Using the National Alzheimer's Coordinating Center's Uniform Data Det (NACC-UDS) to identify cohorts of elders with normal cognition (N = 7,663) and mild cognitive impairment (MCI, N = 4,114), we compared overall cognitive composite and domain specific sub-scores and their progression over time between diabetic and non-diabetic subjects.

Diabetes was more common among those with MCI (14.7%) than among subjects who were cognitively normal (11.7%). In subjects who were cognitively normal, baseline cognitive composite scores, attention, and executive function sub-scores were lower in diabetics than non-diabetics (by 0.098, 0.066, and 0.015 points, respectively). Over time, cognitive composite score showed subtle worsening in non-diabetics (0.025 points every 6 months), with an additional worsening of 0.01 points every 6 months in diabetics compared to non-diabetics. In the MCI groups, baseline cognitive composite as well as attention and executive domain sub-scores were lower in diabetics than non-diabetics (by 0.078, 0.092, and 0.032 points, respectively). Over time, cognitive composite (by 0.103 points every 6 months) and all domain specific sub-scores showed subtle worsening in non-diabetics, but diabetics had significantly slower worsening than non-diabetics on both cognitive composite (by 0.028 points) and domain specific sub-scores.

Among elders, diabetes may be associated with lower cognitive performance, primarily in non-memory domains. However it is not associated with continued worsening, suggesting a static deficit with minimal memory involvement. This data suggest that diabetes may contribute more to a vascular profile of cognitive impairment than a profile more typical of AD.

Systematic review and random effect meta-analysis.

Observational (cross-sectional and longitudinal cohort) studies comparing low and high vitamin D status and interventions comparing vitamin D supplementation with a control group were included in the review and meta-analysis.

Studies including adults and older adults without a dementia diagnosis were included.

Medline (PubMed), AMED, Psych INFO, and Cochrane Central databases were searched for articles until August 2016. The Newcastle-Ottawa Scale and Physiotherapy Evidence Database assessed methodological quality of all studies.

Twenty-six observational and three intervention studies (n = 19–9,556) were included in the meta-analysis. Low vitamin D was associated with worse cognitive performance (OR = 1.24, CI = 1.14–1.35) and cognitive decline (OR = 1.26, CI = 1.09–1.23); with cross-sectional yielding a stronger effect compared to longitudinal studies. Vitamin D supplementation showed no significant benefit on cognition compared with control (SMD = 0.21, CI = −0.05 to 0.46).

Observational evidence demonstrates low vitamin D is related to poorer cognition; however, interventional studies are yet to show a clear benefit from vitamin D supplementation. From the evidence to date, there is likely a therapeutic age window relevant to the development of disease and therefore vitamin D therapy. Longitudinal lifespan studies are necessary to depict the optimal timing and duration in which repletion of vitamin D may protect against cognitive decline and dementia in aging, to better inform trials and practice towards a successful therapy.

Prospective study with baseline data collection from 2012 to 2014 and 1 year of follow-up.

Women's Health Initiative participants living in the United States.

Ambulatory women aged 63 to 99 (N = 5,545).

Minutes of MVPA per day measured using an accelerometer, functional status measured using the Short Physical Performance Battery (SPPB), fall risk factors assessed using a questionnaire, fall injuries assessed in a telephone interview, incident falls ascertained from fall calendars.

Incident rate ratios (IRRs) revealed greater fall risk in women in the lowest quartile of MVPA compared to those in the highest (IRR = 1.18, 95% confidence interval = 1.01–1.38), adjusted for age, race and ethnicity, and fall risk factors. Fall rates were not significantly associated with MVPA in women with high SPPB scores (9–12) or one or fewer falls in the previous year, but in women with low SPPB scores (≤ 8) or a history of frequent falls, fall rates were higher in women with lower MVPA levels than in those with higher levels (interaction P < .03 and < .001, respectively). Falls in women with MVPA above the median were less likely to involve injuries requiring medical treatment (9.9%) than falls in women with lower MVPA levels (13.0%) (P < .001).

These findings indicate that falls are not more common or injurious in older women who engage in higher levels of MVPA. These findings support encouraging women to engage in the amounts and types of MVPA that they prefer. Older women with low physical function or frequent falls with low levels of MVPA are a high-risk group for whom vigilance about falls prevention is warranted.

Retrospective chart review.

An ED with universal screening for suicide risk.

Eligible charts included a random sample of adults (≥18 years) who screened positive for suicidal ideation (SI) in past 2 weeks and/or a suicide attempt (SA) within the past 6 months. Visit dates were from May 2014 to September 2016.

A total of 800 charts were reviewed, with oversampling of older adults. Of the 200 older adults sampled, fewer older adults compared to younger adults (n = 600) had a chief complaint involving psychiatric behavior (53% vs 70%) or self-harm behavior (26% vs 36%). Although a higher number of older adults (93%) had documentation of current SI compared to younger adults (79%), fewer older adults (17%) reported SA in the past 2 weeks compared to younger adults (23%). Of those with a positive suicide screen who were discharged home, less than half of older adults received a mental health evaluation during their visit (42%, 95% CI 34–52) compared to 66% (95% CI 61–70) of younger adults who met the same criteria. Similarly, fewer older, than younger, adult patients with current SI/SA received referral resources (34%; 95% CI 26–43; vs 60%; 95% CI 55–65).

Significantly fewer suicidal older adult patients who were discharged home received a mental health evaluation when compared to similar younger adults. These findings highlight an important area for improvement in the treatment of older adults at risk for suicide.

Retrospective cohort study using health administrative databases.

Ontario, Canada.

41,351 individuals with cognitive impairment or dementia, aged 66+ years newly admitted to nursing home between 2011 and 2014.

PIMs were defined with 2015 Beers Criteria and included antipsychotics, H₂-receptor antagonists, benzodiazepines, and drugs with strong anticholinergic properties. Medication information was obtained at nursing home admission and in the subsequent 180 days. Multivariable Cox proportional-hazards models were used to assess the impact of frailty status (determined by a 72-item frailty index) on the hazard of starting and discontinuing PIMs.

At admission, 44% of residents with cognitive impairment or dementia were on a PIM and prevalence varied by frailty (38.7% non-frail, 42.8% pre-frail, and 48.1% frail, P < .001). Following admission, many residents discontinued PIMs (23.5% for antipsychotics, 49.3% benzodiazepines, 32.2% anticholinergics, and 30.9% H₂-receptor antagonists). However, PIMs were also introduced with 10.9% newly started on antipsychotics, benzodiazepines (10.1%), anticholinergics (6.6%), and H₂-receptor antagonists (1.2%). After adjustment for other characteristics, frail residents had a similar risk of PIM discontinuation as non-frail residents except for anticholinergics (HR = 1.21, 95% CI 1.06–1.39) but were more likely to be newly prescribed benzodiazepines (HR = 1.32, 95% CI 1.20–1.44), antipsychotics (HR = 1.36, 1.23–1.49), and anticholinergics (HR = 1.34, 95% CI 1.20–1.50).

Many residents with cognitive impairment or dementia enter nursing homes on PIMs. PIMs are more likely to be started in frail individuals following admission. Interventions to support deprescribing of PIMs should be implemented targeting frail individuals during the transition to nursing home.

Cross-sectional analysis of data from a longitudinal prospective study.

Health, aging, and body composition study performed in the USA.

2,597 community-dwelling adults aged 70–79.

A semi-quantitative, interviewer-administered, 108-item food frequency questionnaire designed to estimate dietary intake. Poor appetite was defined as the report of a moderate, poor, or very poor appetite in the past month and was compared with good or very good appetite.

The mean age of the study sample was 74.5 ± 2.8 years; 48.2% were men, 37.7% were black, and 21.8% reported a poor appetite. After adjustment for total energy intake and potential confounders (including biting/chewing problems), participants with a poor appetite had a significantly lower consumption of protein and dietary fiber, solid foods, protein rich foods, whole grains, fruits, and vegetables, but a higher consumption of dairy foods, fats, oils, sweets, and sodas compared to participants with very good appetite. In addition, they were less likely to report consumption of significant larger portion sizes.

Older adults reporting a poor appetite showed a different dietary intake pattern compared to those with (very) good appetite. Better understanding of the specific dietary intake pattern related to a poor appetite in older adults can be used for nutrition interventions to enhance food intake, diet variety, and diet quality.

Systematic review and meta-analysis.

Pubmed, Scopus, CINHAL, and Physiotherapy Evidence Database (PEDro) were searched to May 26, 2016.

Older adult population and at-risk adults.

Randomized controlled trials analyzing the effect of tai chi versus other treatments on risk of falls.

The incidence rate ratio (IRR) for falls incidence and hazard ratio (HR) for time to first fall.

The search strategy identified 891 potentially eligible studies, of which 10 met the inclusion criteria. There was high-quality evidence of a medium protective effect for fall incidence over the short term (IRR = 0.57; 95% CI = 0.46, 0.70) and a small protective effect over the long term (IRR = 0.87; 95% CI = 0.77, 0.98). Regarding injurious falls, we found very low-quality evidence of a medium protective effect over the short term (IRR = 0.50; 95% CI = 0.33, 0.74) and a small effect over the long term (IRR = 0.72; 95% CI = 0.54, 0.95). There was no effect on time to first fall, with moderate quality of evidence (HR = 0.98; 95% CI = 0.69, 1.37).

In at-risk adults and older adults, tai chi practice may reduce the rate of falls and injury-related falls over the short term (<12 months) by approximately 43% and 50%, respectively. Tai chi practice may not influence time to first fall in these populations. Due to the low quality of evidence, more studies investigating the effects of tai chi on injurious falls and time to first fall are required.

