- Email: [email protected]
Assessing Alzheimer's disease patients' quality of life: Discrepancies between patient and caregiver perspectives
Alzheimer’s & Dementia - (2015) 1-11
Assessing Alzheimer’s disease patients’ quality of life: Discrepancies between patient and caregiver perspectives Sandrine Andrieua,b,*, Nicola Coleya,b, Yves Rollanda,c, Christelle Canteta,c, Catherine Arnauda,b, Sophie Guyonneta,c, Fati Nourhashemia,c, Alain Granda,b, Bruno Vellasa,c, and the PLASA group1 a Inserm-Toulouse University UMR1027, Toulouse, France CHU Toulouse, Department of Epidemiology and Public Health, Toulouse, France c Gerontopole, CHU Toulouse, Department of Geriatric Medicine, Toulouse, France b
Abstract
Introduction: Quality of life (QOL) is an important dimension to consider in Alzheimer’s disease (AD), but few large-scale studies have analyzed self and caregiver reports of patient QOL. Methods: Patient QOL was evaluated in a cohort of 574 AD patients with the QOL-AD scale over 2 years. Results: Caregiver reports of patient QOL were lower at baseline than self reports. Older patient age was associated with overestimation of QOL by caregivers, whereas neuropsychiatric inventory score and caregiver burden were associated with underestimation. Activities of daily living limitation, depressive symptoms, and caregiver burden were systematically associated with poorer QOL, whereas caregiver relationship and apathy were associated with poorer QOL only for self reports or caregiver reports, respectively. Cognitive function and professional care were not associated with QOL. Self-rated patient QOL did not change over time, whereas disease severity markers and caregiver-rated patient QOL declined. Discussion: It is important to assess both self and caregiver ratings when assessing patient QOL. Ó 2015 The Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
Keywords:
Alzheimer’s disease; Quality of life; QOL-AD; Caregivers; Proxy report; Longitudinal
1. Introduction Alzheimer’s disease (AD) is a chronic disease that requires care for several years and remains one the most frequent disabling diseases in aging populations [1,2]. It is a major cause of decreased quality of life (QOL) in older adults, and some studies have shown that more
There are conflicts of interest to disclose. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of the French Ministry of Health. The study sponsor had no role in the study. 1 See acknowledgments for the list of the PLASA study group members. *Corresponding author. Tel.: 133-5-61-14-59-50; Fax: 133-5-62-2642-40. E-mail address: [email protected]
than the disease itself, it is its disability-related impact that affects QOL [3]. Simple measures of cognitive or functional decline are not sufficient to address the complexity of AD, and QOL is a useful additional dimension to consider as a broader outcome measure which may represent more “clinically significant” changes or benefits to the patient and caregiver [4]. QOL is a broad conceptual field, linked to physical health as well as psychological state, level of independence, social relationships, personal beliefs, and the subject’s relationship with the specifics of their environment [5]. Researchers [6,7] and regulatory agencies [8] emphasize the need to take into account patient QOL in the evaluation of new drugs but stress the need for further validation work to use it as an end point in clinical trials. For example, little is known about the natural history of QOL in AD, the determinants
http://dx.doi.org/10.1016/j.jalz.2015.09.003 1552-5260/Ó 2015 The Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
2
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
of high and low QOL, or the effects of medication on QOL [9]. Caregiver ratings are frequently used as part of multidimensional evaluations of AD patients, in particular to measure patients’ functional status and neuropsychiatric symptoms, and caregivers may also be asked to rate patients’ QOL [4]. However, because QOL is subjective and defined as an individual perception, it should ideally be assessed from the subject’s own perspective [10,11], but declining cognitive function makes this harder to do in AD [10]. Many QOL scales have been proposed in AD [12], including the QOL-AD scale which requires both the patient and caregiver to rate the same domains of patient QOL [10]. Several previous studies [13–17] have reported discrepancies between caregiver and self reports of AD patients’ QOL, and some have identified factors associated with divergent ratings [15–19]. Most studies were small scale and targeted populations coming from particular settings, such as residential care homes [20], or from therapeutic trials, thus providing a selected sample with an abnormally low rate of comorbidities [21,22]. Furthermore, very few studies have examined longitudinal changes in patient and proxy QOL reports over .1 year of follow-up [18]. The primary objective of this multicentre study was to explore discrepancies between self and caregiver reports of patient QOL in a large population of well-characterized community-dwelling AD patients at baseline and during 2 years of follow-up and to determine factors associated with disagreement at baseline. The secondary objective was to assess factors explaining a low level of self- or caregiver-reported patient QOL at baseline. 2. Methods This study includes AD patients and their primary caregivers enrolled in the Plan de soin et d’aide dans la maladie d’Alzheimer (PLASA) study, a randomized trial that tested the efficacy of a standardized care plan versus usual dementia care on functional decline over 2 years of follow-up. PLASA is described in detail elsewhere [23]. 2.1. Subjects Briefly, a total of 1131 AD patients and their caregivers were recruited in a French nationwide network of 50 memory centers. To be included, subjects had to meet the following inclusion criteria: diagnosis of probable or possible AD according to the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria [24], mini-mental status examination (MMSE) [25] score between 12 and 26, community-dwelling, looked after by a well-identified informal caregiver, and not participating in any other research program. In the present analysis, we included only the 574 patients randomly assigned to the intervention group to be able to study the determinants of
QOL and patient-proxy agreement because this group underwent more comprehensive evaluations than the usual care group (as part of the intervention). The intervention had no significant effect on any of the primary or secondary efficacy measures, including QOL. The study was funded by the Ministry of Health and was approved by the Institutional Review Board and ethics committee of Toulouse University. Written informed consent was obtained from all patients and their caregivers. This trial was registered in clinicaltrials.gov (identifier: NCT00480220). 2.2. Measurements Patients were evaluated every 6 months in memory centers in general or university hospitals. 2.2.1. QOL measure Patient QOL was assessed by the QOL-AD scale [10], a 13-item questionnaire designed to provide patient and caregiver reports of the patient’s QOL which covers various domains (physical health, energy, mood, living situation, memory, family, marriage, friends, self as a whole, ability to do chores, ability to do things for fun, financial situation, and QOL as a whole). Items are scored on a 4-point Likert scale ranging from 1 (poor) to 4 (excellent). Total scores range from 13 to 52 with higher scores indicating better QOL. As proposed by Logsdon [10], missing items were replaced by the mean score of the remaining items if no more than two items were missing; otherwise, the score was considered missing. 2.2.2. Other measures Sociodemographic data (age, sex, educational level, and living arrangements), time since diagnosis of dementia as declared by the caregiver, comorbidities, use of support services, and assistance in the home (home help, supervision, meal delivery service, and nursing care) were recorded using a standardized questionnaire. Number of medications used and use of anti-AD drugs (i.e., cholinesterase inhibitors, N-Methyl-D-aspartate receptor antagonists) were recorded. Dementia severity was evaluated based on assessments of cognition (MMSE) [25], function (activities of daily living [ADL] [26], instrumental activities of daily living [IADL] [27]), and behavioral disorders (neuropsychiatric inventory [NPI]) [28]. Nutritional status was evaluated with the mini nutritional assessment (MNA) [29], and caregiver burden was evaluated using the Zarit Burden Interview [30]. 2.3. Data analysis Baseline mean total scores and individual domain scores were calculated for patient- and caregiver-rated QOL when a score was available for both members of the dyad. To address the extent of agreement between patient and caregiver reports, Spearman and intraclass correlation coefficients (ICCs) were computed for each item at
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
baseline. Correlations between total scores were also evaluated via the Pearson correlation coefficient and the ICC. Differences between mean patient and caregiver scores for total QOL-AD score and individual domains were tested using paired Student’s t-tests or nonparametric Wilcoxon tests. Baseline patient and caregiver characteristics of dyads in agreement for the total QOL-AD score were compared with those of dyads in which patient and caregiver scores were not in agreement. Disagreement was separated into caregivers who scored lower and caregivers who scored higher than their patient. Statistical significance was assessed with Fisher (analysis of variance) or Kruskal-Wallis tests for quantitative variables and c2 tests for qualitative variables. Variables with a P value of ,.2, and patient age and sex, were entered into a multivariate multinomial logistic regression model, and a backward stepwise selection procedure was used to identify factors that were significantly and independently associated with disagreement (underestimation or overestimation of patient QOL by caregivers compared with self-ratings). Prospective changes over time in self- and caregiverreported patient QOL were analyzed using a three-level repeated measures linear-mixed regression model with random intercept and random linear slope to account for the within-subject correlations due to repeated observations and correlations within clusters (centers) [31]. We took into account all available data during follow-up (i.e., until the end of the study or until the participant dropped out, was institutionalized, or there was a change in the caregiver). Determinants of poorer self- and caregiver-reported patient QOL at baseline were assessed using two separate analyses. In the absence of established cut-offs, a low level of QOL was defined as the lowest quartile, corresponding to a score ,31.0 for patient reports and 27.1 for caregiver reports. First, in a bivariate analysis, baseline differences between subjects with a poorer level of QOL and subjects with higher QOL were tested using t-tests for continuous variables and c2 tests for categorical variables. Variables were eligible for inclusion in the multivariate analysis if they were associated with a low level of QOL in bivariate analyses with P .20. For the multivariate analysis, a series of logistic regression models were performed. Our initial hypothesis was that after taking into account the severity of the disease, caregiver characteristics would influence the patient’s level of QOL. Thus, for both patient- and caregiver-rated patient QOL, we first modeled the association between a poorer level of QOL and patient characteristics (model 1). We then performed an analysis including only caregiver characteristics (model 2). Finally, we fit a global model which included both patient and caregiver characteristics that were significantly associated with QOL in models 1 and 2 (model 3). We used a backward stepwise selection procedure to retain the factors independently associated with the level of QOL. We present adjusted odds ratios
3
(OR) and 95% confidence intervals (CIs) for significant variables. Model fit was examined using the HosmerLemeshow goodness-of-fit test [32]. Subjects with missing data for any of the variables included in the models were excluded from this analysis. All P values were based on two-sided tests and were considered statistically significant at P , .05. Analyses were conducted using SAS 9.1 software (SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Population characteristics At baseline, the 574 patients had a mean age of 79.6 years (standard deviation [SD], 5.9) and 66.6% were women. On average, they had been diagnosed with AD for 1.22 years (1.44). Their mean MMSE score was 19.5 (3.9), and 76.5% were treated with cholinesterase inhibitors or memantine before inclusion in the study. Most patients showed limitations in more than two IADLs (64.2%) and 45.1% had at least one basic ADL limitation. Behavioral disorders were frequent in this population, with 93.8% subjects presenting at least one NPI symptom (mean NPI score was 17.7 [16.4]). A third (32.6%) of patients lived at home alone. Caregivers were mostly female (66.2%) with a mean age of 64.1 years (13.7). They were mainly the spouse (47.2%) or child (44.6%) of the patient. 3.2. Agreement between self and caregiver reports of patient QOL at baseline At baseline, 93.0% of patients (n 5 534) provided a selfrating of their QOL, and 91.6% of caregivers (n 5 526) provided a proxy rating of their patient’s QOL. This resulted in 501 dyads (87.3%) with total scores for both self- and caregiver-reported QOL. Caregiver reports of patient QOL were significantly lower than patient self-reports for the total mean QOL-AD score (31.03 [5.35] vs. 34.54 [5.29]; P , .001), and for all items except living situation (Table 1). The mean difference between caregiver and patient total scores (3.51) was .0.5 SD (of the caregiver’s score), which is a widely recognized definition of clinically meaningful difference in QOL [33]. Correlations between AD patients and caregiver scores for the individual items and for the total score were low (Spearman/Pearson correlations between 0.10 and 0.53; intraclass correlations between 0.08 and 0.49; Table 1). For total QOL-AD scores, 50% of caregivers underestimated patient QOL compared with self-ratings, whereas 12% of caregivers overestimated patient QOL, and 38% agreed with patient ratings (Fig. 1). Agreement for individual items of the QOL-AD scale ranged from 41% to 56% of dyads, and caregivers were more likely to underestimate rather than overestimate patient QOL, as compared with self-ratings, for all items.
