Depression in the Family of Patients With Dementia in Korea

Abstract

Introduction:

Dementia is the leading cause of disability worldwide in the elderly individuals. Although prior studies have examined psychiatric symptoms in dementia caregivers, few studies have examined physician-diagnosed depression in the family caregiver of a patient with dementia.

Methods:

We used data from 457 864 respondents from the Korea Community Health Survey. We used logistic regression to examine the relationship between physician-diagnosed depression and cohabitation with a patient with dementia.

Results:

Cohabitation with a patient with dementia (1.2% of the Korean population) was significantly associated with physician-diagnosed depression. The significance remained in females when the data were stratified by sex. A significant association also occurred among males with low family income.

Conclusion:

To reduce the burden of dementia, we need a management policy that includes the caregiver as well as the patient with dementia. In particular, political management for the vulnerable population, male caregiver in low-income family and female caregiver in high-income family, should be prepared.

Keywords

dementia caregiver depression community health survey

Introduction

Dementia is the leading cause of disability among the elderly individuals and is a major contributor to mortality as well as a challenge to the health care system worldwide.¹ In 2010, an estimated 35.6 million people were living with dementia worldwide, and these numbers are expected to double approximately every 20 years to 65.7 million in 2030 and 115.4 million in 2050.²

Dementia is an umbrella term to describe a variety of diseases and conditions that develop when nerve cells in the brain die or no longer function normally.³ Dementia includes Alzheimer’s disease (AD), vascular dementia (VD), dementia with Lewy bodies, frontotemporal lobar degeneration, and Parkinson’s disease. The early signs of AD, which is the most common type of dementia, include sporadic memory loss and subtle behavioral changes. Alzheimer’s disease gradually progresses until, often after a decade or more, the individual is unable to speak or comprehend language, can no longer control his or her bowels, and requires assistance with all aspects of personal care.⁴

No cure for dementia currently exists, and the number of individuals with AD and other types of dementia continues to grow.⁵ The home-based care of individuals with dementia is an important topic to address. The burden on caregivers is known to be higher among those who care for a patient with dementia compared to those who care for patients with other medical illnesses.⁶ Home-based caregivers play an important role in the management of dementia, and several studies have demonstrated that this role has a strong impact on the physical and mental health of the caregiver. In particular, caregivers of patients with dementia have poor sleep, exhaustion, and depression.^7,8 Given the impending worldwide dementia epidemic, caregiver burden has become one of the main concerns in the dementia clinic as well as for researchers and health care policy makers.⁹

Meanwhile, depression is one of the most prevalent issues globally.¹⁰ According to Blazer,¹¹ Pinquart et al,¹² and Wilkins et al,¹³ depression is associated with a significant increased risk of functional decline and all-cause mortality in older adults. Poor health and impaired functioning are also found to be associated with depressive symptoms, so depression places a considerable burden on both society and individual.¹⁴

Depression of family caregivers of people with dementia is well known in Western countries.¹⁵ An obvious association between dementia caregiving and negative effects on psychological health has been demonstrated in numerous studies.^16,17

In Korea, the prevalence of dementia has increased sharply as the overall population ages. Previous studies have reported that the prevalence of dementia is between 7.2% and 8.1% among the urban Korean elderly population.^18,19 In addition, the prevalence of AD and VD has been estimated to range from 4.2% to 6.5% and 1.3% to 2.5%, respectively, in the Korean population.^20
–22 Many studies have examined the health of the caregivers of patients with dementia in Korea. For example, Han et al⁹ have analyzed the association between the burden on the caregivers of patients with dementia and the caregiver’s social supports, and both Kang et al²³ and Lim et al²⁴ have reported on the factors that contribute to the burden on the caregivers of patients with dementia. Shin et al²⁵ compared the caregiver burden between those who cared for patients with Parkinson’s disease and dementia and those who had patients with AD, and Lee et al²⁶ studied stress among caregivers of patients with dementia in Korea. These studies, however, are limited by small sample sizes and do not represent the entire Korean population. The purpose of this study, therefore, was to analyze the association between physician-diagnosed depression and cohabitation with a patient with dementia using national representative data. A secondary purpose of this study was to examine this association stratified by sex and economic status.

Methods

Dataset

The data used in this study came from the Korea Community Health Survey (KCHS), which was administered by the Korean Centers for Disease Control and Prevention in 2011 and 2012. The KCHS standardized questionnaire was developed jointly by the Korea Centers for Disease Control and Prevention staff, a working group of health indicators standardization subcommittee, and 16 metropolitan cities and provinces with 253 regional sites. The questionnaire covers a variety of topics related to health behaviors and prevention, which is used to assess the prevalence of personal health practices and behaviors related to the leading causes of disease, including smoking, alcohol use, drinking and driving, high-blood pressure control, physical activity, weight control, quality of life (European Quality of Life-5 Dimensions, European Quality of Life-Visual Analogue Scale, Korean Instrumental Activities of Daily Living), medical service, accident, injury, and so on. The microdata (in the form of Statistical Analysis Software [SAS] files) and analytic guidelines can be downloaded from the KCHS website (http://KCHS.cdc.go.kr/) in Korean.²⁷

Participant

Data were gathered for 458 147 individuals (229 226 in 2011 and 228 921 in 2012). Data from both years were integrated and sampling weights were incorporated for the analysis. The final analysis used data from 457 864 individuals after excluding 283 people for whom information on physician-diagnosed depression and/or cohabitation with a patient with dementia was unavailable (209 were missing information about depression, 77 were missing information about cohabitation with a patient with dementia, and 3 were missing both). Our study did not need to address any ethical concerns because the Community Health Survey data are secondary data that are available in the public domain and do not contain private information.