Meta-analysis of randomized controlled trials.

PubMed, Cochrane Library, Medline, EMBASE, Web of Science, and clinicaltrials.gov of selected randomized controlled trials and previous systematic reviews were searched.

Cognitively impaired persons.

Information was extracted on study characteristics, quality assessment, and outcomes. Outcome measures included BDs, sleep quality, and depression.

Nine randomized controlled trials were examined. The results showed that light therapy has a moderate effect on BD (g = −0.61) and depression (g = −0.58) and a small effect on total sleep time at night (g = 0.25). Subgroup analysis indicated that a light intensity of 2,500 lux or greater has a greater effect on depression than an intensity of less than 2,500 lux (P = .03), and the low risk of bias in blinding was superior to the RCTs deemed to be of high or unclear risk of bias in blinding in terms of BD (P = .02).

Light therapy can relieve BD, improve sleep quality, and alleviate symptoms of depression for cognitively impaired persons.

The ENabling Reduction of low-Grade Inflammation in SEniors (ENRGISE) Pilot Study was a multicenter, double-blind, placebo-controlled randomized pilot trial of two interventions to reduce IL-6 levels.

Five university-based research centers.

Target enrollment was 300 men and women aged 70 and older with an average plasma IL-6 level between 2.5 and 30 pg/mL measured twice at least 1 week apart. Participants had low to moderate physical function, defined as self-reported difficulty walking one-quarter of a mile or climbing a flight of stairs and usual walk speed of less than 1 m/s on a 4-m usual-pace walk.

Participants were randomized to losartan, omega-3 fish oil (ω-3), combined losartan and ω-3, or placebo. Randomization was stratified depending on eligibility for each group. A titration schedule was implemented to reach a dose that was safe and effective for IL-6 reduction. Maximal doses were 100 mg/d for losartan and 2.8 g/d for ω-3.

IL-6, walking speed over 400 m, physical function (Short Physical Performance Battery), other inflammatory markers, safety, tolerability, frailty domains, and maximal leg strength were measured.

Results from the ENRGISE Pilot Study will provide recruitment yields, feasibility, medication tolerance and adherence, and preliminary data to help justify a sample size for a more definitive randomized trial.

The ENRGISE Pilot Study will inform a larger subsequent trial that is expected to have important clinical and public health implications for the growing population of older adults with low-grade chronic inflammation and mobility limitations.

Retrospective cohort study.

English Longitudinal Study of Ageing.

Adults aged 50 and older.

Cross-sectional associations between self-reported (n = 7,865) and objective hearing measures (n = 6,902) and dementia were examined using multinomial-logistic regression. The longitudinal association between self-reported hearing at Wave 2 (2004/05) and cumulative physician-diagnosed dementia up to Wave 7 (2014/15) was modelled using Cox proportional hazards regression.

After adjustment for potential confounders, in cross-sectional analysis, participants who had self-reported or objective moderate and poor hearing were more likely to have a dementia diagnosis than those with normal hearing (self-reported: odds ratio OR = 1.6, 95% CI = 1.1–2.4 moderate hearing; OR = 2.6, 95% CI = 1.7–3.9 poor hearing, objective: OR = 1.6, 95% CI = 1.0–2.8 moderate hearing; OR = 4.4, 95% CI = 1.9–9.9 poor hearing). Longitudinally, the hazard of developing dementia was 1.4 (95% CI = 1.0–1.9) times as high in individuals who reported moderate hearing and 1.6 (95% CI = 1.1–2.0) times as high in those who reported poor hearing.

Older adults with hearing loss are at greater risk of dementia than those with normal hearing. These findings are consistent with the rationale that correction of hearing loss could help delay the onset of dementia, or that hearing loss itself could serve as a risk indicator for cognitive decline.

Cohort study.

Uppsala Longitudinal Study of Adult Men, Sweden.

Swedish men (n = 1,104) at a mean age of 71 (range 69.4–74.1) were investigated, 369 of whom were evaluated for independent aging 16 years later, at a mean age of 87 (range 84.8–88.9).

A questionnaire was used to obtain information on lifestyle, including education, living conditions, and physical activity. Adherence to a Mediterranean-like diet was assessed according to a modified Mediterranean Diet Score derived from 7-day food records. Cardiovascular risk factors were measured. Independent aging at a mean age of 87 was defined as lack of diagnosed dementia, a Mini-Mental State Examination score of 25 or greater, not institutionalized, independence in personal activities of daily living, and ability to walk outdoors alone. Complete survival data at age 85 were obtained from the Swedish Cause of Death Register.

Fifty-seven percent of the men survived to age 85, and 75% of the participants at a mean age of 87 displayed independent aging. Independent aging was associated with never smoking (vs current) (odds ratio (OR) = 2.20, 95% confidence interval (CI) = 1.05–4.60) and high (vs low) adherence to a Mediterranean-like diet (OR = 2.69, 95% CI = 1.14–6.80). Normal weight or overweight and waist circumference of 102 cm or less were also associated with independent aging. Similar associations were observed with survival.

Lifestyle factors such as never smoking, maintaining a healthy diet, and not being obese at age 71 were associated with survival and independent aging at age 85 and older in men.

Semistructured interviews paired with a qualitative analytical approach informed by Social Constructivist theory.

Inpatient care units in three hospitals. Purposive sampling was used to maximize variability in hospitals, units within hospitals, and staff on those units.

Clinicians (hospitalists, nurses, therapists, social workers, case managers) involved in evaluation and decision-making regarding postacute care (N = 25).

Central themes related to clinician evaluation and discharge decision-making.

Clinicians described pressure to expedite evaluation and discharge decisions, resulting in the use of SNFs as a “safety net” for older adults being discharged from the hospital. The lack of hospital-based clinician knowledge of SNF care practices, quality, or patient outcomes resulted in lack of a standardized evaluation process or a clear primary decision-maker.

Hospital clinician evaluation and decision-making about postacute care in SNFs may be characterized as rushed, without a clear system or framework for making decisions and uninformed by knowledge of SNF or patient outcomes in those discharged to SNFs. This leads to SNFs being used as a “safety net” for many older adults. As hospitals and SNFs are increasingly held jointly accountable for outcomes of individuals transitioning between hospitals and SNFs, novel solutions for improving evaluation and decision-making are urgently needed.

Semistructured interviews focusing on discharge planning and nursing facility selection, including how facilities were chosen, who was involved, and what factors were important in decision-making.

14 SNFs in five cities across the United States.

Newly admitted, previously community-dwelling SNF residents (N = 98) and their family members.

Semistructured interviews were qualitatively coded to identify underlying themes.

Most respondents reported receiving only a list of SNF names and addresses from discharge planners and that hospital staff were minimally involved. Proximity to home and prior experience with the facility most often influenced choice of SNF. Most respondents reported being satisfied with their placement, although many stated that they would have been willing to travel further to another SNF were it recommended. Many reported feeling rushed and unprepared, stating that they did not know where or how to get help.

SNF placement is a stressful transition, occurring when people are physically vulnerable and with limited guidance from discharge planners. Therefore, most people select a facility based on its location, perhaps because they are provided with no other information. Given Centers for Medicare and Medicaid Services' proposed changes to the discharge planning process, this research highlights the value of providing people and family caregivers with quality data and assistance in interpreting it.

Systematic review.

Randomized controlled trials (RCTs) recruiting individuals with AD and published during the last 10 years in 16 major general medicine, neurology, psychiatry, and geriatric psychiatry journals and RCTs registered on clinicaltrials.gov and currently enrolling individuals with AD.

Individuals with AD.

Outcome measures adopted by the included studies.

Only 21.4% of the included studies identified through the bibliographic searches had measures of NPSs as a primary outcome. Only 17.7% of the studies retrieved on clinicaltrials.gov made a specific effort to test the effect of pharmacological or nonpharmacological interventions on NPSs.

These findings show how rarely previous and current research on AD has considered NPSs as primary research targets. Although these symptoms are widely recognized as the most-stressful and -challenging manifestations of dementia, they are addressed much less often than other research targets.

Prospective trial with sequential introduction of screening and interventional processes.

Two tertiary referral hospitals in Australia.

Individuals aged 65 and older presenting to the emergency department (ED) and not requiring immediate resuscitation (N = 3,905).

Formal ED delirium screening algorithm and use of a risk warning card with a recommended series of actions for the prevention and management of delirium during the subsequent admission

Delirium diagnosis at hospital discharge, proportion discharged to new assisted living arrangements, in-hospital complications (use of sedation, falls, aspiration pneumonia, death), hospital length of stay.

Participants with a positive risk screen were significantly more likely (relative risk = 6.0, 95% confidence interval = 4.9–7.3) to develop delirium, and the proportion of at-risk participants with a positive screen was constant across three study phases. Delirium detection rate in participants undergoing the final intervention (Phase 3) was 12.1% (a 2% absolute and 17% relative increase from the baseline rate) but this was not statistically significant (P = .29), and a similar relative increase was seen over time in participants not receiving the intervention

A risk screening and warning or action card intervention in the ED did not significantly improve rates of delirium detection or other important outcomes.

Prospective cohort study.

Forty clinical centers in the United States.

Women aged 65 and older (mean age 69.8) enrolled in the Women's Health Initiative Clinical Trials with gait speed, timed chair stand, grip strength, and self-reported recreational PA data assessed at baseline (1993–98) and follow-up Years 1, 3, and 6 (N = 5,092).

Mixed-effects linear regression models were used to determine the association between time-varying PA and change in each PPM. Potential interactions between time-varying PA and age (<70, ≥70) were also tested.