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
4
Table 1 Differences between patient and caregiver ratings of patient QOL at baseline (based on complete patient-caregiver pairs for each domain) Domain Physical health Energy Mood Living situation Memory Family Marriage Friends Self as a whole Ability to do chores Ability to do things for fun Finances Life as a whole QOL-AD score
Complete pairs, n (%) 563 (98.1) 563 (98.1) 563 (98.1) 563 (98.1) 562 (97.9) 562 (97.9) 521 (90.8) 518 (90.2) 546 (95.1) 533 (92.9) 518 (90.2) 541 (94.3) 550 (95.8) 501 (87.3)
Patient score, mean (SD) 2.61 (0.67) 2.52 (0.73) 2.55 (0.66) 2.97 (0.61) 1.84 (0.70) 3.18 (0.74) 2.99 (0.69) 2.91 (0.71) 2.54 (0.67) 2.66 (0.74) 2.39 (0.79) 2.71 (0.69) 2.64 (0.70) 34.54 (5.29)
Caregiver score, mean (SD) 2.51 (0.75) 2.12 (0.80) 2.34 (0.70) 2.93 (0.62) 1.34 (0.52) 2.89 (0.79) 2.71 (0.75) 2.71 (0.79) 2.32 (0.84) 2.29 (0.87) 1.91 (0.80) 2.56 (0.86) 2.29 (0.75) 31.03 (5.35)
Difference, mean (SD) y
0.10 (0.88) 0.40 (0.93)z 0.21 (0.82)z 0.04 (0.81) 0.51 (0.83)z 0.28 (0.90)z 0.28 (0.72)z 0.21 (0.87)z 0.21 (0.91)z 0.37 (0.95)z 0.48 (0.93)z 0.14 (0.99)y 0.35 (0.92)z 3.51 (5.90)z
Correlation*
ICC
0.2647 0.2664 0.2743 0.1444 0.1003 0.2678 0.5279 0.3346 0.2627 0.3081 0.3098 0.1707 0.2059 0.3849
0.2488 0.2580 0.2810 0.1239 0.0841 0.3086 0.4884 0.3075 0.2633 0.2926 0.3140 0.1896 0.1992 0.3675
Abbreviations: QOL, quality of life; SD, standard deviation; ICC, intraclass correlation coefficient; AD, Alzheimer’s disease. *Spearman correlation coefficients for individual items; Pearson correlation coefficient for total score. P values are ,.05 for all correlation coefficients. y Two-sided Student paired t-test/nonparametric Wilcoxon test P , .01. z Two-sided Student paired t-test/nonparametric Wilcoxon test P , .001.
3.3. Factors associated with patient and caregiver agreement Significant differences between dyads in agreement for the total QOL-AD score and those in which patient and caregiver scores were not in agreement were observed for both patient (IADL, ADL, MNA, one-leg balance, and NPI total score and certain items) and caregiver (level of subjective burden) characteristics (Table 2). In the multivariate multinomial logistic regression model, older patient age (OR, 1.08; 95% CI, 1.02–1.14) was significantly associated with overestimation of QOL by caregivers compared with patient
self-ratings, and NPI total score (OR, 1.02; 95% CI, 1.00– 1.04) and higher caregiver burden (low-moderate burden: OR, 1.67; 95% CI, 1.02–2.72; moderate-severe burden: OR, 1.99; 95% CI, 0.93–4.23) were significantly associated with underestimation of QOL by caregivers compared with patient self-ratings (Table 2). 3.4. Two-year changes in patient- and caregiver-reported patient QOL During the 2 years of follow-up, 43 patients died and 127 dropped out for other reasons. Patients who remained in the
Fig. 1. Percentage (of complete pairs) of patient-caregiver dyads in agreement or disagreement for individual items and total score of the QOL-AD scale at baseline. Abbreviation: QOL-AD, quality of life-Alzheimer’s disease.
Table 2 Factors associated with disagreement between self and caregiver ratings of patient quality of life for the total score of the QOL-AD scale at baseline (unadjusted and multivariate analyses) Unadjusted analysis Characteristics
Agreement (n 5 189)
Patient score , caregiver score (n 5 62)
Patient score . caregiver score (n 5 250)
78.9 (5.6) 127 (67.2)
79.7 (7.2) 38 (61.3)
79.6 (5.7) 166 (66.4)
49 (25.9) 73 (38.6) 34 (18.0) 33 (17.5) 1.2 (1.5) 150 (79.4) 4.0 (2.2) 19.6 (3.9) 69 (36.5) 80 (44.2) 45 (24.7) 13.4 (13.4)
15 (24.2) 22 (35.5) 10 (16.1) 15 (24.2) 1.1 (1.3) 43 (69.4) 4.0 (2.4) 20.3 (4.0) 21 (33.9) 29 (47.5) 16 (27.6) 12.2 (13.0)
54 (21.7) 110 (44.2) 58 (23.3) 27 (10.8) 1.3 (1.4) 192 (76.8) 4.2 (2.5) 19.5 (3.9) 122 (48.8) 138 (56.6) 86 (36.0) 21.0 (17.6)
64 (33.9) 107 (56.6) 18 (9.5) 67 (35.5) 30 (16.0)
19 (30.7) 36 (58.1) 7 (11.3) 15 (24.2) 5 (8.2)
73 (29.2) 139 (55.6) 38 (15.2) 101 (40.9) 50 (20.0)
64.3 (13.9) 119 (63.0)
66.7 (13.1) 37 (59.7)
64.6 (13.8) 169 (67.6)
93 (49.2) 80 (42.3) 16 (8.5)
34 (54.8) 23 (37.1) 5 (8.1)
126 (50.4) 106 (42.4) 18 (7.2)
92 (53.8) 61 (35.7) 18 (10.5)
37 (63.8) 12 (20.7) 9 (15.5)
82 (36.3) 102 (45.1) 42 (18.6)
P .416 .687 .077
.295 .269 .656 .387 .013 .036 .040 ,.001 .453
OR (95% CI), patient score , caregiver score
P
OR (95% CI), patient score . caregiver score
P
1.08 (1.02–1.14)
.012
1.03 (0.99–1.07)
.109
0.99 (0.96–1.01)
.331
1.02 (1.00–1.04)
.022
1 0.63 (0.29–1.38) 1.79 (0.64–5.00)
.248 .269
1 1.67 (1.02–2.72) 1.99 (0.93–4.23)
.040 .075
.046 .080 .472 .396 .927
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
Patient characteristics Age, mean (SD), y Gender, women, n (%) Educational level Elementary or no formal, n (%) Primary school certificate, n (%) Secondary education, n (%) Tech/high school certificate or higher, n (%) Years since diagnosis, mean (SD) Use of AD medication, n (%) Number of other medications, mean (SD) MMSE score, mean (SD) Limitation in 1 ADL, n (%) Limitation in .2 IADLs, n (%) MNA score 23.5, n (%) NPI total score, mean (SD) Living arrangements Living alone, n (%) With spouse, n (%) Other, n (%) Abnormal one-leg balance, n (%) Autonomy allowance, n (%) Caregiver characteristics Age, mean (SD), y Gender, women, n (%) Relationship with patient, n (%) Spouse Child Other Subjective level of burden, ZBI, n (%) Absent to low (score 0–20) Low to moderate (score 21–40) Moderate to severe (score .40)
Multivariate (adjusted) analysis*
,.001
5
Abbreviations: QOL, quality of life; AD, Alzheimer disease; OR, odds ratio; CI, confidence interval; SD, standard deviation; MMSE, mini-mental status examination (scores range from 0 to 30, with higher scores indicating better function); ADL, activities of daily living (scores range from 0 to 6, with higher score indicating better function, dichotomized as 0 vs. 1 limitation); IADL, instrumental activities of daily living (scores range from 0 to 8, with higher scores indicating better function, dichotomized as 0 vs. .2 limitations); MNA, mini nutritional assessment (score range from 0 to 30, with higher score in better nutritional status, with a score 23.5 indicating that the patient is at risk of malnutrition or is malnourished); NPI, neuropsychiatric inventory (score range from 0 to 144, with higher score indicating more severe symptoms); ZBI, Zarit Burden Interview (scores range from 0 to 88, with higher scores indicating higher levels of burden, scores 20 represent low-or-no burden; scores of 21–40 represent low-to-moderate burden, and scores .40 represent moderate-to-severe burden). NOTE. Global P values in the final model were: Patient age (P 5 .031); NPI total score (P 5 .007); and caregiver burden (P 5 .027). Bold text indicates P ,.05. *Adjusted odds ratios are shown for variables that remained in the final multivariate multinomial model. Variables with a P value of ,.2, and patient age, sex, and level of education, were entered into the initial model, and a backward stepwise selection procedure was used to identify factors that were significantly and independently associated with underestimation or overestimation of patient QOL by caregivers compared with self-ratings. Variables that did not remain in the final model (following backward stepwise selection) were level of education, level of functional autonomy (IADL, ADL), nutritional status (MNA), receipt of autonomy allowance, and one-leg balance.
6
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
study were significantly younger and had better cognitive, functional, and neuropsychiatric scores at baseline, as well as higher self- and caregiver-reported QOL. They were also more likely to live with a spouse at baseline and their caregivers reported significantly lower levels of burden at baseline. Mean patient self-reported QOL remained stable over the 2 years of follow-up, (34.55 [standard error, 0.48] at baseline and 34.50 [0.51] at 2 years, P 5 .857), whereas mean caregiver-reported patient QOL decreased from 30.75 (0.36) to 28.64 (0.40) (P , .001; Table 3). In this period, there was significant progression of disease severity, as measured by cognitive, functional, and behavioral measures (Table 3). 3.5. Factors associated with a low level of QOL at baseline Table 4 shows patient and caregiver characteristics associated with poorer patient QOL at baseline. The results of the multivariate logistic regression models adjusted for covariates showed that several factors linked to AD severity were associated with QOL, regardless of who (patient or caregiver) made the assessment; having at least one basic ADL limitation and the presence of depressive symptoms according to the NPI were both associated with poorer QOL. The level of subjective caregiver burden was also independently associated with poorer QOL whether it was self rated or caregiver rated. Furthermore, caregivers tended to rate QOL lower for female patients compared with male patients; and for selfreported QOL, patients cared for by their children were more likely to declare poorer QOL than those cared for by their spouses. The presence of apathy was independently associated with poorer QOL for caregiver-reported scores only. Neither the level of cognitive impairment nor the number of comorbidities was associated with patient QOL. The use of support services or home assistance was also not associated with the level of QOL.
4. Discussion We observed discrepancies between caregiver and self ratings of patient QOL at baseline and during 2-years of follow-up in a population of community-dwelling AD patients who were evaluated every 6 months. In cross-sectional analyses, we found that caregivers systematically underestimated patient QOL compared with patient self-ratings for the total QOL-AD score and for most of the domains covered by the scale. This result is consistent with previous findings reported in the literature [13–17]. Discrepancies between patient and caregiver ratings at baseline were not associated with patient cognitive or functional status, nor with the patient-caregiver relationship. However, increasing frequency and severity of neuropsychiatric symptoms, and a higher level of caregiver burden, were significantly and independently associated with underestimation of patient QOL by caregivers compared with selfratings. Caregivers may consider the effects of burden and neuropsychiatric symptoms (which may affect them more than the patient) on their own QOL when rating those of the patient. Our results are consistent with recent findings from smaller studies demonstrating an association between caregiver psychosocial well-being (particularly level of burden and depression) and discrepancies in patient QOL ratings [15–19] and confirm that this association remains even after adjustment for numerous patient characteristics (including living alone) in a population of patients who were exclusively community dwelling. An original finding of our study was that increasing patient age was associated with overestimation of patient QOL by caregivers compared with patients, suggesting that they may take into account age in their ratings. During the 2 years of follow-up, there was a progressive worsening of dementia in the AD patients included in this study, based on cognitive, functional, and behavioral assessments. Despite this worsening, mean self-rated patient QOL did not change during this time, whereas caregiver ratings of patient QOL significantly declined. This finding
Table 3 Two-year changes in disease severity parameters and self- and caregiver-reported patient QOL with linear mixed model (n 5 574) Parameter Patient characteristics Basic activities of daily living (ADL score) Instrumental activities of daily living (IADL/8 score) Cognitive function (mini-mental status examination score) Behavioral disturbances (NPI score) Self-reported QOL score Caregiver evaluations Subjective burden (ZBI score) Caregiver-reported QOL score
Baseline, mean (SE)
12 mo, mean (SE)
24 mo, mean (SE)
Slope*, 12 mo, mean (SE)
Slope*, 24 mo, mean (SE)
P
5.35 (0.06) 4.26 (0.17)
4.79 (0.07) 3.18 (0.17)
4.21 (0.09) 2.40 (0.18)
20.56 (0.04) 21.08 (0.08)
21.15 (0.06) 21.86 (0.11)
,.001 ,.001
19.44 (0.19)
17.90 (0.23)
15.75 (0.29)
21.54 (0.15)
23.69 (0.23)
,.001
18.26 (1.20) 34.55 (0.48)
19.05 (1.17) 34.52 (0.48)
19.85 (1.26) 34.50 (0.51)
0.80 (0.39) 20.02 (0.13)
1.59 (0.77) 20.05 (0.25)
.040 .857
24.14 (0.92) 30.75 (0.36)
26.62 (0.93) 29.70 (0.35)
29.10 (1.06) 28.64 (0.40)
2.48 (0.35) 21.05 (0.14)
4.95 (0.70) 22.11 (0.28)
,.001 ,.001
Abbreviations: QOL, quality of life; SE, standard error; ADL, activities of daily living; IADL, instrumental activities of daily living; NPI, neuropsychiatric inventory; ZBI, Zarit Burden Interview. *Slope estimated from the mixed model.
