Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 1.
Published in final edited form as: Depress Anxiety. 2020 Jun 2;37(9):908–915. doi: 10.1002/da.23059

Health Care Use and Costs in Adult Cancer Patients with Anxiety and Depression

Brent T Mausbach 1,2, Gabrielle Decastro 3, Richard B Schwab 2,4, Maria Tiamson-Kassab 1,2, Scott A Irwin 5
PMCID: PMC7484454  NIHMSID: NIHMS1614237  PMID: 32485033

Abstract

Objective:

Anxiety and depression are common in individuals with cancer and may impact healthcare service use and costs in this population. This study examined the effects of anxiety alone, depression alone, and comorbid anxiety and depressive disorder on health care use and costs among patients with cancer.

Method:

This was a retrospective cohort analysis of administrative data of patients aged 18 or older with an ICD-9 diagnosis of cancer. Key outcomes were any visit to emergency department (ED), any inpatient hospitalization, length of hospital stays, and annual healthcare costs one year from cancer diagnosis.

Results:

A total of 13,426 patients were included. Relative to patients with neither anxiety nor depression, those with anxiety alone, depression alone, or comorbid anxiety and depression were more likely to experience an ED visit and be hospitalized. Length of hospital stays were also longer and annual healthcare costs were significantly higher in all three clinical groups.

Conclusions:

Cancer patients with anxiety and depression were at greater risk for ED visits and hospitalizations, experienced longer hospital stays, and accrued higher healthcare costs. Future researchers should determine whether screening and treating comorbid anxiety and depression may decrease healthcare utilization and improve turnover well-being among cancer patients.

Keywords: Costs and cost analyses, healthcare administrative claims, mental health, Psycho-oncology, oncology

1. INTRODUCTION

Depression and anxiety are common psychiatric conditions affecting patients diagnosed with cancer, with meta-analyses suggesting a prevalence of 8–16% for depressive disorders (Mitchell, Ferguson, Gill, Paul, & Symonds, 2013), which is notable relative to the 5% 1-year prevalence of depression in the general population (Regier et al., 1993). With regard to anxiety, 19% of patients with cancer show clinical levels of anxiety and another 22.6% have subclinical symptoms (Linden, Vodermaier, Mackenzie, & Greig, 2012). Depression and anxiety are some of the most pressing public health issues in the United States (Hasin, Goodwin, Stinson, & Grant, 2005), with depression being the third leading cause of disease burden in the world (World Health Organization, 2006). Thus, while cancer presents many physical challenges, the high prevalence of depression and anxiety highlight that patients with cancer experience burdens beyond those brought about by cancer itself.

While the direct costs of treating depression and anxiety are substantial (Marciniak et al., 2005; Thomas & Morris, 2003), they may also be associated with indirect costs via increased healthcare utilization. In older Health Maintenance Organization (HMO) patients, depressed patients have 38% more outpatient healthcare visits and 61% higher healthcare charges relative to non-depressed patients (Unutzer et al., 1997). Depressed primary care patients have significantly more annual healthcare visits and higher rates of specialist referrals and radiologic tests than non-depressed patients (Luber et al., 2000). Older depressed primary care patients have significantly more annual outpatient healthcare visits, laboratory tests, radiologic procedures and scans, and have a higher number of consultations with medical, surgical, and other specialists than non-depressed patients (Luber et al., 2001). Significantly less is known about the relation between anxiety and healthcare use, but evidence suggests a similar trend. For example, individuals with anxiety disorders are significantly more likely to visit primary care compared to those without anxiety (Bijl & Ravelli, 2000).

In cancer patients, those with an ICD-9 chart diagnosis of a depressive disorder make 76% more annual healthcare visits, are at 145% increased odds of experiencing an emergency department (ED) visit, 81% increased risk of experiencing an inpatient hospitalization, and 103% increased risk of a 30-day rehospitalization relative to those without a diagnosis (Mausbach & Irwin, 2017). Those with any depressive disorder are at least twice as likely to use ED and medical inpatient hospital services than those without a depressive disorder (Himelhoch, Weller, Wu, Anderson, & Cooper, 2004). However, less is clear about the link between anxiety and healthcare use in cancer patients. In hematologic cancer patients, diagnosis of any mood, anxiety, or adjustment disorder is associated with longer length of hospital stay (Prieto et al., 2002). Another study demonstrated greater risk for ED visits in patients with cancer and comorbid anxiety compared to those without anxiety (Lebel, Tomei, Feldstain, Beattie, & McCallum, 2013). However, others have found no association between levels of anxiety and self-reported health care utilization (Sarkar et al., 2015).

The financial costs associated with anxiety and depression may also be substantial in patients with cancer. Cancer patients with a chart diagnosis of depression have annual healthcare costs more than double those of non-depressed patients, with higher charges occurring in major healthcare categories of ambulatory care, emergency department charges, and inpatient hospital settings (Mausbach, Yeung, Bos, & Irwin, 2018).

