Abstract
Introduction:
Serious illness is a life-limiting condition negatively impacting daily function, quality of life, or excessively straining caregivers. Over one million older seriously ill adults undergo major surgery annually, and national guidelines recommend that palliative care be available to all seriously ill patients. However, palliative care needs in elective surgical patients are incompletely described. Understanding baseline caregiving needs and symptom burden among seriously ill older surgical patients could inform interventions to improve outcomes.
Methods:
Using Health and Retirement Study data (2008–2018) linked to Medicare claims, we identified patients ≥66 years who met an established serious illness definition from administrative data and underwent major elective surgery using AHRQ criteria. Descriptive analyses were performed for preoperative patient characteristics, including: unpaid caregiving (no or yes); pain (none/mild or moderate/severe); and depression (no, CES-D <3, or yes, CES-D ≥3). Multivariable regression was performed to examine the association between unpaid caregiving, pain, depression, and in-hospital outcomes, including hospital days (days admitted between discharge date and one-year post-discharge), in-hospital complications (no or yes), and discharge destination (home or non-home).
Results:
Of 1,343 patients, 55.0% were female and 81.6% non-Hispanic White. Mean age was 78.0 (SD 6.8); 86.9% had ≥2 comorbidities. Before admission, 27.3% of patients received unpaid caregiving. Pre-admission pain and depression was 42.6% and 32.8%, respectively. Baseline depression was significantly associated with non-home discharge (OR 1.6, 95% CI 1.2–2.1, p=0.003) while baseline pain and unpaid caregiving needs were not associated with in-hospital or post-acute outcomes in multivariable analysis.
Conclusions:
Prior to elective surgery, older adults with serious illness have high unpaid caregiving needs and prevalence of pain and depression. Baseline depression alone was associated with discharge destination. These findings highlight opportunities for targeted palliative care interventions throughout the surgical encounter.
Keywords: older adult, serious illness, elective surgery, caregiving, symptom burden, palliative care
Introduction:
Serious illness is defined as a potentially life-limiting condition negatively impacting daily function, quality of life, or excessively straining caregivers.1 Palliative care is an approach to care that prioritizes management of physical and psychological symptoms and improving quality of life for seriously ill individuals and their families. National consensus-based guidelines recommend palliative care for seriously ill patients and their families at all stages of illness.2 Further, insufficient numbers of clinicians trained in palliative care mean that interventions must be targeted to provide the greatest benefit to those in greatest need and should focus on outcomes valued by patients.3 Seriously ill patients have reported pain and symptom management and not being a burden to family as top priorities at the end of life.4 Addressing unpaid caregiving, pain, and depression requires close examination of caregiving needs and symptom burden in this population.
Previous studies evaluating symptom burden in both older and seriously ill adults have found high prevalence of pain and depression among community-dwelling older adults and nursing home residents ranging from 36–55% for pain and 5%−30% for depression.5–10 These symptoms have been associated with caregiver burden and increased hospitalization, emergency department visits, and nursing home admissions.1,11,12 In surgical populations, preoperative pain and depression have been associated with adverse postoperative outcomes, including persistent pain and increased length of stay and complications.13–16 Palliative care access and interventions have been shown to reduce pain and depression and ease caregiver burden.17–19 However, underlying symptom prevalence and caregiver burden are incompletely described in seriously ill adults undergoing elective surgery.
It is estimated that over one million older seriously ill adults experience major surgery each year.20,21 National data on caregiving needs, pain, and depression in these patients is lacking despite palliative symptom assessment being a surgical quality indicator in seriously ill patients.22 Accordingly, the American College of Surgeons has incorporated assessment for palliative care needs as a mandatory component of preoperative vulnerability screening for older surgical patients, many of whom are seriously ill.23 This information is critical in designing palliative care interventions to improve care for these vulnerable patients and potentially reduce excess healthcare utilization. In this retrospective cohort study, we aimed to measure unpaid caregiving needs and prevalence of pain and depression in older seriously ill adults undergoing elective surgery and to examine their association with in-hospital and post-acute outcomes, including hospital days, in-hospital complications, and discharge destination. Our hypothesis was that older seriously ill patients with unpaid caregiving needs and pain and depression at baseline would have more hospital days and higher rates of in-hospital complications and non-home discharge.