Table 1 presents the demographic information from the 457 864 participants who were included in the analysis stratified by sex. The proportion of individuals who reported cohabitation with a patient with dementia was 1.2% across both sexes. The average age of males surveyed was 51.2 years, which was 1.5 years older on average than female respondents. The dominant employment type was site worker (44.4%) among male respondents and unemployed/homemaker (49.1%) among female respondents. The majority (38.5%) of males reported their highest level of education as university graduate or higher. In contrast, the majority of female respondents had a low education level (elementary school: 34.4%). The proportion of respondents with physician-diagnosed depression was 1.2% among males and 3.6% among females. The overall prevalence of depression across all survey respondents was 2.5%.

Table 1.

Study Population Characteristics.

							(n, %)
		Total		Male		Female	P Value
Cohabitation with a patient with dementia	Yes	5360	1.2	2398	1.2	2962	1.2	.752
Cohabitation with a patient with dementia	No	452 504	98.8	203 421	98.8	249 083	98.8	.752
Age	Mean, SD	51.2	16.9	50.4	16.3	51.9	17.3	<.001
Employment status	Office worker	142 517	31.2	68 604	33.4	73,913	29.4
	Site worker	145 625	31.8	91 276	44.4	54,349	21.6
	Unemployed/homemaker	169 190	37.0	45 642	22.2	123 548	49.1
Highest level of education	Elementary school	121 878	26.7	35 365	17.2	86 513	34.4	<.001
	Middle school	53 307	11.7	25 185	12.3	28 122	11.2
	High school	133 453	29.2	65 815	32.0	67 638	26.9
	University or higher	148 334	32.5	79 066	38.5	69 268	27.5
Marital status	Single	66 204	14.5	36 550	17.8	29 654	11.8	<.001
	Living with spouse	315 579	69.0	152 992	74.4	162 587	64.6
	Separated/divorced/widowed	75 746	16.6	16 134	7.8	59 612	23.7
Family income level	First quartile (highest)	112 621	26.4	54 115	28.2	58 506	25.0	<.001
	Second quartile	96 212	22.6	45 313	23.6	50,899	21.7
	Third quartile	107 002	25.1	48 525	25.2	58 477	25.0
	Fourth quartile (lowest)	110 734	26.0	44 264	23.0	66 470	28.4
Number of family members	1	47 860	10.5	14 519	7.1	33 341	13.2	<.001
	2	145 088	31.7	69 005	33.5	76 083	30.2
	3	96 092	21.0	45 498	22.1	50 594	20.1
	4	112 384	24.6	52 541	25.5	59 843	23.7
	5+	56 440	12.3	24 256	11.8	32 184	12.8
Insurance type	Medical aid	15 768	3.4	5877	2.9	9891	3.9	<.001
Insurance type	National Health Insurance	441 973	96.6	199 892	97.1	242 081	96.1
Self-reported health condition	Excellent	180 117	39.3	93 099	45.2	87 018	34.5	<.001
	Average	179 979	39.3	78 140	38.0	101 839	40.4
	Poor	97 706	21.3	34 550	16.8	63 156	25.1
Self-reported stress	Very often	15 216	3.3	6662	3.2	8554	3.4	<.001
	Often	105 871	23.2	45 920	22.3	59 951	23.8
	Occasionally	243 744	53.3	109 361	53.2	134 383	53.4
	Seldom	92 483	20.2	43 719	21.3	48 764	19.4
Smoking habit	Never	287 942	62.9	49 530	24.1	238 412	94.6	<.001
	Past	73 395	16.0	67 811	33.0	5584	2.2
	Current	96 452	21.1	88 442	43.0	8010	3.2
Alcohol use	No	156 969	34.3	44 404	21.6	112 565	44.7	<.001
Alcohol use	Yes	300 844	65.7	161 401	78.4	139 443	55.3
Hours of sleep per night	Less than 6 hours	71 318	15.6	28 597	13.9	42 721	17.0	<.001
	6 hours	130 080	28.4	61 957	30.1	68 123	27.1
	7 hours	145 982	31.9	67 205	32.7	78 777	31.3
	8 hours	90 408	19.8	39 516	19.2	50 892	20.2
	More than 9 hours	19 624	4.3	8407	4.1	11 217	4.5
Underlying chronic disease^a	Yes	167 733	36.6	68 578	33.3	99 155	39.3	<.001
Underlying chronic disease^a	No	290 131	63.4	137 241	66.7	152 890	60.7
Physician-diagnosed depression	Yes	11 422	2.5	2424	1.2	8998	3.6	<.001
Physician-diagnosed depression	No	446 442	97.5	203 395	98.8	243 047	96.4
Survey year	2011	229 059	50.0	102 958	50.0	126 101	50.0	.959
Survey year	2012	228 805	50.0	102 861	50.0	125 944	50.0
Total		457 864	100	205 819	45.0	252 045	55.0

^aUnderlying chronic disease includes hypertension, diabetes, dyslipidemia, and arthritis.

Interesting Variable

The variable of interest in this study was the existence of a cohabitating patient with dementia in the family. We defined the family caregiver of a patient with dementia as any individual who resided with a patient having dementia, a definition that is consistent with a previous study.²⁸ The Community Health Survey asked each individual “Do you live with a dementia patient in your family?” and the response choices were binary (yes or no). The dependent variable in this study was the existence of physician-diagnosed depression in the respondent. The Community Health Survey asked “Have you ever been diagnosed with depression by a doctor?” and the response choices were binary (yes or no).