Despite previous research examining depressive and anxiety disorders alone in patients with cancer, there appears to be a gap in the literature on the role of comorbid depression and anxiety in raising healthcare use in patients with cancer. Further, many studies focus only on healthcare use or healthcare costs, thereby failing to study both use and costs in the same population of participants. This effectively leaves a gap in determining the role of comorbid psychiatric conditions in increasing both use and costs in the greater cancer population. The purpose of this study was to investigate whether psychiatric diagnoses contributed to increased health care use and costs during the year following cancer diagnosis. The following hypotheses were examined: 1) patients with anxiety alone or depression alone would experience greater annual healthcare use and accrue greater annual costs relative than those without anxiety or depression, 2) patients with comorbid anxiety and depression would experience greater annual healthcare use and costs relative to those without anxiety or depression, 3) patients with comorbid anxiety and depression would experience greater annual healthcare use and costs relative to those with anxiety alone or depression alone.

2. MATERIALS AND METHODS

2.1. Study Design and Sample

The sample consisted of 13,426 adults with an International Classification of Diseases, Ninth Revision (ICD-9) diagnosis of cancer occurring between January 1 and December 31, 2014 at the University of California, San Diego (UCSD) healthcare system in San Diego, California. Data were obtained from electronic medical records (EMRs) which included ICD-9 diagnostic codes, encounter history, demographic information, and hospital and emergency department admissions information. Only those who were 18 years of age or older and had at least one health care claim within one year of cancer diagnosis were included. The study was approved by the UC San Diego Institutional Review Board.

2.2. Measures

2.2.1. Demographic information.

Patient age, gender, and race/ethnicity were extracted from patients’ electronic medical records (EMR).

2.2.2. Health care service use.

Patients’ EMR were reviewed for any emergency department visits and inpatient hospitalizations during the year following their cancer diagnoses. Length of hospital stays and use of radiation and chemotherapy services were also extracted from EMR data.

2.2.3. Health care costs.

The charges incurred for relevant services in the UC San Diego Healthcare System were provided by the UC San Diego Health cost-accounting system. Included charges were total annual healthcare charges, annual outpatient (ambulatory) office charges, ED charges, and inpatient hospital charges.

2.2.4. Anxiety diagnosis.

Anxiety diagnosis was extracted from patients’ EMR. Based on previous studies (Gupta & Knapp, 2014; Scherrer et al., 2010), those with an ICD-9 diagnostic code of any anxiety disorder (300.00, 300.01, 300.02, 300.2, 300.23, 309.81, 309.24) in the year 2014 were coded as having anxiety.

2.2.5. Depression diagnosis.

Depression diagnosis was extracted from patients’ EMR. Consistent with prior studies (Druss, Rosenheck, & Sledge, 2000; Greenberg, Fournier, Sisitsky, Pike, & Kessler, 2015; Luber et al., 2000; Luber et al., 2001), those with an ICD-9 diagnostic code of any depressive disorder (296.2, 296.3, 300.4, 309.0, 309.1, or 311) in the year 2014 were coded as having depression.

2.2.6. Medical comorbidity.

The Deyo-Charlson Comorbidity Index (DCCI) score (Deyo, Cherkin, & Ciol, 1992) was calculated for each patient. The DCCI is a valid predictor of physician visits, hospitalization use, and mortality (Charlson, Szatrowski, Peterson, & Gold, 1994; Charlson, Pompei, Ales, & MacKenzie, 1987). The DCCI assigns a value of ‘2’ to patients experiencing cancers. Because all participants in this study had cancer, the value of “2” was removed from each patient’s overall score. Additionally, because the impact of metastatic cancer on health care use was of interest in this study, the value of “6” was subtracted from each overall score and a separate variable for metastasis (1 = yes; 0 = no) was created and included in the analyses.

2.2.7. Insurance.

Insurance coverage was extracted from EMR for inclusion as a covariate in analyses. Insurance was coded into three types: Self-pay, Medicare/Medi-Cal, and Private Insurance. For the analyses, two dummy variables were created (i.e., Medicare and private insurance), with “Self-Pay” as the reference category.

2.3. Statistical analyses

Emergency department and hospitalization outcomes were analyzed using multivariate logistic regression analyses, adjusting for gender, race/ethnicity, DCCI score, metastatic status, any receipt of chemotherapy, any receipt of radiation, and insurance status. Four mental health groups were created: 1) Neither depression nor anxiety (reference group), 2) anxiety only, 3) depression only, and 4) both anxiety and depression. Three dummy variables (i.e., anxiety only, depression only, and both anxiety and depression) were used in the regression analyses to examine the impact of depression and anxiety diagnoses on health care use, with neither anxiety nor depression used as the reference group. Odds ratios were examined as measures of effect size and represented the increased risk for service use (i.e., ED visits or hospitalizations) in patients with psychiatric diagnoses versus those without.

