An automatic robust Bayesian approach to principal component regression.

Principal component regression uses principal components (PCs) as regressors. It is particularly useful in prediction settings with high-dimensional covariates. The existing literature treating of Bayesian approaches is relatively sparse. We introduce a Bayesian approach that is robust to outliers in both the dependent variable and the covariates. Outliers can be thought of as observations that are not in line with the general trend. The proposed approach automatically penalises these observations so that their impact on the posterior gradually vanishes as they move further and further away from the general trend, corresponding to a concept in Bayesian statistics called whole robustness. The predictions produced are thus consistent with the bulk of the data. The approach also exploits the geometry of PCs to efficiently identify those that are significant. Individual predictions obtained from the resulting models are consolidated according to model-averaging mechanisms to account for model uncertainty. The approach is evaluated on real data and compared to its nonrobust Bayesian counterpart, the traditional frequentist approach and a commonly employed robust frequentist method. Detailed guidelines to automate the entire statistical procedure are provided. All required code is made available, see ArXiv:1711.06341.

Factors involved in sleep efficiency: a population-based study of community-dwelling elderly persons.

STUDY OBJECTIVES: Research indicates that sleep efficiency below 80% substantially increases mortality risk in elderly persons. The aim of this study was to identify factors that would best predict poor sleep efficiency in the elderly, and to determine whether associations between these factors and sleep efficiency were similar for men and women and for younger and older elderly persons. METHODS: A total of 2468 individuals aged 65-96 years (40.7% men) participated. They were recruited via random generation of telephone numbers according to a geographic sampling strategy. The participants agreed to have health professionals visit their home and to answer structured interview questions. Sleep efficiency was calculated based on interview responses. Descriptive statistics and logistic regressions were conducted. RESULTS: The factors most strongly associated with sleep efficiency below 80% were pain, nocturia, sleep medication use, and awakening from bad dreams. Some factors varied by sex: women aged 75 years and older or who had an anxiety disorder were more likely to have sleep efficiency below 80%, whereas being single or having painful illness raised the likelihood for men only. Except for sex, all the factors that showed associations with sleep efficiency affected younger and older elderly persons similarly. CONCLUSIONS: Poor sleep efficiency is prevalent among elderly persons. The results shed new light on factors associated with poor sleep efficiency, highlighting the presence of sex differences and that certain factors make no significant contribution, such as typically proscribed sleep hygiene behaviors, mood disorders, and illness in general.

Sleep Environment and Insomnia in Elderly Persons Living at Home.

The aim of this study was to draw a portrait of the sleep environment of elderly persons living in private households and to determine its relationship with the presence of insomnia. A sample of 599 individuals aged 70 years and older responded to questions about the comfort of their pillow and mattress and the noise level and brightness of their bedroom at night and in the morning. They were also asked whether or not they shared their bed or bedroom with a sleep partner. The Insomnia Severity Index was used to assess insomnia severity. Over 40% of the study participants were using a pillow that was not very comfortable, and almost 30% said that their bedroom was not completely quiet. Binomial logistic regression results revealed that two variables were significantly associated with insomnia symptoms: a pillow rated as moderately comfortable to very uncomfortable and a bedroom that was not completely quiet. No other sleep environment characteristics considered in this study were associated with the risk of insomnia. These results indicate that a nonnegligible proportion of the elderly population endures a suboptimal sleep environment. Although it is difficult to predict the real impact of changes to the sleep environment, this study supports the proposal that simple, minor changes to the bedroom can promote sleep in the elderly.

Sleep cognitions associated with anxiety and depression in the elderly.

PURPOSE: The objective of this study was to identify the maladaptive sleep-related cognitions most often maintained by the elderly, according to the presence or absence of anxiety and mood disorders. The presence of dysfunctional sleep-related beliefs and attitudes at bedtime in asymptomatic, depressive, and anxious seniors was thus compared. The second objective was to verify the relationships between various dysfunctional cognitions and mental disorders. METHOD: The sample in this study consisted of 2,759 participants aged 65 years and over, with a mean age of 73.8 years. They were recruited through a method of random generation of telephone numbers according to a sampling strategy based on geographic location. After the goal of the study was explained to them, the participants agreed to have health professionals visit their home and to answer questions in a 1.5-hour-long structured interview (after signing a consent form). RESULTS: Depressive and anxious seniors adopt dysfunctional sleep-related cognitions in higher proportions than asymptomatic older persons. Once we had controlled for the other factors, we were able to specifically link two sleep-related beliefs and all the sleep-related attitudes studied to the probability of being anxious or depressive. CONCLUSION: The clarifications obtained will make it possible to improve detection, assessment, and intervention processes regarding anxiety or mood disorders, by pinpointing the most direct link between each of the dysfunctional cognitions and the two types of mental disorders, and not just the link to sleep problems.

Sleep problems in anxious and depressive older adults.