Methods:
Data source
The Health and Retirement Study (HRS) is a nationally representative longitudinal cohort study that collects surveys from respondents ≥51 years biennially and conducts a proxy ‘exit’ interview upon the death of participating respondents. The Medicare Provider Analysis and Review data includes claims data from Medicare beneficiaries who received inpatient hospital services. Among HRS respondents, 78–84% authorize linkage to their Medicare claims data.24
Study cohort
Using HRS data from 2008–2018 linked to Medicare claims, we identified patients ≥66 years with a one-year minimum of continuous fee-for-service data who met established criteria for serious illness in surgery (defined using administrative data as having one of the following conditions: vulnerable elder, advanced cancer and ≥one hospitalization in prior year, oxygen-dependent pulmonary disease, heart failure diagnosis with any all-cause hospitalization and at least two ED visits in past six months, cirrhosis with any Childs-Turcotte-Pugh class or Model for End-Stage Liver Disease score, end-stage renal disease on dialysis or eligible for dialysis, dementia with impaired daily function and at least one hospitalization in prior year, frailty, trauma patients with severe traumatic brain injury or critical injury, and nursing home residents) and had undergone major surgery as defined by International Classification of Diseases codes using Agency for Healthcare Research and Quality Healthcare Cost and Utilization Project Major Therapeutic procedure classification.20,25 Patients who had a trauma admission (defined by ICD codes and meeting the definition for the serious illness condition) during the same encounter as a qualifying surgical procedure and patients who had undergone non-elective (urgent or emergent, defined by inpatient admission type code) surgery were excluded. Patients who had more than one surgical procedure during the study period were included by each qualifying admission. Patients with missing data (n=61) were excluded (Figure 1). This study was approved by the Icahn School of Medicine at Mount Sinai Institutional Review Board.
Figure 1.
Flow diagram for patient cohort
Study variables
The following patient demographic, comorbidity, and health status variables were derived from the HRS survey preceding surgical admission: age (66–74, 75–84, ≥85 years), sex (female, male), race/ethnicity (non-Hispanic White, non-Hispanic Black, and Hispanic), education (<high school, high school, some college/college, graduate school), annual household wealth (quartiles), marital status (married and living with other(s), not married and living with other(s), not married and living alone), functional dependence (dependence in activities of daily living and dependence in independent activities of daily living, dichotomized as yes vs no), and cognition (dichotomized as no dementia, Modified Telephone Interview for Cognitive Status (TICS-M) score ≥12 vs probable dementia, TICS-M score <12),26 and time between surgical admission and preceding HRS survey (months). For the race variable, the non-Hispanic others category was excluded for low sample size in compliance with the Centers for Medicare and Medicaid cell suppression policy.27 Charlson comorbidity index (dichotomized as <2 vs ≥ 2) and serious illness-defining conditions were collected from Medicare claims except for American Society of Anesthesiologists Classification, which cannot be extracted from claims data.
Unpaid caregiving, pain, and depression were collected from the HRS. Unpaid caregiving was dichotomized as no or yes. Pain severity in the preceding year was assessed using the questions, ‘Are you often troubled with pain?’ and ‘How bad is the pain most of the time?’ Response options included none, mild, moderate, and severe, and baseline pain was dichotomized as none/mild vs moderate/severe. Depression was measured using the Center for Epidemiologic Studies Depression Scales (CES-D), scored from 0–8, with higher scores being worse, and baseline depression was dichotomized as no, CES-D<3 vs yes, CES-D≥3.28,29
Outcomes were collected from Medicare claims and included hospital days (count measure of total days admitted to hospital between date of discharge from surgical admission and one-year post-discharge), in-hospital complications (composite measure of any one or more of the following complications: pneumonia, myocardial infarction, cardiac arrest, pulmonary failure, cerebral infarction, acute renal failure, and surgical site infection; dichotomized as yes vs no), and discharge destination (defined using patient discharge status code, dichotomized as home, including home/self-care and home care of organized home health service organization, vs non-home, including acute rehab facility, skilled or unskilled nursing facility, long term care facility).