Independent Variables

The independent variables included the survey year, sex, age, employment status, education level, marital status, family income, number of family members, insurance type, self-reported health condition, self-reported stress, smoking habit, alcohol use, number of hours of sleep per night, and the existence of an underlying chronic disease in the respondent. Age was analyzed as a continuous variable. The employment variable was divided into 3 categories: office worker, site worker, and unemployed/homemaker. The education-level variable was divided into 4 categories: graduation from university or higher, graduation from high school, graduation from middle school, and graduation from elementary school. The marital status variable was divided into 3 categories: single, married living with spouse, and separated/divorced/widowed. We used square root scale for family income, which divides household income by the square root of household size.²⁹ The family income variable was divided into 4 quartiles adjusted by family size. The number of family members variable was divided into 5 groups: 1,2,3,4 and 5+. The medical insurance-type variable was divided into 2 categories: National Health Insurance (NHI) and Medical aid. Universal coverage with NHI for Koreans was achieved in 1989, and Medical aid contributes to the social welfare and health of citizens with a low economic status, and it helps them receive medical services at minimal or no cost. In 2010, 96.4% of all citizens were covered by NHI and 3.6% by Medical aid.³⁰ The Medical aid program classifies beneficiaries into 2 categories, type 1 and 2, on the basis of being incapable (those younger that 18 or older than 65 years of age, or disabled) or capable of working, respectively.³¹ The self-reported health condition variable was divided into 3 categories: excellent, average, and poor. The self-reported stress variable was divided into 4 categories: very often, often, occasionally, and seldom. The smoking habit variable was divided into 3 categories: never, past, and current smoker. The alcohol use variable was divided into 2 categories: no and yes. Because sleep duration is independently associated with depression, we divided this variable into 5 categories by the number of hours of sleep per night: less than 6 hours, 6 hours, 7 hours, 8 hours, and more than 9 hours.³² The variable about the existence of an underlying disease considered individuals who were diagnosed with hypertension, diabetes, dyslipidemia, or any type of arthritis by a doctor. The survey year variable was divided into 2 categories: 2011 and 2012. All variables had statistically significant associations with physician-diagnosed depression except the survey year variable.

Statistical Analysis

The data analysis was conducted using the χ² test and logistic regression in SAS 9.4 (SAS Institute, Inc.; Cary, North Carolina). The data were analyzed over the entire population and then stratified by sex and family income level. Results with a P value <.05 were considered significant.

Results

Table 2 presents the results of the univariate analyses that examined the association between physician-diagnosed depression and each variable by sex. The results of the chi-square test revealed that cohabitation with a patient with dementia was significantly associated with physician-diagnosed depression. The percentage of respondents with physician-diagnosed depression among those who lived with a patient with dementia was twice that of those who did not live with a patient with dementia (5.0% vs 2.5%, respectively). The average age of the respondents with physician-diagnosed depression (56.0 years) was higher than that of nondepressed respondents. In addition, the prevalence of physician-diagnosed depression was twice as high in respondents who identified their employment status as unemployed/homemaker as those who identified their employment status as an office worker (1.4%) or a site worker (1.9%). The prevalence of depression was higher in the lower income groups (4th quartile: 4.1%) and 3 times as high in respondents with Medical aid compared with NHI (7.7% vs 2.3%, respectively). Physician-diagnosed depression was also higher among those respondents who reported poor health (6.5%) and more stress (10.8%). In addition, physician-diagnosed depression was higher among respondents who reported too little sleep (<6 hours, 4.7%) as well as too much sleep (>9 hours, 4.5%). Respondents who reported an underlying chronic disease had a 4.0% prevalence of physician-diagnosed depression.

Table 2.

Associations Between Physician-Diagnosed Depression and Subject Demographics.