PURPOSE: The objective of this study was to identify the sleep problems most often encountered by the elderly according to the presence or absence of anxiety and mood disorders. The aim was also to determine whether groups of anxious, depressive, and asymptomatic individuals differ in relation to sleep onset latency; awakenings at night or early in the morning; subjective quality of sleep; taking of sleep medication; and daytime sleepiness. METHODS: Structured interviews based on the DSM-IV-TR were administered to a sample of 2,759 seniors aged 65 years and older at the participants' home by health professionals. RESULTS: Awakening was found to be the most common disturbance. Increased sleep onset latency was the second most frequent sleep difficulty. Taking more than 30 minutes to fall asleep was associated with the likelihood of meeting the diagnostic criteria for an anxiety disorder, and even reduced the risk of meeting the diagnostic criteria for a mood disorder rather than an anxiety disorder. Awakenings were associated with the probability of suffering from an anxiety disorder or a mood disorder. Quality of sleep, as perceived by the elderly, was not found to be associated with the probability of suffering from a mental disorder. CONCLUSION: These findings should help to facilitate the practitioner's diagnosis and add further nuances to be considered when encountering symptoms of an anxious or depressive appearance. All of these data also add fuel to the ongoing debate about whether anxiety and depression are one or two distinct categories of disorders.

The relationship between sleep habits, anxiety, and depression in the elderly.

PURPOSE: The objective of this study is to determine which sleep-related behaviors are most often used by the elderly according to the presence or absence of anxiety and mood disorders. In particular, we are attempting to determine whether these behaviors are associated with the probability of suffering from a mental disorder. The behaviors being examined in the present study are taking naps, television watching or reading at bedtime, physical exercise at bedtime, relaxing activities at bedtime, and caffeine consumption in the evening. METHODS: The sample in this study consists of 2,759 participants aged 65 and over, with a mean age of 73.8. They were recruited through a method of random generation of telephone numbers according to a sampling strategy based on geographic location. After the goal of the study was explained to them, the participants agreed to have health professionals visit their home and to answer questions in an hour-and-a-half-long structured interview (after signing a consent form). RESULTS: Taking naps is the activity most often practiced by the elderly. Watching television and reading at bedtime are also frequent practices among them. The probabilities of suffering from anxiety are greater if the person never or rarely consumes caffeine after 6 pm, if the individual takes naps during the day, or if the person practices relaxation before bedtime. Television watching, reading, and physical exercise before bedtime are activities that are not associated with the probability of suffering from a mental disorder. CONCLUSION: It would be beneficial for research to be conducted to support the findings on behavioral differences between depressive and anxious seniors so that these behaviors can become further indicators of the presence of mental disorders.

Trends and determinants of antiresorptive drug use for osteoporosis among elderly women.

AIM: It has been established that women who have had a first osteoporotic fracture are at a significantly greater risk of future fractures. Effective antiresorptive treatments (ART) are available to reduce this risk, yet little information is available on trends in ART drug use among the elderly. The objective is to estimate the rate ratio (RR) of having an ART prescription filled among elderly women and its relation to selected determinants from 1995 through 2001. METHOD: A cohort design was used. Through random sampling, we selected 40% of the women aged 70 years and older listed in the Régie de l'assurance maladie du Québec (RAMQ) health database. The women were grouped into four cohorts (for 1995, 1996, 1998 and 2000). January 1 was established as the index date within each cohort (1995, 1996, 1998 and 2000). The dependent variable was the RR of having at least one prescription of ART drugs filled during the year following the index date among women with and without prior use. ART users were divided in two groups: bone-specific drugs (bisphosphonates, calcitonin, raloxifen) and HRT (hormone replacement therapy). The independent variable was whether or not (control) there had been an osteoporotic-related fracture. The RR was determined for having at least one prescription of bone-specific drugs or of HRT filled during the year following the index date using a Cox regression adjusted for age, chronic disease score (CDS) and prior bone mineral density (BMD) test. RESULTS: Crude rates of BMD testing (per 500 person-years) ranged from 20.4 (1995) to 41.1 (2000) in women who had had an osteoporotic-related fracture, and from 4.4 to 15.3 in controls. The crude rate of women (per 100 person-years) who had had an osteoporotic-related fracture and who took at least one bone-specific drug during follow-up ranged from 1.9 in 1995 to 31 in 2000 among those with prior osteoporotic-related fracture, and from 0.5 in 1995 to 11 in 2000 for controls; the corresponding figures for HRT ranged 6.7 in 1995 to 13 in 2000, and from 8.4 in 1995 to 11 in 2000 respectively. BMD test is the only major factor affecting the adjusted RR of having a prescription filled for bone-specific drugs (RR of 10.44; 6.91-15.79 in 1995 and RR of 3.68; 3.30-4.10 in 2000) or HRT (RR of 2.08; 1.64-2.64 in 1995 and RR of 1.44; 1.17-1.77 in 2000), particularly among women who had not had prior use. The fact of having a fracture status does significantly affect the RR of having at least one bone-specific drug prescription filled only among women without prior use (RR of 1.71; 1.26-2.33 in 1996 and RR of 1.77; 1.44-2.19 in 2000). The fact of being younger did not affect the RR of having at least one prescription of bone-specific drugs filled, but being younger increased the RR of filling a prescription of HRT. CONCLUSIONS: Significant change was seen over time in the number of BMD tests ordered and ART use. Effective osteoporosis interventions are not optimal in the treatment of elderly women in a Canadian health-care system who have had an osteoporotic fracture, given that approximately 25% of women who had had an osteoporotic-related fracture were users of ART.