Data analysis
We performed descriptive analyses for patient variables and in-hospital and post-acute outcomes. Bivariate analyses were performed to examine the association between patient characteristics, including unpaid caregiving, pain, depression, and in-hospital and post-acute outcomes using two-sided t-tests for the continuous outcome and chi-square tests for the categorical outcomes.
Multivariable analysis was conducted using negative binomial regression for hospital days and logistic regression for in-hospital complications and discharge destination. We examined the association between unpaid caregiving, pain, depression, and each outcome in unadjusted models. We subsequently examined the associations between unpaid caregiving, pain, depression,5–8,10 and each outcome, controlling for age, sex, race/ethnicity, education, annual household wealth, marital status, functional dependence, cognition, Charlson comorbidity index, and time between surgical admission and preceding HRS survey. These covariates were all selected for inclusion in the model based on clinical relevance and previous literature, including work from our group.30–37 To account for clustering within individuals, we included clustered standard errors at the patient level in our models.
We performed a series of sensitivity analyses. First, we examined the association between baseline pain and depression in our seriously ill elective surgical cohort given prior reports of concurrent pain and depression in patients.38 Next, we conducted a mediation analysis to examine whether functional dependence acted as a mediator in the associations between unpaid caregiving, pain, and depression and the outcomes.39 We followed with examining the association between unpaid caregiving, pain, and depression and non-home discharge in separate adjusted models; for the model with pain, functional dependence was excluded from the model. Lastly, we developed additional models for each outcome to test the interaction between 1) marital status and unpaid caregiving and 2) time between surgical admission and preceding HRS survey and unpaid caregiving, pain, and depression. All statistical analyses were performed using Stata 16 (College Station, TX).
Results:
We identified 1,343 patients who met criteria for serious illness in surgery and had undergone major elective surgery between 2008 and 2018. Mean time between surgical admission and the preceding HRS survey was 11 months (SD 6.6). Mean age was 78.0 years (SD 6.8) (Table 1). Most (55.0%) were female and non-Hispanic White (81.6%). Nearly four-fifths (79.6%) reported a high school education or less, and 57.3% were married and living with other(s). Multimorbidity was present in 86.9% of patients, and 81.1% had functional dependence in activities of daily living. The most common characteristics of serious illness were vulnerable elder (53.4%), frailty (39.9%), and heart failure (30.8%). Nearly half (48.3%) of patients had more than one qualifying serious illness, and 24.3% had more than two. Prior to elective surgery, 27.3% of patients received unpaid caregiving. Baseline pain and depression were 42.6% and 32.8%, respectively. Of patients with baseline pain, 45.5% had baseline depression, and 56.6% of patients with baseline depression also had baseline pain; prevalence of one symptom was significantly associated with that of the other (p<0.001).