We compared length of hospital stays and health care costs for our four groups using generalized linear models with a log‐link function and gamma distribution. In this approach, a generalized linear model with a gamma distribution and a log‐link function estimates length of hospital stays and total annual health care costs only for participants who have values >0. Following the initial analysis of total costs, we conducted follow‐up analyses to evaluate subsets of cost type (i.e., ambulatory, ED, and inpatient hospital costs).

3. Results

3.1. Sample characteristics.

Demographic and clinical information for the sample is presented in Table 1. Patients were between the ages of 18 and 101 years (M = 61.7 years, SD = 14.5) and 53.3% were female. The majority were White and carried private insurance, with 19.4% having undergone radiation and 13.3% chemotherapy. A total of 23% of the sample had a diagnosis of anxiety, depression, or both.

Table 1.

Demographic and clinical information for the sample.

Total Sample
(N = 13,426)
Neither Depression nor Anxiety
(N = 10,325)
Depression Only
(N = 1,006)
Anxiety Only
(N = 1,043)
Anxiety and Depression
(N = 1,052)
Age (years), M (SD) 61.7 (14.6) 62.2 (14.6) 63.9 (13.9) 58.1 (14.7) 58.7 (14.0)
Female, n (%) 7,156 (53.3) 5,247 (50.8) 568 (56.5) 652 (62.5) 689 (65.5)
Race/Ethnicity
 Caucasian
 Hispanic/Latino
 Black/Afr-Am
 Asian/Pacif Islndr
 Other race/ethnicity

8,629 (64.3)
1,938 (14.4)
586 (4.4)
1,134 (8.4)
1,139 (8.5)

6,485 (62.8)
1,486 (14.4)
433 (4.2)
961 (9.3)
960 (9.3)

705 (70.1)
128 (12.7)
58 (5.8)
52 (5.2)
63 (6.3)

699 (67.0)
164 (15.7)
47 (4.5)
67 (6.4)
66 (6.3)

740 (70.3)
160 (15.2)
48 (4.6)
54 (5.1)
50 (4.8)
Charlson score, M (SD) 0.62 (1.43) 0.47 (1.24) 1.27 (2.00) 0.83 (1.53) 1.24 (1.96)
Metastatic Disease, n (%) 3,306 (24.6) 2,306 (22.3) 270 (26.8) 346 (33.2) 384 (36.5)
Any chemotherapy, n (%) 1,790 (13.3) 1,036 (10.0) 192 (19.1) 266 (25.5) 296 (28.1)
Any radiation, n (%) 2,607 (19.4) 2,022 (19.6) 163 (16.2) 196 (7.5) 826 (8.7)
Insurance, n (%)
 Private
 Medicare/Medicaid
 Self-pay

8,310 (61.9)
3,932 (29.3)
1,184 (8.8)

6,527 (63.2)
2,769 (26.8)
1,029 (10.0)

512 (50.9)
441 (43.8)
53 (5.3)

676 (64.8)
303 (29.1)
64 (6.1)

595 (56.6)
419 (39.8)
38 (3.6)

3.2. Risk for emergency department visit.

A total of 3,222 (24.0%) patients had at least one ED visit during the year following their diagnosis. Of these, there were 1,850 with neither anxiety nor depression, 403 with anxiety alone, 401 with depression alone, and 568 with both anxiety and depression. Results of the service use analyses are presented in Table 2. As seen, all three depression/anxiety groups were significantly more likely to visit an ED when compared to the reference group. Specifically, patients diagnosed with anxiety alone or depression alone were over twice the risk of visiting an ED (Anxiety alone: OR = 2.34, 95% CI = 2.03 - 2.70; Depression alone: OR = 2.26, 95% CI = 1.96 - 2.60). Patients with a diagnosis of both anxiety and depression were nearly four times as likely to visit an ED (OR = 3.87, 95% CI = 3.36 - 4.46).

Table 2.

Model estimates for risk of ED visit and Hospitalization

ED Visit Hospitalization
B (SE) OR (95% CI) p-value B (SE) OR (95% CI) p-value
Age −.004 (.002) .996 (.993-.999) .021 −.013 (.002) .987 (.983-.990) <.001
Female −.059 (.045) .943 (.863-1.030) .190 −.291 (.044) .747 (.685-.815) <.001
Asian .234 (.079) 1.264 (1.082-1.477) .003 .118 (.080) 1.125 (.962-1.316) .140
Black .851 (.096) 2.343 (1.939-2.831) <.001 .522 (.099) 1.685 (1.389-2.045) <.001
Hispanic .240 (.063) 1.272 (1.123-1.440) <.001 .589 (.060) 1.802 (1.601-2.028) <.001
Other Race −.237 (.090) .789 (.661-.941) .009 −.249 (.089) .780 (.655-.928) .005
Charlson .220 (.014) 1.246 (1.212-1.282) <.001 .260 (.015) 1.297 (1.260-1.334) <.001
Metastases .472 (.051) 1.603 (1.451-1.770) <.001 .850 (.049) 2.340 (2.125-2.576) <.001
Any Chemotherapy .794 (.059) 2.211 (1.969-2.484) <.001 .726 (.059) 2.067 (1.842-2.320) <.001
Any Radiation −.104 (.058) .901 (.805-1.009) .072 −.164 (.057) .849 (.759-.949) .004
Private Insurance .064 (.086) 1.066 (.901-1.261) .459 .005 (.081) 1.005 (.857-1.179) .949
Medicare .666 (.092) 1.946 (1.625-2.329) <.001 .593 (.088) 1.809 (1.521-2.150) <001
Depression Only .817 (.074) 2.264 (1.957-2.620) <.001 .584 (.076) 1.793 (1.544-2.083) <.001
Anxiety Only .851 (.073) 2.343 (2.029-2.704) <.001 .704 (.074) 2.021 (1.748-2.337) <.001
Depression + Anxiety 1.354 (.072) 3.873 (3.361-4.463) <.001 .989 (.074) 2.689 (2.327-3.107) <.001