Table 4 Patient and caregiver characteristics associated with poorer self- or caregiver-reported patient QOL at baseline Self-reported patient QOL, n 5 534
Characteristics
Quartiles 1–3 (QOL-AD score 31)
Lowest quartile (QOL-AD score 27.1)
Quartiles 1–3 (QOL-AD score .27.1)
Adjusted OR (95% CI)z
79.3 (6.2) 92 (70.2)
79.5 (5.8) 262 (65.0)
79.5 (5.2) 104 (75.9)*
79.4 (6.1) 244 (62.7)*
2.0 (1.2–3.4)
.010
27 (20.6) 60 (45.8) 22 (16.8) 22 (16.8) 1.5 (1.5)* 4.1 (2.2) 109 (83.2)* 19.0 (3.8) 74 (56.5)* 76 (59.4)* 44 (36.1) 18 (13.7)
97 (24.1) 163 (40.6) 87 (21.6) 55 (13.7) 1.2 (1.4)* 4.1 (2.4) 299 (74.2)* 19.8 (3.9) 158 (39.2)* 184 (48.9)* 118 (30.5) 53 (13.2)
37 (27.0) 56 (40.9) 31 (22.6) 13 (9.5) 1.5 (1.5)* 4.2 (2.4) 108 (78.8) 18.6 (3.9)* 87 (63.5)* 84 (64.1)* 49 (37.4) 28 (20.4)*
92 (23.8) 155 (40.1) 73 (18.9) 67 (17.3) 1.1 (1.4)* 4.1 (2.3) 297 (76.4) 19.9 (3.8)* 141 (36.3)* 172 (46.6)* 107 (28.8) 47 (12.1)*
2.0 (1.2–3.3)
.005
24 (18.5) 13 (10.0) 50 (38.5) 76 (58.9)* 73 (56.2) 9 (6.9) 89 (68.5)* 28 (21.7) 56 (43.1) 29 (22.5) 33 (26.6) 52 (40.3)*
85 (21.1) 38 (9.4) 161 (40.0) 178 (44.3)* 194 (48.1) 50 (12.4) 224 (55.7)* 81 (20.1) 181 (44.9) 96 (23.9) 90 (22.9) 112 (27.9)*
41 (29.9)* 16 (11.7) 69 (50.4)* 91 (66.4)* 81 (59.1)* 12 (8.8) 108 (78.8)* 42 (30.9)* 75 (54.7)* 39 (28.9) 43 (32.6)* 52 (38.2)*
65 (16.8)* 36 (9.3) 137 (35.3)* 163 (42.2)* 183 (47.2)* 47 (12.1) 194 (50.1)* 67 (17.3)* 155 (40.0)* 81 (20.9) 81 (21.5)* 105 (27.1)*
2.3 (1.4–3.7)
,.001
2.3 (1.3–4.0)
.003
46 (35.1) 67 (51.2) 18 (13.7)
122 (30.3) 229 (56.8) 52 (12.9)
45 (32.9) 67 (48.9) 25 (18.3)
122 (31.4) 221 (56.8) 46 (11.8)
62.1 (13.2)* 90 (68.7)
65.2 (13.9)* 258 (64.0)
53 (40.5) 65 (49.6) 13 (9.9)
209 (51.9) 166 (41.2) 28 (6.9) * 173 (49.3)
61.8 (14.6)* 92 (67.2) * 56 (40.9) 70 (51.1) 11 (8.0) * 22 (18.6)
65.6 (13.3)* 250 (64.3) * 205 (52.7) 154 (39.6) 30 (7.7) * 197 (56.1)
* 43 (36.1)
Adjusted OR (95% CI)y
2.0 (1.3–3.1)
1.8 (1.1–2.8)
P
.002
.012
1 1.8 (1.1–2.8) 2.0 (0.9–4.4)
— .015 .111
1
—
1
P
7
— (Continued )
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
Patient characteristics Age, mean (SD), y Gender, women, n (%) Educational level, n (%) Elementary or no formal Primary school certificate Secondary education Tech/high school certificate or higher Time elapsed since diagnosis, mean (SD), y Number of medications, mean (SD) Use of AD medication, n (%) MMSE score, mean (SD) Limitation in at least one ADL, n (%) Limitation in at least three IADLs, n (%) MNA score 23.5, n (%) History of osteoarthritis, n (%) Behavioral symptoms, NPI items, n (%) Delusions Hallucinations Agitation Depression Anxiety Elation/euphoria Apathy Disinhibition Irritability Aberrant motor behavior Sleep Appetite and eating disorder Living arrangements, n (%) Living alone With spouse Other Caregiver characteristics Age, mean (SD), y Gender, women, n (%) Relationship with patient, n (%) Spouse Child Other Subjective level of burden, ZBI, n (%) Absent to low (score 0–20)
Lowest quartile (QOL-AD score ,31)
Caregiver-reported patient QOL, n 5 526
,.001 ,.001 3.8 (2.1–6.7) 7.2 (3.6–14.4) 119 (33.9) 35 (10.0) 59 (50.0) 37 (31.4) .527 .006 1.2 (0.7–1.9) 2.3 (1.3–4.3) 135 (38.5) 43 (12.3) 46 (38.7) 30 (25.2)
P Adjusted OR (95% CI)y Quartiles 1–3 (QOL-AD score 31) Characteristics
Low to moderate (score 21–40) Moderate to severe (score .40)
P Adjusted OR (95% CI)z Lowest quartile (QOL-AD score 27.1)
Quartiles 1–3 (QOL-AD score .27.1)
Caregiver-reported patient QOL, n 5 526 Self-reported patient QOL, n 5 534
Lowest quartile (QOL-AD score ,31)
Table 4 Patient and caregiver characteristics associated with poorer self- or caregiver-reported patient QOL at baseline (Continued )
Abbreviations: QOL, quality of life; AD, Alzheimer disease; OR, odds ratio; CI, confidence interval; MMSE, mini-mental status examination (scores range from 0 to 30 with higher scores indicating better function); ADL, activities of daily living (scores range from 0 to 6 with higher score indicating better function, dichotomized as 0 vs. 1 limitation); IADL, instrumental activities of daily living (scores range from 0 to 8 with higher scores indicating better function, dichotomized as 0 vs. .2 limitations); MNA, mini nutritional assessment (score range from 0 to 30, with higher score in better nutritional status, with a score 23.5 indicating that the patient is at risk of malnutrition or is malnourished); NPI, neuropsychiatric inventory (score range from 0 to 144 with higher score indicating more severe symptoms); ZBI, Zarit Burden Interview (scores range from 0 to 88, with higher scores indicating higher levels of burden, scores 20 represent low-or-no burden; scores of 21–40 represent low-to-moderate burden, and scores .40 represent moderate-to-severe burden). NOTE. Bold text indicates P ,.05. *P , .05. y Adjusted odds ratios for poorer quality of life are shown for variables that remained in the final multivariate logistic regression model following the stepwise selection procedure. For self-rated QOL analyses, model 3 initially included limitations in ADL, behavioral troubles (depression, elation/euphoria), caregiver-patient relationship, and the level of subjective burden experienced by the caregiver. z Adjusted odds ratios for poorer quality of life are shown for variables that remained in the final multivariate logistic regression model following the stepwise selection procedure. For caregiver-rated QOL, model 3 initially included patient sex, limitations in ADL, behavioral troubles (delusion, depression, and apathy), caregiver-patient relationship, and the level of subjective burden experienced by the caregiver.
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
8
confirms results from smaller studies of trajectories of selfrated QOL over time in community-dwelling patients with dementia [18,34–36]. It is important to understand this increasing discrepancy between the caregiver and the patient and to consider which QOL assessment must be taken into account in the evaluation of the patient’s situation. The lack of change in patients’ self-reported QOL over time could be caused by patients having reduced abilities to judge their own difficulties due to increased cognitive impairment. It has been suggested that only patients with an MMSE score .10 can reliably report their own QOL [10]. As only 16% of patients had MMSE scores ,10 at 2 years in our study, the lack of change in patient self-reported QOL in our study may not be primarily due to cognitive impairment. Another explanation could be that AD patients adapt, through coping mechanisms, to the new difficulties that they encounter [37], and so may not perceive any QOL decline. Some initiatives that have attempted to better understand the meaning of QOL for patients with dementia have revealed that these patients share a need to maintain a sense of usefulness in their lives which could contribute to their QOL [38]. There may also be reasons related to the caregiver which could explain the differences in patient and caregiver ratings of the patient’s QOL over time. For example, the caregiver’s ratings may be influenced by their own expectations, mood, burden of care, and their relationship with the patient [10,39]. The decline of caregiver-rated patient QOL over time, in contrast to patient self-ratings, could also suggest a worsening of the caregiver’s perception of the patient’s situation. Such misunderstandings of the patient’s situation may bring about mismanagement of the disease [40]. In terms of factors associated with patient QOL, our study was in agreement with previous findings, suggesting that functional impairment [20] and neuropsychiatric symptoms [41,42] play an important role in determining patient QOL but not cognitive status [42,43]. Patient age was not associated with QOL in our study. This is in line with some other studies, in particular those with relatively homogeneous populations in terms of age [22]. We also showed that patient QOL, whether rated by the patient or by the caregiver, was independently affected by the level of caregiver burden. Other studies have also found caregiver burden, and caregiver depression, to be linked to patient QOL [10], with a seemingly greater effect on caregiverratings than on self-ratings, thus raising the issue of the direction of causality between caregiver burden and his/her rating of the patient’s QOL. Interventions aimed at decreasing caregiver burden could provide benefits for the caregiver, but also for the patient. Among the strengths of the present study are the detailed characterization of clinical parameters by experts in AD, using standard outcome measures which are widely used in clinical trials, as well as the assessment of patient QOL from both the patient’s and caregiver’s point of view in a large community-dwelling sample. The scale we used is well
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
adapted for longitudinal studies of AD patients, despite progression of disease symptoms [42], and is not overly dependent on functional status, unlike some generic QOL scales. Consequently, most of the subjects in our study were able to self-rate their QOL. Our patients were relatively homogenous in terms of age, in contrast to numerous other studies with much variation in subject age, which can create large differences in QOL because individuals’ expectations vary according to the period of life [44]. However, several limitations should be considered in the interpretation of our results. Our study was based on a sample of diagnosed AD patients with access to health care systems, rather than a representative sample of the general population. Thus, our results may not be generalizable to all patients with AD. The inclusion of caregivers took place after the diagnosis of AD, and consequently after the beginning of their caregiving activities. Further selection bias may have arisen through study dropouts due to death or institutionalization. We can hypothesize that patients with the most stable management were overrepresented and so perceived QOL decreased little in our cohort even though patients’ needs increased and caregivers were more involved in caregiving over time. 5. Conclusion Further analyses are required to study the temporal relationship between the level of caregiver burden and patient QOL, identify determinants of increasing patient-caregiver disagreement about patient QOL over time, and better assess QOL as a valuable end point to measure the impact of future therapeutic or nontherapeutic interventions. It remains difficult to determine which type of rating (self or proxy) of AD patients’ QOL provides the most accurate picture. However, given the subjective nature of QOL, it is of upmost importance to take into account the patient’s own rating. Acknowledgments Authors’ contributions: S.A., F.N., S.G., and B.V. contributed to the conception and design. S.G., S.A., and F.N. did the data acquisition. S.A., C.C., N.C., Y.R., and C.A. did the data analysis and interpretation. S.A., N.C., and A.G. did the drafting of the article. All authors contributed to the critical review of the article and final approval of the version to be published. B.V. obtained the funding and contributed to the study supervision. All authors take responsibility for the entire article. The PLASA study was supported by the French Ministry of Health grant: PHRC 02-006-01. The authors thank Helene Grandjean for her helpful advice concerning the analysis and article preparation and Clarisse Ventelon for her initial participation in this analysis. The authors are indebted to the investigators from the following hospitals participating in the PLASA study: University hospital: Angers (Dr. Berrut, Dr Barre), Bordeaux (Pr. Rainfray, Dr Harston), Brest (Dr Gentric),
9
Grenoble (Pr. Franco, Dr. Couturier), Lille (Dr Roche, Dr Huvent), Lyon (Pr Bonnefoy), Marseille (Dr. Michel), Montpellier (Pr. Jeandel), Montpellier (Pr. Touchon, Dr. Portet, Dr Lerouge), Nice (Dr. Brocker, Dr Guerin), Nice (Pr. Robert), Paris (Pr Teillet, Dr Lechowski), Paris (Pr. Belmin, Dr Pariel-Madjlessi), Paris (Pr. Rigaud, Dr Latour), Paris (Pr Verny, Dr Marquis, Dr Brihier), Paris (Pr Legrain, Dr Girard), Reims (Pr Blanchard, Dr Kack), Rennes (Pr Jouanny, Dr Ledu), Rouen (Pr Chassagne, Dr Levasseur), and Rouen (Pr Hannequin, Dr Dugny). General hospital: Albi (Dr Quincon), Ales (Dr Peju), Annecy (Dr Picot), Bar Le Duc (Dr Tagu, Dr De Guio), Carcassonne (Dr Tannie), Carvin (Dr Taillez), Chambery (Dr Declippeleir), Champcueil (Dr Maugourd, Dr Baptiste), Dieppe (Dr Pesque), Elbeuf (Dr Simon), Grasse (Dr Ribiere), Lannemezan (Dr Bordes), Lavaur (Dr De Pemille), Le Havre (Dr Landrin-Dutot, Dr Olivier), Lens (Dr Senechal, Dr Fournier), Nice (Dr Giordana), N^ımes (Dr Strubel), Niort (Dr Chaumier), Paris (Dr DurandGasselin), Pau (Dr. De La Fourniere, Dr Sauvanier—Geriatric Unit), Plaisir (Dr Bessey, Dr Drunat), Roubaix (Dr Forzy), Rouen (Dr Moynot, Dr Denis), Saint Dizier (Dr Aubertin), Sezanne (Dr Quignard), Valenciennes (Dr Leurs), Vannes (Dr Le Provost), Villejuif (Dr Feteanu), and Wasquehal (Dr. Frigard, Dr Idiri).