Table 1.
Characteristics of Older Seriously Ill Patients undergoing Elective Surgery in HRS linked to Medicare data from 2008–2018 (N=1,343)
| Characteristic (%) | Entire Cohort (%) |
|---|---|
| Age (yrs) | |
| 66–74 | 476 (35.4) |
| 75–84 | 611 (45.5) |
| ≥85 | 256 (19.1) |
| Gender | |
| Female | 739 (55.0) |
| Male | 604 (45.0) |
| Race/Ethnicity | |
| Non-Hispanic White | 1,096 (81.6) |
| Non-Hispanic Black | 170 (12.67) |
| Hispanic | 77 (5.7) |
| Education | |
| <High school | 322 (24.0) |
| High school | 746 (55.6) |
| Some college/college | 179 (13.3) |
| Graduate degree | 96 (7.1) |
| Household Wealth | |
| 1st quartile | 264 (19.7) |
| 2nd quartile | 348 (25.9) |
| 3rd quartile | 344 (25.6) |
| 4th quartile | 387 (28.8) |
| Marital status | |
| Married and living with spouse | 770 (57.3) |
| Not married and living with other(s) | 177 (13.2) |
| Not married and living alone | 396 (29.5) |
| Charlson comorbidity index | |
| <2 | 176 (13.1) |
| ≥2 | 1,167 (86.9) |
| Functional dependence | |
| Activities of daily living | 1,089 (81.1) |
| Independent activities of daily living |
1,025 (76.3) |
| Cognition | |
| No dementia | 1,234 (91.9) |
| Probable dementia | 109 (8.1) |
| Qualifying serious illness | |
| Vulnerable elder | 784 (58.4) |
| Advanced cancer | 234 (17.4) |
| Pulmonary disease | 116 (8.6) |
| Heart failure | 413 (30.8) |
| Cirrhosis | 76 (5.7) |
| End stage renal disease | 156 (11.6) |
| Dementia | 131 (9.7) |
| Frailty | 536 (39.9) |
| Nursing home resident | 42 (3.1) |
| >1 serious illness | 649 (48.3) |
| >2 serious illnesses | 326 (24.3) |
| Unpaid caregiving | |
| No | 977 (72.8) |
| Yes | 366 (27.2) |
| Pain | |
| None/mild | 771 (57.4) |
| Moderate/severe | 572 (42.6) |
| Depression | |
| No | 903 (67.2) |
| Yes | 440 (32.8) |
In our cohort, mean hospital days within one year post-discharge from surgical admission was 5.4 days (SD 7.6) (Table 2). Of the qualifying serious illnesses, patients with advanced cancer (6.5 vs 5.2, p=0.01), heart failure (7.0 vs 4.7, p<0.001), end stage renal disease (8.0 vs 5.1, p<0.001), dementia (7.2 vs 5.2, p=0.004), and frailty (6.7 vs 4.6, p<0.001) had significantly more hospital days in the year following discharge from surgical admission than those without the respective serious illness.
Table 2.
Unadjusted associations between characteristics of older seriously ill surgical patients and in-hospital outcomes using HRS-Medicare data from 2008–2018
| Characteristic (%) | Hospital days (mean, SD) | p-value | In-hospital complications (n=549) (40.9%) | p-value | Non-home discharge (n=482) (35.9%) | p-value |
|---|---|---|---|---|---|---|
| Age (yrs) | 0.64 | 0.32 | <0.001 | |||
| 66–74 | 5.4 (8.0) | 184 (38.7) | 132 (27.7) | |||
| 75–84 | 5.4 (7.6) | 263 (43.0) | 220 (36.0) | |||
| >85 | 5.7 (7.2) | 102 (39.8) | 130 (50.8) | |||
| Gender | 0.01 | 0.04 | 0.01 | |||
| Female | 4.9 (6.0) | 284 (38.4) | 289 (39.1) | |||
| Male | 6.0 (9.2) | 265 (43.9) | 193 (32.0) | |||
| Race | 0.001 | 0.99 | 0.39 | |||
| Non-Hispanic White | 5.2 (7.3) | 447 (40.8) | 401 (36.6) | |||
| Non-Hispanic Black | 5.7 (8.0) | 70 (41.2) | 53 (31.2) | |||
| Hispanic | 7.6 (10.8) | 32 (41.6) | 28 (36.4) | |||
| Education | 0.21 | 0.01 | 0.61 | |||
| <High school | 6.1 (8.2) | 151 (46.9) | 111 (34.5) | |||
| High school | 5.3 (7.6) | 305 (40.9) | 278 (37.3) | |||
| College/some college | 4.9 (6.8) | 57 (31.8) | 63 (35.2) | |||
