Excess costs of depression among a population-based older adults in Singapore

doi:10.21203/rs.3.rs-3144328/v1

Download PDF

Research Article

Excess costs of depression among a population-based older adults in Singapore

https://doi.org/10.21203/rs.3.rs-3144328/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

In Singapore, there is currently limited evidence on the economic burden of depression among the older population.

Objective

We aim to evaluate the impact of depression on healthcare expenditures and productivity loss among older adults with chronic disease in Singapore.

Methods

Using the data from the Well-being of the Singapore Elderly study (WiSE), a 2011 national representative survey of older adults aged 60 years and above, 2510 respondents were included in this study. The sample comprised 44% male and 56% female respondents with 75% respondents aged between 60–74 years old. Healthcare utilisation data were obtained from respondents and healthcare cost was tabulated by multiplying each service unit (i.e., consultations per minutes, number of visits) by the unit cost price. These services include care from polyclinic doctors, private general practitioners (GP), restructured hospital doctors/healthcare workers and inpatient care. We modelled the relationship of healthcare cost and depression using two-part models (probit & generalised linear model - loglink and gamma distribution).

Results

For total healthcare expenditure, older adults with both depression and chronic disease were associated with an average annual incremental cost of $7940 (95% CI 1490–14400; p = 0.016), compared to those without these conditions. They were also associated with an average incremental cost of ($257, 95% CI: 38.7–475; p = 0.021) for primary care. Likewise, in the case of specialist outpatient clinics, they had an average incremental cost of $970 (95% CI 163–1780; p = 0.018). However, for inpatient setting, the average incremental cost of $6180 (95% CI -418–12800; p = 0.066). was not significant. Additionally, older adults with depression and chronic diseases contribute to an annual productivity loss of $676 (95% CI 346–1010; p < 0.001).

Discussion

This study provides evidence that there are significant incremental costs associated with depression amongst the elderly in Singapore’s primary healthcare setting. The increased somatic presentations among the depressed elderly and the underdiagnosis of depression in primary care may contribute to higher utilisation of healthcare resources which entail higher expenditures. This is one of the first studies to look at the cost of depression using a representative sample of Singapore elders and taking into account the multi-ethnic nature of the population. Analyses were restricted to a cross-sectional design, and do not permit any conclusions on causality and data relied heavily on the accuracy of self-report utilisation on health care services which was subjected to recall bias.

Conclusion

Depression was found to inflate the total healthcare expenditure among older adults with chronic disease by two-fold. This study provides evidence that there are significant incremental costs associated with depression among the older adults with chronic disease, demonstrating a need for more resources to improve systematic and clinical care for depressed older adults with chronic disease.

Depression

healthcare cost

chronic diseases

older adults.

Depression is a common mental disorder characterised by persistent sadness, decreased energy, and loss of interest or pleasure in activities that were previously enjoyable¹. It is a significant public health issue that impacts 350 million people worldwide, predominantly older adults, with studies estimating that the prevalence ranges from 8–20% among the older adults living in the community^2–4. Depression can lead to undesirable consequences such as functional decline and decreased quality of life which further increase the disability among the older population^2,3. In Singapore, the prevalence of depression among the older adults was 3.7% in 2016².

Depression is found to be associated with increased healthcare utilisation among older people^5–8. Among chronically ill patients, depression and psychological distress were among the strongest predictors of hospitalisations and physician visits⁹. Depression is also associated with high direct and indirect costs to society, leading to a significant economic burden to many countries. A systematic review reported that the annual average direct excess costs for a patient with depression ranged from US $1000-$2500, depending on the modes of health care financing and provision¹⁰. Another study suggested that mental disorders were associated with higher healthcare cost and utilisation among patients who have chronic diseases. The mean 3-year adjusted healthcare costs for those with a mental disorder were estimated to be approximately 1.7 times more than patients with chronic diseases without mental disorder¹¹.

Depression incurs substantial indirect costs that are associated with decreased work functioning, absence from work, impaired productivity, and even poor job retention across many occupations¹². According to the World Health Organisation (WHO), it was estimated that depression and anxiety cost the global economy US$ 1 trillion in productivity loss annually¹³.

The magnitude of healthcare expenditures and productivity loss has been examined among people with depression and chronic medical conditions independently in many studies^{5–8,12,14,15}, but few have examined the impact of depression on healthcare cost among the older adults with existing chronic medical conditions. There is also a lack of evidence on the economic burden of depression in Singapore^16,17. Therefore, we aim to evaluate the association of depression on healthcare expenditures and productivity loss among the older adult population with chronic diseases in Singapore.