												(n, %)
		Both Sexes					Male					Female
		Diagnosed depression		Nondiagnosed		P Value	Diagnosed Depression		Nondiagnosed		P Value	Diagnosed Depression		Nondiagnosed		P Value
Cohabitation with a patient with dementia	Yes	267	5.0	5093	95.0	<.001	66	2.8	2332	97.3	<.001	201	6.8	2761	93.2	<.001
	No	11 155	2.5	441 349	97.5		2358	1.2	201 063	98.8		8797	3.5	240 286	96.5
Age	Mean, SD	56	15.3	51.1	16.9	<.001	54.9	16.8	50.4	16.3	<.001	56.3	14.8	51.7	17.3	<.001
Employment status	Office worker	1987	1.4	140 530	98.6	<.001	435	0.6	68 169	99.4	<.001	1552	2.1	72 361	97.9	<.001
	Site worker	2694	1.9	142 931	98.2		783	0.9	90 493	99.1		1911	3.5	52 438	96.5
	Unemployed or homemaker	6732	4.0	162 458	96.0		1204	2.6	44 438	97.4		5528	4.5	118 020	95.5
Highest level of education	Elementary school	4612	3.8	117 266	96.2	<.001	578	1.6	34 787	98.4	<.001	4034	4.7	82 479	95.3	<.001
	Middle school	1897	3.6	51 410	96.4		416	1.7	24 769	98.4		1481	5.3	26 641	94.7
	High school	2973	2.2	130 480	97.8		736	1.1	65 079	98.9		2237	3.3	65 401	96.7
	University or higher	1915	1.3	146 419	98.7		694	0.9	78 372	99.1		1221	1.8	68 047	98.2
Marital status	Single	1091	1.7	65 113	98.4	<.001	536	1.5	36 014	98.5	<.001	555	1.9	29 099	98.1	<.001
	Living with spouse	7016	2.2	308 563	97.8		1461	1.0	151 531	99.1		5555	3.4	157 032	96.6
	Separated/divorced/widowed	3305	4.4	72 441	95.6		426	2.6	15 708	97.4		2879	4.8	56 733	95.2
Family income level	1st quartile (highest)	1809	1.6	110 812	98.4	<.001	377	0.7	53 738	99.3	<.001	1432	2.5	57 074	97.6	<.001
	2nd quartile	1617	1.7	94 595	98.3		327	0.7	44 986	99.3		1290	2.5	49 609	97.5
	3rd quartile	2688	2.5	104 314	97.5		564	1.2	47 961	98.8		2124	3.6	56 353	96.4
	4th quartile (lowest)	4573	4.1	106 161	95.9		997	2.3	43 267	97.8		3576	5.4	62 894	74.6
Number of family members	1	1923	4.0	45 937	96.0	<.001	339	2.3	14 180	97.7	<.001	1584	4.8	31 757	95.3	<.001
	2	4372	3.0	140 716	97.0		1006	1.5	67 999	98.5		3366	4.4	72 717	95.6
	3	2174	2.3	93 918	97.7		493	1.1	45 005	98.9		1681	3.3	48 913	96.7
	4	1793	1.6	110 591	98.4		368	0.7	52 173	99.3		1425	2.4	58 418	97.6
	5+	1160	2.1	55 280	97.9		218	0.9	24 038	99.1		942	2.9	31 242	97.1
Insurance type	Medical aid	1212	7.7	14 556	92.3	<.001	330	5.6	5547	94.4	<.001	882	8.9	9009	91.1	<.001
Insurance type	National Health Insurance	10 206	2.3	431 767	97.7		2094	1.1	197 798	99.0		8112	3.4	233 969	96.7
Self-reported health condition	Excellent	1473	0.8	178 644	99.2	<.001	407	0.4	92 692	99.6	<.001	1066	1.2	85 952	98.8	<.001
	Average	3642	2.0	176 337	98.0		736	0.9	77 404	99.1		2906	2.9	98 933	97.2
	Poor	6305	6.5	91 401	93.6		1280	3.7	33 270	96.3		5025	8.0	58 131	92.0
Self-reported stress	Very often	1647	10.8	13 569	89.2	<.001	391	5.9	6271	94.1	<.001	1256	14.7	7298	85.3	<.001
	Often	5060	4.8	100 811	95.2		1014	2.2	44 906	97.8		4046	6.8	55 905	93.3
	Occasionally	3582	1.5	240 162	98.5		764	0.7	108 597	99.3		2818	2.1	131 565	97.9
	Seldom	1112	1.2	91 371	98.8		251	0.6	43 468	99.4		861	1.8	47 903	98.2
Smoking habit	Never	8448	2.9	279 494	97.1	<.001	525	1.1	49 005	98.9	<.001	7923	3.3	230 489	96.7	<.001
	Past	1278	1.7	72 117	98.3		909	1.3	66 902	98.7		369	6.6	5215	93.4
	Current	1691	1.8	94 761	98.3		988	1.1	87 454	98.9		703	8.8	7307	91.2
Alcohol use	No	5648	3.6	151 321	96.4	<.001	916	2.1	43 488	97.9	<.001	4732	4.2	107 833	95.8	<.001
Alcohol use	Yes	5773	1.9	295 071	98.1		1507	0.9	159 894	99.1		4266	3.1	135 177	96.9
Hours of sleep per night	Less than 6 hours	3320	4.7	67 998	95.3	<.001	574	2.0	28 023	98.0	<.001	2746	6.4	39 975	93.6	<.001
	6 hours	2624	2.0	127 456	98.0		528	0.9	61 429	99.2		2096	3.1	66 027	96.9
	7 hours	2506	1.7	143 476	98.3		545	0.8	66 660	99.2		1961	2.5	76 816	97.5
	8 hours	2072	2.3	88 336	97.7		494	1.3	39 022	98.8		1578	3.1	49 314	96.9
	More than 9 hours	875	4.5	18 749	95.5		279	3.3	8128	96.7		596	5.3	10 621	94.7
Underlying chronic disease^a	Yes	6769	4.0	160 964	96.0	<.001	1270	1.9	67 308	98.2	<.001	5499	5.6	93 656	94.5	<.001
Underlying chronic disease^a	No	4653	1.6	285 478	98.4		1154	0.8	136 087	99.2		3499	2.3	149 371	97.7
Survey year	2011	5802	2.5	223 257	97.5	<.001	1218	1.2	101 740	98.8	.824	4584	3.6	121 517	96.4	.078
Survey year	2012	5620	2.5	223 185	97.5		1206	1.2	101 655	98.8		4414	3.5	121 530	96.5
Sex	Male	2424	1.2	203 395	98.8	.096
Sex	Female	8998	3.6	243 047	96.4
Total		11 422	2.5	446 442	97.5		2424	1.2	203 395	98.8		8998	3.6	243 047	96.4

^aUnderlying chronic disease includes hypertension, diabetes, dyslipidemia, and arthritis.