RESEARCH IN CONTEXT
1. Systematic review: Alzheimer’s disease (AD) patients’ quality of life (QOL) is often rated by caregivers, but such ratings are generally divergent with self-reports. Discrepancies may be related to patient and caregiver characteristics, but previous studies have been limited by small sample sizes, selected study populations, and incomprehensive patient evaluations. Furthermore, little is known about the natural history of QOL in AD. 2. Interpretation: In our study, caregivers consistently underestimated AD patients’ QOL compared with that of patients themselves, and patient self-ratings did not change over time, despite worsening of dementia symptoms, although caregiver ratings significantly declined. Patient age and neuropsychiatric symptoms and caregiver burden were associated with discrepancies. 3. Future directions: Caregiver ratings of patient QOL may be biased. Studies using solely proxy reports of patient QOL are to be discouraged, except in severe dementia. Temporal relationships between level of caregiver burden and patient QOL and the evolution of patient–caregiver disagreement should be further explored.
10
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
References [1] Aguero-Torres H, Fratiglioni L, Guo Z, Viitanen M, von Strauss E, Winblad B. Dementia is the major cause of functional dependence in the elderly: 3-year follow-up data from a population-based study. Am J Public Health 1998;88:1452–6. [2] Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Global prevalence of dementia: A Delphi consensus study. Lancet 2005;366:2112–7. [3] Netuveli G, Wiggins RD, Hildon Z, Montgomery SM, Blane D. Functional limitation in long standing illness and quality of life: Evidence from a national survey. BMJ 2005;331:1382–3. [4] Banerjee S, Smith SC, Lamping DL, Harwood RH, Foley B, Smith P, et al. Quality of life in dementia: More than just cognition. An analysis of associations with quality of life in dementia. J Neurol Neurosurg Psychiatry 2006;77:146–8. [5] WHO World Organization Group. Development of the WHOQOL: Rationale and current status. Int J Ment Health 1994;23:24–56. [6] Vellas B, Andrieu S, Sampaio C, Coley N, Wilcock G. Endpoints for trials in Alzheimer’s disease: A European task force consensus. Lancet Neurol 2008;7:436–50. [7] Trinh NH, Hoblyn J, Mohanty S, Yaffe K. Efficacy of cholinesterase inhibitors in the treatment of neuropsychiatric symptoms and functional impairment in Alzheimer disease: A meta-analysis. JAMA 2003;289:210–6. [8] EMEA. European Medicines Agency. Committee for Medicinal Products for Human Use (CHMP) draft guideline on medicinal products for the treatment of Alzheimer’s disease and other dementias. Doc. Ref. CPMP/EWP/553/95 Rev. 1 2007; Available from: http://www. ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/ 2009/09/WC500003565.pdf. [9] Banerjee S. Commentary on “Health economics and the value of therapy in Alzheimer’s disease.” Quality of life in dementia: Development and use of a disease-specific measure of health-related quality of life in dementia. Alzheimers Dement 2007;3:166–71. [10] Logsdon RG, Gibbons LE, McCurry SM, Teri L. Assessing quality of life in older adults with cognitive impairment. Psychosom Med 2002; 64:510–9. [11] Scholzel-Dorenbos CJ, Ettema TP, Bos J, Boelens-van der Knoop E, Gerritsen DL, Hoogeveen F, et al. Evaluating the outcome of interventions on quality of life in dementia: Selection of the appropriate scale. Int J Geriatr Psychiatry 2007;22:511–9. [12] Ready RE, Ott BR. Quality of life measures for dementia. Health Qual Life Outcomes 2003;1:11. [13] Jonsson L, Andreasen N, Kilander L, Soininen H, Waldemar G, Nygaard H, et al. Patient- and proxy-reported utility in Alzheimer disease using the EuroQoL. Alzheimer Dis Assoc Disord 2006;20:49–55. [14] Yasuda N, Zimmerman S, Hawkes WG, Gruber-Baldini AL, Hebel JR, Magaziner J. Concordance of proxy-perceived change and measured change in multiple domains of function in older persons. J Am Geriatr Soc 2004;52:1157–62. [15] Schulz R, Cook TB, Beach SR, Lingler JH, Martire LM, Monin JK, et al. Magnitude and causes of bias among family caregivers rating Alzheimer disease patients. Am J Geriatr Psychiatry 2013;21:14–25. [16] Zucchella C, Bartolo M, Bernini S, Picascia M, Sinforiani E. Quality of life in Alzheimer disease: A comparison of patients’ and caregivers’ points of view. Alzheimer Dis Assoc Disord 2015;29:50–4. [17] Huang HL, Weng LC, Tsai YH, Chiu YC, Chen KH, Huang CC, et al. Predictors of self- and caregiver-rated quality of life for people with dementia living in the community and in nursing homes in northern Taiwan. Int Psychogeriatr 2015;27:825–36. [18] Conde-Sala JL, Turro-Garriga O, Garre-Olmo J, Vilalta-Franch J, Lopez-Pousa S. Discrepancies regarding the quality of life of patients with Alzheimer’s disease: A three-year longitudinal study. J Alzheimers Dis 2014;39:511–25.
[19] Zhao H, Novella JL, Drame M, Mahmoudi R, Barbe C, di Pollina L, et al. Factors associated with caregivers’ underestimation of quality of life in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 2012;33:11–7. [20] Hoe J, Hancock G, Livingston G, Orrell M. Quality of life of people with dementia in residential care homes. Br J Psychiatry 2006; 188:460–4. [21] Aisen PS, Schafer KA, Grundman M, Pfeiffer E, Sano M, Davis KL, et al. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: A randomized controlled trial. JAMA 2003;289:2819–26. [22] Wlodarczyk JH, Brodaty H, Hawthorne G. The relationship between quality of life, mini-mental state examination, and the instrumental activities of daily living in patients with Alzheimer’s disease. Arch Gerontol Geriatr 2004;39:25–33. [23] Nourhashemi F, Andrieu S, Gillette-Guyonnet S, Giraudeau B, Cantet C, Coley N, et al. Effectiveness of a specific care plan in patients with Alzheimer’s disease: Cluster randomised trial (PLASA study). BMJ 2010;340:c2466. [24] McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984;34:939–44. [25] Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98. [26] Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. JAMA 1963;185:914–9. [27] Lawton MP, Brody EM. Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 1969; 9:179–86. [28] Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The neuropsychiatric inventory: Comprehensive assessment of psychopathology in dementia. Neurology 1994;44:2308–14. [29] Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of the elderly: The mini nutritional assessment as part of the geriatric evaluation. Nutr Rev 1996;54(1 Pt 2):S59–65. [30] Zarit SH, Reever KE, Bach-Peterson J. Relatives of the impaired elderly: Correlates of feelings of burden. Gerontologist 1980; 20:649–55. [31] Singer JD. Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. J Educ Behav Stat 1998; 23:323–55. [32] Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. , New York, NY: John Wiley & Sons; 2000. . 118. [33] Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation. Med Care 2003;41:582–92. [34] Bosboom PR, Alfonso H, Almeida OP. Determining the predictors of change in quality of life self-ratings and carer-ratings for communitydwelling people with Alzheimer disease. Alzheimer Dis Assoc Disord 2013;27:363–71. [35] Clare L, Woods RT, Nelis SM, Martyr A, Markova IS, Roth I, et al. Trajectories of quality of life in early-stage dementia: Individual variations and predictors of change. Int J Geriatr Psychiatry 2014; 29:616–23. [36] Tatsumi H, Nakaaki S, Torii K, Shinagawa Y, Watanabe N, Murata Y, et al. Neuropsychiatric symptoms predict change in quality of life of Alzheimer disease patients: A two-year follow-up study. Psychiatry Clin Neurosci 2009;63:374–84. [37] Pearlin LI, Schooler C. The structure of coping. J Health Soc Behav 1978;19:2–21. [38] Whitehouse PJ, George D. Banking on stories for healthier cognitive ageing. Lancet 2009;373:1166–7. [39] Scholzel-Dorenbos CJ, van der Steen MJ, Engels LK, Olde Rikkert MG. Assessment of quality of life as outcome in dementia
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11 and MCI intervention trials: A systematic review. Alzheimer Dis Assoc Disord 2007;21:172–8. [40] Grand A, Grand-Filaire A, Bocquet H, Clement S. Caregiver stress: A failed negotiation? A qualitative study in South West France. Int J Aging Hum Dev 1999;49:179–95. [41] Wetzels RB, Zuidema SU, de Jonghe JF, Verhey FR, Koopmans RT. Determinants of quality of life in nursing home residents with dementia. Dement Geriatr Cogn Disord 2010;29:189–97.
11
[42] Selwood A, Thorgrimsen L, Orrell M. Quality of life in dementia—A one-year follow-up study. Int J Geriatr Psychiatry 2005;20:232–7. [43] Gow AJ, Whiteman MC, Pattie A, Whalley L, Starr J, Deary IJ. Lifetime intellectual function and satisfaction with life in old age: Longitudinal cohort study. BMJ 2005;331:141–2. [44] Hoe J, Katona C, Roch B, Livingston G. Use of the QOL-AD for measuring quality of life in people with severe dementia—the LASER-AD study. Age Ageing 2005;34:130–5.