The WiSE study was a cross-sectional study conducted among Singapore residents who are aged 60 years and above. A random representative sample was selected from a national registry that contains the names and socio-demographic details such as the age, gender, ethnicity, and addresses of all residents in Singapore. Respondents were approached at their residential address provided by the registry, and interviews were conducted using an online Computer Assisted Personal Interviewing (CAPI) application. The survey data was weighted to the 2011 resident population. For confidentiality, all personal identifiers were removed before analysis. Written informed consent was obtained from respondents. When respondents were unable to answer the questions or unsure of their answers, information may be collected from their nominated informant as proxy. Methodologies were described in detail elsewhere². The study was approved by the National University of Singapore Institutional Review Board (IRB), National Healthcare Group Domain Specific Review Board and Singhealth Centralised IRB.

Assessment of depression

The Geriatric Mental State (GMS) examination was used in the assessment of depression and comprised a semi-structured interview with a rating section that assess psychopathology and cognition.The Automated Geriatric Examination for Computer Assisted Taxonomy (AGECAT) was used to diagnose depression. The validity of AGECAT has been established with an agreement of 0.88 (Kappa score) with the psychiatrist’s diagnoses of depression. Details were explained in previous studies^2,17.

Data collection

Socio-demographics data such as age, gender, ethnicity, marital status and education were collected. Data on common medical conditions including history of heart trouble (myocardial infarction, cardiac failure, and valvular heart disease), stroke and diabetes were also collected. Heart trouble, stroke, and diabetes were defined as a dichotomous variable for answering ‘yes’ to the question of whether a doctor had ever told them that they had any of the conditions.

Healthcare utilisation data were obtained from respondents and their informants using an adapted version of the Client Service Receipt Inventory (CSRI)^2,17. The CSRI was shown to be a well-validated scale that has been widely adopted in 10/66 population-based studies among the older adult population¹⁸. Respondents were asked if they used specific services during the three months before the interview. Details on the number of visits, average time spent on visits, time spent on travelling and out-of-pocket expenses were collected. These services include care from polyclinic doctors, private general practitioners (GP), restructured hospital doctors/healthcare workers, and inpatient care.

Cost calculation

Healthcare cost was calculated by multiplying each service unit (i.e., consultations per minute, number of visits) by the unit cost price. The 3-month healthcare expenditure for each service was multiplied by four to obtain annualised cost. Due to sparse local data, extrapolation of United Kingdom (UK) unit cost was used to estimate the unit cost of selected health services in Singapore (primary care doctor, restructured hospital doctor, and healthcare workers). Ratios were generated using data from the World Health Organisation Choosing Interventions that are Cost-Effective (WHO-CHOICE) database^17,19. The ratios were then applied to with the UK’s unit cost of each selected services to generate Singapore unit costs. Reliable sources such as the Unit Cost of Health and Social Care 2013^17,20 and NHS Reference Costs^17,21 were used. Unit cost per bed day was calculated for inpatient care. These calculations were based on the assumption that the unit costs for health services were fixed and had remained unchanged between both countries. Figures were converted to local currency. Average out-of-pocket expenses were used if UK unit cost data were not available. This method was adopted by previous population-based cost evaluation studies using the WiSE study^17,18. Appendix Table 1 presents the unit costs for health services in Singapore. Cost will be reported in Singapore Dollars in this paper.

Table 1

Distribution of socio-demographic characteristics by chronic disease and depression, n = 2510
Characteristic	No depression and no chronic diseases (n = 1,364)		No depression and has chronic diseases (n = 975)		Depression and no chronic diseases (n = 74)		Depression and chronic diseases (n = 97)
	n	Weighted %	n	Weighted %	n	Weighted %	n	Weighted %	P
Age group
60–74	846	78.9	530	70.0	42	60.3	53	62.1	< 0.0001
75–84	309	16.1	295	24.0	23	34.2	28	31.3
85+	209	4.98	150	6.08	9	5.49	16	6.62
Gender
Male	573	41.1	468	50.4	18	29.0	31	37.4	0.0027
Female	791	58.9	507	49.6	56	71.0	66	62.6
Ethnicity
Chinese	623	86.4	332	79.5	17	67.8	14	55.2	< 0.0001
Malay	406	8.38	273	10.3	28	20.1	29	18.8
Indian	318	4.05	353	8.28	29	12.1	52	21.5
Others	17	1.15	17	1.99	0	0	2	4.54
Marital status
Single	87	9.12	35	5.87	3	2.32	4	6.34	0.309
Married	808	64.1	573	64.5	34	65.0	47	64.2
Divorced/Separated/Widowed	469	26.8	367	29.6	37	32.7	46	29.4
Education
None	255	15.0	193	18.2	24	21.5	27	32.7	0.0968
Up to primary	662	47.5	502	51.1	33	50.8	44	40.4
Completed Secondary	296	24.6	187	19.5	9	13.0	18	15.6
Completed tertiary	151	13.0	93	11.2	8	14.6	8	11.2
Employment
Paid Work	451	39.3	208	25.9	13	19.0	14	23.5	< 0.0001
Unemployed	18	1.75	10.0	1.20	2	1.53	2	1.07
Homemaker	418	25.7	318	26.7	31	40.1	39	33.4
Retired	477	33.3	439	46.2	28	39.4	42	42.1
Weighted row percentages included.