| Graduate degree | 5.2 (7.4) | 36 (37.5) | 30 (31.2) | |||
| Household Wealth | 0.40 | 0.01 | 0.11 | |||
| 1st quartile | 5.9 (8.8) | 116 (43.9) | 108 (40.9) | |||
| 2nd quartile | 5.9 (8.2) | 143 (41.1) | 132 (37.9) | |||
| 3rd quartile | 5.6 (8.6) | 146 (42.4) | 113 (32.9) | |||
| 4th quartile | 4.5 (4.9) | 144 (37.2) | 129 (33.3) | |||
| Marital status | 0.59 | 0.77 | 0.11 | |||
| Married and living with spouse | 5.5 (7.7) | 320 (41.6) | 238 (30.9) | |||
| Not married and living with other(s) | 5.9 (9.0) | 73 (41.2) | 62 (35.0) | |||
| Not married and living alone | 5.0 (6.8) | 156 (39.4) | 182 (46.0) | |||
| Qualifying Serious Illness | ||||||
| Vulnerable elder | 0.29 | 0.34 | <0.001 | |||
| Yes | 5.2 (7.1) | 312 (39.8) | 324 (41.3) | |||
| No | 5.7 (8.4) | 237 (42.4) | 158 (28.3) | |||
| Advanced cancer | 0.01 | 0.92 | <0.001 | |||
| Yes | 6.5 (9.3) | 95 (40.6) | 57 (24.4) | |||
| No | 5.2 (7.2) | 454 (40.9) | 425 (38.3) | |||
| Pulmonary disease | 0.58 | 0.27 | 0.35 | |||
| Yes | 5.8 (7.1) | 53 (45.7) | 37 (31.9) | |||
| No | 5.4 (7.7) | 496 (40.4) | 445 (36.3) | |||
| Heart failure | <0.001 | <0.001 | 0.18 | |||
| Yes | 7.0 (10.1) | 198 (47.9) | 159 (38.5) | |||
| No | 4.7 (6.1) | 351 (37.7) | 323 (34.7) | |||
| Cirrhosis | 0.82 | 0.35 | 0.005 | |||
| Yes | 5.2 (6.0) | 35 (46.1) | NR | |||
| No | 5.4 (7.7) | 514 (40.6) | NR | |||
| ESRD | <0.001 | 0.02 | 0.21 | |||
| Yes | 8.0 (11.7) | 77 (49.4) | 63 (40.4) | |||
| No | 5.1 (6.9) | 472 (39.8) | 419 (35.3) | |||
| Dementia | 0.004 | 0.52 | <0.001 | |||
| Yes | 7.2 (9.9) | 57 (43.5) | 79 (60.3) | |||
| No | 5.2 (7.3) | 492 (40.6) | 403 (33.3) | |||
| Frailty | <0.001 | 0.07 | <0.001 | |||
| Yes | 6.67 (9.4) | 235 (43.8) | 267 (49.8) | |||
| No | 4.6 (6.1) | 314 (38.9) | 215 (26.6) | |||
| Nursing home resident | 0.19 | 0.71 | <0.001 | |||
| Yes | 7.0 (7.6) | NR | NR | |||
| No | 5.4 (7.6) | NR | NR | |||
| Charlson comorbidity index | <0.001 | <0.001 | 0.33 | |||
| <2 | 3.5 (3.6) | 48 (27.3) | 69 (39.2) | |||
| ≥2 | 5.7 (8.0) | 501 (42.9) | 413 (35.4) | |||
| Functional dependence | ||||||
| ADLs | 0.09 | 0.25 | <0.001 | |||
| Yes | 5.2 (7.3) | 437 (40.1) | 359 (33.0) | |||
| No | 6.2 (9.0) | 112 (44.1) | 123 (48.4) | |||
| IADLs | 0.06 | 0.12 | <0.001 | |||
| Yes | 5.2 (7.4) | 407 (39.7) | 326 (31.8) | |||
| No | 6.1 (8.2) | 142 (44.7) | 156 (49.1) | |||
| Cognition | 0.15 | 0.19 | 0.004 | |||
| No dementia | 5.3 (7.5) | 498 (40.4) | 429 (34.8) | |||
| Probable dementia | 6.4 (8.7) | 51 (46.8) | 53 (48.6) | |||
| Unpaid caregiving | 0.05 | 0.24 | <0.001 | |||
| No | 5.2 (7.3) | 390 (39.9) | 318 (32.6) | |||
| Yes | 6.0 (8.6) | 159 (43.4) | 164 (44.8) | |||
| Pain | 0.89 | 0.27 | 0.004 | |||
| None/mild | 5.4 (7.7) | 325 (42.2) | 252 (32.9) | |||
| Moderate/severe | 5.4 (7.6) | 224 (39.2) | 230 (40.2) | |||
| Depression | 0.002 | 0.10 | <0.001 | |||
| No | 5.0 (6.7) | 355 (39.3) | 282 (31.2) | |||
| Yes | 6.4 (9/3) | 194 (44.1) | 200 (45.5) |
NR, not reportable due to small sample size; ADL, activity of daily living; IADL, independent activity of daily living.
Forty percent of patients experienced in-hospital complications. Patients with heart failure (47.9% vs 37.7%, p<0.01) and end stage renal disease (49.4% vs 39.8%, p=0.02) were significantly more likely to have in-hospital complications compared to those without the respective serious illness.