Assessing Alzheimer’s disease patients’ quality of life: Discrepancies between patient and caregiver perspectives Sandrine Andrieua,b,*, Nicola Coleya,b, Yves Rollanda,c, Christelle Canteta,c, Catherine Arnauda,b, Sophie Guyonneta,c, Fati Nourhashemia,c, Alain Granda,b, Bruno Vellasa,c, and the PLASA group1 a Inserm-Toulouse University UMR1027, Toulouse, France CHU Toulouse, Department of Epidemiology and Public Health, Toulouse, France c Gerontopole, CHU Toulouse, Department of Geriatric Medicine, Toulouse, France b
Abstract
Introduction: Quality of life (QOL) is an important dimension to consider in Alzheimer’s disease (AD), but few large-scale studies have analyzed self and caregiver reports of patient QOL. Methods: Patient QOL was evaluated in a cohort of 574 AD patients with the QOL-AD scale over 2 years. Results: Caregiver reports of patient QOL were lower at baseline than self reports. Older patient age was associated with overestimation of QOL by caregivers, whereas neuropsychiatric inventory score and caregiver burden were associated with underestimation. Activities of daily living limitation, depressive symptoms, and caregiver burden were systematically associated with poorer QOL, whereas caregiver relationship and apathy were associated with poorer QOL only for self reports or caregiver reports, respectively. Cognitive function and professional care were not associated with QOL. Self-rated patient QOL did not change over time, whereas disease severity markers and caregiver-rated patient QOL declined. Discussion: It is important to assess both self and caregiver ratings when assessing patient QOL. Ó 2015 The Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
Keywords:
Alzheimer’s disease; Quality of life; QOL-AD; Caregivers; Proxy report; Longitudinal
1. Introduction Alzheimer’s disease (AD) is a chronic disease that requires care for several years and remains one the most frequent disabling diseases in aging populations [1,2]. It is a major cause of decreased quality of life (QOL) in older adults, and some studies have shown that more
There are conflicts of interest to disclose. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of the French Ministry of Health. The study sponsor had no role in the study. 1 See acknowledgments for the list of the PLASA study group members. *Corresponding author. Tel.: 133-5-61-14-59-50; Fax: 133-5-62-2642-40. E-mail address: [email protected]
than the disease itself, it is its disability-related impact that affects QOL [3]. Simple measures of cognitive or functional decline are not sufficient to address the complexity of AD, and QOL is a useful additional dimension to consider as a broader outcome measure which may represent more “clinically significant” changes or benefits to the patient and caregiver [4]. QOL is a broad conceptual field, linked to physical health as well as psychological state, level of independence, social relationships, personal beliefs, and the subject’s relationship with the specifics of their environment [5]. Researchers [6,7] and regulatory agencies [8] emphasize the need to take into account patient QOL in the evaluation of new drugs but stress the need for further validation work to use it as an end point in clinical trials. For example, little is known about the natural history of QOL in AD, the determinants
http://dx.doi.org/10.1016/j.jalz.2015.09.003 1552-5260/Ó 2015 The Alzheimer’s Association. Published by Elsevier Inc. All rights reserved.
2
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
of high and low QOL, or the effects of medication on QOL [9]. Caregiver ratings are frequently used as part of multidimensional evaluations of AD patients, in particular to measure patients’ functional status and neuropsychiatric symptoms, and caregivers may also be asked to rate patients’ QOL [4]. However, because QOL is subjective and defined as an individual perception, it should ideally be assessed from the subject’s own perspective [10,11], but declining cognitive function makes this harder to do in AD [10]. Many QOL scales have been proposed in AD [12], including the QOL-AD scale which requires both the patient and caregiver to rate the same domains of patient QOL [10]. Several previous studies [13–17] have reported discrepancies between caregiver and self reports of AD patients’ QOL, and some have identified factors associated with divergent ratings [15–19]. Most studies were small scale and targeted populations coming from particular settings, such as residential care homes [20], or from therapeutic trials, thus providing a selected sample with an abnormally low rate of comorbidities [21,22]. Furthermore, very few studies have examined longitudinal changes in patient and proxy QOL reports over .1 year of follow-up [18]. The primary objective of this multicentre study was to explore discrepancies between self and caregiver reports of patient QOL in a large population of well-characterized community-dwelling AD patients at baseline and during 2 years of follow-up and to determine factors associated with disagreement at baseline. The secondary objective was to assess factors explaining a low level of self- or caregiver-reported patient QOL at baseline. 2. Methods This study includes AD patients and their primary caregivers enrolled in the Plan de soin et d’aide dans la maladie d’Alzheimer (PLASA) study, a randomized trial that tested the efficacy of a standardized care plan versus usual dementia care on functional decline over 2 years of follow-up. PLASA is described in detail elsewhere [23]. 2.1. Subjects Briefly, a total of 1131 AD patients and their caregivers were recruited in a French nationwide network of 50 memory centers. To be included, subjects had to meet the following inclusion criteria: diagnosis of probable or possible AD according to the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association criteria [24], mini-mental status examination (MMSE) [25] score between 12 and 26, community-dwelling, looked after by a well-identified informal caregiver, and not participating in any other research program. In the present analysis, we included only the 574 patients randomly assigned to the intervention group to be able to study the determinants of
QOL and patient-proxy agreement because this group underwent more comprehensive evaluations than the usual care group (as part of the intervention). The intervention had no significant effect on any of the primary or secondary efficacy measures, including QOL. The study was funded by the Ministry of Health and was approved by the Institutional Review Board and ethics committee of Toulouse University. Written informed consent was obtained from all patients and their caregivers. This trial was registered in clinicaltrials.gov (identifier: NCT00480220). 2.2. Measurements Patients were evaluated every 6 months in memory centers in general or university hospitals. 2.2.1. QOL measure Patient QOL was assessed by the QOL-AD scale [10], a 13-item questionnaire designed to provide patient and caregiver reports of the patient’s QOL which covers various domains (physical health, energy, mood, living situation, memory, family, marriage, friends, self as a whole, ability to do chores, ability to do things for fun, financial situation, and QOL as a whole). Items are scored on a 4-point Likert scale ranging from 1 (poor) to 4 (excellent). Total scores range from 13 to 52 with higher scores indicating better QOL. As proposed by Logsdon [10], missing items were replaced by the mean score of the remaining items if no more than two items were missing; otherwise, the score was considered missing. 2.2.2. Other measures Sociodemographic data (age, sex, educational level, and living arrangements), time since diagnosis of dementia as declared by the caregiver, comorbidities, use of support services, and assistance in the home (home help, supervision, meal delivery service, and nursing care) were recorded using a standardized questionnaire. Number of medications used and use of anti-AD drugs (i.e., cholinesterase inhibitors, N-Methyl-D-aspartate receptor antagonists) were recorded. Dementia severity was evaluated based on assessments of cognition (MMSE) [25], function (activities of daily living [ADL] [26], instrumental activities of daily living [IADL] [27]), and behavioral disorders (neuropsychiatric inventory [NPI]) [28]. Nutritional status was evaluated with the mini nutritional assessment (MNA) [29], and caregiver burden was evaluated using the Zarit Burden Interview [30]. 2.3. Data analysis Baseline mean total scores and individual domain scores were calculated for patient- and caregiver-rated QOL when a score was available for both members of the dyad. To address the extent of agreement between patient and caregiver reports, Spearman and intraclass correlation coefficients (ICCs) were computed for each item at
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
baseline. Correlations between total scores were also evaluated via the Pearson correlation coefficient and the ICC. Differences between mean patient and caregiver scores for total QOL-AD score and individual domains were tested using paired Student’s t-tests or nonparametric Wilcoxon tests. Baseline patient and caregiver characteristics of dyads in agreement for the total QOL-AD score were compared with those of dyads in which patient and caregiver scores were not in agreement. Disagreement was separated into caregivers who scored lower and caregivers who scored higher than their patient. Statistical significance was assessed with Fisher (analysis of variance) or Kruskal-Wallis tests for quantitative variables and c2 tests for qualitative variables. Variables with a P value of ,.2, and patient age and sex, were entered into a multivariate multinomial logistic regression model, and a backward stepwise selection procedure was used to identify factors that were significantly and independently associated with disagreement (underestimation or overestimation of patient QOL by caregivers compared with self-ratings). Prospective changes over time in self- and caregiverreported patient QOL were analyzed using a three-level repeated measures linear-mixed regression model with random intercept and random linear slope to account for the within-subject correlations due to repeated observations and correlations within clusters (centers) [31]. We took into account all available data during follow-up (i.e., until the end of the study or until the participant dropped out, was institutionalized, or there was a change in the caregiver). Determinants of poorer self- and caregiver-reported patient QOL at baseline were assessed using two separate analyses. In the absence of established cut-offs, a low level of QOL was defined as the lowest quartile, corresponding to a score ,31.0 for patient reports and 27.1 for caregiver reports. First, in a bivariate analysis, baseline differences between subjects with a poorer level of QOL and subjects with higher QOL were tested using t-tests for continuous variables and c2 tests for categorical variables. Variables were eligible for inclusion in the multivariate analysis if they were associated with a low level of QOL in bivariate analyses with P .20. For the multivariate analysis, a series of logistic regression models were performed. Our initial hypothesis was that after taking into account the severity of the disease, caregiver characteristics would influence the patient’s level of QOL. Thus, for both patient- and caregiver-rated patient QOL, we first modeled the association between a poorer level of QOL and patient characteristics (model 1). We then performed an analysis including only caregiver characteristics (model 2). Finally, we fit a global model which included both patient and caregiver characteristics that were significantly associated with QOL in models 1 and 2 (model 3). We used a backward stepwise selection procedure to retain the factors independently associated with the level of QOL. We present adjusted odds ratios
3
(OR) and 95% confidence intervals (CIs) for significant variables. Model fit was examined using the HosmerLemeshow goodness-of-fit test [32]. Subjects with missing data for any of the variables included in the models were excluded from this analysis. All P values were based on two-sided tests and were considered statistically significant at P , .05. Analyses were conducted using SAS 9.1 software (SAS Institute Inc., Cary, NC, USA). 3. Results 3.1. Population characteristics At baseline, the 574 patients had a mean age of 79.6 years (standard deviation [SD], 5.9) and 66.6% were women. On average, they had been diagnosed with AD for 1.22 years (1.44). Their mean MMSE score was 19.5 (3.9), and 76.5% were treated with cholinesterase inhibitors or memantine before inclusion in the study. Most patients showed limitations in more than two IADLs (64.2%) and 45.1% had at least one basic ADL limitation. Behavioral disorders were frequent in this population, with 93.8% subjects presenting at least one NPI symptom (mean NPI score was 17.7 [16.4]). A third (32.6%) of patients lived at home alone. Caregivers were mostly female (66.2%) with a mean age of 64.1 years (13.7). They were mainly the spouse (47.2%) or child (44.6%) of the patient. 3.2. Agreement between self and caregiver reports of patient QOL at baseline At baseline, 93.0% of patients (n 5 534) provided a selfrating of their QOL, and 91.6% of caregivers (n 5 526) provided a proxy rating of their patient’s QOL. This resulted in 501 dyads (87.3%) with total scores for both self- and caregiver-reported QOL. Caregiver reports of patient QOL were significantly lower than patient self-reports for the total mean QOL-AD score (31.03 [5.35] vs. 34.54 [5.29]; P , .001), and for all items except living situation (Table 1). The mean difference between caregiver and patient total scores (3.51) was .0.5 SD (of the caregiver’s score), which is a widely recognized definition of clinically meaningful difference in QOL [33]. Correlations between AD patients and caregiver scores for the individual items and for the total score were low (Spearman/Pearson correlations between 0.10 and 0.53; intraclass correlations between 0.08 and 0.49; Table 1). For total QOL-AD scores, 50% of caregivers underestimated patient QOL compared with self-ratings, whereas 12% of caregivers overestimated patient QOL, and 38% agreed with patient ratings (Fig. 1). Agreement for individual items of the QOL-AD scale ranged from 41% to 56% of dyads, and caregivers were more likely to underestimate rather than overestimate patient QOL, as compared with self-ratings, for all items.