Significan, dose-response associations between PA and improvements in all PPMs were observed over the 6 years of follow-up after adjusting for important covariates. High PA groups (≥1,200 metabolic equivalent (MET)-min/wk) had stronger grip strength (0.48 kg greater; P < .01), more chair stands (0.35 more; P < .001), and faster gait speeds (0.06 m/s faster; P < .001) than sedentary women (<100 MET-min/wk). Higher PA levels were associated with a greater increase in chair stands over time in women aged 70 and older (P < .001) than in those younger than 70 (P_{interaction for age} = .01).

In postmenopausal women, maintaining high PA levels over time is associated with better lower extremity function. These data support the view that regular PA plays an important role in maintaining functional status during aging in older women.

Cross-sectional longitudinal study.

The third and fourth waves of the community-representative, five-county, 10-year Duke Established Populations for Epidemiologic Studies of the Elderly study, carried out in a health service-rich area.

Cognitively intact survivors of the third wave (N = 1,740), who provided blood samples for standard laboratory work.

Biomarkers (n = 28) were measured to develop a cumulative deficit laboratory test–based FI (Duke FI) derived from standard laboratory tests: SMAC-24 chemistry panel, high-density lipoprotein cholesterol panel, and complete blood count. Information was gathered on scales assessing intrinsic characteristics (personal locus of control, life satisfaction, self-esteem, depressive symptomatology) and extrinsic characteristics (support received from and provided to family and friends, stressful life events, neighborhood disadvantage).

The newly developed Duke FI had content, construct, concurrent, and predictive validity. In addition to sex, race, and income, the Duke FI was associated at the intrinsic level with locus of control, self-esteem, life satisfaction, and depressive symptomatology (each P < .01) and at the extrinsic level with provision (P < .01) and with receipt of instrumental help (P < .10), social stressors (P < .03), and neighborhood disadvantage (P < .01) in unadjusted analysis; race fully explained neighborhood disadvantage.

Intrinsic (personality) characteristics and personally close extrinsic characteristics (contacts with family and friends, personal stressors) are associated with laboratory test–based frailty, as is neighborhood disadvantage, although in this accessible, health service–rich environment, race fully explained association with neighborhood disadvantage, suggesting that interventions to reduce frailty in residents in such an environment should pay particular attention to characteristics that immediately affect the individual.

A 24-week, randomized, double-blind, placebo-controlled, parallel-arm, proof-of-concept study.

Five centers in the United States.

Community-dwelling adults (N = 40) aged 65 and older with gait speed between 0.4 and 1.0 m/s over 4 m and an appendicular skeletal muscle index of 7.25 kg/m² or less for men and 5.67 kg/m² or less for women.

Intravenous bimagrumab 30 mg/kg (n = 19) or placebo (n = 21).

Change from baseline in thigh muscle volume (TMV), subcutaneous and intermuscular fat, appendicular and total lean body mass, grip strength, gait speed, and 6-minute walk distance (6MWD).

Thirty-two (80%) participants completed the study. TMV increased by Week 2, was sustained throughout the treatment period, and remained above baseline at the end of study in bimagrumab-treated participants, whereas there was no change with placebo treatment (Week 2: 5.15 ± 2.19% vs −0.34 ± 2.59%, P < .001; Week 4: 6.12 ± 2.56% vs 0.16 ± 3.42%, P < .001; Week 8: 8.01 ± 3.70% vs 0.35 ± 3.32%, P < .001; Week 16: 7.72 ± 5.31% vs 0.42 ± 5.14%, P < .001; Week 24: 4.80 ± 5.81% vs −1.01 ± 4.43%, P = .002). Participants with slower walking speed at baseline receiving bimagrumab had clinically meaningful and statistically significantly greater improvements in gait speed (mean 0.15 m/s, P = .009) and 6MWD (mean 82 m, P = .022) than those receiving placebo at Week 16. Adverse events in the bimagrumab group included muscle-related symptoms, acne, and diarrhea, most of which were mild in severity and resolved by the end of study.

Treatment with bimagrumab over 16 weeks increased muscle mass and strength in older adults with sarcopenia and improved mobility in those with slow walking speed.

Cross-sectional.

Community.

Community-dwelling older adults (N = 482).

Polypharmacy, defined as use of five or more medications and a cohort-specific alternate definition of eight or more medications, was examined. Velocity (cm/s) measured quantitatively during NW and WWT conditions.

The 164 participants (34%) with polypharmacy of five or more medications were older (77.0 ± 6.6 vs 76.0 ± 6.4) and more likely to have hypertension, congestive heart failure, diabetes mellitus, myocardial infarction, and higher body mass index (BMI) and to have fallen within the last year than the remaining 318 without polypharmacy and walked 6 cm/s slower (P = .004) during NW and 4 cm/s slower during WWT (P = .07), adjusting for age, sex, and education. Group differences were not statistically significant after adjusting for comorbidities. Prevalence of polypharmacy of eight or more medications was 10%. This group walked 11 cm/s slower during NW (P < .001) and 8.6 cm/s slower during WWT (P = .01) than those without polypharmacy, adjusted for age, sex, and education. Participants taking eight or more medications had slower NW (8.5 cm/s; P = .01), and WWT (6.9 cm/s; P = .07), compared to those without polypharmacy, adjusting for comorbidities. Adjustments for BMI, high-risk drugs, falls, and comorbidities yielded slower NW (9.4 cm/s, P = .005) and WWT (7.9 cm/s, P = .04 among those with polypharmacy compared to those without polypharmacy).

These results suggest an association between polypharmacy and locomotion that medical comorbidities only partly explained.

Prospective cohort study.

Health, Aging and Body Composition Study.

Black (n = 850) and white (n = 1,352) men and women with a mean age of 73.5 ± 2.85 at baseline.

Mixed models were used to assess the association between the COMT genotype and 6-m walk time, cross-sectionally and longitudinally over 10 years. Models were assessed unstratified and stratified according to race because allele distributions were different between white and black participants.

There was a significant U-shaped association between COMT genotype and 6-m walk time: those with higher (Val/Val) and lower (Met/Met) dopamine slowed more over 10 years (0.22 ± 0.02 seconds per visit and 0.23 ± 0.02 seconds per visit, respectively) than those with the intermediate (Met/Val) dopamine (0.20 ± 0.02 seconds per visit) (P = .005). Stratified results showed a significant relationship in black (P = .01) but not white (P = .15) participants.

These findings indicate a role of dopaminergic regulation of gait speed in community-dwelling older adults and of prefrontal cortex involvement in gait performance. Future work should investigate the molecular integrity of dopaminergic networks and gait changes over time and structural changes in the brain with COMT and gait decline in older adults.

Longitudinal cohort study.

Health and Retirement Study (HRS).

Nationally representative sample of 16,673 Medicare hospitalizations of 8,559 community-dwelling older adults from 2000 to 2012.

The main outcome was total Medicare costs in the year after hospital discharge, assessed according to Medicare claims data. The main predictor was functional impairment (level of difficulty or dependence in activities of daily living (ADLs)), determined from HRS interview preceding hospitalization. Multivariable linear regression was performed, adjusted for age, race, sex, income, net worth, and comorbidities, with clustering at the individual level to characterize the association between functional impairment and costs of postacute care.

Unadjusted mean Medicare costs for 1 year after discharge increased with severity of impairment in a dose-response fashion (P < .001 for trend); 68% had no functional impairment ($25,931), 17% had difficulty with one ADL ($32,501), 7% had dependency in one ADL ($39,928), and 8% had dependency in two or more ADLs ($45,895). The most severely impaired participants cost 77% more than those with no impairment; adjusted analyses showed attenuated effect size (33% more) but no change in trend. Considering costs attributable to comorbidities, only three conditions were more expensive than severe functional impairment (lymphoma, metastatic cancer, paralysis).

Functional impairment is associated with greater Medicare costs for postacute care and may be an unmeasured but important marker of long-term costs that cuts across conditions.

Retrospective electronic record review.

Academic medical center ED.

Individuals aged 65 and older seen in the ED between January 1, 2013, and September 30, 2015.

Two fall definitions were applied (individually and together) to the cohort: an ICD-9-based definition and a chief complaint definition. Admission rates and 30-day mortality (per encounter) were measured for each definition.

Twenty-three thousand eight hundred eighty older adult visits occurred during the study period. Using the most-inclusive definition (ICD-9 code or chief complaint indicating a fall), 4,363 visits (18%) were fall related. Of these visits, 3,506 (80%) met the ICD-9 definition for a fall-related visit, and 2,664 (61%) met the chief complaint definition. Of visits meeting the chief complaint definition, 857 (19.6%) were missed when applying the ICD-9 definition alone. Encounters missed using the ICD-9 definition were less likely to lead to an admission (42.9%, 95% confidence interval (CI) = 39.7–46.3%) than those identified (54.4%, 95% CI = 52.7–56.0%).

Identifying individuals in the ED who have fallen based on diagnosis codes underestimates the true burden of falls. Individuals missed according to the code-based definition were less likely to have been admitted than those who were captured. These findings call attention to the value of using chief complaint information to identify individuals who have fallen in the ED—for research, clinical care, or policy reasons.

Cross-sectional multicenter study.

Acute geriatric units in Belgium.

Team members of different professional backgrounds.

Perceptions of interprofessional teamwork among team members of 55 acute geriatric units in Belgium were measured using a survey covering collaborative practice and experience, managerial coaching and open team culture, shared reflection and decision-making, patient files facilitating teamwork, members’ belief in the power of teamwork, and members’ comfort in reporting incidents. Cluster analysis was used to determine types of interprofessional teamwork. Professions and clusters were compared using analysis of variance.