Productivity loss was calculated using the human capital approach²² by multiplying the total number of visits (absent from work) and time spent on visits by the hourly income. The median national gross income ²³ and average national working hours²⁴ for 2013 published by the Ministry of Manpower (MOM) were used to estimate productivity loss. The median income in 2013 was $3700 (including Employer CPF contributions)²³, and the average national working hours for a week was 45.3 hours²⁴. We acknowledge that majority of the older population may not be in the workforce, labour participation rate was taken into account in the calculation of productivity loss among older adults. The labour participation rate was reported to be 67.1% for those aged between 55–64 and 23.8% for 65 years and above as published by MOM Research and Statistics Department’s “Singapore Workforce 2013” report²⁵. Primary care expenditure was derived from polyclinic doctors and GP services. Specialists’ Outpatient Clinics (SOC) expenditure was derived from visits to specialists in the restructured hospital. Total health care expenditure was derived from the sum of primary health care, SOC and inpatient costs.

Statistical analysis

All estimates were analysed using survey weights to adjust for oversampling, non-response and post-stratification according to age and ethnicity of the Singapore older adult population to ensure a better representation of the population. The dependent variables were healthcare expenditures (total healthcare, primary care, SOC & inpatient) and productivity loss. The primary independent variable of interest was depression with chronic disease. Categorical variables were expressed in frequencies and percentages, by depression status. Continuous variables were expressed in mean ± SD. Pearson’s Chi-Square test was used to compare categorical variables. Age was categorised into three groups: 60–74, 74–85 and 85 years and above with 60–74 years old being the reference group. Ethnicity was categorised into four groups: Chinese (reference group), Malay, Indian and Others. Marital status was categorised into three groups: married (reference group), single and divorced/separated/widowed. Education level was categorised into four groups: no education (reference group), up to primary level, completed secondary level and completed tertiary level. Income was categorised into four groups: paid work, unemployed, homemaker and retired. Gender and medical condition were dichotomised. Participants who reported yes to any of the following conditions – heart trouble, stroke and diabetes were considered to have chronic disease in the analysis. Incomplete data on demographics and medical conditions were not included in the analysis.

We used two-part models to estimate annual health care expenditures and productivity loss associated with depression and chronic medical conditions, controlling for individual characteristics as well as accounting for sampling design. The two-part models are widely used in health economics and health services research when the outcome of interest has a large number of zero outcomes and a group of nonzero outcomes that are discrete or highly skewed^26,27. Before deciding on the two-part models, histograms were plotted to show the sample distribution of cost data. The histogram was presented in Appendix Fig. 1.

Given that the distribution of costs in the dataset was skewed with many zeros, and both assumptions of normality and homoscedasticity of residuals were violated, we applied a probit model for the first part of the model to predict the probability of incurring any healthcare expenditures. For the second part, we modelled the positive cost using a generalised linear model (GLM) with the log link and gamma distribution. Individual components of the total healthcare expenditures were modelled to look at the cost of depression in the older adult population with chronic disease in different sectors of the healthcare system in Singapore: primary care, specialist outpatient clinics, inpatient. Socio-demographic factors were adjusted in the multivariable regression.

Statistical analysis was performed with STATA ver16 (STATA Corp, College Station, Tx, USA). The ‘twopm’ command was used to execute the two-part models in Stata ²⁸. Two-sided p-values less than 0.05 were considered statistically significant.

Two thousand five hundred ten respondents were included in this study. The sample comprised 44% male and 56% female respondents, with 75% respondents aged between 60–74 years old. The prevalence of depression was 3.7%, and the prevalence of chronic diseases was 35.5%. Table 1 shows the descriptive statistics of the sample stratified into four groups – no depression and no chronic diseases (reference group); no depression and has chronic diseases; has depression and no chronic diseases; has both depression and chronic diseases. Age group, gender, ethnicity and employment status were shown to be statistically different between the four groups. For age, a higher proportion of respondents aged 60–74 was in the reference group (p < 0.0001). A higher proportion of respondents aged 75–84 were shown more likely to be depressed. There were a higher proportion of females in both the depression group with and without chronic diseases (p = 0.0027). Malays and Indians were more likely to be depressed and have chronic diseases (p < 0.0001). A higher proportion of the respondents who were employed with paid work were in the reference group (p < 0.0001). Marital status and education were not shown to be statistically different between the four groups. The weighted percentages for each variable are presented in Table 1

Total healthcare care expenditure

Table 2 presents the factors associated with the total annual healthcare expenditure. After adjusting for socio-demographic factors, the average incremental cost of chronic disease alone was $3780 (95% CI 1820–5370; p < 0.001), followed by depression only ($1480,95% CI -853–3810; p = 0.214), though not statistically significant. Having both chronic diseases and depression was associated with an average annual incremental cost of $7940 (95% CI 1490–14400; p = 0.016). In the multivariable regression model, those of Indian ethnicity ($1330, 95% CI 59.3–2590; p = 0.040) and those above 85 years of age ($5620, 95% CI 1060–10200; p = 0.016) were associated with higher healthcare expenditures compared to the reference group.