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
4
Table 1 Differences between patient and caregiver ratings of patient QOL at baseline (based on complete patient-caregiver pairs for each domain) Domain Physical health Energy Mood Living situation Memory Family Marriage Friends Self as a whole Ability to do chores Ability to do things for fun Finances Life as a whole QOL-AD score
Complete pairs, n (%) 563 (98.1) 563 (98.1) 563 (98.1) 563 (98.1) 562 (97.9) 562 (97.9) 521 (90.8) 518 (90.2) 546 (95.1) 533 (92.9) 518 (90.2) 541 (94.3) 550 (95.8) 501 (87.3)
Patient score, mean (SD) 2.61 (0.67) 2.52 (0.73) 2.55 (0.66) 2.97 (0.61) 1.84 (0.70) 3.18 (0.74) 2.99 (0.69) 2.91 (0.71) 2.54 (0.67) 2.66 (0.74) 2.39 (0.79) 2.71 (0.69) 2.64 (0.70) 34.54 (5.29)
Caregiver score, mean (SD) 2.51 (0.75) 2.12 (0.80) 2.34 (0.70) 2.93 (0.62) 1.34 (0.52) 2.89 (0.79) 2.71 (0.75) 2.71 (0.79) 2.32 (0.84) 2.29 (0.87) 1.91 (0.80) 2.56 (0.86) 2.29 (0.75) 31.03 (5.35)
Difference, mean (SD) y
0.10 (0.88) 0.40 (0.93)z 0.21 (0.82)z 0.04 (0.81) 0.51 (0.83)z 0.28 (0.90)z 0.28 (0.72)z 0.21 (0.87)z 0.21 (0.91)z 0.37 (0.95)z 0.48 (0.93)z 0.14 (0.99)y 0.35 (0.92)z 3.51 (5.90)z
Correlation*
ICC
0.2647 0.2664 0.2743 0.1444 0.1003 0.2678 0.5279 0.3346 0.2627 0.3081 0.3098 0.1707 0.2059 0.3849
0.2488 0.2580 0.2810 0.1239 0.0841 0.3086 0.4884 0.3075 0.2633 0.2926 0.3140 0.1896 0.1992 0.3675
Abbreviations: QOL, quality of life; SD, standard deviation; ICC, intraclass correlation coefficient; AD, Alzheimer’s disease. *Spearman correlation coefficients for individual items; Pearson correlation coefficient for total score. P values are ,.05 for all correlation coefficients. y Two-sided Student paired t-test/nonparametric Wilcoxon test P , .01. z Two-sided Student paired t-test/nonparametric Wilcoxon test P , .001.
3.3. Factors associated with patient and caregiver agreement Significant differences between dyads in agreement for the total QOL-AD score and those in which patient and caregiver scores were not in agreement were observed for both patient (IADL, ADL, MNA, one-leg balance, and NPI total score and certain items) and caregiver (level of subjective burden) characteristics (Table 2). In the multivariate multinomial logistic regression model, older patient age (OR, 1.08; 95% CI, 1.02–1.14) was significantly associated with overestimation of QOL by caregivers compared with patient
self-ratings, and NPI total score (OR, 1.02; 95% CI, 1.00– 1.04) and higher caregiver burden (low-moderate burden: OR, 1.67; 95% CI, 1.02–2.72; moderate-severe burden: OR, 1.99; 95% CI, 0.93–4.23) were significantly associated with underestimation of QOL by caregivers compared with patient self-ratings (Table 2). 3.4. Two-year changes in patient- and caregiver-reported patient QOL During the 2 years of follow-up, 43 patients died and 127 dropped out for other reasons. Patients who remained in the
Fig. 1. Percentage (of complete pairs) of patient-caregiver dyads in agreement or disagreement for individual items and total score of the QOL-AD scale at baseline. Abbreviation: QOL-AD, quality of life-Alzheimer’s disease.
Table 2 Factors associated with disagreement between self and caregiver ratings of patient quality of life for the total score of the QOL-AD scale at baseline (unadjusted and multivariate analyses) Unadjusted analysis Characteristics
Agreement (n 5 189)
Patient score , caregiver score (n 5 62)
Patient score . caregiver score (n 5 250)
78.9 (5.6) 127 (67.2)
79.7 (7.2) 38 (61.3)
79.6 (5.7) 166 (66.4)
49 (25.9) 73 (38.6) 34 (18.0) 33 (17.5) 1.2 (1.5) 150 (79.4) 4.0 (2.2) 19.6 (3.9) 69 (36.5) 80 (44.2) 45 (24.7) 13.4 (13.4)
15 (24.2) 22 (35.5) 10 (16.1) 15 (24.2) 1.1 (1.3) 43 (69.4) 4.0 (2.4) 20.3 (4.0) 21 (33.9) 29 (47.5) 16 (27.6) 12.2 (13.0)
54 (21.7) 110 (44.2) 58 (23.3) 27 (10.8) 1.3 (1.4) 192 (76.8) 4.2 (2.5) 19.5 (3.9) 122 (48.8) 138 (56.6) 86 (36.0) 21.0 (17.6)
64 (33.9) 107 (56.6) 18 (9.5) 67 (35.5) 30 (16.0)
19 (30.7) 36 (58.1) 7 (11.3) 15 (24.2) 5 (8.2)
73 (29.2) 139 (55.6) 38 (15.2) 101 (40.9) 50 (20.0)
64.3 (13.9) 119 (63.0)
66.7 (13.1) 37 (59.7)
64.6 (13.8) 169 (67.6)
93 (49.2) 80 (42.3) 16 (8.5)
34 (54.8) 23 (37.1) 5 (8.1)
126 (50.4) 106 (42.4) 18 (7.2)
92 (53.8) 61 (35.7) 18 (10.5)
37 (63.8) 12 (20.7) 9 (15.5)
82 (36.3) 102 (45.1) 42 (18.6)
P .416 .687 .077
.295 .269 .656 .387 .013 .036 .040 ,.001 .453
OR (95% CI), patient score , caregiver score
P
OR (95% CI), patient score . caregiver score
P
1.08 (1.02–1.14)
.012
1.03 (0.99–1.07)
.109
0.99 (0.96–1.01)
.331
1.02 (1.00–1.04)
.022
1 0.63 (0.29–1.38) 1.79 (0.64–5.00)
.248 .269
1 1.67 (1.02–2.72) 1.99 (0.93–4.23)
.040 .075
.046 .080 .472 .396 .927
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
Patient characteristics Age, mean (SD), y Gender, women, n (%) Educational level Elementary or no formal, n (%) Primary school certificate, n (%) Secondary education, n (%) Tech/high school certificate or higher, n (%) Years since diagnosis, mean (SD) Use of AD medication, n (%) Number of other medications, mean (SD) MMSE score, mean (SD) Limitation in 1 ADL, n (%) Limitation in .2 IADLs, n (%) MNA score 23.5, n (%) NPI total score, mean (SD) Living arrangements Living alone, n (%) With spouse, n (%) Other, n (%) Abnormal one-leg balance, n (%) Autonomy allowance, n (%) Caregiver characteristics Age, mean (SD), y Gender, women, n (%) Relationship with patient, n (%) Spouse Child Other Subjective level of burden, ZBI, n (%) Absent to low (score 0–20) Low to moderate (score 21–40) Moderate to severe (score .40)
Multivariate (adjusted) analysis*
,.001
5
Abbreviations: QOL, quality of life; AD, Alzheimer disease; OR, odds ratio; CI, confidence interval; SD, standard deviation; MMSE, mini-mental status examination (scores range from 0 to 30, with higher scores indicating better function); ADL, activities of daily living (scores range from 0 to 6, with higher score indicating better function, dichotomized as 0 vs. 1 limitation); IADL, instrumental activities of daily living (scores range from 0 to 8, with higher scores indicating better function, dichotomized as 0 vs. .2 limitations); MNA, mini nutritional assessment (score range from 0 to 30, with higher score in better nutritional status, with a score 23.5 indicating that the patient is at risk of malnutrition or is malnourished); NPI, neuropsychiatric inventory (score range from 0 to 144, with higher score indicating more severe symptoms); ZBI, Zarit Burden Interview (scores range from 0 to 88, with higher scores indicating higher levels of burden, scores 20 represent low-or-no burden; scores of 21–40 represent low-to-moderate burden, and scores .40 represent moderate-to-severe burden). NOTE. Global P values in the final model were: Patient age (P 5 .031); NPI total score (P 5 .007); and caregiver burden (P 5 .027). Bold text indicates P ,.05. *Adjusted odds ratios are shown for variables that remained in the final multivariate multinomial model. Variables with a P value of ,.2, and patient age, sex, and level of education, were entered into the initial model, and a backward stepwise selection procedure was used to identify factors that were significantly and independently associated with underestimation or overestimation of patient QOL by caregivers compared with self-ratings. Variables that did not remain in the final model (following backward stepwise selection) were level of education, level of functional autonomy (IADL, ADL), nutritional status (MNA), receipt of autonomy allowance, and one-leg balance.
6
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
study were significantly younger and had better cognitive, functional, and neuropsychiatric scores at baseline, as well as higher self- and caregiver-reported QOL. They were also more likely to live with a spouse at baseline and their caregivers reported significantly lower levels of burden at baseline. Mean patient self-reported QOL remained stable over the 2 years of follow-up, (34.55 [standard error, 0.48] at baseline and 34.50 [0.51] at 2 years, P 5 .857), whereas mean caregiver-reported patient QOL decreased from 30.75 (0.36) to 28.64 (0.40) (P , .001; Table 3). In this period, there was significant progression of disease severity, as measured by cognitive, functional, and behavioral measures (Table 3). 3.5. Factors associated with a low level of QOL at baseline Table 4 shows patient and caregiver characteristics associated with poorer patient QOL at baseline. The results of the multivariate logistic regression models adjusted for covariates showed that several factors linked to AD severity were associated with QOL, regardless of who (patient or caregiver) made the assessment; having at least one basic ADL limitation and the presence of depressive symptoms according to the NPI were both associated with poorer QOL. The level of subjective caregiver burden was also independently associated with poorer QOL whether it was self rated or caregiver rated. Furthermore, caregivers tended to rate QOL lower for female patients compared with male patients; and for selfreported QOL, patients cared for by their children were more likely to declare poorer QOL than those cared for by their spouses. The presence of apathy was independently associated with poorer QOL for caregiver-reported scores only. Neither the level of cognitive impairment nor the number of comorbidities was associated with patient QOL. The use of support services or home assistance was also not associated with the level of QOL.
4. Discussion We observed discrepancies between caregiver and self ratings of patient QOL at baseline and during 2-years of follow-up in a population of community-dwelling AD patients who were evaluated every 6 months. In cross-sectional analyses, we found that caregivers systematically underestimated patient QOL compared with patient self-ratings for the total QOL-AD score and for most of the domains covered by the scale. This result is consistent with previous findings reported in the literature [13–17]. Discrepancies between patient and caregiver ratings at baseline were not associated with patient cognitive or functional status, nor with the patient-caregiver relationship. However, increasing frequency and severity of neuropsychiatric symptoms, and a higher level of caregiver burden, were significantly and independently associated with underestimation of patient QOL by caregivers compared with selfratings. Caregivers may consider the effects of burden and neuropsychiatric symptoms (which may affect them more than the patient) on their own QOL when rating those of the patient. Our results are consistent with recent findings from smaller studies demonstrating an association between caregiver psychosocial well-being (particularly level of burden and depression) and discrepancies in patient QOL ratings [15–19] and confirm that this association remains even after adjustment for numerous patient characteristics (including living alone) in a population of patients who were exclusively community dwelling. An original finding of our study was that increasing patient age was associated with overestimation of patient QOL by caregivers compared with patients, suggesting that they may take into account age in their ratings. During the 2 years of follow-up, there was a progressive worsening of dementia in the AD patients included in this study, based on cognitive, functional, and behavioral assessments. Despite this worsening, mean self-rated patient QOL did not change during this time, whereas caregiver ratings of patient QOL significantly declined. This finding
Table 3 Two-year changes in disease severity parameters and self- and caregiver-reported patient QOL with linear mixed model (n 5 574) Parameter Patient characteristics Basic activities of daily living (ADL score) Instrumental activities of daily living (IADL/8 score) Cognitive function (mini-mental status examination score) Behavioral disturbances (NPI score) Self-reported QOL score Caregiver evaluations Subjective burden (ZBI score) Caregiver-reported QOL score
Baseline, mean (SE)
12 mo, mean (SE)
24 mo, mean (SE)
Slope*, 12 mo, mean (SE)
Slope*, 24 mo, mean (SE)
P
5.35 (0.06) 4.26 (0.17)
4.79 (0.07) 3.18 (0.17)
4.21 (0.09) 2.40 (0.18)
20.56 (0.04) 21.08 (0.08)
21.15 (0.06) 21.86 (0.11)
,.001 ,.001
19.44 (0.19)
17.90 (0.23)
15.75 (0.29)
21.54 (0.15)
23.69 (0.23)
,.001
18.26 (1.20) 34.55 (0.48)
19.05 (1.17) 34.52 (0.48)
19.85 (1.26) 34.50 (0.51)
0.80 (0.39) 20.02 (0.13)
1.59 (0.77) 20.05 (0.25)
.040 .857
24.14 (0.92) 30.75 (0.36)
26.62 (0.93) 29.70 (0.35)
29.10 (1.06) 28.64 (0.40)
2.48 (0.35) 21.05 (0.14)
4.95 (0.70) 22.11 (0.28)
,.001 ,.001
Abbreviations: QOL, quality of life; SE, standard error; ADL, activities of daily living; IADL, instrumental activities of daily living; NPI, neuropsychiatric inventory; ZBI, Zarit Burden Interview. *Slope estimated from the mixed model.