The overall response rate was 60%. Of the 890 respondents, 71% were nursing professionals, 20% other allied health professionals, 5% physicians, and 4% logistic and administrative staff. More than 70% of respondents scored highly on interprofessional teamwork competencies, consultation, experiences, meetings, management, and results. Fewer than 55% scored highly on items about shared reflection and decision-making, reporting incidents from a colleague, and patient files facilitating interprofessional teamwork. Nurses in this study rated shared reflection and decision-making lower than physicians on the same acute geriatric units (P < .001). Using the mean score on each of the six areas, four clusters that differed significantly in all areas were identified using hierarchical cluster analysis and scree plot analysis (P < .001).

Interprofessional teamwork in acute geriatric units is satisfactory, but shared reflection and decision-making needs improvement. Four types of interprofessional teamwork are identified and can be used to benchmark the teamwork of individual teams.

Cross-sectional.

Nutrition and Health Survey in Taiwan (NAHSIT), 2014–2016.

Noninstitutionalized Taiwanese nationals aged 65 years and older enrolled in the NAHSIT (N = 923).

Dietary intake was assessed using a 79-item food-frequency questionnaire (FFQ). Presence of 5 frailty phenotypes was determined using modified Fried criteria and are summed into a frailty score. Using data from the NAHSIT (2014–15), reduced rank regression was used to find a dietary pattern that explained maximal degree of variation of the frailty scores. Logistic regression models were used to estimate the association between frailty and dietary pattern. The findings were validated with data from 2016.

The derived dietary pattern was characterized with a high consumption of fruit, nuts and seeds, tea, vegetables, whole grains, shellfish, milk, and fish. The prevalence of frailty was 7.8% and of prefrailty was 50.8%, defined using the modified Fried criteria. Using data from the NAHSIT (2014–15), the dietary pattern score showed an inverse dose-response relationship with prevalence of frailty and pre-frailty. Individuals in the second dietary pattern tertile were one-third as likely to be frail as those in the first tertile (adjusted odds ratio (aOR) = 0.32, 95% confidence interval (CI) = 0.12−0.85), and those in the third tertile were 4% as likely to be frail as those in the first tertile (aOR = 0.04, 95% CI = 0.01−0.18). The dietary pattern score estimated using FFQ data from the NAHSIT 2016 was also significantly and inversely associated with frailty.

Individuals with a dietary pattern with more phytonutrient-rich plant foods, tea, omega-3-rich deep-sea fish, and other protein-rich foods such as shellfish and milk had a reduced prevalence of frailty. Further research is necessary to confirm these findings and investigate whether related dietary interventions can reduce frailty in older adults.

Retrospective cohort study

Taiwan National Health Insurance Research Database.

Individuals with newly diagnosed PD (N = 182,747)

Participants were followed from January 1, 2000, to December 31, 2010. Participants were assigned to dental prophylaxis, intensive periodontal treatment, tooth extraction, or no treatment, according to International Classification of Diseases codes and PD treatment codes. The incidence rate of dementia of the groups was compared. The association between PD and dementia was analyzed using Cox regression, with adjustments for age, sex, monthly income, residential urbanicity, and comorbidities.

The incidence of dementia was significantly higher in the group with PD that did not receive treatment (0.76% per year) and in the group that had teeth extracted (0.57% per year) than in the group that underwent intensive PD treatment (0.35% per year) and the group that received dental prophylaxis (0.39% per year) (P < .001). After adjusting for confounders, the Cox proportional hazards model revealed a higher risk of dementia in the group with PD who did not undergo treatment (hazard ratio (HR) = 1.14, 95% confidence interval (CI) = 1.04–1.24) and the group that had teeth extracted (HR = 1.10, 95% CI = 1.04–1.16) than in the group that received dental prophylaxis.

Subjects who had more severe PD or did not receive periodontal treatment were at greater risk of developing dementia.

Case-control study.

The Age and Thrombosis—Acquired and Genetic risk factors in the Elderly Study, a two-center study conducted in the Netherlands and the United States from 2008 to 2011.

Individuals aged 70 and older with a first-time deep venous thrombosis in the leg or pulmonary embolism (n = 401) and controls aged 70 and older (n = 431) without a history of thrombosis. Exclusion criteria were active malignancy and severe cognitive disorders.

The thrombotic risk associated with functional impairment, defined as impairment in two or more activities of daily living (ADLs), impaired mobility (inability to walk outside), sedentary lifestyle (≥20 h/d sleeping or sitting), and low handgrip strength (<15th percentile), was assessed. Odds ratios (ORs) adjusted for age, sex, and study center with 95% confidence intervals (95% CI) and population attributable risks (PAR) were calculated.

Risk of venous thrombosis was 2.9 times greater (OR = 2.9, 95% CI = 1.6–5.3) in individuals with impairment in ADLs, three times as great (OR = 3.0, 95% CI = 1.9–4.7) in those with impaired mobility, four times as great (OR = 4.0, 95% CI = 2.5–6.3) in those with a sedentary life style, and 2.3 times as great (OR = 2.3, 95% CI = 1.5–3.4) in those with weak handgrip strength. PARs were 8% for ADL disability, 13% for inability to walk outside for 15 minutes, 29% for sedentary lifestyle, and 13% for weak hand grip strength.

In individuals aged 70 and older, functional impairments are a major risk factor for venous thrombosis. These findings may help providers caring for older people be more aware of venous thrombosis risk.

Observational, longitudinal study.

Tertiary academic Alzheimer's Disease Centers funded by the National Institute on Aging.

Research volunteers aged 50 and older with two to six annual visits; 884 were taking PPIs at every visit, 1,925 took PPIs intermittently, and 7,677 never reported taking PPIs. All had baseline normal cognition or MCI.

Multivariable Cox regression analyses evaluated the association between PPI use and annual conversion of baseline normal cognition to MCI or dementia or annual conversion of baseline MCI to dementia, controlling for demographic characteristics, vascular comorbidities, mood, and use of anticholinergics and histamine-2 receptor antagonists.

Continuous (always vs never) PPI use was associated with lower risk of decline in cognitive function (hazard ratio (HR) = 0.78, 95% confidence interval (CI) =0.66–0.93, P = .005) and lower risk of conversion to MCI or AD (HR = 0.82, 95% CI = 0.69–0.98, P = .03). Intermittent use was also associated with lower risk of decline in cognitive function (HR = 0.84, 95% CI = 0.76–0.93, P = .001) and risk of conversion to MCI or AD (HR = 0.82, 95% CI = 0.74–0.91, P = .001). This lower risk was found for persons with normal cognition or MCI.

Proton pump inhibitors were not associated with greater risk of dementia or of AD, in contrast to recent reports. Study limitations include reliance on self-reported PPI use and lack of dispensing data. Prospective studies are needed to confirm these results to guide empirically based clinical treatment recommendations.

Observasional registry.

Prospective multicenter population-based registry of acute stroke patients treated with reperfusion therapies in Catalonia, Spain (Sistema Online d'Informació de l'Ictus Agut).

Individuals treated only with IVT (N = 3,231; 1,189 (36.8%) aged ≥80).

Symptomatic intracranial hemorrhage, mortality, and favorable outcome (modified Rankin Scale (mRS) score = 0–2) at 3 months were evaluated according to hospital characteristics. Treating hospitals were classified in three categories: comprehensive stroke centers (CSCs), primary stroke centers (PSCs), and community hospitals operating a telestroke system (TS). First individuals aged 80 and older were compared with those younger than 80, and then participants aged 80 and older were focused on.

Participants aged 80 and older had significantly higher baseline National Institute of Health Stroke Scale (NIHSS) scores, longer onset to treatment times, and worse outcomes than younger participants. For participants aged 80 and older, 90-day mortality was 23.2%, with 38.7% having favorable outcomes at 3 months. Symptomatic intracranial hemorrhage (SICH; Safe Implementation of Thrombolysis in Stroke-MOnitoring STudy definition) was observed in 4.7% of subjects. None of the risk factors differed significantly between participants treated in different types of hospitals. Basal stroke severity measured according to NIHSS score was not significantly different either. The three different types of hospitals achieved similar outcomes, although the TS and PSC hospitals had significantly higher proportions of SICH (6.3% and 6.3%, respectively) than the CSC (3.2%).

Older adults with acute stroke treated with IVT had similar outcomes regardless of hospital characteristics.

Cross-sectional study based on retrospective hospital records review.

Geriatric acute care ward.

Individuals aged 65 and older (N = 226; mean age 81.1), with (n = 104) and without (n = 122) a diagnosis of HF.

Information was collected on demographic characteristics, comorbidities, and laboratory and echocardiographic data. The relationship between hemoglobin and NT-proBNP was evaluated using linear regression models adjusted for potential confounders.

A negative association was found between NT-proBNP and hemoglobin (β = −0.226, P < .001). The regression coefficient was −0.114 (P = .04) in the subsample with HF and −0.191 (P < .001) in the subsample without HF. After adjustment for potential confounders, the inverse association between hemoglobin and NT-proBNP was confirmed in the whole sample (β = −0.182, P < .001), in those with HF (β = −0.136, P = .007), and in those without HF (β = −0.165, P = .003).

Hemoglobin concentration should be taken into account in the interpretation of NT-proBNP in hospitalized older adults, especially those without HF.

Retrospective cohort.

Women's Health Initiative.

Older women who reported at least 2 years of bisphosphonate use in 2008–09 (N = 5,120).

Exposure data were from a current medications inventory. Outcomes (hip, clinical vertebral, wrist or forearm, any clinical fracture) were ascertained annually. Using multivariate Cox proportional hazards models, the association between duration of bisphosphonate use (3–5, 6–9, 10–13 years) and fracture was estimated, using 2 years as the referent group.