Table 2

Incremental effects of total annual healthcare expenditure by chronic disease and depression using two-part regression model
Adjustments Variables	Unadjusted		Demographic		Social Economic Status
Adjustments Variables	β	SE	β	SE	β	SE
No depression & no chronic diseases No depression and has chronic diseases Depression and no chronic diseases Depression and chronic diseases	Ref 4120* 1170 8170	Ref (1020) (743) (2810)	Ref 3860*** 1490 8540*	Ref (1020) (1110) (3660)	Ref 3780*** 1480 7940*	Ref (997) (1190) (3290)
Gender Male Female			Ref -1000	Ref (671)	Ref -825	Ref (636)
Age Group 60–74 75–84 85+			Ref 1010 6580*	Ref (608) (2600)	Ref 813 5620*	Ref (667) (2320)
Ethnicity Chinese Malay Indian Others			Ref 261 1520* 857	Ref (644) (706) (1530)	Ref 307 1330* 511	Ref (675) (646) (1280)
Marital Status Married Not Married Divorced/Separated			Ref -699 -568	Ref (867) (472)	Ref -1120 -697	Ref (604) (472)
Education None Up to Primary level Secondary level Tertiary level					Ref -564 823 43.2	Ref (578) (1090) (846)
Employment Paid Work Unemployed Homemaker Retired					Ref 3310 904 1530*	Ref (3310) (596) (686)

Primary care expenditures

Table 3 presents the predictors of annual primary care expenditures based on the two-part models. After adjusting for socio-demographics, chronic diseases were associated with an average incremental cost of $107 (95% CI: 34.3–179, p = 0.004), followed by depression only ($188, 95% CI: -52.1– 429; p = 0.125), and having both depression and chronic diseases ($257, 95% CI: 38.7–475; p = 0.021). Also, Indian respondents were shown to be associated with higher primary care expenditures ($109, 95% CI: 20.6–197; p = 0.016) in the multivariable regression model.

Table 3

Incremental effects of annual primary care expenditure by chronic disease and depression using two-part regression model
Adjustments Variables	Unadjusted		Demographic		Social Economic Status
Adjustments Variables	β	SE	β	SE	β	SE
No depression & no chronic diseases No depression and has chronic diseases Depression and no chronic diseases Depression and chronic diseases	Ref 107* 144 335**	Ref (42.8) (105) (125)	Ref 101** 160 275*	Ref (37.5) (117) (116)	Ref 107** 188 257*	Ref (37.0) (123) (111)
Gender Male Female			Ref 51.0	Ref (40.5)	Ref 14.8	Ref (38.9)
Age Group 60–74 75–84 85+			Ref -31.9 27.8	Ref (36.8) (64.4)	Ref -41.1 15.8	Ref (41.0) (67.0)
Ethnicity Chinese Malay Indian Others			Ref 45.5 99.0* 294	Ref (65.0) (44.8) (282)	Ref 35.1 109* 371	Ref (55.8) (44.9) (326)
Marital Status Married Not Married Divorced/Separated			Ref -41.6 -23.0	Ref (84.6) (50.6)	Ref -36.3 -32.6	Ref (82.0) (46.8)
Education None Up to Primary level Secondary level Tertiary level					Ref -99.7 -121 -132	Ref (93.5) (99.2) (104)
Employment Paid Work Unemployed Homemaker Retired					Ref 228 27.7 -1.55	Ref (145) (61.2) (43.2)