In the analysis stratified by sex, the prevalence of physician-diagnosed depression among male respondents who resided with a patient with dementia was 2.8%. Meanwhile, the prevalence of physician-diagnosed depression among female respondents who resided with a patient with dementia was 6.8%, which represented a prevalence approximately twice that of female respondents who did not reside with a patient with dementia (3.5%). The average age of physician-diagnosed males with depression was higher than nondepressed males (54.9 vs 54.0 years, respectively). The average age of female respondents who were diagnosed with depression was 56.3 years (nondepressed females 51.7). In both sexes, the percentage of people with physician-diagnosed depression, who reported their employment status as unemployed/homemaker was higher than for males in the other job groups (male: 2.6%; female: 2.3%). The highest prevalence of physician-diagnosed depression was also found among female respondents in the lowest income group (4.5%) and in those who reported living alone (4.8%) as well as those who reported a current smoking habit (8.8%) or slept less than 6 hours per night (6.4%).

Table 3 reports the logistic regression results for the association between physician-diagnosed depression and cohabitation with a patient with dementia for all variables. For both male and female respondents, those who resided with a patient with dementia had increased odds of physician-diagnosed depression compared to those who did not reside with a patient with dementia (odds ratio [OR] 1.33, 95% confidence interval [CI] 1.09-1.61, P value = .004). Based on these results, respondents who reported their employment status as office worker or site worker had significantly lower odds of physician-diagnosed depression than those who reported their employment status as unemployed/homemaker (office worker: OR 0.58, 95% CI 0.54-0.63, P value ≤ .001; site worker: OR 0.67, 95% CI 0.62-0.72, P value < .001). Similarly, respondents who reported their highest level of education as high school or middle school had significantly higher odds of diagnosed depression than those whose highest level of education was university graduate or higher (high school: OR 1.34, 95% CI 1.23-1.48, P value < .001; middle school: OR 1.53, 95% CI 1.36-1.72, P value < .001). Respondents who reported their family income level in the third or fourth quartile, which represented the lowest family income levels, had significantly higher odds of physician-diagnosed depression than those who reported their family income level in the first quartile (third quartile: OR 1.15, 95% CI 1.06-1.25, P value = .001; fourth quartile: OR 1.16, 95% CI 1.05-1.27, P value = .004). Respondents with Medical aid had significantly higher odds of physician-diagnosed depression (OR 1.45, 95% CI 1.31-1.61, P value < .001) than respondents with NHI. Respondents who were past or current smokers (past smoker: OR 1.43, 95% CI 1.28-1.60, P value < .001; current smoker: OR 1.33, 95% CI 1.20-1.49, P value < .001) also had significantly higher odds of physician-diagnosed depression than respondents who were nonsmokers.

Table 3.

Adjusted Odds Ratios and 95% CI for the Existence of Physician-Diagnosed Depression for Each Variable by Sex.

		Both sexes, n = 457 864			Male, n = 205 819			Female, n = 252 045
		OR	95% CI	P Value	OR	95% CI	P Value	OR	95% CI	P Value
Cohabitation with a patient with dementia	Yes	1.33	1.09-1.61	.004	1.26	0.87-1.82	.214	1.33	1.06-1.67	.013
Cohabitation with a patient with dementia	No	1.00			1.00			1.00
Age	Per 1 year older	0.99	0.99-0.99	<.001	1.00	1.00-1.01	.269	0.99	0.98-0.99	<.001
Employment status	Office worker	0.58	0.54-0.63	<.001	0.56	0.46-0.67	<.001	0.63	0.58-0.69	<.001
	Site worker	0.67	0.62-0.72	<.001	0.58	0.50-0.69	<.001	0.8	0.73-0.87	<.001
	Unemployed or homemaker	1.00			1.00			1.00
Highest level of education	Elementary school	0.98	0.86-1.11	.709	0.58	0.47-0.73	<.001	1.17	1.00-1.37	.05
	Middle school	1.53	1.36-1.72	<.001	0.94	0.75-1.17	.568	1.74	1.52-2.00	<.001
	High school	1.34	1.23-1.47	<.001	0.98	0.83-1.15	.796	1.44	1.29-1.60	<.001
	University or higher	1.00			1.00			1.00
Marital status	Single	1.00			1.00			1.00
	Living with spouse	0.96	0.86-1.06	.398	0.47	0.38-0.58	<.001	1.45	1.26-1.67	<.001
	Separated/divorced/widowed	1.13	1.00-1.28	.059	0.78	0.62-0.99	.038	1.6	1.36-1.88	<.001
Family income level	1st quartile (highest)	1.00			1.00			1.00
	2nd quartile	0.99	0.91-1.09	.858	1.09	0.90-1.32	.369	0.96	0.86-1.06	.382
	3rd quartile	1.15	1.06-1.25	.001	1.26	1.05-1.52	.012	1.09	0.99-1.20	.074
	4th quartile (lowest)	1.16	1.05-1.27	.004	1.24	1.01-1.53	.045	1.09	0.98-1.22	.125
Number of family members	1	1.23	1.08-1.39	.001	1.20	0.91-1.57	.201	1.18	1.02-1.35	.022
	2	1.23	1.11-1.36	<.001	1.27	1.01-1.59	.044	1.15	1.03-1.28	.016
	3	1.07	0.96-1.19	.202	1.05	0.84-1.32	.672	1.03	0.92-1.15	.629
	4	0.94	0.85-1.05	.278	0.87	0.69-1.10	.253	0.96	0.85-1.07	.452
	5+	1.00			1.00			1.00
Insurance type	Medical aid	1.45	1.31-1.61	<.001	1.67	1.37-2.03	<.001	1.35	1.20-1.52	<.001
Insurance type	National Health Insurance	1.00			1.00			1.00
Self-reported health condition	Excellent	1.00			1.00			1.00
	Average	1.93	1.77-2.11	<.001	1.92	1.61-2.30	<.001	1.94	1.75-2.14	<.001
	Poor	4.31	3.91-4.76	<.001	4.57	3.74-5.59	<.001	4.19	3.74-4.69	<.001
Self-reported stress	Very often	6.86	6.07-7.75	<.001	7.7	6.01-9.87	<.001	6.63	5.77-7.62	<.001
	Often	3.46	3.12-3.82	<.001	4.02	3.27-4.93	<.001	3.31	2.95-3.71	<.001
	Occasionally	1.30	1.17-1.44	<.001	1.47	1.19-1.81	<.001	1.25	1.11-1.4	<.001
	Seldom	1.00			1.00			1.00
Smoking habit	Never	1.00			1.00			1.00
	Past	1.43	1.28-1.60	<.001	1.21	1.02-1.43	.03	1.48	1.27-1.73	<.001
	Current	1.33	1.20-1.49	<.001	1.00	0.85-1.17	.972	1.92	1.70-2.17	<.001
Alcohol use	No	1.00			1.00			1.00
Alcohol use	Yes	0.94	0.89-1.01	.078	0.75	0.65-0.87	<.001	1.01	0.95-1.09	.700
Hours of sleep per night	Less than 6 hours	1.48	1.37-1.60	<.001	1.30	1.09-1.55	.003	1.54	1.41-1.68	<.001
	6 hours	1.06	0.98-1.14	.182	0.94	0.79-1.11	.468	1.10	1.01-1.20	.037
	7 hours	1.00			1.00			1.00
	8 hours	1.22	1.12-1.33	<.001	1.26	1.05-1.51	.013	1.20	1.09-1.32	<.001
	More than 9 hours	1.93	1.71-2.17	<.001	2.24	1.79-2.81	<.001	1.79	1.56-2.05	<.001
Underlying chronic disease^a	Yes	1.53	1.43-1.65	<.001	1.50	1.30-1.74	<.001	1.56	1.44-1.70	<.001
Underlying chronic disease^a	No	1.00			1.00			1.00
Survey year	2011	1.02	0.96-1.08	.53	0.97	0.86-1.09	.578	1.04	0.97-1.10	.262
Survey year	2012	1.00			1.00			1.00
Sex	Male	0.34	0.31-0.38	<.001
Sex	Female	1.00

