Follow-up study on subjective quality of life for elderly persons with dementia using the Japanese version of Dementia Quality of Life Instrument (JDQoL)

PURPOSE: There are few longitudinal studies on the subjective quality of life for elderly persons living in their own homes. The purpose of this study was to clarify the validity and reliability of the Japanese version of the Dementia Quality of Life Scale (JDQoL) used in a follow-up survey as part of a longitudinal study in Japan. METHODS: A baseline study was conducted from November 2002 to January 2003, and a year later a follow-up study was conducted from October 2003 to January 2004. The subjects included 72 (19 men and 53 women) elderly persons with dementia living in their own homes. Sixty-six were diagnosed with vascular dementia (VD) and 6 had senile dementia of Alzheimer's type (SDAT). The follow-up study included 60 subjects (VD: 56 and SDAT: 4); _10 of the original subjects were hospitalized, one died, and one was unable to attend daycare because of deterioration. Evaluation was conducted using the JDQoL, MiniMental State Examination (MMSE), and Geriatric Depression Scale (GDS). RESULTS: For the MMSE, the average score of the subjects in the follow-up study was 20.87 (+/- 4.80), which was significantly higher than the score of subjects that were excluded from the follow-up study (17.82 +/- 5.65). At baseline, reliability coefficient was demonstrated by Cronbach's alpha value of 0.744 approximately 0.886. One year later, Cronbach's alpha was 0.723 approximately 0.872. Among the subjects in the follow-up study, MMSE scores decreased significantly compared with the baseline study. In the subscales of the JDQoL, Negative Feelings (reversal item) significantly improved and Feeling of Belonging significantly decreased after one year (p<0.05). Each subscale of the JDQoL had significant correlation with the GDS at baseline and follow-up. Scores on the GDS at baseline had significant correlation with Self Esteem, Positive Feelings, Negative Feelings, and Feeling of Belonging of the JDQoL at follow-up (0.320 to 0.504; p<0.05). CONCLUSION: We demonstrated that the reliability and validity of the JDQoL was comparable to the English version of DQoL. Subjective QOL did not decline for all sub-scales, Negative Feelings in the JDQoL significantly improved compared to baseline.     

Dementia Care Mapping-Japanese version(DCM-J) as a research evaluation method for measuring quality of life among elderly patients with dementia: reliability and validity of Well-being and Ill-being value

AIM: Dementia Care Mapping is a behavior evaluation tool that can be used to measure and improve the quality of life of elderly patients with dementia. However, the reliability and validity of the Dementia Care Mapping-Japanese version (DCM-J) has not yet been established. Therefore, the purpose of this research was to clarify the reliability and validity of the Well-being and Ill-being (WIB) value of the DCM-J as a method for evaluating quality of life. METHODS: The study was conducted from April 1, 2005 to June 30, 2006. The participants included 130 elderly patients (men 31, women 99, average age 82.65+/-7.69 years) who were given a diagnosis of dementia. We established inter-rater reliability during a parallel observation method and also used test-retest for reliability. The correlation between the WIB value of the DCM-J and the Japanese Quality of Life Inventory for Elderly with Dementia (QOL-D) was used to establish criterion-related validity. RESULTS: Forty-nine (37.7%) subjects were given a diagnosis of dementia of Alzheimer's type, 80 (61.5%) had vascular dementia and 1 (0.8%) had dementia with Lewy bodies. The results showed correlation between the WIB value and social withdrawal measured by the behavioral category code (BCC) on the DCM-J and the three subscales of QOL-D: "interacting with surroundings", "expressing self", and "experiencing minimum negative behaviors". There was good internal consistency among these items. The interclass correlation coefficient was 82.32 (+/-5.85) for the WIB value of the DCM-J. The correlation coefficient of the retest, administered one week later, was 0.836 (p=0.001). The WIB value was significantly correlated with three sub-scales of QOL-D, and the correlation coefficient was greater than 0.53. CONCLUSION: We demonstrated that the WIB value of the DCM-J has good inter-rater reliability and test re-test reliability and criterion-related validity. In this study, the WIB value was shown to have similar reliability to the WIB value of the original DCM. Furthermore, our results suggest that the DCM-J could be useful for evaluating quality of life among elderly Japanese patients with dementia.     

Factors attenuating the validity of the Geriatric Depression Scale in a dementia population.