Table 4

Incremental effects of tertiary expenditure by chronic disease and depression using two-part regression model
Adjustments Variables	Unadjusted		Demographic		Social Economic Status
Adjustments Variables	β	SE	β	SE	β	SE
A. Inpatient expenditure
No depression & no chronic diseases No depression and has chronic diseases Depression and no chronic diseases Depression and chronic diseases	Ref 3360*** 175 6850*	Ref (984) (420) (2820)	Ref 3240** 14.7 6050*	Ref (996) (354) (3010)	Ref 3240*** 30.3 6180	Ref (955) (380) (3370)
Gender Male Female			Ref -988	Ref (703)	Ref -1520	Ref (953)
Age Group 60–74 75–84 85+			Ref 1000 7870*	Ref (601) (3230)	Ref 558 7040*	Ref (728) (3310)
Ethnicity Chinese Malay Indian Others			Ref 291 2030* 1010	Ref (703) (955) (1910)	Ref 398 2280* 740	Ref (841) (1150) (1600)
Marital Status Married Not Married Divorced/Separated			Ref -2090*** -1090	Ref (562) (621)	Ref -2220*** -1270	Ref (636) (702)
Education None Up to Primary level Secondary level Tertiary level					Ref -729 846 -554	Ref (812) (1270) (1260)
Employment Paid Work Unemployed Homemaker Retired					Ref - 1640 1580*	Ref - (1030) (791)
B. Outpatient expenditure
No depression & no chronic diseases No depression and has chronic diseases Depression and no chronic diseases Depression and chronic diseases	Ref 651*** 849 986*	Ref (141) (601) (412)	Ref 612*** 917 954*	Ref (140) (664) (400)	Ref 597*** 1030 970*	Ref (123) (754) (411)
Gender Male Female			Ref -69.9	Ref (120)	Ref 10.8	Ref (145)
Age Group 60–74 75–84 85+			Ref 187 240	Ref (159) (253)	Ref 170 201	Ref (151) (266)
Ethnicity Chinese Malay Indian Others			Ref -209* 108 399	Ref (99.3) (120) (366)	Ref -166 123 262	Ref (106) (119) (339)
Marital Status Married Not Married Divorced/Separated			Ref 127 -18.6	Ref (300) (129)	Ref -30.6 7.87	Ref (224) (129)
Education None Up to Primary level Secondary level Tertiary level					Ref -67.7 101 224	Ref (191) (223) (244)
Employment Paid Work Unemployed Homemaker Retired					Ref 1080 -31.7 217	Ref (944) (166) (146)

Tertiary (inpatient and outpatient) expenditures

The average increment cost of chronic diseases for annual inpatient expenditures was $3240 (95% CI 1370–5110, p = 0.001) after adjusting socio-demographics and comorbidities (Table 4A). Respondents who were 85 years old and above were associated with higher inpatient expenditure ($7040, 95% CI 548–13500, p = 0.034) compared to the 60–74 years old. Indians were also associated with higher inpatient expenditure ($2280, 95% CI 16.3–4540, p = 0.048). Respondents who were single had lower inpatient expenditure compared to those who were married. Having depression and chronic diseases contributed to an average incremental cost of $6180 (95% CI -418–12800; p = 0.066). The average incremental cost of chronic diseases on annual SOC expenditures was $597 (95% CI 355–839, p < 0.001) after adjusting for socio-demographics (Table 4B). The average increment cost of depression alone was $1030 (95% CI -445–2510; p = 0.171). Having both depression and chronic diseases had an average incremental cost of $970 (95% CI 163–1780; p = 0.018) in specialist outpatient clinics.

Productivity loss

Table 5 presents the predictors of annual productivity loss. Having chronic disease alone was associated with an average incremental annual productivity loss of $394 (95% CI 216–571, p < 0.001) after adjusting for socio-demographics and comorbidities. In the multivariable regression model, depression alone contributes tfo an annual productivity loss of $343 (95% CI 9.91–676, p = 0.044). Having both depression and chronic diseases contributed to an annual productivity loss of $676 (95% CI 346–1010; p < 0.001).

Table 5

Incremental effects of productivity loss by chronic disease and depression using two-part regression model
Adjustments Variables	Unadjusted		Demographic		Social Economic Status
Adjustments Variables	β	SE	β	SE	β	SE
No depression & no chronic diseases No depression and has chronic diseases Depression and no chronic diseases Depression and chronic diseases	Ref 346* 278 699*	Ref (104) (144) (210)	Ref 376*** 323 * 663***	Ref (96.0) (163) (177)	Ref 394*** 343* 676***	Ref (90.6) (170) (169)
Gender Male Female			Ref 71.5	Ref (74.2)	Ref 131	Ref (92.5)
Age Group 60–74 75–84 85+			Ref -197** -75.8	Ref (65.1) (91.2)	Ref -132* 7.62	Ref (63.7) (104)
Ethnicity Chinese Malay Indian Others			Ref -50.5 164 332	Ref (71.2) (86.7) (300)	Ref -40.0 158 219	Ref (69.1) (86.1) (253)
Marital Status Married Not Married Divorced/Separated			Ref -175 -182*	Ref (119) (71.2)	Ref -232* -160*	Ref (96.4) (70.0)
Education None Up to Primary level Secondary level Tertiary level					Ref 28.8 215 156	Ref (76.6) (129) (109)
Employment Paid Work Unemployed Homemaker Retired					Ref 462 -97.9 -97.7	Ref (374) (124) (98.0)

