Care costs and caregiver burden for older persons with dementia in Taiwan

Transcription

1 Care costs and caregiver burden for older persons with dementia in Taiwan Li-Jung Elizabeth Ku Department of Public Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan 2017/4/28 ADI 2017, Kyoto 1

2 Introduction Prevalence of dementia in Taiwan The disease prevalence is clearly on the rise, from 0.7% in 1996 to 4.84% in 2012 (Chien et al., 2008; Sun et al.,, 2014) Cost-of-illness (COI) studies provide estimates about the economic impact of diseases and offer data to assist decision makers for planning and financing of health systems A systematic review of 28 COI studies of dementia found disease stage an important determinant of costs (Quentin et al., 2011) Costs more than doubled from mild to severe dementia Large variation found between studies due to setting and cultural characteristics 2

3 Introduction Literature: predictors of caregiver burden Patient characteristics associated with higher burden include: behavioral and psychological symptoms (Bergvall, 2011) functional disability (ADL) (Wolfs, 2012) Caregiver characteristics associated with higher burden include: Female (Poysti, 2012) Younger age (Pinquart and Sörensen, 2006) Co-resident family members (Kim, 2012) Social support Higher support is negatively related to caregiver burden (Clyburn, 2000) 3

4 Objective To investigate dual burdens of caring for persons with dementia among Chinese caregivers: Financial/ Objective burden Psychological / Subjective burden Predictors of care costs and caregiver burden for older persons with dementia in Taiwan Are the predictors for dual burdens similar or different? 4

5 Materials and Methods Methods Participants (dyads): Persons with dementia and their family caregivers Setting and time periods NCKU Hospital dementia clinic Baseline survey:2013/11~2015/4 (N=231 dyads) Follow-up survey: 2015/11 ~2016/6 (N=167 dyads) Inclusion criteria for persons with dementia (pwd) 286 dementia patient from a NCKU dementia clinic Must been living in the community and had an informal caregiver Neuropsychological and functional status exams for pwd 5

6 Inclusion criteria for caregivers: ages >= 18 years, who was fluent in either mandarin or Taiwanese excluded paid caregivers excluded a caregiver for less than 12 months Telephone interview using structured questionnaires Measures of Caregivers Sociodemographic characteristics Living arrangement Economic status Social support Methods 2 Formal support: use of any paid long-term care service Family support: another family caregiver for the same pwd 6

7 Materials and Methods Subjective burden Zarit burden Interview Item # and descriptions 1 Not enough time for myself 2 Stressed for caring & other responsibilities 3 Relative s dependence on you 4 Strained by relative 5 Health decrease 6 Lack of privacy 7 Lack of social life 8 Unable to care much longer 9 Lost control of life 10 Leave the care to someone else 11 Uncertain about what to do 12 Overall feeling of burden All 12 items on the ZBI are rated on a 5-point scale from 0 (not at all) to 4 (extremely) A higher ZBI summary score indicates a higher level of caregiver burden 7

8 Objective burden Measures of care costs Direct medical costs (NHI) Direct medical costs (outof-pocket) Direct social care costs Informal care costs

9 Materials and Methods Statistical analyses Descriptive analysis of study sample at two waves Bivariate analyses to identify significant predictors to enter into multivariate models of burden/cost Mixed models using data from both waves Costs models used log dependent variables after BoxCox tests Burden model assumed the Gaussian distribution Included random intercept and chose variance component for covariance structures 9

10 Result Table 1. Sample characteristics of dementia patients and caregivers Variables Baseline(n=231) 1 year(n=167) Patients characteristics Age: mean (range) mean/n (range/%) 80.0 (65-101) mean/n (range/%) 81.5 (66-102) Gender: N, % female 138 (59.7) 104 (62.3) Number of children 5.6 (0-14) 5.6 (0-14) Education: No education 57 (24.7) 46 (27.5) Elementary school 108 (46.8) 72 (43.1) Junior high school 66 (28.5) 49 (29.3) or above ADL score 69.1 (0-100) 63.8 (0-100) NPI score 18 (0-100) 16.9 (0-97) Dementia severity Mild 102 (44.1) 62 (37.1) Moderate 88 (38.1) 63 (37.7) Severe 41 (17.8) 42 (25.2) ADL= Activities of Daily Living, higher scores indicating less need for support; NPI= Neuropsychiatric Inventory, higher scores indicate greater impairment. 10