OBJECTIVE: The validity of the Geriatric Depression Scale (GDS) in cognitively impaired patients has been questioned. We investigated possible factors (memory loss, dementia severity, unawareness of illness) attenuating the validity of the GDS in patients with dementia. PATIENTS: Eighty-three patients who met research diagnostic criteria for "probable Alzheimer's disease." Subjects with major depressive disorder were excluded. Dementia severity ranged from mild to moderate. SETTING: Outpatient clinics, including institutional settings and private research settings. MEASUREMENTS: Depression--GDS; Hamilton Depression Scale. Memory--Wechsler Memory Scale; Benton Visual Retention Test. Dementia severity--Mini-Mental State Examination. Self-awareness of cognitive deficits--Difference score between a self-report memory questionnaire and an informant-rated memory questionnaire. RESULTS: Multiple regression analysis revealed that Hamilton scores were the major predictor of GDS scores. Memory scores and self-awareness scores were also significant predictors. Dementia severity scores were not a significant predictor. CONCLUSIONS: The GDS is a valid measure of mild-to-moderate depressive symptoms in Alzheimer patients with mild-to-moderate dementia. However, Alzheimer patients who disavow cognitive deficits also tend to disavow depressive symptoms, and the GDS should be used with caution in such patients. Finally, the argument that memory impairment precludes accurate self-report of recent mood is negated by our finding that many patients accurately reported depressive symptoms and that worse memory was associated with more self-reported depressive symptoms.     

Effects of dementia on perceived daily pain in home-dwelling elderly people: a population-based study

BACKGROUND: Pain is a significant problem in the elderly, but the impact of dementia on perceived pain has not been studied in population-based study settings. OBJECTIVES: To analyse the prevalence of daily pain and analgesic use among home-dwelling older people with and without dementia. DESIGN: A cross-sectional population-based survey. SETTING: Population of Kuopio city, Finland. SUBJECTS: 523 home-dwelling subjects aged 75 years and older. METHODS: Structured clinical examination and interview. RESULTS: Prevalence rates for any pain, any daily pain, pain every day interfering with routine activities, and daily pain at rest were significantly lower in those subjects with dementia (43%, 23%, 19% and 4%, respectively) compared to those subjects without dementia (69%, 40%, 36% and 13%, respectively). The subjects with dementia were less likely to use analgesics (33%) than the non-demented (47%). CONCLUSION: Dementia was related to a lower prevalence of reported pain and analgesic use among home-dwelling elderly people.     

Computer-aided visuospatial cognitive-performance test

The validity of a newly developed computer-aided "Visuospatial Cognitive-Performance Test" (VCP-Test) was examined in two populations, 68 normal elderly volunteers (mean age, 69.5% years) and 25 patients with dementia (mean age, 68.2 years). None of the cases in the two groups had either motor impairment or ataxia. The VCP-Test was a kind of vigilance task in which a subject is required to tap the key corresponding to the local change-over of the image on the computer display as quickly as possible. After 40 trials, consecutive data of performance efficiency were automatically displayed on the LCD display and printed out. A significant negative correlation between the VCP-Test and age in normal elderly population was recognized. Moreover, the relationship between the VCP-Test and the Hasegawa Dementia Scale (HDS), and the Mini-Mental State (MMS) was also significant in patients with dementia. In conclusion, the VCP-Test was considered to be useful for geriatric neurobehavioral functional assessment and for the detection of dementia.     

Depression as prognostic factor: 6 months follow-up in a geriatric institution

Depression in the elderly and especially in the demented elderly is common, but it is unclear to what extent depression affects the prognosis of dementia. We performed a 6 months follow-up on the elderly to evaluate the effect of depression on the prognosis of dementia. Among the residents in a Korean geriatric institution, a total of 141 elderly residents were included in this study. Those residents who had geriatric depression scale (GDS) scores over 20 were considered as having depression. The subjects were assessed at baseline and at 6 months later with the Korean Mini-Mental State Examination (K-MMSE), Clinical Dementia Rating (CDR) and Barthel index (BI). For all the 141 subjects, the changes of the K-MMSE, CDR and BI scores were first compared between the depressed and non-depressed groups. Then, for the 58 depressed subjects subdivided into three groups (36 subjects in the non-demented group, 14 subjects in the Alzheimer disease (AD) group and 8 subjects in the vascular dementia (VaD) group), the changes of the scores among the three groups were compared. For all subjects, the changes of the K-MMSE and CDR scores were more severe in the depressed group than in the non-depressed group. For the 58 depressed subjects, the changes were significantly different among the three groups (p=0.02 for the K-MMSE, p<0.001 for the CDR), and the changes were the most severe in the VaD group. For the elderly, depression has a significant influence on the cognitive deterioration. Moreover, the effect of depression on the cognitive dysfunction and on the prognosis of dementia is more severe in the demented elderly, and especially for the VaD patients.     