Our study shows that depression significantly impacts healthcare expenditure among the older adults with chronic diseases in Singapore. Having depression doubles the healthcare cost among older adults with chronic diseases. The prevalence of older adults aged 65 years and above was 13.7% in Singapore in 2018. This is a 5% increase over the last decade, and the numbers are expected to double by 2030²⁹. Additionally, patients with chronic conditions are more likely to develop psychiatric disorders such as depression or anxiety³⁰. We found that having depression was associated with higher healthcare expenditure among older adults. This finding was consistent with current literature that chronic diseases lead to increased medical costs around the world^31,32. In our study, older adults with both chronic diseases and depression were associated with an average incremental cost of $7940 compared to those without these conditions (Appendix Fig. 2). Although having depression alone increases healthcare costs by $1480, having depression and chronic diseases double the total healthcare cost ($7940 vs. $3780). This finding is consistent with a study done in Canada in 2019 that reported the 3-year adjusted healthcare costs for patients with chronic diseases was 1.7 times higher when the patient had a mental disorder¹¹. For primary care (Polyclinic and GP), the additional effect of depression on healthcare cost was 2.4 times of older adults with chronic diseases ($257 vs $107). While having depression had 1.7 times the outpatient cost compared to those with chronic diseases only ($1030 vs. $597), this was not statistically significant. Our findings are consistent with previous studies in the United States, where depression had no significant effect on the frequency of hospital admission or length of stay^5,33.

There are several reasons to explain the impact of depression on healthcare expenditures among older adults with chronic diseases in the primary care setting. Studies have shown that depressed older patients had a significantly higher frequency of “non-specific” medical complaints such as fatigue, dizziness, headache and pain compared to the non-depressed older patients^5,34,35. These complaints are three to four times more than the non-depressed group⁵. Depressed patients in the general health setting also tend to have more somatic presentations than depressed patients in the specialised mental health sector ³⁴. Furthermore, the diagnosis of depression was infrequent among primary care setting, with studies suggesting that half of all depressed patients go undiagnosed in primary care ^5,36,37. Depressed patients are also more likely to continue seeking medical attention. The increased somatic presentations among the depressed older adults and the underdiagnosis of depression in primary care may contribute to higher utilisation of healthcare resources which entail higher expenditures.

Depression was associated with an average incremental annual productivity loss of $676 among older adults with chronic diseases. Productivity loss was estimated to be 1.7 times higher ($676 vs. $394) with depression in older adults with chronic diseases. This finding is consistent with current literature that shows positive productivity loss associated with depression. ^12,16,38 However, compared to a local study that reported a productivity loss of USD $3063 (SD = 11,763) due to medical leave related to depressive disorder¹⁶; the findings are likely to be underestimated. Firstly, our study did not account for productivity losses from presenteeism (reduced productivity at paid work) – this is a huge component of productivity loss. Secondly, absenteeism was calculated based on respondents seeking healthcare and did not account for those who were absent but did not seek healthcare nor would we estimate the productivity loss from presentism. It is recommended in the literature to use a measurement instrument including questions about both paid and unpaid productivity³⁹.

Strengths and limitations

This study is the first to look at additional costs of depression among older adults using a representative Singapore epidemiological study adjusting for the multi-ethnic nature of the population. Validated instruments were used in the diagnosis of depression (AGECAT) and the measurement of healthcare utilisation (CSRI). The advantage of the bottom-up approach to estimate the cost of depression in this study provided greater details about how cost differs in different health sectors in Singapore. Future studies could look at quantifying the cost of depression from the top down using individual cost data from Mediclaim through the Ministry of Health. This method would mitigate recall bias but would be limited to services and medications from Medisave claims. Medisave is a national savings scheme that helps individuals set aside some part of their income to pay for their medical expenses.

One of the study’s limitations was that the cost data relied heavily on the accuracy of self-reported utilisation of health care services, which is subject to recall bias. Furthermore, the study’s cross-sectional design did not allow us to establish the temporal relation of depression. Hence, no conclusions on causal interference of the variables considered can be drawn. The 66% non-response rate with little medical information on the non-respondents further limits the generalisability of results to the population. Extrapolating from a 3-month healthcare utilisation period can also lead to imprecise estimation of the annual healthcare expenditures. Also, several assumptions were made while calculating the unit costs of healthcare due to the lack of information available in Singapore. However, this practice has been widely accepted and adopted in the estimation of economic cost^17,18,40. The annual healthcare expenditure was likely to be an underestimation as this study did not include utilisation from private specialists, healthcare workers, and pharmaceuticals. Previous studies^29,41 have shown that medication made up one of the largest components in the higher cost of depression.