On average participants were 80 years old and were followed for 3.7 ± 1.2 years. There were 127 hip, 159 wrist or forearm, 235 clinical vertebral, and 1,313 clinical fractures. In multivariate-adjusted analysis, 10 to 13 years of bisphosphonate use was associated with higher risk of any clinical fracture than 2 years of use (hazard ratio (HR) = 1.29, 95% confidence interval (CI) = 1.07–1.57). This association persisted in analyses limited to women with a prior fracture (HR = 1.30, 95% CI = 1.01–1.67) and women with no history of cancer (HR = 1.36, 95% CI = 1.10–1.68). The association of 10 to 13 years of use, compared with 2 years of use, was not statistically significant for hip (HR = 1.66, 95% CI = 0.81–3.40), clinical vertebral (HR = 1.65, 95% CI = 0.99–2.76), or wrist fracture (HR = 1.16, 95% CI = 0.67–2.00).

In older women at high risk of fracture, 10 to 13 years of bisphosphonate use was associated with higher risk of any clinical fracture than 2 years of use. These results add to concerns about the benefit of very long-term bisphosphonate use.

Prospective cohort study.

Two academic medical centers.

Adults aged 70 and older undergoing major noncardiac surgery (N = 560).

Plasma CRP was measured using enzyme-linked immunosorbent assay. Delirium was assessed from Confusion Assessment Method (CAM) interviews and chart review. Delirium duration was measured according to number of hospital days with delirium. Delirium feature severity was defined as the sum of CAM-Severity (CAM-S) scores on all postoperative hospital days. Generalized linear models were used to examine independent associations between CRP (preoperatively and POD2 separately) and delirium incidence, duration, and feature severity; prolonged hospital length of stay (LOS, >5 days); and discharge disposition.

Postoperative delirium occurred in 24% of participants, 12% had 2 or more delirium days, and the mean ± standard deviation sum CAM-S was 9.3 ± 11.4. After adjusting for age, sex, surgery type, anesthesia route, medical comorbidities, and postoperative infectious complications, participants with preoperative CRP of 3 mg/L or greater had a risk of delirium that was 1.5 times as great (95% confidence interval (CI) = 1.1–2.1) as that of those with CRP less than 3 mg/L, 0.4 more delirium days (P < .001), more-severe delirium (3.6 CAM-S points higher, P < .001), and a risk of prolonged LOS that was 1.4 times as great (95% CI = 1.1–1.8). Using POD2 CRP, participants in the highest quartile (≥235.73 mg/L) were 1.5 times as likely to develop delirium (95% CI = 1.0–2.4) as those in the lowest quartile (≤127.53 mg/L), had 0.2 more delirium days (P < .05), and had more severe delirium (4.5 CAM-S points higher, P < .001).

High preoperative and POD2 CRP were independently associated with delirium incidence, duration, and feature severity. CRP may be useful to identify individuals who are at risk of developing delirium.

Open-label, protocolized treatment study.

Medical centers in Pittsburgh, St Louis, and Toronto.

Older adults with major depression (N = 168).

Serum levels of the bone resorption marker C-terminal cross-linking telopeptide of type 1 collagen (CTX) and the bone formation marker procollagen type 1 N propeptide (P1NP) were assayed before and after 12 weeks of treatment with venlafaxine. Whether CTX and P1NP changes were associated with depression remission and duration of depression and genetic polymorphisms in the serotonin transporter (5HTTLPR) and 1B receptor (HTR1B) were also examined.

CTX increased and P1NP decreased during venlafaxine treatment, a profile consistent with accelerated bone loss. Two individual-level clinical variables were correlated with bone turnover; participants whose depression did not go into remission had higher CTX levels, and those with chronic depression had lower P1NP levels. HTR1B genotype predicted P1NP change, whereas 5HTTLPR genotype was unrelated to either biomarker.

Bone turnover markers change with antidepressant treatment in a pattern that suggests accelerated bone loss, although the clinical significance of these changes is unclear. These data are preliminary and argue for a larger, controlled study to confirm whether antidepressants are harmful to bone metabolism and whether certain individuals might be at increased risk.

Retrospective study.

United States.

Medicare beneficiaries aged 66 and older who died from July to December 2011.

Transitions between healthcare settings (e.g., hospital, skilled nursing facility, inpatient hospice, home hospice, home without hospice) in the last 6 months of life. A count variable for number of transitions was summarized, and Sankey diagrams were produced to illustrate the sequences of healthcare transitions. Multivariable analyses were used to identify factors associated with likelihood of having four or more transitions.

More than 80% decedents (n = 556,437) had at least one transition within the last 6 months of life; 218,731 had four or more transitions within the last 6 months of life. The most-frequent transition pattern (19.3% of all decedents; n = 127,435) was home to hospital, back to home or skilled nursing facility, to hospital again, and then to settings other than hospital, ending with four or more transitions. The average number of transitions in the last 6 months of life varied substantially across states, ranging from 1.8 in Alaska to 3.1 in New Jersey. Transitions became more intensive for decedents approaching death. In multivariable analyses, women, blacks, individuals younger than 85, and individuals without dementia were more likely to have four or more transitions (all P < .05).

Approximately one-third of the Medicare beneficiaries who died in 2011 had four or more transitions within their last 6 months of life. Identifying interventions that can facilitate care transitions consistent with beneficiaries’ preferences is warranted.

Cross-sectional.

Mail and electronic survey.

PCSs (N = 280; mean age 72 ± 8).

Cancer history, falls, and frailty status (robust, prefrail, frail) using traditionally defined and obese phenotypes.

Current (37%) or past (34%) ADT users were more than twice as likely to have fallen in the previous year as never users (15%) (P = .002). ADT users had twice as many recurrent falls (P < .001) and more fall-related injuries than unexposed men (P = .01). Current (43%) or past (40%) ADT users were more likely to be classified as prefrail or frail than never users (15%) (P < .001), and the prevalence of combined obese frailty + prefrailty was even greater in current (59%) or past (62%) ADT users than never users (25%) (P < .001). Traditional and obese frailty significantly increased the likelihood of reporting falls in the previous year (odds ratio (OR) = 2.15, 95% CI = 1.18–3.94 and OR = 2.97, 95% CI = 1.62–5.58, respectively) and was also associated with greater risk of recurrent falls (OR = 3.10, 95% CI = 1.48–6.5 and OR = 3.99, 95% CI = 1.79–8.89, respectively).

Current and past exposure to ADT is linked to higher risk of falls and frailty than no treatment. PCSs should be appropriately counseled on fall prevention strategies, and approaches to reduce frailty should be considered.

Multicenter retrospective analysis of multiple data systems to study victims of elder mistreatment in the greater Chicago metropolitan area.

Five teaching hospitals with Level 1 trauma centers.

Individuals aged 60 and older treated for physical and sexual abuse between 2000 and 2011.

History of revictimization was based on hospital admission histories, Adult Protective Services records, and self-report. Death records were also linked to participant files.

Fifty-eight individuals (52.3%) out of 111 cases suffering physical abuse had documented histories of revictimization. Based on multivariable models, individuals who were female, widowed, diagnosed with dementia, and returning to the home where the perpetrator lived or visited were substantially more likely to be revictimized. Revictimized individuals were more likely to be assaulted through unarmed force by a proximal relative, in particular a husband, boyfriend, child, or child-in-law. Based on hospital records, only 57% of community-dwelling cases had their abuse reported to Adult Protective Services or the police, and only 26.6% had Adult Protective Services investigations on record.

Better screening that connects victims of abuse with community services, police action, and alternative residential options is important in reducing the risk of revictimization and connecting individuals with resources that can improve their safety at home, regardless of whether it is in the community or a residential facility.

To evaluate intervention effects on tertiary physical performance outcomes.

The Lifestyle Interventions and Independence for Elders (LIFE) was a multi-centered, single-blind randomized trial of older adults.

Eight field centers throughout the United States.

1635 adults aged 78.9 ± 5.2 years, 67.2% women at risk for mobility disability (Short Physical Performance Battery [SPPB] <10).

Moderate PA including walking, resistance and balance training compared with HE consisting of topics relevant to older adults.

Grip strength, SPPB score and its components (balance, 4 m gait speed, and chair-stands), as well as 400 m walking speed.

Total SPPB score was higher in PA versus HE across all follow-up times (overall P = .04) as was the chair-stand component (overall P < .001). No intervention effects were observed for balance (overall P = .12), 4 m gait speed (overall P = .78), or grip strength (overall P = .62). However, 400 m walking speed was faster in PA versus HE group (overall P =<.001). In separate models, 29% of the rate reduction of major mobility disability in the PA versus HE group was explained by change in SPPB score, while 39% was explained by change in the chair stand component.

Lower extremity performance (SPPB) was significantly higher in the PA compared with HE group. Changes in chair-stand score explained a considerable portion of the effect of PA on the reduction of major mobility disability–consistent with the idea that preserving muscle strength/power may be important for the prevention of major mobility disability.

Observational study.

All counties in the United States.

All Medicare-certified home health agencies included in the Centers for Medicare & Medicaid Services Home Health Compare system.

County-specific availability of home health care, defined as the number of available home health agencies that provided services to a given county per 100,000 population aged ≥18 years.