Abbreviation: OR, odds ratio; CI, confidence interval.

^aUnderlying chronic disease includes hypertension, diabetes, dyslipidemia, and arthritis.

When the data were stratified by sex, a significant association between physician-diagnosed depression and cohabitation with a patient with dementia remained among females (OR 1.33, 95% CI 1.06-1.67, P value = .013) but not for males (OR 1.26, 95% CI 0.87-1.82, P value = .214) . Based on the results for all adjusted variables, male respondents who reported elementary school as their highest level of education had significantly lower odds for diagnosed depression than males who graduated from university or higher (OR 0.58, 95% CI 0.47-0.73, P value < .001). Female respondents who reported high school or middle school as their highest level of education had significantly higher odds of physician-diagnosed depression than those who graduated from university or higher (high school: OR 1.44, 95% CI 1.29-1.60, P value < .001; middle school: OR 1.74, 95% CI 1.52-2.00, P value < .001). Males who reported their employment status as office worker or site worker had significantly lower odds of physician-diagnosed depression (office worker: OR 0.56, 95% CI 0.46-0.67, P value < .001; site worker: OR 0.58, 95% CI 0.50-0.69, P value < .001) than males who reported their employment status as unemployed/homemaker. Female respondents who reported their employment status as office worker or site worker also had significantly lower odds of physician-diagnosed depression (office worker: OR 0.63, 95% CI 0.58-0.69, P value < .001; site worker: OR 0.80, 95% CI 0.73-0.87, P value < 0.001) than those who reported their employment status as unemployed/homemaker. In addition, males who lived with their married spouse had significantly lower odds of physician-diagnosed depression (OR 0.47, 95% CI 0.38-0.58, P value < 0.001) than males who were single. The odds of physician-diagnosed depression also was higher for female respondents who lived with their married spouse (OR 1.45, 95% CI 1.26-1.67, P value < .001) compared with female respondents who were single.

Table 4 reports the logistic regression results for all variables stratified by family income level. Among respondents who reported the lowest family income level (fourth quartile), males who resided with a patient having dementia had significantly higher odds of physician-diagnosed depression than males who did not reside with a patient with dementia (OR 1.67, 95% CI 1.08-2.59, P value = .022). In contrast, among respondents who reported the highest family income level (1st quartile), females who resided with a patient having dementia had increased odds of diagnosed depression compared with females who did not reside with a patient having dementia (OR 2.37, 95% CI 1.36-4.14, P value = .002).

Table 4.

Adjusted Odds Ratios and 95% CI for the Existence of Physician-Diagnosed Depression for Each Variable by Family Income Level.^a

			1 Q (highest)			2 Q			3 Q			4 Q (lowest)
		OR	95% CI	P Value	OR	95% CI	P Value	OR	95% CI	P Value	OR	95% CI	P Value
Male
Cohabitation with a patient with dementia	Yes	1.60	0.54-4.74	.394	0.84	0.31-2.29	.737	0.69	0.29-1.65	.405	1.67	1.08-2.59	.022
Cohabitation with a patient with dementia	No	1.00			1.00			1.00			1.00
Female
Cohabitation with a patient with dementia	Yes	2.37	1.36-4.14	.002	1.07	0.61-1.86	.826	0.99	0.98-1.00	.253	1.23	0.90-1.68	.196
Cohabitation with a patient with dementia	No	1.00			1.00			1.00			1.00

Abbreviation: OR, odds ratio; CI, confidence interval.