Systematic review of quantitative clinical gait analysis in patients with dementia

INTRODUCTION: Diminished mobility often accompanies dementia and has a great impact on independence and quality of life. New treatment strategies for dementia are emerging, but the effects on gait remains to be studied objectively. In this review we address the general effects of dementia on gait as revealed by quantitative gait analysis. METHODS: A systematic literature search with the (MESH) terms: 'dementia' and 'gait disorders' in Medline, CC, Psychlit and CinaHL between 1980-2002. Main inclusion criteria: controlled studies; patients with dementia; quantitative gait data. RESULTS: Seven publications met the inclusion criteria. All compared gait in Alzheimer's Disease (AD) with healthy elderly controls; one also assessed gait in Vascular Dementia (VaD). The methodology used was inconsistent and often had many shortcomings. However, there were several consistent findings: walking velocity decreased in dementia compared to healthy controls and decreased further with progressing severity of dementia. VaD was associated with a significant decrease in walking velocity compared to AD subjects. Dementia was associated with a shortened step length, an increased double support time and step to step variability. DISCUSSION: Gait in dementia is hardly analyzed in a well-designed manner. Despite this, the literature suggests that quantitative gait analysis can be sufficiently reliable and responsive to measure decline in walking velocity between subjects with and without dementia. More research is required to assess, both on an individual and a group level, how the minimal clinically relevant changes in gait in elderly demented patients should be defined and what would be the most responsive method to measure these changes.     

Vitality Index as a useful tool to assess elderly with dementia

Background: The aim of the present paper was to establish a new objective scale to measure vitality related to activities of daily living in elderly patients with dementia.

Methods:

Methods: Cross-sectional and longitudinal observational studies were carried out on patients and residents in long-term care facilities. Reliability of the scale was examined by determining test–retest reliability, interrater reliability and internal consistency. Validity of the scale was evaluated with respect to criterion validity and predictive validity. Clinical usefulness of the scale was tested by measuring the time taken to perform scoring and evaluating the changes in the score before and after behavioral therapy for functional urinary incontinence.

Results:

Results: The test–retest correlation coefficient was 0.98. Mean kappa coefficient of Vitality Index was 0.63. Interrater coefficient of variation was 0.14. Cronbach α of the Index was 0.88. Vitality Index was negatively correlated with Geriatric Depression Scale (GDS) ( P < 0.01). A low score was an independent risk factor for survival. Behavioral therapy for functional urinary incontinence increased Activities of Daily Living (ADL) ( P < 0.05) and Vitality Index ( P < 0.05).

Conclusion:

Conclusion: Vitality Index could be a new tool to assess patients with dementia.     

Criterion validity of the cross-cultural cognitive examination in Japan.

Criterion validity of a two-stage Cross-Cultural Cognitive Examination (CCCE) designed for epidemiologic use was evaluated in Japanese subjects by comparison with a physician's DSM-III-R diagnosis of dementia and the Hasegawa Dementia Rating Scale (the standard Japanese instrument similar to the Mini-Mental State Exam). We report on 188 subjects tested in three locations in Japan: Tokyo area, Ise, and Osaka. Subjects ranged in age (50-93 years) and education (1-22 years) and included neurology outpatients, community volunteers, and inpatients. The CCCE was 97.4% specific for dementia, with sensitivity of 88%. The correlation with the Hasegawa scale was significant (r(175) = .8230, p less than .0001). Diagnosis using the CCCE showed good validity when compared with Japanese criteria for dementia. If the instrument could be shown to be reliable and more "culture fair" than the currently available tests, it may be useful in cross-cultural epidemiologic studies of dementia.     