The study included 15,184 Medicare-certified home health agencies that served 97% of U.S. ZIP codes. Between 2002–2003 and 2014–2015, the county-specific number of available home health agencies per 100,000 population aged ≥18 years increased from 14.7 to 21.8 and the median (inter-quartile range) population that was serviced by at least one home health agency increased from 403,605 (890,329) to 455,488 (1,039,328). Considerable geographic variation in the availability of home health care was observed. The West, North East, and South Atlantic regions had lower home health care availability than the Central regions, and this pattern persisted over the study period. Counties with higher median income, a larger senior population, higher rates of households without a car and low access to stores, more obesity, greater inactivity, and higher proportions of non-Hispanic white, non-Hispanic black, and Hispanic populations were more likely to have higher availability of home health care.

The availability of home health care increased nationwide during the study period, but there was much geographic variation.

Retrospective cohort study.

Taiwanese National Health Insurance Research Database.

The study cohort consisted of 786 dementia patients with new PPI usage and 786 matched dementia patients without PPI usage.

The study endpoint was defined as the occurrence of pneumonia. The Cox proportional hazard model was used to estimate the pneumonia risk. Defined daily dose methodology was applied to evaluate the cumulative and dose-response relationships of PPI.

Incidence of pneumonia was higher among patients with PPI usage (adjusted hazard ratio (HR) = 1.89; 95% CI = 1.51–2.37). Cox model analysis also demonstrated that age (adjusted HR = 1.05; 95% CI = 1.03–1.06), male gender (adjusted HR = 1.57; 95% CI = 1.25–1.98), underlying cerebrovascular disease (adjusted HR = 1.30; 95% CI = 1.04–1.62), chronic pulmonary disease (adjusted HR = 1.39; 95% CI = 1.09–1.76), congestive heart failure (adjusted HR = 1.54; 95% CI = 1.11–2.13), diabetes mellitus (adjusted HR = 1.54; 95% CI = 1.22–1.95), and usage of antipsychotics (adjusted HR = 1.29; 95% CI = 1.03–1.61) were independent risk factors for pneumonia. However, usage of cholinesterase inhibitors and histamine receptor-2 antagonists were shown to decrease pneumonia risk.

PPI usage in dementia patients is associated with an 89% increased risk of pneumonia.

Two-wave survey design.

Province of Quebec, Canada.

Lead nurses and physicians at 57 EDs that participated in both the T1 and T2 surveys.

Changes over time in ED geriatric services, observed naturalistically.

Survey questionnaires assessed: ED geriatric services (11 items) and nursing and geriatric staffing resources. Key administrative data indicators for ED bed visits for T1 and T2 for ages 75 and over included: volume of ED visits; length of ED stay; admission rate; and 30-day return visits.

There was a significant overall increase in the number ED geriatric services, from mean 2.8 (SD 2.2) at T1 to mean 6.0 (SD 2.0) at T2. EDs were clustered into 3 groups based on their T1 and T2 geriatric service scores: “early adopters” (n = 12); “late adopters” (n = 27); “non-adopters” (n = 18). Group membership was associated with three T1 variables: availability of a geriatric nurse clinician, a lower ratio of nurses to ED beds, and longer ED stays. There were significant overall increases between T1 and T2 in number of ED bed visits and ED length of stay among those aged 75 and over, decreases in hospitalization rates, but no change in ED return visits. There were no differential changes in the key indicators among the three ED clusters.

Overall, ED geriatric services improved in Quebec from 2006 to 2013/14. EDs with a geriatric nurse clinician, relatively fewer nursing resources, and longer ED stays improved more quickly.

To provide population-based estimates of the risk of developing incident difficulty managing medications and finances in older adults.

A prospective cohort study.

The Health and Retirement Study (HRS), a nationally representative study of older adults.

9,434 participants aged 65 and older who did not need help in managing medications or managing finances in 2002. Follow-up assessments occurred every 2 years until 2012.

The primary outcomes were time to difficulty managing medications and time to difficulty managing finances. Risk factors such as demographics, comorbidities, functional status, and cognitive status were assessed at baseline. Hazard models that considered the competing risk of death were used to estimate both the cumulative incidence of developing difficulty managing medications and finances and to identify potential risk factors. Analyses were adjusted for age, gender, race, marital status, wealth and education.

The 10 years incidence of difficulty increased markedly with age, ranging from 10.3% (95% CI 9.3–11.6) for managing medications and 23.1% (95% CI 21.6–24.7) for managing finances in those aged 65–69, to 38.2% (95% CI 33.4–43.5) for medicines and 69% (95% CI 63.7–74.3) for finances in those over age 85. Women had a higher probability of developing difficulty managing medications and managing finances than men.

This study highlights the importance of preparing older adults for the likelihood they will need assistance with managing their medicines and finances as the risk for having difficulty with these activities over time is substantial.

We examined the relationship between mushroom consumption and incident dementia in a population of elderly Japanese subjects.

Prospective cohort study.

Ohsaki Cohort 2006 Study.

13,230 individuals aged ≥65 years living in Ohsaki City, northeastern Japan.

Daily mushroom consumption, other lifestyle factors, and dementia incidence.

The 5.7 years incidence of dementia was 8.7%. In comparison with participants who consumed mushrooms <1 time/wk, the multi-adjusted HRs (95% CI) for incident dementia among those did so 1–2 times/week and ≥3 times/week were 0.95 (0.81, 1.10) and 0.81 (0.69, 0.95), respectively (P-trend <.01). The inverse association persisted after excluding participants whose dementia event occurred in the first 2 years of follow-up and whose baseline cognitive function was lower. The inverse association did not differ statistically in terms of vegetable consumption (P-interaction = .10).

This cohort study suggests that frequent mushroom consumption is significantly associated with a lower risk of incident dementia, even after adjustment for possible confounding factors.

Retrospective cohort study.

US Medicare- and Medicaid-certified nursing homes, 2011 to 2014.

Individuals admitted to all US nursing homes for PAC, aged ≥65 years, and without prior history of nursing home residence (n = 5,588,702).

Minimum Data Set (MDS) 3.0 admission assessments identified delirium based upon Confusion Assessment Method (CAM) items. Robust Poisson regression was used to calculate adjusted relative risks (aRRs) with 95% confidence intervals (CIs) for death following PAC admission, and for 30-day discharge outcomes including re-hospitalization from PAC, discharge home, and functional improvement.

Delirium was identified in 4.3% of new postacute nursing home admissions. Mortality within 30 days of PAC admission was observed in 16.3% of patients with delirium and 5.8% of patients without delirium (aRR = 2.27, CI = 2.24–2.30). The rate of 30-day readmission from PAC was 21.3% for patients with delirium compared with 15.1% among patients without delirium (aRR = 1.42, 95% CI = 1.40, 1.43). 26.9% of patients with delirium were discharged home within 30 days of admission compared to 52.5% of patients without delirium (aRR = 0.57, 95% CI = 0.57, 0.58). 48.9% of patients with delirium showed functional improvement at discharge compared to 59.9% of patients without delirium (aRR = 0.83, 95% CI = 0.82, 0.83).

Patients with delirium present upon PAC admission were at high risk for mortality and 30-day re-hospitalization and were less likely to have timely discharge to home or to improve in physical function at discharge. Early identification and care planning for individuals with delirium at PAC admission may be essential to improve outcomes.

Cross-sectional analyses of data collected in 1992–96 from a longitudinal cohort. Setting: Research clinic within a community.

Community-dwelling men (n = 630) and women (n = 961), mean age 71.5 years (SD = 10.8), from the Rancho Bernardo Study.

DISH assessed from lateral thoracic and lumbar spine radiographs; self-reported difficulty bending over to the floor, walking 2–3 level blocks, or climbing 1 flight of stairs; performance-based measures of grip strength and chair-stand testing (ability to stand up and sit down in a chair 5 times without using chair arms).

DISH was present in 25.6% of men and 5.5% of women. In age and sex-adjusted models, those with DISH had 1.72-fold increased odds (95% CI: 1.13, 2.62) of self-reported difficulty bending; this remained significant after further adjustment for Cobb angle, weight, stroke, arthritis, and exercise, OR = 1.69, (95% CI: 1.07, 2.66). In fully adjusted multivariate models, those with DISH had worse grip strength, −1.08 kg, P = .01, but did not differ from those without DISH on walking or climbing stairs. In sex-stratified, fully adjusted models, among men only, those with DISH were 2.17-times (95% CI: 1.04, 4.52) more likely to be unable to complete 5 chair stands without using their arms.

DISH was less prevalent in women but affected almost one-quarter of older white men. People with DISH are more likely to experience physical functional impairment, suggesting that DISH has clinical correlations and is not an incidental radiographic finding.

Nationally representative cohort study.

Health and Retirement Study.

7,492 U.S. adults aged ≥65 years.

Grip strength was measured by a hand dynamometer and classified as normal (≥16 kg for female; ≥26 kg for male) and weak. Gait speed was assessed over a 98.5-inch walk and classified as non-slow (≥0.60 m/s for female; ≥0.52 m/s for male) and slow.

Over an average follow-up time of 6.0 years, 1,870 (25.0%) participants died. After adjustment for socio-demographic, behavioral, and clinical measures, elevated SBP (≥150 mmHg) and DBP (≥90 mmHg) was associated with a 24% (95% CI, 7–43%) and 25% (95% CI, 5–49%) higher mortality among participants with normal grip strength. In contrast, elevated SBP and DBP was associated with a 6% (95% CI, 31 to −27%) and a 16% (95% CI, 46 to −26%) lower mortality among those with weak grip strength (P-values of interactions: both=.07). The inverse relations between BP with death were most pronounced among slow walkers with weak grip strength. The HRs of elevated SBP and DBP for death was 0.85 (95% CI, 0.56–1.29) and 0.53 (95% CI, 0.30–0.96), respectively, and was substantially different from non-slow walkers with normal grip strength (HR = 1.24 and 1.15, respectively; P-values of interactions: both <.001). Therefore, associations of BP with death varied modestly by gait speed.

Grip strength modified the association of BP with death. Combination of grip strength and gait speed has incremental value for modifying the association of BP with death.

Retrospective cohort study using hospice patient electronic medical record data.

Large, national hospice provider.

Data from 32,605 hospice patients who received routine hospice care from 2009 to 2014 were analyzed.

Descriptive statistics were calculated for utilization measures for each type of provider and by location of care. Frequency and duration of service contacts were standardized to a 1 week period and pairwise comparisons were used to detect differences in care provided between the three settings.

Minimal differences were found in overall intensity of service contacts across settings, however, the mix of services were different for patients living at home versus nursing home versus assisted living facility. Overall, more nurse care was provided at the beginning and end of the hospice episode; intensity of aide care services was higher in the middle portion of the hospice episode. Nearly 43% of the sample had hospice stays less than 2 weeks and up to 20% had stays greater than 6 months.

There are significant differences between characteristics of hospice patients in different settings, as well as the mix of services they receive. Medicare hospice payment methodology was revised starting in 2016. While the new payment structure is in greater alignment with the U shape distribution of services, it will be important to evaluate the impact of the new payment methodology on length of stay and mix of services by different providers across settings of care.

Pragmatic randomized, open label comparative trial of AChEI's currently available in the United States.

Four memory care practices within four healthcare systems in the greater Indianapolis area.

Eligibility criteria included older adults with a diagnosis of possible or probable Alzheimer's disease (AD) who were initiating treatment with an AChEI. Participants were required to have a caregiver to complete assessments, access to a telephone, and be able to understand English. Exclusion criteria consisted of a prior severe adverse event from AChEIs.

Participants were randomized to one of three AChEIs in a 1:1:1 ratio and followed for 18 weeks.

Caregiver-reported adherence, defined as taking or not taking study medication, and caregiver-reported adverse events, defined as the presence of an adverse event.

196 participants were included with 74.0% female, 30.6% African Americans, and 72.9% who completed at least twelfth grade. Discontinuation rates after 18 weeks were 38.8% for donepezil, 53.0% for galantamine, and 58.7% for rivastigmine (P = .063) in the intent to treat analysis. Adverse events and cost explained 73.1% and 25.4% of discontinuation. No participants discontinued donepezil due to cost. Adverse events were reported by 81.2% of all participants; no between-group differences in total adverse events were statistically significant.

This pragmatic comparative trial showed high rates of adverse events and cost-related non-adherence with AChEIs. Interventions improving adherence and persistence to AChEIs may improve AD management. Trial Registration: Clinicaltrials.gov: NCT01362686 (https://clinicaltrials.gov/ct2/show/NCT01362686).

Secondary analysis.

One hundred sixteen general practices in the Netherlands.

Community-dwelling individuals aged 70 to 78 (N = 2,994).

Nurse-led multidomain intervention targeting cardiovascular risk factors to prevent dementia.

The associations between participant baseline sociodemographic (age, sex, education), clinical (medical history, disability, cardiovascular risk), neuropsychiatric (depressive symptoms (Geriatric Depression Scale-15), and cognitive (Mini-Mental State Examination)) characteristics and dropout from the trial and nonadherence to the trial intervention were explored using multilevel logistic regression models.

Older age, poorer cognitive function, more symptoms of depression, and greater disability were the most important determinants of dropout of older people. The presence of cardiovascular risk factors was not associated with dropout but was associated with nonadherence. Being overweight was a risk factor for nonadherence, whereas people with high blood pressure or a low level of physical exercise adhered better to the intervention. The association between poorer cognitive function and symptoms of depression and dropout was stronger in the control group than in the intervention group, and vice versa for increased disability.

In a large dementia prevention trial with 6-year follow-up, dropout was associated with older age, poorer cognitive function, symptoms of depression, and disability at baseline. These findings can help to guide the design of future dementia prevention trials in older adults. The associations found between cardiovascular risk factors and nonadherence need to be confirmed in other older populations receiving cardiovascular prevention interventions.

Longitudinal study.

Hispanic Established Population for the Epidemiologic Study of the Elderly survey conducted in the southwestern of United States (Texas, Colorado, Arizona, New Mexico, and California).

Four hundred thirty-two Mexican Americans aged 75 and older with normal or high cognitive function at baseline.

Socio-demographic factors, living arrangement, type of household, social support, financial strain, self-reported medical conditions, Mini-Mental State Examination (MMSE), depressive symptoms, activities of daily living (ADLs), and Short Physical Performance Battery. Life-space assessment (LSA) during the past 4 weeks was assessed during in-home interview. Scores ranged from 0 (daily restriction to the bedroom) to 120 (daily trips outside of their own town without assistance) and categorized as 0 to 20, 21 to 40, 41 to 60, 61 to 80, and 81 to 120. Because of the small sample size in the category of 81 to 120, the two highest categories were combined into a single group.

The mean LSA score and MMSE score of participants at baseline was 44.6 (Standard Deviation [SD], 20.7) and 25.7 (SD, 3.2), respectively. Mixed Model analyses showed that participants in the highest life-space category (≥61) experienced slower rates of cognitive decline over time compared to participants in the lowest category (0 to 20) (β = 1.03, Standard Error [SE] = 0.29, P = 0.0004), after adjusting for all covariates.

Greater life-space mobility at baseline was predictor of slower rates of cognitive decline over 5 years in older Mexican Americans.

Retrospective cohort study

Australia. Administrative claims data from the Australian Government Department of Veterans' Affairs were used.

Male Vietnam veterans aged 55 to 65 at baseline (2001–02) with no preexisting dementia diagnosis (N = 15,612).

The association between PTSD and dementia was assessed over 12 years of follow-up. Dementia was identified as a hospital diagnosis, dementia record in service disability data, or dispensing of medicines for dementia. Cox-proportional hazards models were used, with age as the time-scale. Results were stratified according to baseline antipsychotic use.

No greater risk of dementia was observed with PTSD. In veterans who received antipsychotics, dementia risk was significantly higher than in those who did not (hazard ratio (HR) = 2.1, 95% confidence interval (CI) = 1.4–3.3). Dementia risk was significantly greater in veterans hospitalized for PTSD who received antipsychotics (HR = 2.2, 95% CI = 1.1–4.6) and veterans without PTSD who received antipsychotics (HR = 4.3, 95% CI = 2.1–8.6) than in those without PTSD with no antipsychotic use.

Antipsychotic use may be a contributor to dementia risk. These findings should be interpreted with caution because the study design was observational. Further research using prospective study designs in settings where diagnostic data, cognitive function, and disease severity are available are required.

Retrospective cohort study.

Medicare fee-for-service (FFS) claims and skilled nursing facility (SNF) Minimum Data Set 3.0 assessments.

Medicare beneficiaries with 90 days or more of no contact with a HCF before a hospital admission without a CDI diagnosis. Participants were separated into two cohorts: discharged to the community and discharged to a SNF.

Risk of HCF-acquired CDI associated with CDI prevalence at the index facility measured according to 30-day rehospitalization with a discharge diagnosis of CDI or diagnosis in the SNF after admission. Hospital and SNF CDI prevalence were categorized into three groups: 0% and above and below the median value for facilities with greater than 0% prevalence.

Of 817,900 eligible individuals, there were 553,423 admissions in the first cohort (discharged to the community) and 315,109 in the second (discharged to a SNF). In the first cohort, the risk of HCF-acquired CDI was higher for individuals admitted to hospitals with CDI prevalence less than the median (relative risk (RR) = 1.58, 95% confidence interval (CI) = 1.18–2.12) and greater than the median (RR = 2.56, 95% CI = 1.91–3.45) than for those with no CDI. In the second cohort, the risk of HCF-acquired CDI was greater for individuals admitted to a hospital (RR = 1.89, 95% CI = 1.49–2.39) and a SNF (RR = 1.48, 95% CI = 1.31–1.67) with CDI prevalence greater than the median.

The risk of HCF-acquired CDI is greater for noninfected individuals admitted to hospitals and SNFs with a high prevalence of CDI.

Population-based, cross-sectional study using Resident Assessment Instrument Minimum Data Set 2.0 linked to health administrative data.

Ontario, Canada.

All 9,298 nursing home residents with advanced dementia who died between June 1, 2010, and March 31, 2013; were aged 66 and older at time of death; and received at least one MQB in their last year of life.

Prevalence of eight classes of MQBs (e.g., lipid-lowering agents, antidementia drugs) used in the last 120 days and last week of life.

Of older nursing home residents with advanced dementia who received at least one MQB in the last year of life, 8,027 (86.3%) received them in the last 120 days and 4,180 (45.0%) in the last week of life. The most commonly prescribed MQB were antidementia (63.6%) and lipid-lowering agents (47.8%). Severe cognitive impairment (adjusted odds ratio (aOR) = 1.19, 95% confidence interval (CI) = 1.07–1.33, P = .002) and fewer signs and symptoms of health instability (aOR = 1.58, 95% CI = 1.44–1.74, P < .001) were associated with MQB use into the last week of life. Seeing a neurologist or psychiatrist was associated with less likelihood of MQB use in the last week of life.

Many nursing home residents with advanced dementia are dispensed MQBs in the last week of life. Given that MQBs may cause more harm than benefit in this vulnerable population, it is important for physicians to actively reassess the role of all medications toward the end of life.

There were three stages in this study: development of instrument items and validity testing, a feasibility pilot study, and a cross-sectional trial to establish construct and criterion validity and reliability.

One 116-bed assisted living facility (ALF).