^aAdjusted for age, employment status, education level, marital status, number of family members, insurance type, self-reported health condition, self-reported stress, smoking habit, alcohol use, hours of sleep per night, underlying chronic disease, and survey year.

Discussion

In the current study, we analyzed 457 864 participants, 5360 of whom resided with a family member with dementia, to assess whether family members who provide care for a family member with dementia are at a higher risk of physician-diagnosed depression than those who do not reside with a family member with dementia. The results of this study have revealed that female caregivers who reside with a family member with dementia are at increased risk of physician-diagnosed depression. This increase in risk was not observed in male caregivers who reside with a family member with dementia.

This result is consistent with the results of previous studies. According to a systematic review by Cuijpers,³³ the relative risk of depressive disorder among dementia caregivers is 2.80 to 38.68, compared with the general population. The studies included in this systematic review, however, had small sample sizes and therefore were not representative of the Korean population. The finding in the current study that the prevalence of depression is higher in females than males is consistent with a previous study by Mahoney et al.³⁴ In particular, males who reside with their married spouse have decreased odds (ie, OR < 1.00) of physician-diagnosed depression compared with single males, whereas females who reside with their married spouse have increased odds for physician-diagnosed depression compared with single females.

This would be due to the role difference between male and female in family. According to Brodaty et al¹⁵ the largest proportion of those caregivers was spouses, followed by children and children-in-law, mostly female. By the Pearlin et al³⁵ model of caregiver strain, this role difference would make a result of this study. In Korea, the role of female member on family care is further dominant than male because of Confucian culture. So the difference of caregiving burden might have influenced to this result.

Further differences emerged when the data were stratified by family income level. The results of the current study indicate that financial status differentially impacts the prevalence of depression among dementia caregivers. The risk of physician-diagnosed depression among males who resided with a patient with dementia was highest among those who reported a low family income level. In contrast, however, the risk of physician-diagnosed depression was highest in females who reported a high family income level. These results may be explained by the traditional gender roles in Korea, in which males provide primarily economic support and females provide primarily home-based support. So, males in low-income family would have financial stress for cost of caregiving than who are in high-income family. In the case of female, it would need to further study for interpretation. One of the possible reasons for more depression in high-income family female caregiver comparing with low-income family female is difference of opportunity cost of caregiving between high-income family female and low-income family female. Future studies should investigate the additional factors that impact the risk of physician-diagnosed depression among male versus female dementia caregivers.

Limitations

The current study has several limitations. First, the exact relationship between the patient with dementia and the caregiver is unknown. Most older adults with dementia receive assistance from their spouse, but adult children also serve as the caregiver when the spouse is no longer alive or is unavailable to provide assistance.³⁶ According to a study by Andren et al,³⁷ the perceived health of the caregiver varies with the level of burden on the caregiver. This burden is not equally shared by all family members, and therefore the relationship with the patient with dementia may have an important impact on perceived health.^37,38

A second limitation of this study is that no information about the severity or duration of dementia in the sick family member was available. Participants in this study may have been influenced by the severity of dementia in the patient. Because this study analyzed the association between depression and cohabitation with a family member with dementia rather than factors that affect the caregiver, the results of this study would have a unique interpretation for the Korean population.

A third limitation of this study is that we used survey data rather than medical records to confirm a diagnosis of depression. This self-report of physician-diagnosed depression may introduce recall bias to the data. This survey, however, was conducted by a trustworthy national institution, the Korean Centers for Disease Control and Prevention, using skilled interviewers and a computer-based system. We do not suspect, therefore, that the self-report of physician-diagnosed depression has a significant effect on the results of this study.

Last, this study evaluated the effects only for caregivers who were family member of a patient with dementia. The effect on the other caregiver is also very interesting and need to be evaluated. But in our dataset, we cannot distinguish caregivers who are doing it as a job or else formal caregivers. In further study, it would be necessary to study the impact of dementia caregiving on the formal caregiver.

Conclusion

The results of this study confirm an association between physician-diagnosed depression and cohabitation with a patient with dementia. In particular, high-income females and low-income males who reside with a patient with dementia are at a significantly increased risk of diagnosed depression. To reduce the burden of dementia, we need a management policy that includes the caregiver as well as the patient with dementia. This policy should focus in particular on female caregivers as well as male caregivers who have a low-family income level.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Hurd

Martorell

Langa

. Monetary costs of dementia in the United States. N Engl J Med. 2013;369(5):489–490.

Prince

Bryce

Albanese

Wimo

Ribeiro

Ferri

. The global prevalence of dementia: a systematic review and metaanalysis. Alzheimers Dement. 2013;9(1):63–75, e62.

Alzheimer’s

2013 Alzheimer’s disease facts and figures. Alzheimers Dement. 2013;9(2):208–245.

Sloane

Zimmerman

Suchindran

. The public health impact of Alzheimer’s disease, 2000-2050: potential implication of treatment advances. Annu Rev Public Health. 2002;23:213–231.

McConnell

. Improving dementia care through behavioral interventions. N C Med J. 2014;75(5):355–358.

Schoenmakers

Buntinx

DeLepeleire

. Supporting the dementia family caregiver: the effect of home care intervention on general well-being. Aging Ment Health. 2010;14(1):44–56.

Sansoni

Anderson

Varona

Varela

. Caregivers of Alzheimer’s patients and factors influencing institutionalization of loved ones: some considerations on existing literature. Ann Ig. 2013;25(3):235–246.