The prevalence of dementia in older people in an urban population of Korea: the Seoul study

OBJECTIVES: To estimate prevalence of dementia and its subtypes in older people in Seoul, a metropolitan area of Korea, and compare these findings with estimates reported for other populations. DESIGN: The study employed a two-stage design for case identification. Initially, the Mini-Mental State Examination in the Korean version (MMSE-KC) of the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) assessment packet was administered to all participants. Two hundred seventeen persons sampled from three levels of performance on MMSE-KC underwent the second-stage clinical evaluation based on the Korean Version of the CERAD assessment packet. SETTING: The study was conducted in an urban community setting. PARTICIPANTS: Six hundred forty-three persons aged 65 and over participated in the study. MEASUREMENTS: Dementia was defined using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnostic features of dementia. RESULTS: Dementia prevalence ranged from 2.6% in persons aged 65 to 69 to 32.6% in persons aged 85 and older. Age-standardized prevalence was 8.2% for dementia, 5.4% for Alzheimer's disease, and 2.0% for vascular dementia. The prevalence estimates, which excluded very mild cases (clinical dementia rating index 0.5), were approximately 5.3% for dementia and 4.3% for Alzheimer's disease. CONCLUSION: The prevalence of dementia in older people in Seoul appears to be somewhat lower than in rural areas of Korea. Considering the difficulties involved in establishing a diagnostic threshold for dementia, actual differences in dementia prevalence between Asian populations are probably minimal.     

Self-recognition in Alzheimer's disease: a mirror and video study.

Twenty-nine subjects with Senile Dementia of the Alzheimer Type (SDAT) were tested for self-recognition when confronted with their reflection in a mirror and with a noncontingent video image. All subjects at stage 5 (moderately severe cognitive decline) on the Global Deterioration Scale (GDS) showed self-recognition in the mirror condition, whereas 25% of GDS stage 6 subjects (severe cognitive decline) did not self-recognize. Only seven subjects showed signs of self-recognition on video (six at GDS 5, one at GDS 6). The results confirm two hypotheses: the first, based on studies with infants, is that contingency cues in the mirror condition facilitate self-recognition; the second is that SDAT patients may show "regression" of at least some cognitive functions.     

Development of a brief communication ability scale for Japanese demented elderly

We developed a brief scale to evaluate communication ability of the demented elderly. This scale assesses not only abilities related to overall communication such as verbal function, judgment and emotional function, but also non-verbal communication such as eye-contact, nodding and smiling. The scale places little burden on the demented elderly subject and takes only a few minutes to perform, even if the dementia is severe. We evaluated 106 demented elderly residents of nursing homes using this brief communication ability scale, and the following results were obtained. The validity of this scale was confirmed by the high correlation coefficient between this scale and the formal caregiver questionnaire scores concerning communication ability, and the high-correlation coefficient between this scale and intellectual functions (r = -0.904), emotional functions (r = -0.841) and motor functions (r = -0.679) of dementia syndromes rating scale (Gottfries, Br?ne, Steen scale; GBS scale), Hasegawa's Dementia Scale-Revised (HDS-R) (r = 0.625) and the Mini-Mental State Examination (MMS) (r = 0.733). The reliability of this scale was confirmed by the high interrater reliability coefficient of 0.828, test-retest reliability coefficient of 0.940 and Cronbach alpha coefficient of 0.938. These results indicate that the new scale is useful in the assessment of communication ability among the demented elderly.     

Association of diabetes mellitus and dementia: The Rotterdam Study

Summary Dementia and non-insulin-dependent diabetes mellitus (NIDDM) are highly prevalent disorders in the elderly. Diabetes has repeatedly been reported to affect cognition, but its relation with dementia is uncertain. We therefore studied the association between diabetes and dementia in the Rotterdam Study, a large population-based study in the elderly. Of 6330 participants, aged 55 to 99 years old, complete information on diabetes and presence of dementia was available. Diabetes was diagnosed as use of anti-diabetes medication or random or post-load serum glucose over 11 mmol/l. Dementia was diagnosed through a stepped approach, including a sensitive screening of all participants and a comprehensive diagnostic work-up. Diabetes was present in 724 (11.4 %) subjects. Of the 265 dementia patients 59 (22.3 %) had diabetes. Multiple logistic regression analyses, adjusting for age and sex differences, revealed a positive association between diabetes and dementia (odds ratio: 1.3, 95 % confidence interval: 1.0–1.9). In particular, strong associations were found between dementia and diabetes treated with insulin (odds ratio: 3.2, 95 % confidence interval: 1.4–7.5). The relation was strongest with vascular dementia, but was also observed with Alzheimer's disease. These associations were independent of educational attainment, smoking, body mass index, atherosclerosis, blood pressure and antihypertensive drug treatment, and could not be explained by clinical cerebral infarctions. The results suggest that NIDDM is associated with dementia. Alzheimer's disease may be more frequent in elderly diabetic patients treated with insulin. [Diabetologia (1996) 39: 1392–1397] Received: 20 May 1996 and in revised form: 22 July 1996     