Note. B = Unstandardized coefficient; SE = Standard Error; OR = Odds ratio; 95% CI = 95% Confidence Interval.

Post-hoc pairwise comparisons were also made to evaluate if patients with comorbid anxiety and depression were significantly more likely to visit an ED compared to those with anxiety or depression alone. Results indicated that those with comorbid anxiety and depression were significantly more likely to experience an ED visit than those with anxiety alone (Wald = 35.57, df = 1, p < .001; OR = 1.74; 95% CI = 1.45-2.08) or depression alone (Wald = 32.81, df = 1, p < .001; OR = 1.73, 95% CI = 1.44-2.09).

3.3. Risk for hospitalization.

A total of 3,441 (25.6%) patients had at least one hospitalization. Of these, 2,114 had neither anxiety nor depression, 416 had anxiety alone, 383 had depression alone, and 528 had both anxiety and depression. Compared to the reference group, patients with anxiety alone or depression alone were over 70% more likely to experience a hospitalization in the year following their diagnoses (Anxiety alone: OR = 1.72, 95% CI = 1.75 - 2.34; Depression alone: OR = 1.71, 95% CI =1.54 - 2.08). Patients with comorbid anxiety and depression were more than 2.5 times as likely to be hospitalized (OR = 2.53, 95% CI = 2.33 - 3.11) than the reference group.

Post-hoc analyses indicated that patients with comorbid anxiety and depression were significantly more likely to be hospitalized compared to those with anxiety alone (Wald = 14.73, df = 1, p < .001; OR = 1.44, 95% CI = 1.19-1.73) or depression alone (Wald = 17.80, df = 1, p < .001; OR = 1.51, 956% CI = 1.25-1.83).

3.4. Length of hospital visits.

The mean length of stay for patients with neither anxiety nor depression was 11.89 days (95% CI = 11.38-12.42), compared to 15.91 days (95% CI = 14.51-17.46) for those with depression alone, 15.36 days (95% CI = 14.00-16.85) for those with anxiety alone, and 20.41 days (95% CI = 18.84-22.12) for those with comorbid anxiety and depression. Patients with anxiety alone experienced significantly longer hospital stays relative to those with neither anxiety nor depression [exp(B) = 1.34; 95% CI = 1.20-1.51; p < .001], as did those with depression alone [exp(B) = 1.30, 95% CI = 1.16-1.45; p < .001]. Finally, patients with comorbid anxiety and depression had significantly longer lengths of stay relative to the reference group [exp(B) = 1.73, 95% CI = 1.56-1.91; p < .001). Pairwise comparisons of the comorbid group with the anxiety alone and depression alone groups indicated that the comorbid group experienced significantly longer stays compared to those with depression alone [exp(B) = 1.29, 95% CI = 1.12-1.47; p < .001] and those with anxiety alone [exp(B) = 1.33, 95% CI = 1.17-1.52; p < .001].

3.5. Healthcare costs.

Estimated annual costs by healthcare category are presented in Table 3. Patients with anxiety alone [exp(B) = 1.77, 95% CI = 1.61-1.94; p < .001], depression alone [exp(B) = 1.38, 95% CI = 1.25-1.52; p < .001], and comorbid anxiety and depression [exp(B) = 2.61, 95% CI = 2.37-2.87; p < .001] had significantly higher total annual healthcare costs relative to those with neither anxiety nor depression. Depression alone [exp(B) = 1.44, 95% CI = 1.35-1.52; p < .001], anxiety alone [exp(B) = 1.54, 95% CI = 1.46-1.64; p < 001], and comorbid anxiety and depression [exp(B) = 1.99, 95% CI = 1.88-2.11; p < .001] were each associated with significantly higher ambulatory costs. ED costs were also significantly higher in patients with depression alone [exp(B) = 1.26, 95% CI = 1.13-1.40; p < .001], anxiety alone [exp(B) = 1.29, 95% CI = 1.16-1.44; p < .001], and comorbid anxiety and depression [exp(B) = 1.57, 95% CI = 1.43-1.73; p < .001], when compared to the reference group. Finally, inpatient hospital costs were significantly higher in patients with depression alone [exp(B) = 1.21, 95% CI = 1.09-1.35; p = .001], anxiety alone [exp(B) = 1.31, 95% CI = 1.18-1.46; p < .001] and comorbid anxiety and depression [exp(B) = 1.78, 95% CI = 1.62-1.97; p < .001].