Table 4 Patient and caregiver characteristics associated with poorer self- or caregiver-reported patient QOL at baseline Self-reported patient QOL, n 5 534
Characteristics
Quartiles 1–3 (QOL-AD score 31)
Lowest quartile (QOL-AD score 27.1)
Quartiles 1–3 (QOL-AD score .27.1)
Adjusted OR (95% CI)z
79.3 (6.2) 92 (70.2)
79.5 (5.8) 262 (65.0)
79.5 (5.2) 104 (75.9)*
79.4 (6.1) 244 (62.7)*
2.0 (1.2–3.4)
.010
27 (20.6) 60 (45.8) 22 (16.8) 22 (16.8) 1.5 (1.5)* 4.1 (2.2) 109 (83.2)* 19.0 (3.8) 74 (56.5)* 76 (59.4)* 44 (36.1) 18 (13.7)
97 (24.1) 163 (40.6) 87 (21.6) 55 (13.7) 1.2 (1.4)* 4.1 (2.4) 299 (74.2)* 19.8 (3.9) 158 (39.2)* 184 (48.9)* 118 (30.5) 53 (13.2)
37 (27.0) 56 (40.9) 31 (22.6) 13 (9.5) 1.5 (1.5)* 4.2 (2.4) 108 (78.8) 18.6 (3.9)* 87 (63.5)* 84 (64.1)* 49 (37.4) 28 (20.4)*
92 (23.8) 155 (40.1) 73 (18.9) 67 (17.3) 1.1 (1.4)* 4.1 (2.3) 297 (76.4) 19.9 (3.8)* 141 (36.3)* 172 (46.6)* 107 (28.8) 47 (12.1)*
2.0 (1.2–3.3)
.005
24 (18.5) 13 (10.0) 50 (38.5) 76 (58.9)* 73 (56.2) 9 (6.9) 89 (68.5)* 28 (21.7) 56 (43.1) 29 (22.5) 33 (26.6) 52 (40.3)*
85 (21.1) 38 (9.4) 161 (40.0) 178 (44.3)* 194 (48.1) 50 (12.4) 224 (55.7)* 81 (20.1) 181 (44.9) 96 (23.9) 90 (22.9) 112 (27.9)*
41 (29.9)* 16 (11.7) 69 (50.4)* 91 (66.4)* 81 (59.1)* 12 (8.8) 108 (78.8)* 42 (30.9)* 75 (54.7)* 39 (28.9) 43 (32.6)* 52 (38.2)*
65 (16.8)* 36 (9.3) 137 (35.3)* 163 (42.2)* 183 (47.2)* 47 (12.1) 194 (50.1)* 67 (17.3)* 155 (40.0)* 81 (20.9) 81 (21.5)* 105 (27.1)*
2.3 (1.4–3.7)
,.001
2.3 (1.3–4.0)
.003
46 (35.1) 67 (51.2) 18 (13.7)
122 (30.3) 229 (56.8) 52 (12.9)
45 (32.9) 67 (48.9) 25 (18.3)
122 (31.4) 221 (56.8) 46 (11.8)
62.1 (13.2)* 90 (68.7)
65.2 (13.9)* 258 (64.0)
53 (40.5) 65 (49.6) 13 (9.9)
209 (51.9) 166 (41.2) 28 (6.9) * 173 (49.3)
61.8 (14.6)* 92 (67.2) * 56 (40.9) 70 (51.1) 11 (8.0) * 22 (18.6)
65.6 (13.3)* 250 (64.3) * 205 (52.7) 154 (39.6) 30 (7.7) * 197 (56.1)
* 43 (36.1)
Adjusted OR (95% CI)y
2.0 (1.3–3.1)
1.8 (1.1–2.8)
P
.002
.012
1 1.8 (1.1–2.8) 2.0 (0.9–4.4)
— .015 .111
1
—
1
P
7
— (Continued )
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
Patient characteristics Age, mean (SD), y Gender, women, n (%) Educational level, n (%) Elementary or no formal Primary school certificate Secondary education Tech/high school certificate or higher Time elapsed since diagnosis, mean (SD), y Number of medications, mean (SD) Use of AD medication, n (%) MMSE score, mean (SD) Limitation in at least one ADL, n (%) Limitation in at least three IADLs, n (%) MNA score 23.5, n (%) History of osteoarthritis, n (%) Behavioral symptoms, NPI items, n (%) Delusions Hallucinations Agitation Depression Anxiety Elation/euphoria Apathy Disinhibition Irritability Aberrant motor behavior Sleep Appetite and eating disorder Living arrangements, n (%) Living alone With spouse Other Caregiver characteristics Age, mean (SD), y Gender, women, n (%) Relationship with patient, n (%) Spouse Child Other Subjective level of burden, ZBI, n (%) Absent to low (score 0–20)
Lowest quartile (QOL-AD score ,31)
Caregiver-reported patient QOL, n 5 526
,.001 ,.001 3.8 (2.1–6.7) 7.2 (3.6–14.4) 119 (33.9) 35 (10.0) 59 (50.0) 37 (31.4) .527 .006 1.2 (0.7–1.9) 2.3 (1.3–4.3) 135 (38.5) 43 (12.3) 46 (38.7) 30 (25.2)
P Adjusted OR (95% CI)y Quartiles 1–3 (QOL-AD score 31) Characteristics
Low to moderate (score 21–40) Moderate to severe (score .40)
P Adjusted OR (95% CI)z Lowest quartile (QOL-AD score 27.1)
Quartiles 1–3 (QOL-AD score .27.1)
Caregiver-reported patient QOL, n 5 526 Self-reported patient QOL, n 5 534
Lowest quartile (QOL-AD score ,31)
Table 4 Patient and caregiver characteristics associated with poorer self- or caregiver-reported patient QOL at baseline (Continued )
Abbreviations: QOL, quality of life; AD, Alzheimer disease; OR, odds ratio; CI, confidence interval; MMSE, mini-mental status examination (scores range from 0 to 30 with higher scores indicating better function); ADL, activities of daily living (scores range from 0 to 6 with higher score indicating better function, dichotomized as 0 vs. 1 limitation); IADL, instrumental activities of daily living (scores range from 0 to 8 with higher scores indicating better function, dichotomized as 0 vs. .2 limitations); MNA, mini nutritional assessment (score range from 0 to 30, with higher score in better nutritional status, with a score 23.5 indicating that the patient is at risk of malnutrition or is malnourished); NPI, neuropsychiatric inventory (score range from 0 to 144 with higher score indicating more severe symptoms); ZBI, Zarit Burden Interview (scores range from 0 to 88, with higher scores indicating higher levels of burden, scores 20 represent low-or-no burden; scores of 21–40 represent low-to-moderate burden, and scores .40 represent moderate-to-severe burden). NOTE. Bold text indicates P ,.05. *P , .05. y Adjusted odds ratios for poorer quality of life are shown for variables that remained in the final multivariate logistic regression model following the stepwise selection procedure. For self-rated QOL analyses, model 3 initially included limitations in ADL, behavioral troubles (depression, elation/euphoria), caregiver-patient relationship, and the level of subjective burden experienced by the caregiver. z Adjusted odds ratios for poorer quality of life are shown for variables that remained in the final multivariate logistic regression model following the stepwise selection procedure. For caregiver-rated QOL, model 3 initially included patient sex, limitations in ADL, behavioral troubles (delusion, depression, and apathy), caregiver-patient relationship, and the level of subjective burden experienced by the caregiver.
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
8
confirms results from smaller studies of trajectories of selfrated QOL over time in community-dwelling patients with dementia [18,34–36]. It is important to understand this increasing discrepancy between the caregiver and the patient and to consider which QOL assessment must be taken into account in the evaluation of the patient’s situation. The lack of change in patients’ self-reported QOL over time could be caused by patients having reduced abilities to judge their own difficulties due to increased cognitive impairment. It has been suggested that only patients with an MMSE score .10 can reliably report their own QOL [10]. As only 16% of patients had MMSE scores ,10 at 2 years in our study, the lack of change in patient self-reported QOL in our study may not be primarily due to cognitive impairment. Another explanation could be that AD patients adapt, through coping mechanisms, to the new difficulties that they encounter [37], and so may not perceive any QOL decline. Some initiatives that have attempted to better understand the meaning of QOL for patients with dementia have revealed that these patients share a need to maintain a sense of usefulness in their lives which could contribute to their QOL [38]. There may also be reasons related to the caregiver which could explain the differences in patient and caregiver ratings of the patient’s QOL over time. For example, the caregiver’s ratings may be influenced by their own expectations, mood, burden of care, and their relationship with the patient [10,39]. The decline of caregiver-rated patient QOL over time, in contrast to patient self-ratings, could also suggest a worsening of the caregiver’s perception of the patient’s situation. Such misunderstandings of the patient’s situation may bring about mismanagement of the disease [40]. In terms of factors associated with patient QOL, our study was in agreement with previous findings, suggesting that functional impairment [20] and neuropsychiatric symptoms [41,42] play an important role in determining patient QOL but not cognitive status [42,43]. Patient age was not associated with QOL in our study. This is in line with some other studies, in particular those with relatively homogeneous populations in terms of age [22]. We also showed that patient QOL, whether rated by the patient or by the caregiver, was independently affected by the level of caregiver burden. Other studies have also found caregiver burden, and caregiver depression, to be linked to patient QOL [10], with a seemingly greater effect on caregiverratings than on self-ratings, thus raising the issue of the direction of causality between caregiver burden and his/her rating of the patient’s QOL. Interventions aimed at decreasing caregiver burden could provide benefits for the caregiver, but also for the patient. Among the strengths of the present study are the detailed characterization of clinical parameters by experts in AD, using standard outcome measures which are widely used in clinical trials, as well as the assessment of patient QOL from both the patient’s and caregiver’s point of view in a large community-dwelling sample. The scale we used is well
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
adapted for longitudinal studies of AD patients, despite progression of disease symptoms [42], and is not overly dependent on functional status, unlike some generic QOL scales. Consequently, most of the subjects in our study were able to self-rate their QOL. Our patients were relatively homogenous in terms of age, in contrast to numerous other studies with much variation in subject age, which can create large differences in QOL because individuals’ expectations vary according to the period of life [44]. However, several limitations should be considered in the interpretation of our results. Our study was based on a sample of diagnosed AD patients with access to health care systems, rather than a representative sample of the general population. Thus, our results may not be generalizable to all patients with AD. The inclusion of caregivers took place after the diagnosis of AD, and consequently after the beginning of their caregiving activities. Further selection bias may have arisen through study dropouts due to death or institutionalization. We can hypothesize that patients with the most stable management were overrepresented and so perceived QOL decreased little in our cohort even though patients’ needs increased and caregivers were more involved in caregiving over time. 5. Conclusion Further analyses are required to study the temporal relationship between the level of caregiver burden and patient QOL, identify determinants of increasing patient-caregiver disagreement about patient QOL over time, and better assess QOL as a valuable end point to measure the impact of future therapeutic or nontherapeutic interventions. It remains difficult to determine which type of rating (self or proxy) of AD patients’ QOL provides the most accurate picture. However, given the subjective nature of QOL, it is of upmost importance to take into account the patient’s own rating. Acknowledgments Authors’ contributions: S.A., F.N., S.G., and B.V. contributed to the conception and design. S.G., S.A., and F.N. did the data acquisition. S.A., C.C., N.C., Y.R., and C.A. did the data analysis and interpretation. S.A., N.C., and A.G. did the drafting of the article. All authors contributed to the critical review of the article and final approval of the version to be published. B.V. obtained the funding and contributed to the study supervision. All authors take responsibility for the entire article. The PLASA study was supported by the French Ministry of Health grant: PHRC 02-006-01. The authors thank Helene Grandjean for her helpful advice concerning the analysis and article preparation and Clarisse Ventelon for her initial participation in this analysis. The authors are indebted to the investigators from the following hospitals participating in the PLASA study: University hospital: Angers (Dr. Berrut, Dr Barre), Bordeaux (Pr. Rainfray, Dr Harston), Brest (Dr Gentric),
9
Grenoble (Pr. Franco, Dr. Couturier), Lille (Dr Roche, Dr Huvent), Lyon (Pr Bonnefoy), Marseille (Dr. Michel), Montpellier (Pr. Jeandel), Montpellier (Pr. Touchon, Dr. Portet, Dr Lerouge), Nice (Dr. Brocker, Dr Guerin), Nice (Pr. Robert), Paris (Pr Teillet, Dr Lechowski), Paris (Pr. Belmin, Dr Pariel-Madjlessi), Paris (Pr. Rigaud, Dr Latour), Paris (Pr Verny, Dr Marquis, Dr Brihier), Paris (Pr Legrain, Dr Girard), Reims (Pr Blanchard, Dr Kack), Rennes (Pr Jouanny, Dr Ledu), Rouen (Pr Chassagne, Dr Levasseur), and Rouen (Pr Hannequin, Dr Dugny). General hospital: Albi (Dr Quincon), Ales (Dr Peju), Annecy (Dr Picot), Bar Le Duc (Dr Tagu, Dr De Guio), Carcassonne (Dr Tannie), Carvin (Dr Taillez), Chambery (Dr Declippeleir), Champcueil (Dr Maugourd, Dr Baptiste), Dieppe (Dr Pesque), Elbeuf (Dr Simon), Grasse (Dr Ribiere), Lannemezan (Dr Bordes), Lavaur (Dr De Pemille), Le Havre (Dr Landrin-Dutot, Dr Olivier), Lens (Dr Senechal, Dr Fournier), Nice (Dr Giordana), N^ımes (Dr Strubel), Niort (Dr Chaumier), Paris (Dr DurandGasselin), Pau (Dr. De La Fourniere, Dr Sauvanier—Geriatric Unit), Plaisir (Dr Bessey, Dr Drunat), Roubaix (Dr Forzy), Rouen (Dr Moynot, Dr Denis), Saint Dizier (Dr Aubertin), Sezanne (Dr Quignard), Valenciennes (Dr Leurs), Vannes (Dr Le Provost), Villejuif (Dr Feteanu), and Wasquehal (Dr. Frigard, Dr Idiri).