Role of functional performance in diagnosis of dementia in elderly people with low educational level living in Southern Italy

BACKGROUND AND AIMS: Diagnosis of dementia is often difficult in subjects with low educational level. Our aim was to evaluate the role of functional performance and the possibility of preferring scores of activities of daily living (ADL) and instrumental activities of daily living (IADL) in screening elderly people for diagnosis of dementia in a rural population of Southern Italy with a very high percentage of non-educated subjects. METHODS: a random sample of 300 residents, out of 1089 subjects over 60 years of age living in San Marcellino (Caserta, Campania), received door-to-door visit for information about their medical history, with clinical evaluation of general geriatric conditions, including the cumulative illness rating scale (CIRS). Dementia was diagnosed if subjects had a Clinical Dementia Rating score (CDR) > or = 1 and according to the criteria of DSMIV, but not according to scores on the Mini Mental State Examination (MMSE), ADL and IADL. Two hundred and nineteen normal subjects (NS) and 75 patients with dementia (DP) were evaluated. RESULTS: in NS, their mean age- and education-corrected MMSE score was 22.15 (lower than the normal cut-off value of 23.8) and 12.60 in DP (p<0.0001). In NS, the mean ADL score was higher than in DP (5.53 vs 2.64, p<0.0001); only age was correlated with ADL scores (coeff=-0.44, t=-4.557, p<0.0001). Assuming age as covariate, ADL scores highly differentiated DP from NS (F(1, 289)=26.083, p<0.0001). In both sexes, mean IADL scores were higher in NS than in DP (4.46 vs 1.80 in men, p<0.0001; 6.85 vs 2.31 in women, p<0.0001). Age and education did not influence IADL scores in men, but age greatly affected performance in women. IADL scores clearly differentiated NS from DP. In NS, a positive correlation was evident between ADL and IADL scores (r=0.234, p<0.0005), but neither scores correlated with the MMSE scores, even when correlation was performed separately for men and women. In DP, a strong correlation was observed between ADL and IADL scores (r=0.709, p<0.0001) and significant correlations were also evident between the scores of MMSE and both ADL (r=0.492,p<0.0001) and IADL (r=0.398, p<0.0004). CONCLUSIONS: in a rural community with a high prevalence of non-educated subjects, cognitive impairment is related to education, whereas independent functioning is limited mainly to age and not to cognition, if the latter remains (relatively) unimpaired. These results point to the importance of an "ecological" approach to the evaluation of elderly people, particularly those living in small rural communities, where education and the social environment may give rise to difficulties in diagnosis of dementia. The assessment of functional autonomy by ADL and IADL scales may be a better screening tool in diagnosing dementia than the MMSE scores.     

Factor Structure of the Geriatric Depression Scale in Cognitively Impaired Older Adults

This study explores the factor structure of the 30-item Geriatric Depression Scale (GDS) obtained from a sample of elderly adults who were diagnosed with some level of cognitive impairment. Principle components analysis obtained a four-factor solution utilizing all 30 GDS items. The derived factors were labeled Dysphoria, Meaninglessness, Apathy, and Cognitive Impairment. Although differences in factor structure were noted, some commonalities exist between this and former factor solutions conducted on healthy elderly populations. Further validation of this factor structure in the cognitively impaired elderly may provide another point of validation during clinical assessment and research when attempting to distinguish between geriatric depression, the onset of dementia, or a combination of these conditions.     

Development and application of the extended scale for dementia.

As part of an interdisciplinary study of organic dementia, a psychologic test for assessing the degree of dementia--the Extended Scale for Dementia--was developed through the expansion and rescoring of the original Mattis Dementia Scale. Statistical analyses of the 23 test items resulted in a scoring scheme which includes the "weighting" of items for scoring purposes. The test was successfully administered to 90 subjects from 6 hospitals in the London (Ontario) region. With use of the Extended Scale, it was possible to discriminate between dementia and non-dementia groups of psychogeriatric inpatients and to correlate the findings closely with those of another measure of the degree of dementia, viz, the London Psychogeriatric Rating Scale (Ment.). Dementia patients who were retested after 6-month and 12-month intervals showed a significant decline in scores. No significant scoring differences were noted between males and females or between Alzheimer dementia and multi-infarct dementia.     