Table 3.

Estimated annual healthcare cost estimates by mental health status

Group Total Charges
US$
[95% CI]
Ambulatory Charges
US$
[95% CI]
ED Charges
US$
[95% CI]
Inpatient Hospital Charges
US$
[95% CI]
Neither Anxiety nor Depression $75,373
[$73,235-$77,574]
$1,710
[$1,678-$1,743]
$7,777
[$7,429-$8,141]
$122,440
[$117,351-$127,749]
Anxiety only $133,046
[$121,543-$145,638]
$2,640
[$2,500-$2,788]
$10.034
[$9,124-$11,035]
$160,383
[$145,829-$176,389]
Depression only $103,849
[$94,749-$113,822]
$2,455
[$2,322-$2,595]
$9,761
[$8,841-$10,776]
$148,517
[$134,476-$164,024]
Anxiety + Depression $196,799
[$179,746-$215,470]
$3,399
[$3,220-$3,588]
$12,196
[$11,240-$13,234]
$218,403
[$200,550-$237,844]

A secondary analysis compared healthcare use and costs for patients diagnosed with depression, anxiety, or comorbid depression and anxiety before cancer vs after cancer. Of all patients with a depression diagnosis, 279 (13.6%) had a history of depression prior to being diagnosed with cancer. Of all patients with an anxiety diagnosis, 272 (13.0%) had a diagnosis prior to being diagnosed with cancer. Using logistic regression analyses, we examined if having a diagnosis of anxiety or depression before cancer was associated with significant differences in healthcare use and cost compared to being diagnosed with depression or anxiety after cancer diagnosis. Results indicated that having depression diagnosed after cancer was not associated with significant change in ED use (p = .572), hospitalization (p = .862), or cost (p = .99) vs being diagnosed with depression before cancer. Being diagnosed with anxiety before cancer was also not associated with any significant change in ED use (p = .161), hospitalization (p = .151) or cost (p = .750). Being diagnosed with both conditions after cancer was not associated with significant change ED use (p = .084) or hospitalization (p = .617). However, patients diagnosed with comorbid anxiety and depression after cancer had significantly higher healthcare costs relative to those diagnosed before cancer (mean difference = $78,460; p < .001).

4. Discussion

Patients with cancer who were diagnosed with either anxiety or depression alone were at an increased risk for health care use, specifically ED visits and hospitalizations. However, those with both diagnoses were at a dramatically increased risk, with over half having at least one ED visit and at least one hospitalization. This study replicates and extends previous findings (Luber et al., 2001; Mausbach & Irwin, 2017) by demonstrating increased health care use in not only cancer patients with depression, but also those with either anxiety alone or comorbid anxiety and depression. In addition to increased use, patients with anxiety, depression, or combination of these conditions experienced significantly longer hospital stays relative to those with neither condition. Those with comorbid anxiety and depression experienced hospital stays nearly 72% longer than those with neither condition. As discussed elsewhere (Baek et al., 2018; Bueno et al., 2010; Rotter et al., 2010), longer hospital stays can increase risks of opportunistic infections and side effects of medication, as well as have deleterious effects on treatment outcomes and mortality rates. In addition, longer hospital stays are associated with increased medical costs and decreased bed turnover rates (Bueno et al., 2010; Rotter et al., 2010). These results emphasize the need for anxiety and depression screening in those with cancer, and the need for adequate and timely referral to address the needs of cancer patients living with anxiety and/or depression.

Results also indicated that those with anxiety and/or depression had significantly higher annual healthcare costs. Costs among those with comorbid anxiety and depression were double those with neither condition and were significantly higher than those with depression or anxiety alone. These findings emphasize the impact of depression and anxiety not only on the micro level but also the meso and macro levels. Health care use and costs begin at the micro level when the individual experiences symptoms necessitating use of healthcare services. Meso systems are impacted when resources are expended by healthcare systems that prioritize care and ensure everyone receives adequate and timely services. Macro systems are impacted when payors are required to pay for services, and nations are impacted both medically and financially when services are expended to those in need. The implication is that adequately identified anxiety and/or depression can result in expedient and effective treatment of these conditions, resulting in reduced costs over time at multiple levels. Prior research suggests that psychosocial interventions can significantly offset medical costs for people with general mental distress and with comorbid medical illness (Carlson & Bultz, 2004), with 90% of studies showing cost reduction following psychosocial intervention (Chiles, Lambert, & Hatch, 1999, 2002). The majority of mental health interventions are both more effective and more costly than usual care (Jansen, van Zwieten, Coupe, Leemans, & Verdonck-de Leeuw, 2016), and those with cancer and major depression see significant dose dependent reductions in annual healthcare costs as they receive mental health care (Mausbach, Bos, & Irwin, 2018).