Our study population included 861 patients (64.1%) discharged to a non-home destination. Significant differences in rates of non-home discharge were seen between patients who were and were not vulnerable elders (41.3% vs 28.3, p<0.001) as well as those with and without advanced cancer (24.4% vs 38.3%, p<0.001), cirrhosis (not reportable, p=0.005), dementia (60.3% vs 33.3%, p<0.001), and frailty (49.8% vs 26.6%, p<0.001).
In unadjusted models, unpaid caregiving was significantly associated with non-home discharge (OR 1.7, 95% confidence interval (CI) 1.3–2.2, p<0.01) (Table 3), but not hospital days (incidence rate ratio (IRR) 1.1, 95% CI 1.0–1.4, p=0.12) or in-hospital complications (odds ratio (OR) 1.2, 95% CI 0.9–1.5, p=0.27). After adjusting for covariates, unpaid caregiving was not independently associated with any outcome. With the addition of an interaction term between marital status and unpaid caregiving in sensitivity analysis, the interaction between the category of not married and living alone and unpaid caregiving was significant only for the outcome of non-home discharge (p<0.01).
Table 3.
Associations between preoperative caregiving, pain, and depression and in-hospital outcomes among older seriously ill surgical patients using HRS-Medicare data from 2008–2018
| Incidence Rate Ratio (95% CI) | ||||
|---|---|---|---|---|
| Hospital Days | ||||
| Unadjusted | p-value | Adjusted | p-value | |
| Unpaid caregiving | 1.1 (1.0, 1.4) | 0.12 | 0.9 (0.7, 1.2) | 0.62 |
| Moderate/severe pain | 1.0 (0.8, 1.2) | 0.89 | 1.0 (0.8, 1.1) | 0.57 |
| Depression (CES-D ≥3) | 1.3 (1.1, 1.5) | 0.01 | 1.2 (1.0, 1.4) | 0.05 |
| Odds Ratio (95% CI) | ||||
| In-hospital complications | ||||
| Unadjusted | p-value | Adjusted | p-value | |
| Unpaid caregiving | 1.2 (0.9–1.5) | 0.27 | 0.9 (0.6–1.5) |
0.79 |
| Moderate/severe pain | 0.9 (0.7–1.1) | 0.29 | 0.9 (0.7–1.1) | 0.26 |
| Depression (CES-D ≥3) | 1.2 (1.0–1.6) | 0.11 | 1.1 (0.8–1.5) | 0.44 |
| Non-home discharge | ||||
| Unadjusted | p-value | Adjusted | p-value | |
| Unpaid caregiving | 1.7 (1.3–2.2) | <0.001 | 0.6 (0.4–1.1) | 0.10 |
| Moderate/severe pain | 1.4 (1.1–1.8) | 0.01 | 1.2 (1.0–1.6) | 0.11 |
| Depression (CES-D ≥3) | 1.8 (1.4–2.4) | <0.001 | 1.6 (1.2–2.1) | 0.003 |
Adjusted for age, sex, race/ethnicity, education, annual household wealth, marital status, functional dependence, cognition, Charlson comorbidity index, and time between surgical admission and preceding HRS survey
Baseline pain was not associated with increased hospital days (IRR 1.0, 95% CI 0.8–1.2, p=0.89) or in-hospital complications (OR 0.9, 95% CI 0.7–1.1, p=0.29) in unadjusted regression. These associations did not become significant after adjusting for covariates. While baseline pain was independently associated with non-home discharge in an unadjusted model (OR 1.4, 95% CI 1.1–1.8, p=0.01), this relationship was no longer statistically significant in adjusted multivariable analysis (OR 1.2, 95% CI 1.0–1.6, p=0.11).
Baseline depression was significantly associated with increased hospital days (IRR 1.3, 95% CI 1.1–1.5, p=0.01) and non-home discharge (OR 1.8, 95% CI 1.4–2.4, p<0.01) in unadjusted analyses. These associations held in adjusted regression for non-home discharge (OR 1.6, 95% CI 1.2–2.1, p<0.01) but not hospital days (IRR 1.2, 95% CI 1.0–1.4, p=0.05). Meanwhile, it was not significantly associated with in-hospital complications in either unadjusted (OR 1.2, 95% CI 1.0–1.6, p=0.11) or adjusted (p=0.44) regression.