Schulz

Belle

Czaja

McGinnis

Stevens

Zhang

. Long-term care placement of dementia patients and caregiver health and well-being. JAMA. 2004;292(8):961–967.

Han

Jeong

Park

. Effects of social supports on burden in caregivers of people with dementia. Int Psychogeriatr. 2014;26(10):1639–1648.

10.

Kim

Park

Storr

Tran

Juon

. Depression among Asian-American adults in the community: systematic review and meta-analysis. PLoS One. 2015;10(6):e0127760.

11.

Blazer

. Depression in late life: review and commentary. J Gerontol A Biol Sci Med Sci. 2003;58(3):249–265.

12.

Pinquart

Duberstein

Lyness

. Treatments for later-life depressive conditions: a meta-analytic comparison of pharmacotherapy and psychotherapy. Am J Psychiatry. 2006;163(9):1493–1501.

13.

Wilkins

Kiosses

Ravdin

. Late-life depression with comorbid cognitive impairment and disability: nonpharmacological interventions. Clin Interv Aging. 2010;5:323–331.

14.

Fried

Nesse

. The impact of individual depressive symptoms on impairment of psychosocial functioning. PLoS One. 2014;9(2):e90311.

15.

Brodaty

Donkin

. Family caregivers of people with dementia. Dialogues Clin Neurosci. 2009;11(2):217–228.

16.

Campbell

Wright

Oyebode

. Determinants of burden in those who care for someone with dementia. Int J Geriatr Psychiatry. 2008;23(10):1078–1085.

17.

Gaugler

Roth

Haley

Mittelman

. Can counseling and support reduce burden and depressive symptoms in caregivers of people with Alzheimer’s disease during the transition to institutionalization? Results from the New York University caregiver intervention study. J Am Geriatr Soc. 2008;56(3):421–428.

18.

Kim

Jeong

Chun

Lee

. The prevalence of dementia in a metropolitan city of South Korea. Int J Geriatr Psychiatry. 2003;18(7):617–622.

19.

Lee

. The prevalence of dementia in older people in an urban population of Korea: the Seoul study. J Am Geriatr Soc. 2002;50(7):1233–1239.

20.

Park

Jung

. Dementia among the elderly in a rural Korean community. Br J Psychiatry. 1994;164(6):796–801.

21.

Suh

Kim

Cho

. Community study of dementia in the older Korean rural population. Aust N Z J Psychiatry. 2003;37(5):606–612.

22.

Suh

Kim

Cho

. Community study of dementia in the older Korean rural population. Aust N Z J Psychiatry. 2003; 37(5):606–612.

23.

Kang

Myung

. Factors associated with caregiver burden in patients with Alzheimer’s disease. Psychiatry Invest. 2014;11(2):152–159.

24.

Lim

Son

Song

Beattie

. Factors affecting burden of family caregivers of community-dwelling ambulatory elders with dementia in Korea. Arch Psychiatr Nurs. 2008;22(4):226–234.

25.

Shin

Youn

Kim

Lee

Cho

. Caregiver burden in Parkinson disease with dementia compared to Alzheimer disease in Korea. J Geriatr Psychiatry Neurol. 2012;25(4):222–226.

26.

Lee

Kim

. Stress in caregivers of demented people in Korea—a modification of Pearlin and colleagues’ stress model. Int J Geriatr Psychiatry. 2006; 21(8):784–791.

27.

Kang

Kim

. Korea community health survey data profiles. Osong Public Health Res Perspect. 2015;6(3):211–217.

28.

Livingston

Manela

Katona

. Depression and other psychiatric morbidity in carers of elderly people living at home. Br Med J. 1996;312(7024):153–156.

29.

Organisation for Economic Co-operation and Development. Growing unequal? Income distribution and poverty in OECD Countries. Paris, France: Organisation for Economic Co-operation and Development; 2008.

30.

Jang

Sull

Park

Kim

Ohrr

. Association between all-cause mortality and insurance status transition among the elderly population in a rural area in Korea: Kangwha cohort study. Health Policy. 2015;119(5):680–687.

31.

Kim

Lee

Yoo

Park

. The differences in health care utilization between medical aid and health insurance: a longitudinal study using propensity score matching. Plos One. 2015;10(3):e0119939.

32.

Park

Cho

Chang

. Relationships of sleep duration with sociodemographic and health-related factors, psychiatric disorders and sleep disturbances in a community sample of Korean adults. J Sleep Res. 2010;19(4):567–577.

33.

Cuijpers

. Depressive disorders in caregivers of dementia patients: a systematic review. Aging Ment Health. 2005;9(4):325–330.

34.

Mahoney

Regan

Katona

Livingston

. Anxiety and depression in family caregivers of people with Alzheimer disease—The LASER-AD study. Am J Geriatr Psychiatry. 2005;13(9):795–801.

35.

Pearlin

Mullan

Semple

Skaff

. Caregiving and the stress process: an overview of concepts and their measures. Gerontologist. 1990;30(5):583–594.

36.

Schulz

Martire

. Family caregiving of persons with dementia—Prevalence, health effects, and support strategies. Am J Geriatr Psychiatry. 2004;12(3):240–249.

37.

Andren

Elmstahl

. The relationship between caregiver burden, caregivers’ perceived health and their sense of coherence in caring for elders with dementia. J Clin Nurs. 2008;17(6):790–799.

38.

Meuser

Marwit

. A comprehensive, stage-sensitive model of grief in dementia caregiving. Gerontologist. 2001;41(5):658–670.