The strengths of this study were a large sample size and the use of electronic medical records as opposed to self-report to determine health care use and costs. As discussed elsewhere (Mausbach, Yeung, et al., 2018), the use of EMR allows for various components of care and cost to be readily identified and retrievable via health informatics services. Additionally, outcome measures of this study included discrete aspects of health care use such as ED visits and inpatient hospitalizations, inclusive of costs for these services. An additional strength is the focus on hospital length of stays, which have implications for patient well-being and health system burden. Finally, this study emphasized two common psychiatric conditions in cancer; anxiety and depression, and examined the impact of these conditions, separately and combined, on both health care use and costs.

Despite these strengths, this study also had several limitations. There was no information on cancer stage and grade and some patients may have been experiencing greater cancer-related side effects or more aggressive cancer treatments due to later-stage cancers that would prompt emergency department and hospital visits. Future researchers should determine if these factors are related to higher health care use in this population. Although the sample size was a strength, it necessitated the use of chart data to ascertain diagnosis of anxiety and depression. Thus, it was not possible to determine prevalence, incidence, and severity of depression or anxiety. Further, the criteria used to diagnose depression may vary across providers and disciplines, and subclinical symptoms, including data from self-report questionnaires, can be important indicators of distress, including for cost data (Chiu, Lebenbaum, Cheng, de Oliveira, & Kurdyak, 2017). Despite this limitation, it should be noted that the prevalence of depression and anxiety in our study was 15.3% and 15.6%, respectively. These values appear consistent with other studies reporting a depression prevalence of 16% in outpatients with cancer (Walker et al., 2013), with slightly higher prevalence of anxiety in our sample than found in other studies (Mitchell et al., 2011). Some patients may also have depression or anxiety that goes undiagnosed. However, undiagnosed depression has previously been associated with increased healthcare costs in a community sample (Williams, Chung, & Muennig, 2017). Similar results were also found in a large nationally representative sample of health patients in Japan (Yamabe, Liebert, Flores, & Pashos, 2019). In this latter study, undiagnosed depression was associated with greater number of ER visits, greater number of hospitalizations, more healthcare provider visits, higher direct and indirect costs, and higher hospital and ER costs compared to those who were not depressed. Because in our sample the undiagnosed patients were categorized as “not depressed”, it is likely that the true difference in costs between depressed and non-depressed patients is potentially greater than we present in our manuscript. Although there have not been studies examining the impact of undiagnosed anxiety it would seem likely that similar effects emerge in this population.

In conclusion, cancer patients with both anxiety and depression were at greater risk for ED visits and hospitalizations. The presence of anxiety and/or depression also extended hospital length of stay and increased overall healthcare costs. These findings demonstrate the need to examine anxiety and depression together when assessing the relationship between psychiatric comorbidity and health care utilization. Future researchers should examine psychosocial interventions aimed at treating psychological symptoms, which may reduce health care use. This could have positive impacts on the patients themselves such as reduced utilization, costs, and better symptom management, as well as to the healthcare system by potentially decreasing healthcare burden.

Acknowledgments

The project described was partially supported by the National Institutes of Health, Grant UL1TR000100 of CTSA funding prior to August 13, 2015 and Grant UL1TR001442 of CTSA funding beginning August 13, 2015 and beyond. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Additional funding was provided by the National Institute on Aging (NIA) via grant R25 AG043364.

