Variable judgments of decisional capacity in cognitively impaired research subjects.

OBJECTIVES: To compare assessments of the decisional capacity of cognitively impaired patients by research assistants (RAs) and by family caregiver/proxies and to determine whether either or both groups judge capacity differently depending on the specific (hypothetical) research enrollment decision being made. DESIGN: Cross-sectional. SETTING: Three clinics, one each in Ohio, Kentucky, and Illinois. PARTICIPANTS: One hundred forty-nine patients with established dementia diagnoses and their caregiver/proxies. MEASUREMENTS: As part of a longer interview, patients were asked about future enrollment in five hypothetical research projects with varying risks and benefits. After patients reported each decision and reasons for that decision, RAs (using Applebaum and Grisso's four standards) indicated whether the patients were competent to make that decision on their own. In separate interviews, caregiver/proxies were asked for a similar appraisal based on life experience with the patient. RA and proxy judgments were compared. RESULTS: Capacity judgments by RAs and by caregiver/proxies differed according to specific project for most patients. Agreement between RA and caregiver/proxy judgments varied according to project, but agreement was only fair when tested using kappa (range in data 0.21-0.39). Caregiver/proxies appraised 50 patients as competent for all decisions, and RAs assessed 47 as so. Of these, only 24 were the same patients. CONCLUSION: If capacity were assessed anew for each enrollment decision, more potential participants could maintain authority in making those decisions.

Patients dying with dementia: experience at the end of life and impact of hospice care.

To evaluate the impact hospice enrollment has on the terminal care of patients with dementia and describe the symptom burden these patients experience, caregivers of dementia decedents were assessed in a mortality follow-back survey. Patients had been managed through academic outpatient geriatric clinics in Chicago, IL. A total of 135 family members or other knowledgeable informants of persons who had died with dementia were queried about the patient's preferred location of death, location of death, satisfaction with patient care, and assessment of patient's pain and most bothersome symptom. Patients who were enrolled in hospice were significantly more likely to die in their location of choice (odds ratio [OR] 9.67; 95% confidence interval [CI] 2.57-37.0), and less likely to die in the hospital, (OR 0.04; 95% CI 0.01-0.18), compared to nonenrollees. Caregivers of hospice enrollees were more likely to rate patient care as excellent or very good (OR 5.65; 95% CI 2.61-10.34), compared to nonenrollees. Over 50% of caregivers of hospice enrollees and nonenrollees reported patient pain at a moderate intensity or higher on average in the last two weeks of life. No effect was observed between a caregiver reporting pain at a moderate level or higher and whether the patient was enrolled in hospice (OR 0.58; 95% CI 0.22-1.54). Hospice enrollees did not experience improvement in their most bothersome symptom's frequency, severity, or amount of distress caused (P=0.21, 0.39, 0.95, respectively), compared to nonenrollees. These results suggest that dying patients with dementia who are enrolled in hospice programs are more likely to die in their locations of choice and their caregivers are more likely to perceive that care is excellent or good than those who do not obtain hospice services. No effect on pain or other symptoms could be demonstrated in this analysis.

Dementia, diagnostic disclosure, and self-reported health status.

OBJECTIVES: To investigate the general awareness of cognitive impairment in persons with documented dementia, evaluate the subject's recall of a diagnostic disclosure from a physician and their recollection of the discussion, and determine whether this awareness of cognitive impairment or the recall of diagnostic disclosure is associated with poorer self-rated health scores. DESIGN: Secondary data analysis. SETTING: Three university-based clinical referral sites for dementia illnesses. PARTICIPANTS: Convenience sample of 149 patients with a diagnosis of dementia. MEASUREMENTS: Bivariate and logistic regression models with the outcome variables of patient self-report of memory problems, patient report of being told about memory problems by a physician, and self-reported health scores. RESULTS: Ninety-six of 149 (64.4%) subjects reported that they had memory problems, and this report was independently associated with younger age (P=.01) and higher Mini-Mental State Examination score (P=.02). Thirty-nine (26.2%) subjects reported being told by a physician about a diagnosis of dementia or memory problems. This recall was associated with younger age (P<.001), male sex (P=.04), and higher education level (P=.02). African Americans reported poorer self-rated health scores (odds ratio (OR)=2.4, 95% confidence interval (CI)=1.1-5.1). Persons who reported being told by a physician of a diagnosis of dementia were more likely to report poorer self-rated health (OR=2.5, 95% CI 1.1-5.5). CONCLUSION: Further research is needed to elucidate the relationship between self-rated health and dementia specifically focusing on the potentially negative effects of diagnostic disclosure on self-rated health, further identification of factors that contribute to self-rated health in persons with dementia, and the prognostic value of self-rated health for persons with dementia.