In additional sensitivity analyses, functional dependence was a significant mediator between pain and non-home discharge. Adjusted analysis excluding functional dependence found pain to be independently associated with non-home discharge. Meanwhile, time between surgery and HRS survey did not have significant interaction with unpaid caregiving, pain, or depression.
Discussion:
We applied the definition of serious illness in surgery to a national cohort using HRS data linked to Medicare claims and found that older seriously ill adults undergoing elective surgery had high unpaid caregiving needs, pain, and depression at baseline. The stress of surgery heightens the vulnerability of older adults with serious illness,20 and the preoperative evaluation is an important opportunity for symptom and caregiver assessments to facilitate improving symptoms and transitions of care.40,41 The value of preoperative palliative symptom assessment has been recognized as a quality indicator to improve surgical care for seriously ill patients and led to its inclusion as a required vulnerability screen for high-risk older patients in the American College of Surgeons Geriatric Surgery Verification program standards.23,42 Our findings bolster the argument for palliative care integration throughout surgical episodes of care for these patients.
Our study corroborates previous studies demonstrating high symptom burden in seriously ill patients and underscores the importance of symptom assessment, specifically assessment of pain and depression, in a surgical population. A retrospective study of data from the Palliative Care Quality Network found that 30% of patients seen by inpatient palliative care teams self-reported moderate to severe pain and that improved pain assessment was associated with significantly shortened hospital length of stay.43 With greater than two-fifths of patients reporting moderate to severe pain before surgery in our study, we identified a population that would benefit from palliative care interventions, including effective or improved pain management. Though we did not find a significant association between baseline pain and post-acute hospital days, we argue that preoperative pain assessment would still be beneficial to patients, anticipating potential exacerbation incited by surgery, developing intraoperative and postoperative management plans, to improve patient experience and outcomes. Furthermore, this study echoes others which have shown that pain and depression occur together as frequently as 27–65%,38 and worse postoperative pain and increased length of stay in surgical patients with preoperative depression,15,16 Interventions to address depression in the preoperative setting have potential to decrease perioperative healthcare utilization.44
Our analysis provides novel insight into the caregiving needs of older seriously ill surgical patients. Using the National Study on Caregiving, it was estimated that over 75% of seriously ill patients, defined as three or more chronic diseases and a functional limitation in their ability to care for themselves or perform routine daily tasks, received unpaid caregiving.45 In comparison, 27% of our seriously ill surgical patients had unpaid caregivers at baseline. According to the American Association of Retired Persons report, Caregiving in the United States 2020, 29% and 49% of family members and friends providing high-intensity caregiving reported high physical and emotional strain respectively.46 Additionally, caregiving resulted in one or more financial setbacks in nearly half of unpaid caregivers, and high-intensity caregivers have increased out-of-pocket spending compared to those providing lower-intensity caregiving.47 Our findings highlight the need to assess caregivers for caregiver burden and preparedness for anticipated post-discharge tasks.48 While unpaid caregiving was not significantly associated with non-home discharge in adjusted analysis, the significant interaction between unpaid caregiving and those who were not married and living alone found in sensitivity analysis suggests that patients relying on unpaid caregivers outside the home may be more likely to require postoperative care at a facility. Caregiver burden may be worsened following surgery and presents an opportunity for palliative care, which has been associated with improved caregiver satisfaction.49 Further study is warranted to assess for increased caregiving needs after surgery, impact on caregiver quality of life, and need for reinforcement by paid caregivers for patients whose postoperative needs are not able to be met by their existing caregivers. High levels of preoperative caregiving needs in seriously ill surgical patients present opportunities for health policy changes, such as the Recognize, Assist, Include, Support, and Engage Family Caregivers Act, specifically around the time of surgery.50