As Singapore’s population ages, the incremental cost associated with depression among the older adults with chronic diseases will begin to impose increasing pressure on the healthcare system, inflating the cost of health care. Thus, there is a pressing need to prioritise the capabilities of the health care system to manage depression among older adults in Singapore. The findings of this study also provided insights that could impact the clinical practices of primary care physicians. As the first point of contact in our healthcare system, these physicians play a crucial role in diagnosing and managing depression in the older adult population. Therefore, it is recommended that physicians actively look out for non-specific complaints associated with depression, especially among older adults with multiple comorbidities so appropriate treatment can be instituted promptly, abating the rising costs.

Ethics approval and consent to participate

The study was approved by the National University of Singapore Institutional Review Board (IRB), National Healthcare Group Domain Specific Review Board and Singhealth Centralised IRB. Written informed consent was obtained from respondents in the WiSE study. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable

Availability of data and material

Available from corresponding author on reasonable request.

Competing interest

The authors have no relevant financial or non-financial interests to disclose.

Funding

The study was funded by the Ministry of Health, Singapore and the Singapore Millennium Foundation of the Temasek Trust.

Authors’ contributions

JWN wrote the paper, JWN and CC conceived and designed the study, NJW, CC and XYG performed the statistical analysis. MS, SAC, JA and EA collected the data. The authors read and approved the final manuscript.