Footnotes

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

References

  1. Baek H, Cho M, Kim S, Hwang H, Song M, & Yoo S (2018). Analysis of length of hospital stay using electronic health records: A statistical and data mining approach. PLoS One, 13(4), e0195901. doi: 10.1371/journal.pone.0195901 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bijl RV, & Ravelli A (2000). Psychiatric morbidity, service use, and need for care in the general population: results of The Netherlands Mental Health Survey and Incidence Study. American Journal of Public Health, 90(4), 602–607. doi: 10.2105/ajph.90.4.602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bueno H, Ross JS, Wang Y, Chen J, Vidan MT, Normand SL, 𠈆 Krumholz HM (2010). Trends in length of stay and short-term outcomes among Medicare patients hospitalized for heart failure, 1993-2006. JAMA, 303(21), 2141–2147. doi: 10.1001/jama.2010.748 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Carlson LE, & Bultz BD (2004). Efficacy and medical cost offset of psychosocial interventions in cancer care: making the case for economic analyses. Psychooncology, 13(12), 837–849; discussion 850-836. doi: 10.1002/pon.832 [DOI] [PubMed] [Google Scholar]
  5. Charlson M, Szatrowski TP, Peterson J, & Gold J (1994). Validation of a combined comorbidity index. Journal of Clinical Epidemiology, 47(11), 1245–1251. [DOI] [PubMed] [Google Scholar]
  6. Charlson ME, Pompei P, Ales KL, & MacKenzie CR (1987). A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of Chronic Diseases, 40(5), 373–383. [DOI] [PubMed] [Google Scholar]
  7. Chiles JA, Lambert MJ, & Hatch AL (1999). The impact of psychological interventions on medical cost offset: A meta-analytic review. Clinical Psychology-Science and Practice, 6(2), 204–220. doi: DOI 10.1093/clipsy/6.2.204 [DOI] [Google Scholar]
  8. Chiles JA, Lambert MJ, & Hatch AL (2002). Medical cost offset: A review of the impact of psychological interventions on medical utilization over the past three decades. Impact of Medical Cost Offset on Practice and Research: Making It Work for You, 5, 47–56. [Google Scholar]
  9. Chiu M, Lebenbaum M, Cheng J, de Oliveira C, & Kurdyak P (2017). The direct healthcare costs associated with psychological distress and major depression: A population-based cohort study in Ontario, Canada. PLoS One, 12(9), e0184268. doi: 10.1371/journal.pone.0184268 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Deyo RA, Cherkin DC, & Ciol MA (1992). Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. Journal of Clinical Epidemiology, 45(6), 613–619. [DOI] [PubMed] [Google Scholar]
  11. Druss BG, Rosenheck RA, & Sledge WH (2000). Health and disability costs of depressive illness in a major U.S. corporation. The American Journal of Psychiatry, 157(8), 1274–1278. doi: 10.1176/appi.ajp.157.8.1274 [DOI] [PubMed] [Google Scholar]
  12. Greenberg PE, Fournier AA, Sisitsky T, Pike CT, & Kessler RC (2015). The economic burden of adults with major depressive disorder in the United States (2005 and 2010). The Journal of Clinical Psychiatry, 76(2), 155–162. doi: 10.4088/JCP.14m09298 [DOI] [PubMed] [Google Scholar]
  13. Gupta MA, & Knapp K (2014). Cardiovascular and psychiatric morbidity in obstructive sleep apnea (OSA) with insomnia (sleep apnea plus) versus obstructive sleep apnea without insomnia: a case-control study from a Nationally Representative US sample. PLoS One, 9(3), e90021. doi: 10.1371/journal.pone.0090021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hasin DS, Goodwin RD, Stinson FS, & Grant BF (2005). Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on Alcoholism and Related Conditions. Archives of General Psychiatry, 62(10), 1097–1106. doi: 62/10/1097 [pii] 10.1001/archpsyc.62.10.1097 [DOI] [PubMed] [Google Scholar]
  15. Himelhoch S, Weller WE, Wu AW, Anderson GF, & Cooper LA (2004). Chronic medical illness, depression, and use of acute medical services among Medicare beneficiaries. Medical Care, 42(6), 512–521. doi: 00005650-200406000-00003 [pii] [DOI] [PubMed] [Google Scholar]
  16. Jansen F, van Zwieten V, Coupe VM, Leemans CR, & Verdonck-de Leeuw IM (2016). A Review on Cost-Effectiveness and Cost-Utility of Psychosocial Care in Cancer Patients. Asia-Pacific Journal of Oncology Nursing, 3(2), 125–136. doi: 10.4103/2347-5625.182930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lebel S, Tomei C, Feldstain A, Beattie S, & McCallum M (2013). Does fear of cancer recurrence predict cancer survivors’ health care use? Supportive Care in Cancer, 21(3), 901–906. doi: 10.1007/s00520-012-1685-3 [DOI] [PubMed] [Google Scholar]
  18. Linden W, Vodermaier A, Mackenzie R, & Greig D (2012). Anxiety and depression after cancer diagnosis: prevalence rates by cancer type, gender, and age. Journal of Affective Disorders, 141(2-3), 343–351. doi: 10.1016/j.jad.2012.03.025 [DOI] [PubMed] [Google Scholar]
  19. Luber MP, Hollenberg JP, Williams-Russo P, DiDomenico TN, Meyers BS, Alexopoulos GS, & Charlson ME (2000). Diagnosis, treatment, comorbidity, and resource utilization of depressed patients in a general medical practice. International Journal of Psychiaty in Medicine, 30(1), 1–13. [DOI] [PubMed] [Google Scholar]