After a pilot test with 10 residents, a cross-sectional trial was conducted with 55 additional residents.

The Barthel Index and Functional Independence Measure were used to estimate criterion validity. Construct validity was examined using exploratory factor analyses (EFAs).

Disattenuated correlations between the PCPT ALF and other tools were all greater than 0.72, supporting criterion validity. Internal consistency (physical ability, α = 0.95; cognitive support, α = 0.92) and 1-week test–retest reliability (PCPT ALF, P = .93) were high, as was interrater reliability (IRR) (physical ability, 0.99; cognitive support, 1.00). In two EFAs, a one-factor solution accounted for 64.1% of the variance for the physical ability subscale and 63.5% of the variance for the cognitive support subscale.

The findings provide early evidence of the PCPT ALF's validity and reliability. If confirmed, this study's findings may be used in future work to assess the success of interventions to prevent or slow decline in the skills associated with ADL and IADL performance in ALFs.

Survey of experts in MCC practice, research, and policy. Topics were derived from white papers, funding announcements, or funded research projects relating to older adults with MCCs.

Survey conducted through the Health Care Systems Research Network (HCSRN) and Claude D. Pepper Older Americans Independence Centers (OAICs) Advancing Geriatrics Infrastructure and Network Growth Initiative, a joint endeavor of the HCSRN and OAICs.

Individuals affiliated with the HCSRN or OAICs and national MCC experts, including individuals affiliated with funding agencies having MCC-related grant portfolios.

A “top box” methodology was used, counting the number of respondents selecting the top response on a 5-point Likert scale and dividing by the total number of responses to calculate a top box percentage for each of 37 topics.

The highest-ranked research topics relevant to the health and healthcare of older adults with MCCs were health-related quality of life in older adults with MCCs; development of assessment tools (to assess, e.g., symptom burden, quality of life, function); interactions between medications, disease processes, and health outcomes; disability; implementation of novel (and scalable) models of care; association between clusters of chronic conditions and clinical, financial, and social outcomes; role of caregivers; symptom burden; shared decision-making to enhance care planning; and tools to improve clinical decision-making.

Study findings serve to inform the development of a comprehensive research agenda to address the challenges relating to the care of this “high-need, high-cost” population and the healthcare delivery systems responsible for serving it.

Cross-sectional survey.

The Asian American Quality of Life Survey was conducted with 2,614 Asian Americans living in central Texas using questionnaires available in English and six Asian languages.

Asian American Quality of Life Survey participants aged 60 and older (N = 533; mean age = 69.4 ± 6.9).

Participants were asked whether they had insurance that covered the cost of any dental visit, how they would rate their overall oral health status, and whether they had visited a dental clinic for a routine examination in the past 12 months. Information was also collected on sociodemographic and immigration-related variables.

More than 61% of the sample had no dental health insurance, 45% reported that their oral health was fair or poor, and 44% had not used preventive dental care services. A series of logistic regression analyses identified factors posing a significant risk to oral health and dental care. For example, those with limited English proficiency were 3.5 times as likely to lack dental health insurance and 3.2 times as likely to rate their oral health as fair or poor. The odds of not using preventive dental care services were 6.4 times as great in those without dental health insurance.

The overall findings call attention to efforts to promote oral health and dental care in older Asian Americans.

Systematic review and metaanalysis.

Ward based models on geriatrics wards and visiting team based models on orthopaedics wards were included.

Four trials (three European, one U.S.; 973 participants) were identified. Two assessed ward-based, and two assessed team-based interventions.

MEDLINE, EMBASE, CINAHL and PsycINFO databases; Clinicaltrials.gov; and the Central Register of Controlled Trials were searched. Reference lists from full-text articles were reviewed. Incidence of delirium was the primary outcome. Length of stay, delirium severity, institutionalization, long-term cognition and mortality were predefined secondary outcomes. Duration of delirium was included as a post hoc outcome.

There was a significant reduction in delirium overall (relative risk (RR) = 0.81, 95% confidence interval (CI) = 0.69–0.94) in the intervention group. Post hoc subgroup analysis found this effect to be preserved in the team-based intervention group (RR = 0.77, 95% CI = 0.61–0.98) but not the ward-based group. No significant effect was observed on any secondary outcome.

There was a reduction in the incidence of delirium after hip fracture with CGA. This is in keeping with results of non-randomized controlled trials and trials in other populations. Team-based interventions appeared superior in contrast to the Ellis CGA paper, but it is likely that heterogeneity in interventions and population studied affected this.

Review.

All randomized control trials on the management of osteoporosis entered in the Cochrane Library Database that reported mean age were included. Exclusion criteria were also examined. Of 284 randomized control trials identified, 102 were eligible for inclusion.

Older adult trail participants.

Mean age of participants and exclusion criteria used were analyzed.

The mean age of all participants was 64.0, despite the fact that the average age at hip fracture is 83 for women and 84 for men. Overall, the mean age of those presenting with hip fractures is 84.8. Twenty-four (23%) of the 102 trials used older age as an exclusion factor. Other exclusion factors were long time since menopause, impaired cardiac or pulmonary function, dependent in ambulation, any severe comorbidity, dementia or any cognitive impairment, recent history of peptic ulcer disease or erosive gastric disease, uncontrolled hypertension, and psychiatric illness.

These data show a distinct difference between the mean age of participants in studies of the management of osteoporosis and the mean age of those presenting with hip fractures. Given that osteoporosis is the leading cause of hip fractures, this finding could have a significant effect on future studies in this area. It would follow that future research should include a cohort of an age that is more reflective of those most likely to experience the adverse effects of osteoporosis.

Secondary analysis of county human services claims data and claims from an insured population in the same county.

Inpatient and outpatient clinics in Allegheny County, PA.

Adults aged 65 and older in the county who received treatment for a psychiatric diagnosis in 2012 (county base services, n = 1,457; Medicaid, n = 641; Health plan insurance, n = 5,595).

Psychiatric diagnoses were classified using the International Classification of Diseases, 9th revision (ICD-9).

Episodic mood disorders and schizophrenia were more common among county-funded and Medicaid recipients (50–54% vs 34%). Neurotic conditions were more common among older adults with health plan insurance (18% vs 8%). Nearly a quarter of older adults receiving county base services were classified as having “ill-defined and unknown causes of morbidity and mortality,” compared <1% among insured and 6% among Medicaid recipients.

The prevalence of psychiatric conditions among older adults varies by insurance coverage, suggesting a role for social and economic factors associated with safety net coverage as well as system-level differences in delivery of mental health services. Comparing the prevalence of psychiatric diagnoses across insurance types offers insight on social determinants of risk for mental disorders in late life.

Qualitative study using in-depth semi-structured interviews performed from December 2013 to April 2014.

Eight skilled nursing facilities in Connecticut that ranked in the top 10% or bottom 10% in hospitalization rates from 2008 to 2010.

Nursing facility staff members (N = 31).

A multidisciplinary team performed qualitative content analysis. The constant comparative method was used to develop a coding structure and identify themes.

DNH orders were uncommon at low- and high-hospitalizing facilities. Participants reported that they did not interpret these orders literally. A DNH order was not a prohibition against hospitalization but was understood to have a variety of exceptions. These orders functioned primarily as a signal that hospitalization should be questioned and discussed with the family when an acute event occurred.

In-the-moment discussions about hospitalization are still necessary even when a DNH order is in place. Work to reduce potentially burdensome NH–hospital transfers needs to focus not just on eliciting preferences in advance, but also on preparing residents and their families to make the best decisions about hospitalization when the time comes.

Retrospective descriptive cross-sectional study.

Biopsy records were obtained from the archives of four Brazilian referral centers of oral diagnosis between 2000 and 2016.

A total of 45,506 biopsy records of all patients were analyzed, of these 7,259 persons aged 60 and older were selected.

Data such as gender, age, race, anatomical location, and histopathological diagnosis were collected and categorized. Pearson's chi-square test (P < .005) was used to evaluate differences in the frequency of the several groups of oral lesions.

Oral and maxillofacial lesions were diagnosed in 7,259 older people, including 59.4% women (P < .001) and 61.3% white patients (P = .07). The most commonly affected sites were the cheek mucosa (20.3%) and mandible (8.9%) (P < .001). Reactive and inflammatory lesions were the most common lesions, followed by neoplasms. Oral squamous cell carcinoma was the most prevalent neoplasm (83.4%) (P < .001).

Knowledge of oral diseases obtained from biopsy records provides more accurate data about the diagnosis and oral health of elderly patients. These indicators thus support the development of specific health policies for the prevention and treatment of oral and maxillofacial lesions that affect this population.

Dyadic structural equation modeling using the Actor-Partner Interdependence Model.

The 2004 to 2012 waves of the Health and Retirement Study.

Community-dwelling heterosexual married couples aged 65 and older (N = 2,243).

Structural equation modeling with a probit link function was used to estimate associations between men's and women's age, education, health status, prior hospitalization or outpatient surgery in the past 10 years, regular health care provider, and household net assets and their own and their spouses’ probabilities of having an AD.

Individual and spousal ages were each positively associated with AD completion for men and women. Those with higher education were more likely to possess ADs. Women's probabilities of having ADs were also positively associated with husbands’ education. Men whose wives’ were in poor health were less likely to have ADs. Men who were hospitalized or underwent outpatient surgery were more likely to have ADs, as were their wives. Women who had regular sources of health care were more likely to have completed ADs.

AD completion by older married adults is associated with both individual and spousal factors. How spouses influence one another's advance care planning differs by gender. Future research should account for the dyadic nature of advance care planning, as should public education efforts and interventions promoting AD completion.