Empirical assessment of a research advance directive for persons with dementia and their proxies.

OBJECTIVES: To evaluate a research advance directive for persons with established dementia diagnoses and their family caregivers or proxies. DESIGN: Prospective randomized, controlled trial. SETTING: Three clinics, one each in Ohio, Kentucky, and Illinois. PARTICIPANTS: At the end of separate interviews about enrollment choices in five types of hypothetical research projects, 149 persons with established dementia diagnoses and their family proxies were randomized to jointly complete the Planning Ahead Together (PAT) document, a research advance directive (n=69) or to remain in the control group (n=80). INTERVENTION: The directive was assessed at two points: immediately after sample members received naturally occurring invitations to participate in other studies and again 2 years after initial enrollment. MEASUREMENTS: Personal enrollment rates, reported ease of enrollment decision for patients and proxies, and proxy comfort were compared between the experimental and control groups. RESULTS: Forty-one dyads were reinterviewed immediately after consent discussions for other trials. Forty-seven patients and 106 proxies were interviewed at 2-year follow-up. There was no evidence immediately after a trial enrollment opportunity or in the follow-up interview that the research advance directive (PAT) assisted patients or proxies. Enrollment rates, decision ease, and proxy comfort and certainty were similar in the PAT and control groups. CONCLUSION: Patient and proxy experience making hypothetical decisions in the interview may have affected enrollment decisions by the PAT and control groups. Although the low number of recruitment attempts and the natural attrition of the geriatric population limit conclusions about effectiveness that may be drawn from this unique data set, the feasibility of a research advance directive is clearly demonstrated.

Management of noncancer pain in community-dwelling persons with dementia.

OBJECTIVES: To explore the pharmacological treatment of noncancer pain in persons with dementia and identify predictors associated with insufficient analgesia. DESIGN: Cross-sectional analysis of an observational cohort study. SETTING: Academic outpatient geriatric clinic in Chicago, Illinois. PARTICIPANTS: A total of 115 dyads, mostly African American, consisting of community-dwelling persons with dementia and their caregivers. MEASUREMENTS: Patient report of demographics, noncancer pain, function, cognition, and depression. Caregiver report of patient agitation and over-the-counter and prescription medications. RESULTS: Sixty-two of 115 (54%) patients reported pain "on an average day." The caregivers of more than half of persons with dementia who reported pain "on an average day" did not report analgesic use. The majority of caregivers who reported analgesic use reported that patients took a World Health Organization Class I medication. No patients had been prescribed a Class III (strong opioid) drug. Fifty-three of 115 (46%) patients had potentially insufficient analgesia. In the logistic regression, insufficient analgesia was associated with greater age, Mini-Mental State Examination score of less than 10, and impairment in daily functioning. Insufficient analgesia was 1.07 times as likely (95% confidence interval (CI) = 1.01-1.14) for each additional year of age, 3.0 times as likely (95% CI = 1.05-9.10) if the subject had advanced dementia, and 2.5 times as likely (95% CI = 1.01-6.25) if the patient had any impairment in activities of daily living. CONCLUSION: In this convenience sample from a geriatric clinic, many persons with dementia and noncancer pain were not receiving pharmacological treatment. Those at greatest risk for insufficient analgesia were older, had moderate to severe dementia, and experienced impairments in activities of daily living.

Factors associated with self- and caregiver report of pain among community-dwelling persons with dementia.

BACKGROUND: Among cognitively intact patients, self- and caregiver report of patient pain has been associated with worse psychological health for both the patient and caregiver. Little existing research examines factors associated with patient self-report of pain and caregiver report of patient pain among community-dwelling persons with dementia. OBJECTIVE: To identify patient and caregiver factors associated with self- and caregiver report of patient pain among community-dwelling persons with dementia. DESIGN: Cross-sectional analysis of a longitudinal study using structured interviews with dementia patients and their family caregivers. SETTING AND SUBJECTS: Urban outpatient geriatrics clinics affiliated with a university hospital. Dyads composed of dementia patients and their family caregivers were approached prior to routine clinic visits to participate in the study. MEASUREMENTS: Relevant patient measurements included self-report of pain; cognitive, functional, and comorbidity assessments; a screen for depression; and demographic information. Relevant caregiver measurements included their report of patient pain and agitation, screens for depression and strain, and demographic information. RESULTS: Of 115 dyads, 37 patients (32%) and 57 caregivers (53%) reported the patient to be in pain. No patient or caregiver factors were significantly associated with patient self-report of pain. In the univariate analysis, caregiver reports of patient pain were associated with the patient not being depressed (p = .036), caregiver reports of patient agitation (p = .038), higher level of education in the caregiver (p = .029), and caregiver depression (p = .019). In multivariate logistic regression analyses, caregiver depression and patient agitation remained significantly associated with caregiver report of patient pain. CONCLUSIONS: In community-dwelling persons with dementia, self-report of pain was not associated with any other variables measured, suggesting that pain should be assessed through direct self-report and treated accordingly. Caregiver report of patient pain was associated with both caregiver report of patient agitation and caregiver depression. These findings suggest that clinicians may need to routinely assess patient pain, patient agitation, and caregiver depression. More research is needed to understand the relationship between these conditions.