RESEARCH IN CONTEXT
1. Systematic review: Alzheimer’s disease (AD) patients’ quality of life (QOL) is often rated by caregivers, but such ratings are generally divergent with self-reports. Discrepancies may be related to patient and caregiver characteristics, but previous studies have been limited by small sample sizes, selected study populations, and incomprehensive patient evaluations. Furthermore, little is known about the natural history of QOL in AD. 2. Interpretation: In our study, caregivers consistently underestimated AD patients’ QOL compared with that of patients themselves, and patient self-ratings did not change over time, despite worsening of dementia symptoms, although caregiver ratings significantly declined. Patient age and neuropsychiatric symptoms and caregiver burden were associated with discrepancies. 3. Future directions: Caregiver ratings of patient QOL may be biased. Studies using solely proxy reports of patient QOL are to be discouraged, except in severe dementia. Temporal relationships between level of caregiver burden and patient QOL and the evolution of patient–caregiver disagreement should be further explored.
10
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11
References [1] Aguero-Torres H, Fratiglioni L, Guo Z, Viitanen M, von Strauss E, Winblad B. Dementia is the major cause of functional dependence in the elderly: 3-year follow-up data from a population-based study. Am J Public Health 1998;88:1452–6. [2] Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, et al. Global prevalence of dementia: A Delphi consensus study. Lancet 2005;366:2112–7. [3] Netuveli G, Wiggins RD, Hildon Z, Montgomery SM, Blane D. Functional limitation in long standing illness and quality of life: Evidence from a national survey. BMJ 2005;331:1382–3. [4] Banerjee S, Smith SC, Lamping DL, Harwood RH, Foley B, Smith P, et al. Quality of life in dementia: More than just cognition. An analysis of associations with quality of life in dementia. J Neurol Neurosurg Psychiatry 2006;77:146–8. [5] WHO World Organization Group. Development of the WHOQOL: Rationale and current status. Int J Ment Health 1994;23:24–56. [6] Vellas B, Andrieu S, Sampaio C, Coley N, Wilcock G. Endpoints for trials in Alzheimer’s disease: A European task force consensus. Lancet Neurol 2008;7:436–50. [7] Trinh NH, Hoblyn J, Mohanty S, Yaffe K. Efficacy of cholinesterase inhibitors in the treatment of neuropsychiatric symptoms and functional impairment in Alzheimer disease: A meta-analysis. JAMA 2003;289:210–6. [8] EMEA. European Medicines Agency. Committee for Medicinal Products for Human Use (CHMP) draft guideline on medicinal products for the treatment of Alzheimer’s disease and other dementias. Doc. Ref. CPMP/EWP/553/95 Rev. 1 2007; Available from: http://www. ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/ 2009/09/WC500003565.pdf. [9] Banerjee S. Commentary on “Health economics and the value of therapy in Alzheimer’s disease.” Quality of life in dementia: Development and use of a disease-specific measure of health-related quality of life in dementia. Alzheimers Dement 2007;3:166–71. [10] Logsdon RG, Gibbons LE, McCurry SM, Teri L. Assessing quality of life in older adults with cognitive impairment. Psychosom Med 2002; 64:510–9. [11] Scholzel-Dorenbos CJ, Ettema TP, Bos J, Boelens-van der Knoop E, Gerritsen DL, Hoogeveen F, et al. Evaluating the outcome of interventions on quality of life in dementia: Selection of the appropriate scale. Int J Geriatr Psychiatry 2007;22:511–9. [12] Ready RE, Ott BR. Quality of life measures for dementia. Health Qual Life Outcomes 2003;1:11. [13] Jonsson L, Andreasen N, Kilander L, Soininen H, Waldemar G, Nygaard H, et al. Patient- and proxy-reported utility in Alzheimer disease using the EuroQoL. Alzheimer Dis Assoc Disord 2006;20:49–55. [14] Yasuda N, Zimmerman S, Hawkes WG, Gruber-Baldini AL, Hebel JR, Magaziner J. Concordance of proxy-perceived change and measured change in multiple domains of function in older persons. J Am Geriatr Soc 2004;52:1157–62. [15] Schulz R, Cook TB, Beach SR, Lingler JH, Martire LM, Monin JK, et al. Magnitude and causes of bias among family caregivers rating Alzheimer disease patients. Am J Geriatr Psychiatry 2013;21:14–25. [16] Zucchella C, Bartolo M, Bernini S, Picascia M, Sinforiani E. Quality of life in Alzheimer disease: A comparison of patients’ and caregivers’ points of view. Alzheimer Dis Assoc Disord 2015;29:50–4. [17] Huang HL, Weng LC, Tsai YH, Chiu YC, Chen KH, Huang CC, et al. Predictors of self- and caregiver-rated quality of life for people with dementia living in the community and in nursing homes in northern Taiwan. Int Psychogeriatr 2015;27:825–36. [18] Conde-Sala JL, Turro-Garriga O, Garre-Olmo J, Vilalta-Franch J, Lopez-Pousa S. Discrepancies regarding the quality of life of patients with Alzheimer’s disease: A three-year longitudinal study. J Alzheimers Dis 2014;39:511–25.
[19] Zhao H, Novella JL, Drame M, Mahmoudi R, Barbe C, di Pollina L, et al. Factors associated with caregivers’ underestimation of quality of life in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 2012;33:11–7. [20] Hoe J, Hancock G, Livingston G, Orrell M. Quality of life of people with dementia in residential care homes. Br J Psychiatry 2006; 188:460–4. [21] Aisen PS, Schafer KA, Grundman M, Pfeiffer E, Sano M, Davis KL, et al. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: A randomized controlled trial. JAMA 2003;289:2819–26. [22] Wlodarczyk JH, Brodaty H, Hawthorne G. The relationship between quality of life, mini-mental state examination, and the instrumental activities of daily living in patients with Alzheimer’s disease. Arch Gerontol Geriatr 2004;39:25–33. [23] Nourhashemi F, Andrieu S, Gillette-Guyonnet S, Giraudeau B, Cantet C, Coley N, et al. Effectiveness of a specific care plan in patients with Alzheimer’s disease: Cluster randomised trial (PLASA study). BMJ 2010;340:c2466. [24] McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984;34:939–44. [25] Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–98. [26] Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. JAMA 1963;185:914–9. [27] Lawton MP, Brody EM. Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 1969; 9:179–86. [28] Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The neuropsychiatric inventory: Comprehensive assessment of psychopathology in dementia. Neurology 1994;44:2308–14. [29] Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of the elderly: The mini nutritional assessment as part of the geriatric evaluation. Nutr Rev 1996;54(1 Pt 2):S59–65. [30] Zarit SH, Reever KE, Bach-Peterson J. Relatives of the impaired elderly: Correlates of feelings of burden. Gerontologist 1980; 20:649–55. [31] Singer JD. Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. J Educ Behav Stat 1998; 23:323–55. [32] Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. , New York, NY: John Wiley & Sons; 2000. . 118. [33] Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation. Med Care 2003;41:582–92. [34] Bosboom PR, Alfonso H, Almeida OP. Determining the predictors of change in quality of life self-ratings and carer-ratings for communitydwelling people with Alzheimer disease. Alzheimer Dis Assoc Disord 2013;27:363–71. [35] Clare L, Woods RT, Nelis SM, Martyr A, Markova IS, Roth I, et al. Trajectories of quality of life in early-stage dementia: Individual variations and predictors of change. Int J Geriatr Psychiatry 2014; 29:616–23. [36] Tatsumi H, Nakaaki S, Torii K, Shinagawa Y, Watanabe N, Murata Y, et al. Neuropsychiatric symptoms predict change in quality of life of Alzheimer disease patients: A two-year follow-up study. Psychiatry Clin Neurosci 2009;63:374–84. [37] Pearlin LI, Schooler C. The structure of coping. J Health Soc Behav 1978;19:2–21. [38] Whitehouse PJ, George D. Banking on stories for healthier cognitive ageing. Lancet 2009;373:1166–7. [39] Scholzel-Dorenbos CJ, van der Steen MJ, Engels LK, Olde Rikkert MG. Assessment of quality of life as outcome in dementia
S. Andrieu et al. / Alzheimer’s & Dementia - (2015) 1-11 and MCI intervention trials: A systematic review. Alzheimer Dis Assoc Disord 2007;21:172–8. [40] Grand A, Grand-Filaire A, Bocquet H, Clement S. Caregiver stress: A failed negotiation? A qualitative study in South West France. Int J Aging Hum Dev 1999;49:179–95. [41] Wetzels RB, Zuidema SU, de Jonghe JF, Verhey FR, Koopmans RT. Determinants of quality of life in nursing home residents with dementia. Dement Geriatr Cogn Disord 2010;29:189–97.
11
[42] Selwood A, Thorgrimsen L, Orrell M. Quality of life in dementia—A one-year follow-up study. Int J Geriatr Psychiatry 2005;20:232–7. [43] Gow AJ, Whiteman MC, Pattie A, Whalley L, Starr J, Deary IJ. Lifetime intellectual function and satisfaction with life in old age: Longitudinal cohort study. BMJ 2005;331:141–2. [44] Hoe J, Katona C, Roch B, Livingston G. Use of the QOL-AD for measuring quality of life in people with severe dementia—the LASER-AD study. Age Ageing 2005;34:130–5.