  20. Luber MP, Meyers BS, Williams-Russo PG, Hollenberg JP, DiDomenico TN, Charlson ME, & Alexopoulos GS (2001). Depression and service utilization in elderly primary care patients. American Journal of Geriatric Psychiatry, 9(2), 169–176. doi: 10.1176/appi.ajgp.9.2.169 [DOI] [PubMed] [Google Scholar]
  21. Marciniak MD, Lage MJ, Dunayevich E, Russell JM, Bowman L, Landbloom RP, & Levine LR (2005). The cost of treating anxiety: the medical and demographic correlates that impact total medical costs. Depression and Anxiety 21(4), 178–184. doi: 10.1002/da.20074 [DOI] [PubMed] [Google Scholar]
  22. Mausbach BT, Bos T, & Irwin SA (2018). Mental health treatment dose and annual healthcare costs in patients with cancer and major depressive disorder. Health Psychology, 37(11), 1035–1040. doi: 10.1037/hea0000670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Mausbach BT, & Irwin SA (2017). Depression and healthcare service utilization in patients with cancer. Psycho-Oncology, 26(8), 1133–1139. doi: 10.1002/pon.4133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mausbach BT, Yeung P, Bos T, & Irwin SA (2018). Health care costs of depression in patients diagnosed with cancer. Psychooncology, 27(7), 1735–1741. doi: 10.1002/pon.4716 [DOI] [PubMed] [Google Scholar]
  25. Mitchell AJ, Chan M, Bhatti H, Halton M, Grassi L, Johansen C, & Meader N (2011). Prevalence of depression, anxiety, and adjustment disorder in oncological, haematological, and palliative-care settings: a meta-analysis of 94 interview-based studies. The Lancet Oncology, 12(2), 160–174. doi: 10.1016/S1470-2045(11)70002-X [DOI] [PubMed] [Google Scholar]
  26. Mitchell AJ, Ferguson DW, Gill J, Paul J, & Symonds P (2013). Depression and anxiety in long-term cancer survivors compared with spouses and healthy controls: a systematic review and meta-analysis. Lancet Oncology, 14(8), 721–732. doi: 10.1016/S1470-2045(13)70244-4 [DOI] [PubMed] [Google Scholar]
  27. Prieto JM, Blanch J, Atala J, Carreras E, Rovira M, Cirera E, & Gasto C (2002). Psychiatric morbidity and impact on hospital length of stay among hematologic cancer patients receiving stem-cell transplantation. Journal of Clinical Oncology, 20(7), 1907–1917. doi: 10.1200/JCO.2002.07.101 [DOI] [PubMed] [Google Scholar]
  28. Regier DA, Narrow WE, Rae DS, Manderscheid RW, Locke BZ, & Goodwin FK (1993). The de facto US mental and addictive disorders service system. Epidemiologic catchment area prospective 1-year prevalence rates of disorders and services. Archives of General Psychiatry, 50(2), 85–94. [DOI] [PubMed] [Google Scholar]
  29. Rotter T, Kinsman L, James E, Machotta A, Gothe H, Willis J, 𠈆 Kugler J (2010). Clinical pathways: effects on professional practice, patient outcomes, length of stay and hospital costs. The Cochrane Database of Systematic Reviews (3), CD006632. doi: 10.1002/14651858.CD006632.pub2 [DOI] [PubMed] [Google Scholar]
  30. Sarkar S, Sautier L, Schilling G, Bokemeyer C, Koch U, & Mehnert A (2015). Anxiety and fear of cancer recurrence and its association with supportive care needs and health-care service utilization in cancer patients. Journal of Cancer Survivorship, 9(4), 567–575. doi: 10.1007/s11764-015-0434-2 [DOI] [PubMed] [Google Scholar]
  31. Scherrer JF, Chrusciel T, Zeringue A, Garfield LD, Hauptman PJ, Lustman PJ, 𠈆 True WR (2010). Anxiety disorders increase risk for incident myocardial infarction in depressed and nondepressed Veterans Administration patients. American Heart Journal, 159(5), 772–779. doi: 10.1016/j.ahj.2010.02.033 [DOI] [PubMed] [Google Scholar]
  32. Thomas CM, & Morris S (2003). Cost of depression among adults in England in 2000. British Journal of Psychiatry, 183(6), 514–519. [DOI] [PubMed] [Google Scholar]
  33. Unutzer J, Patrick DL, Simon G, Grembowski D, Walker E, Rutter C, & Katon W (1997). Depressive symptoms and the cost of health services in HMO patients aged 65 years and older. A 4-year prospective study. Journal of the American Medical Association, 277(20), 1618–1623. [DOI] [PubMed] [Google Scholar]
  34. Walker J, Holm Hansen C, Martin P, Sawhney A, Thekkumpurath P, Beale C, 𠈆 Sharpe M (2013). Prevalence of depression in adults with cancer: a systematic review. Annals of Oncology, 24(4), 895–900. doi: 10.1093/annonc/mds575 [DOI] [PubMed] [Google Scholar]
  35. Williams SZ, Chung GS, & Muennig PA (2017). Undiagnosed depression: A community diagnosis. SSM Population Health, 3, 633–638. doi: 10.1016/j.ssmph.2017.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. World Health Organization. (2006). Global Burden of Disease Estimates Retrieved June 19, 2006, from http://www.who.int/healthinfo/bodestimates/en/index.html
  37. Yamabe K, Liebert R, Flores N, & Pashos CL (2019). Health-related quality of life outcomes, economic burden, and associated costs among diagnosed and undiagnosed depression patients in Japan. Clinicoeconomics and Outcomes Research, 11, 233–243. doi: 10.2147/CEOR.S179901 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES