia risk, illiteracy is frequently underappreciated in the management of dementia. In this study, we estimated the proportion of dementia cases attributable to illiteracy using the population attributable fraction (PAF), and calculated to what extent the monetary cost of dementia could be saved in the future by reducing illiteracy from the South Korean, Latin American, South Asian/Middle Eastern, and African populations. We collected necessary data from the 2011 United Nations Human Development Report and prevalence studies conducted in these regions. Additional variables not included in the above sources were estimated using a logit model under a "trend scenario"-based assumption. Around 16% of the total number of dementia cases in South Korea in 2015 can be attributed to illiteracy, with this figure predicted to decline to around 2% by 2050. This translates to a saving in dementia care costs of approximately 52 billion USD, providing we are successful in theoretically eradicating illiteracy as of 2015, in the population aged 65 years or under. Likewise, reducing illiteracy to 50% in Latin America, South Asia/The Middle East, and Africa by 2050 could generate further cost savings of between 71 and 244 billion, 13 and 94 billion, and 17 and 78 billion USD, respectively. Even public policies aimed solely at reducing illiteracy in the childhood, adolescent, or middle-aged population could potentially have a role in the primary prevention of dementia. Moving forward, governments will need to address this issue in a purposeful and systematic manner.