A questionnaire investigation of anxiety and depression in early dementia

We report findings on a study of anxiety and depression by questionnaire in 50 patients with mild dementia and 134 control subjects using the Hospital Anxiety and Depression Scale. Thirty-eight percent of patients and 9% of controls had a possible or probable diagnosis of an anxiety disorder. Possible or probable depression was found in 28% of the patients and 3% of the controls. These rates for the patients were above those in normal populations. All patients and control subjects were tested with the Extended Scale for Dementia (ESD). Neither group showed a significant relationship between depression and ESD scores. In the control subjects there was a negative correlation (P less than .006) between anxiety and cognitive scores, one that was not found in the patients.     

Apolipoprotein E epsilon4, Alzheimer's disease, and cognitive performance in elderly Mexican Mestizos

OBJECTIVES: To determine the relationship between apolipoprotein E (APOE) ɛ4 and Alzheimer's disease (AD) in the Mexican Mestizo population, as well as its effects on the cognitive profile of AD and elderly Mestizos without dementia.
DESIGN: Cross-sectional analysis of a cohort study.
SETTING: Evaluations were conducted at the geriatrics clinic of an academic medical hospital in Mexico City.
PARTICIPANTS: Forty-nine elderly subjects with AD and 141 controls selected from a representative sample of Mexican Mestizos aged 65 to 96 who participated in the Prevalence Survey of Dementia in the Mexico City Elderly Population cohort.
MEASUREMENTS: All subjects underwent APOE genotypification and a comprehensive clinical and neuropsychological evaluation.
RESULTS: There were no significant differences in ɛ3 and ɛ4 frequencies between the subjects with and without AD and a tendency toward higher ɛ4 and lower ɛ3 allele frequencies in subjects aged 75 and older. No association was found between APOE ɛ4 and the presence of AD. The age-, sex-, and education-adjusted risk for AD associated with at least one ɛ4 allele was 1.01 (95% confidence interval=0.45–2.23). Performance on a long-term visual memory test was significantly worse in APOE ɛ4 carriers than in APOE ɛ3 carriers only in the group with AD.
CONCLUSION: APOE ɛ4 did not increase the risk for AD in this Mexican Mestizo elderly urban population sample, although the presence of this allele seems to modify its clinical expression.     

Effects of dementia on mortality in the radiation effects research foundation adult health study

BACKGROUND: Although dementia is rarely listed on death certificates, it does contribute to mortality. The predominant immediate causes of death coincident with dementia are pneumonia and cardiovascular diseases. OBJECTIVE: To estimate the impact of dementia on specific mortality risks. METHODS: We applied DSM-III/R criteria for Alzheimer's disease (AD) and vascular dementia (VaD) to 2,172 subjects of the Adult Health Study of the Radiation Effects Research Foundation who were 60 or more years old when examined from 1992 to 1996. The underlying causes of death were compiled from death certificates. We performed a Poisson regression analysis to evaluate specific causes of mortality for which AD or VaD was a significant risk factor. RESULTS: The relative risk of mortality was 2.2 for AD and 2.4 for VaD. Mortality from pneumonia and stroke was elevated for both types of dementia, independent of other medical conditions. AD was also associated with death from trauma. CONCLUSION: Dementia was a predictor of death due to pneumonia, stroke, and trauma among the Japanese elderly. The prevention and early detection of those conditions are important in the medical care and treatment of dementia cases.     

End-of-life care of elderly patients with dementia: A cross-sectional study of family carer decision-making

BackgroundDementia syndromes pose a major worldwide challenge to public health. In terminal stage of dementia, carers are responsible for decision making in end-of-life treatment and there may be multiple factors that contribute to the choice of a palliative or invasive treatment.AimTo identify possible factors that influence the decision-making of family caregivers on implementing invasive or palliative interventions for people with end stage dementia.DesignA structured interview with family caregivers of elderly patients addressing aspects of the following categories: elderly with dementia, caregiver, medical treatment history. Statistical analysis was performed to test whether there was a significant association between the carer's decision (invasive or palliative treatment) and the collected variables.Settingand participantsThe study was conducted in three hospitals in Brazil. Participants were family caregivers of inpatients with end stage dementiaResultsMost of caregivers chose not to perform invasive procedures. The factors with the greatest association with the decision for invasive care were: elderly with tracheostomy, dementia diagnosis for less than 2 years, caregiver's age less than 50 years, history of hospital admission in the last year, affirmation that interviewee would be surprised with the death of the elderly within 1 year and the denial that health care team has already explained about treatment options.ConclusionThere were identified several factors related to the carer, the elderly person and their medical treatment that may influence the choice between palliative and invasive care for the elderly person with dementia.