Pain in community-dwelling persons with dementia: frequency, intensity, and congruence between patient and caregiver report.

To better understand the pain experience of persons with dementia and to describe what factors are related to congruence of pain reports within patient-caregiver dyads, a cohort study enrolled patient-caregiver dyads at a primary care geriatrics clinic. Thirty-two percent of persons with dementia self-report pain "right now." Of these, 65% report slight/mild pain, 27% moderate pain, and 8% severe pain or greater. Fifty-two percent of caregivers report their care recipients with dementia are in some pain "right now." Of these, 52% report slight/mild pain, 30% moderate, and 18% severe pain or greater. Fifty-nine percent of dyads agree on the presence or absence of patient pain. In multivariate analysis of dyadic congruence of pain reports by patient and caregiver factors, only patient factors predicted congruence. The odds of congruence of pain reports increase 3.7 (1.2-12.3) if the patient is male and decrease 0.938 (0.93-0.99) as the patient becomes more agitated. These findings suggest that community-dwelling persons with dementia report less pain than those in the nursing home and caregivers do a fair job of predicting patient pain.

Ethics reporting in publications about research with Alzheimer's disease patients.

Persons with impaired decision-making capacity require special ethical protections during recruitment for and participation in research. To assess how fully basic protections for these persons were reported in the literature, the first structured review of a sample of reports of trials including Alzheimer's subjects was performed in 62 journals between January 1992 and December 1998. Neither institutional review board review nor informed consent was mentioned in 28% of the studies. In 48% of the studies, there was no mention of subject involvement in the consent process or that any potential subjects refused or withdrew. Protections may have been offered and simply not reported in the journal articles. The critical importance of these protections would be demonstrated if editors required that authors provide full documentation of ethical protections when submitting an article for review. These might be briefly reported in the articles but be made available electronically to interested readers. Authors could then specify in detail how they conducted their research involving persons with diminished decision-making capacity.

Are the rules for research with subjects with dementia changing?: views from the field.

Deliberative bodies have recommended additional protections for persons with dementia included in clinical trials. This survey of experienced dementia researchers revealed that 45 to 64% considered that specific ones of these recommendations would increase subject protection, and 40 to 86% considered they would make research less feasible. The real tradeoff between protection and difficulty in conducting research on devastating illnesses needs to be confronted when new regulations in this area are debated.

Barriers to limiting the practice of feeding tube placement in advanced dementia.

OBJECTIVE: In advanced dementia, many difficult decisions may arise as the disease progresses, including whether to use feeding by tube. Several recent articles question the benefit of percutaneous endoscopic gastrostomy (PEG) tube placement in persons with advanced dementia. However, patients with advanced dementia are still referred for PEG tubes. This study attempts to understand more about physician decisions to recommend PEG tube placement in patients with advanced dementia. DESIGN: Mailed survey. SETTING: Random sample of 500 physicians from the American Medical Association Masterfile. PARTICIPANTS: Primary care physicians. MEASUREMENTS: Physician knowledge, beliefs, and self-reported practices of PEG tubes in advanced dementia. RESULTS: Of the 416 eligible participants, 195 completed surveys (response rate of 46.9%). A significant number of physicians believe PEG tubes have the following benefits in advanced dementia: reduce aspiration pneumonia (76.4%), and improve pressure ulcer healing (74.6%), survival (61.4%), nutritional status (93.7%), and functional status (27.1%). Most physicians underestimate 30-day mortality in patients who receive a PEG and more than half of physicians believe PEG tubes in advanced dementia are the standard of care. Also, a majority of physicians believe speech therapists, nurses, and nutritional support teams recommend feeding tubes, which influences their decision to recommend a PEG. Last, almost half of these respondents had a nursing home request PEG tube placement, leading physicians to recommend a PEG. CONCLUSIONS: We find a notable discord between physician opinion, reported practice, and the literature regarding PEG tubes in advanced dementia that reveals some of the barriers to decreasing the referral of these patients for PEG.