Acknowledgements

Not applicable

WHO. Depression and Other Common Mental Disorders. World Health Organization- Global Health Estimates.; 2017.
Subramaniam M, Abdin E, Sambasivam R, et al. Prevalence of Depression among Older Adults-Results from the Well-being of the Singapore Elderly Study. Ann Acad Med Singapore. 2016;45(4):123–33.
Sayers J. The world health report 2001 — Mental health: new understanding, new hope. Bull World Health Organ. 2001;79(11):1085–5.
Chiu M, Lebenbaum M, Cheng J, de Oliveira C, Kurdyak P. The direct healthcare costs associated with psychological distress and major depression: A population-based cohort study in Ontario, Canada. PLoS ONE. 2017;12(9):e0184268–8.
Luber MP, Meyers BS, Williams-Russo PG, et al. Depression and Service Utilization in Elderly Primary Care Patients. Am J Geriatric Psychiatry. 2001;9(2):169–76.
Kong D, Li M, Wang J, Davitt JK, Dong X. The Relationship Between Depressive Symptoms and Health Services Utilization in U.S. Chinese Older Adults. Gerontologist. 2018;59(3):447–55.
Kim H, Park S-M, Jang S-N, Kwon S. Depressive symptoms, chronic medical illness, and health care utilization: findings from the Korean Longitudinal Study of Ageing (KLoSA). Int Psychogeriatr. 2011;23(8):1285–93.
Feng L, Yap KB, Kua EH, Ng TP. Depressive symptoms, physician visits and hospitalization among community-dwelling older adults. Int Psychogeriatr. 2009;21(3):568–75.
de Boer AG, Wijker W, de Haes HC. Predictors of health care utilization in the chronically ill: a review of the literature. Health Policy. 1997;42(2):101–15.
Luppa M, Heinrich S, Angermeyer MC, König HH, Riedel-Heller SG. Cost-of-illness studies of depression: a systematic review. J Affect Disord. 2007;98(1–2):29–43.
Sporinova B, Manns B, Tonelli M, et al. Association of Mental Health Disorders With Health Care Utilization and Costs Among Adults With Chronic Disease. JAMA Netw Open. 2019;2(8):e199910.
Beck A, Crain AL, Solberg LI, et al. Severity of Depression and Magnitude of Productivity Loss. The Annals of Family Medicine. 2011;9(4):305.
WHO. Mental health in the workplace. World Health Organisation. Information sheet Web site. https://www.who.int/mental_health/in_the_workplace/en/. Published 2019. Accessed 20/5/2020, 2020.
Fouad AM, Waheed A, Gamal A, Amer SA, Abdellah RF, Shebl FM. Effect of Chronic Diseases on Work Productivity: A Propensity Score Analysis. J Occup Environ Med. 2017;59(5):480–5.
Rost KM, Meng H, Xu S. Work productivity loss from depression: evidence from an employer survey. BMC Health Serv Res. 2014;14(1):597.
Ho RCM, Mak K-K, Chua ANC, Ho CSH, Mak A. The effect of severity of depressive disorder on economic burden in a university hospital in Singapore. Expert Rev PharmacoEcon Outcomes Res. 2013;13(4):549–59.
Subramaniam M, Abdin E, Vaingankar JA, et al. Comorbid Diabetes and Depression among Older Adults - Prevalence, Correlates, Disability and Healthcare Utilisation. Ann Acad Med Singapore. 2017;46(3):91–101.
Abdin E, Subramaniam M, Achilla E, et al. The Societal Cost of Dementia in Singapore: Results from the WiSE Study. J Alzheimers Dis. 2016;51(2):439–49.
World Health Organisation. Cost effectiveness and strategic planning (WHO-CHOICE). https://www.who.int/choice/cost-effectiveness/en/. Accessed 27 May 2020.
Curtis L. Unit Costs of Health and Social Care. Personal Social Services Research Unit, University of Kent; 2013.
NHS Reference. Costs 2012–2013. London: Department of Health; 2013.
Filipovic I, Walker D, Forster F, Curry AS. Quantifying the economic burden of productivity loss in rheumatoid arthritis. Rheumatology. 2011;50(6):1083–90.
Gross Monthly Income From Work. Labour Force in Singapore. Ministry of Manpower; Manpower Research & Statistics Department; 2020.
Hours Worked. Labour Market Report. Ministry of Manpower; Manpower Research & Statistics Department; 2020.
Singapore Workforce. 2013 Report. Labour Force in Singapore. Ministry of Manpower; Manpower Research & Statistics Department; 2013.
Twomey CD, Baldwin DS, Hopfe M, Cieza A. A systematic review of the predictors of health service utilisation by adults withz mental disorders in the UK. BMJ Open. 2015;5(7):e007575.
Diehr P, Yanez D, Ash A, Hornbrook M, Lin DY. Methods for analyzing health care utilization and costs. Annu Rev Public Health. 1999;20:125–44.
Belotti F, Deb P, Manning WG, Norton EC. twopm: Two-part models. Stata J. 2015;15(1):3–20.
Bock JO, Brettschneider C, Weyerer S, et al. Excess health care costs of late-life depression - Results of the AgeMooDe study. J Affect Disord. 2016;199:139–47.
Turner J, Kelly B. Emotional dimensions of chronic disease. West J Med. 2000;172(2):124–8.
Trogdon JG, Murphy LB, Khavjou OA, et al. Costs of Chronic Diseases at the State Level: The Chronic Disease Cost Calculator. Prev Chronic Dis. 2015;12:E140–0.
Abegunde DO, Mathers CD, Adam T, Ortegon M, Strong K. The burden and costs of chronic diseases in low-income and middle-income countries. The Lancet. 2007;370(9603):1929–38.
Luber MP, Hollenberg JP, Williams-Russo P, et al. Diagnosis, treatment, comorbidity, and resource utilization of depressed patients in a general medical practice. Int J Psychiatry Med. 2000;30(1):1–13.
Von Korff M, Ormel J, Katon W, Lin EH. Disability and depression among high utilizers of health care. A longitudinal analysis. Arch Gen Psychiatry. 1992;49(2):91–100.
Katon W, Von Korff M, Lin E, et al. Distressed high utilizers of medical care. DSM-III-R diagnoses and treatment needs. Gen Hosp Psychiatry. 1990;12(6):355–62.
Simon G, Ormel J, VonKorff M, Barlow W. Health care costs associated with depressive and anxiety disorders in primary care. Am J Psychiatry. 1995;152(3):352–7.
Katzelnick DJ, Kobak KA, Greist JH, Jefferson JW, Henk HJ. Effect of primary care treatment of depression on service use by patients with high medical expenditures. Psychiatr Serv. 1997;48(1):59–64.
Stewart WF, Ricci JA, Chee E, Hahn SR, Morganstein D. Cost of lost productive work time among US workers with depression. JAMA. 2003;289(23):3135–44.
Krol M, Brouwer W. How to Estimate Productivity Costs in Economic Evaluations. PharmacoEconomics. 2014;32(4):335–44.
Picco L, Achilla E, Abdin E, et al. Economic burden of multimorbidity among older adults: impact on healthcare and societal costs. BMC Health Serv Res. 2016;16:173.
Welch CA, Czerwinski D, Ghimire B, Bertsimas D. Depression and Costs of Health Care. Psychosomatics. 2009;50(4):392–401.

No competing interests reported.

AppendixBMC.docx

Download PDF

Editorial decision: Revision requested
02 Jul, 2024
Reviews received at journal
04 Jun, 2024
Reviewers agreed at journal
06 May, 2024
Reviews received at journal
23 Nov, 2023
Reviewers agreed at journal
09 Nov, 2023
Reviewers invited by journal
05 Aug, 2023
Editor assigned by journal
31 Jul, 2023
Editor invited by journal
23 Jul, 2023
Submission checks completed at journal
23 Jul, 2023
First submitted to journal
06 Jul, 2023

You are reading this latest preprint version

Excess costs of depression among a population-based older adults in Singapore

Status:

Version 1

Abstract

Background

Objective

Methods

Results

Discussion

Conclusion

Introduction

Methods

Assessment of depression

Data collection

Cost calculation

Statistical analysis

Results

Total healthcare care expenditure

Primary care expenditures

Tertiary (inpatient and outpatient) expenditures

Productivity loss

Discussion

Strengths and limitations

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1