11 Result Variables Baseline(n=231) 1 year(n=167) Table 1. Cont. mean/n (range/%) mean/n (range/%) Caregiver characteristics Age: mean (range) 61.1 (25-87) 61.9 (26-88) Gender: N, % female 146 (63.2) 111 (66.5) Co-resident with patients 175 (75.8) 120 (71.8) Education No education 12 (5.2) 10 (6.0) Elementary school 34 (14.7) 26 (15.6) Junior high 22 (9.5) 13 (7.8) Senior high 82 (35.5) 61 (36.5) College or above 81 (35.1) 57 (34.1) Relationship with the patient Spouse 88 (38.1) 66 (39.5) Adult children 109 (47.2) 73 (43.7) Children-in-law 29 (12.5) 24 (14.4) Grandchildren 5 (2.2) 4 (2.4) Married 190 (82.3) 139 (83.2) Employed 97 (42.0) 69 (41.3) Economic status Have difficulty 48 (20.8) 42 (25.5) Have enough money. 147 (63.6) 88 (53.3) Have more than enough money 36 (15.6) 35 (21.2) Caregiver burden 16.9(0-44) 16.6(0-43) Social support Formal support 100 (43.3) 83 (49.7) Family support 206 (89.2) 100 (59.9) 11

12 Result Table 2.Mixed models on caregiver burden 12

13 Result Table 3.Mixed models on total costs of care Variables Total costs Patient characteristics Age 0.01 (0.01)* Male (ref: female) (0.10) Number of Children 0.06 (0.03)** Education(ref: no education) Elementary school 0.02 (0.12) Junior high school or above 0.09 (0.14) ADL score (0.00)*** NPI score 0.00 (0.00) Dementia severity (ref: mild) Moderate 0.06 (0.10) Severe (0.13) Caregiver characteristics Age (0.00)*** College education or above (ref: Senior high 0.31 (0.11)*** school and below) Married (0.14)** Employed 0.07 (0.09) 13

14 Table 4.Mixed models on three types of costs of care Variables Result Medical care Social care Informal care Patient characteristics Age ** 0.14*** (0.01) (0.02) (0.03) Education(ref: no education) Elementary school * (0.14) (0.25) (0.59) Junior high school or above ** (0.16) (0.27) (0.65) ADL score -0.00** (0.00) (0.00) (0.01) NPI score (0.00) (0.00) (0.01) Dementia severity (ref: mild) Moderate (0.12) (0.22) (0.44) Severe -0.31** (0.16) (0.29) (0.60) Caregiver characteristics Age *** (0.00) (0.00) (0.02) 14

15 Predictors of caregiver burden are different from predictors of care costs behavioral disturbance was a significant predictor of caregiver burden functional declines measured by ADL affected total costs Discussion Economic status is significantly associated with caregiver burden economic superiority associated with lower burden implying hardship facing poor family caregivers 15

16 Discussion Total care costs/financial burden were positively correlated with patient s age and receiving formal support, but costs were negatively correlated with patient s functional dependence and caregiver s age Two effective strategies to reduce subjective burden: Providing training to caregivers in addressing dementia patient behavioral disturbance and Providing financial assistance to low income caregivers 16

17 Limitation Cost estimates from a single site study are not nationally representative and did not include patients living in institutions Likely to be underestimates as only from NCKU hospitals Voluntary participation by caregivers and selection bias Burn-out caregivers were less likely to be interviewed Dementia patients with more limitations in ADL were more likely to drop out 17

18 Thank You and Any Questions? 18

19 Materials and Methods Materials and Methods Measures of care costs a) Medical costs included costs of inpatient and outpatient visits, out-of-pocket expenditures from 12 months prior to the survey b) Social care costs included adult daycare, respite care, home care, paid domestic help, and transportation costs. c) Informal care costs were valued by the level of contribution to patient care by the caregiver Hours spent assisting with basic ADL, IADL, and supervision Human capital approach (Base-case estimate) Replacement costs approach (Sensitivity analysis) 19