Our study has several limitations. First, our national cohort of older adults with linked fee-for-service Medicare claims may not be representative and may underestimate symptom prevalence and caregiving needs for all seriously ill patients seeking elective surgery.51 Yet higher levels of symptom burden and caregiving needs would only support efforts to perform preoperative symptom and caregiver assessments. Medicare claims data are subject to errors in coding. Functional dependence is a potential confounder in the association between unpaid caregiving and outcomes due to high observed rates of functional dependence in this cohort and significant interaction between functional dependence and caregiving. We therefore recognize that our findings may only be generalizable to seriously ill patients. Next, we categorized unpaid caregiving as a binary variable due to the positive skew in distribution of unpaid caregiving hours seen in our cohort. We interpret these findings conservatively and acknowledge that caregiving hours as a continuous variable could yield additional information about its association with outcomes. Also, the timing of the HRS survey varied relative to the timing of surgical admission by patient; however, we accounted for this variation by controlling for time between preceding HRS survey and elective surgery in our adjusted analyses, as seen in previous studies.30 As we did not correct for multiple comparisons within our cohort, we exercise caution in the interpretation of our reported p-values across several outcomes. Lastly, evaluation of baseline caregiving needs, pain, and depression does not fully capture the potential palliative care needs of seriously ill surgical patients, such as other physical symptoms, including nausea, constipation, shortness of breath, and fatigue, or spiritual distress, which are not well-assessed through HRS surveys. However, the high prevalence of caregiving needs, pain, and depression in this cohort suggest that these are fruitful targets for palliative care, even in resource-constrained environments. Future work will consider postoperative caregiving needs, pain, and depression in older seriously ill patients following elective surgery and compare rates between pre- and postoperative groups.
In conclusion, we identified high unpaid caregiving needs and high prevalence of pain and depression in older seriously ill patients before elective surgery. Baseline depression was found to be significantly associated with discharge to a non-home destination. Our findings support preoperative pain, depression and caregiver assessments to address underlying palliative care needs and potentially improve perioperative outcomes.
Figure 2.
Adjusted associations between preoperative caregiving, pain, and depression and in-hospital and post-acute outcomes among older seriously ill surgical patients using HRS-Medicare data from 2008–2018. Outcomes include hospital days (A), in-hospital complications (B), and non-home discharge (C).
Key Points:
Using a national retrospective cohort, we identified older adults with serious illness undergoing major elective surgery and identified high unpaid caregiving needs and high prevalence of pain and depression prior to surgery
Regression analysis demonstrated that baseline depression was associated with more hospital days and non-home discharge while baseline pain and unpaid caregiving needs were not associated with in-hospital or post-acute outcomes
Why does this matter?
Over one million older seriously ill adults undergo major surgery annually, and many rely on unpaid caregivers to assist with medical tasks and activities of daily living. Seriously ill patients benefit from palliative care but palliative care needs in surgical patients and their caregivers are poorly defined. Understanding baseline caregiving needs, pain, and depression may inform palliative care interventions and policy changes for seriously ill surgical patients and their families.
Acknowledgments:
Sponsor’s Role: This study was supported by NIH R01AG070252.
Funding Source:
Dr. Cooper is supported by NIH R01AG070252.
This work was presented at American College of Surgeons Clinical Congress, San Diego, CA, September 2022.
Footnotes
Conflicts of Interest: Dr. Cooper is supported by NIH R01AG070252.
References:
- 1.Kelley AS, Covinsky KE, Gorges RJ, et al. Identifying Older Adults with Serious Illness: A Critical Step toward Improving the Value of Health Care. Health Serv Res. 2017;52(1):113–131. doi: 10.1111/1475-6773.12479 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.National Consensus Project for Quality Palliative Care. Clinical Practice Guidelines for Quality Palliative Care. 4th ed.; 2018.
- 3.Spetz J, Dudley N. Consensus-Based Recommendations for an Adequate Workforce to Care for People with Serious Illness. J Am Geriatr Soc. 2019;67(S2):S392–S399. doi: 10.1111/jgs.15938 [DOI] [PubMed] [Google Scholar]
- 4.Steinhauser KE, Christakis NA, Clipp EC, McNeilly M, McIntyre L, Tulsky JA. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA. 2000;284(19):2476–2482. doi: 10.1001/jama.284.19.2476 [DOI] [PubMed] [Google Scholar]
- 5.Zivin K, Llewellyn DJ, Lang IA, et al. Depression among older adults in the United States and England. American Journal of Geriatric Psychiatry. 2010;18(11):1036–1044. doi: 10.1097/JGP.0b013e3181dba6d2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kamal AH, Nipp RD, Bull J, Stinson CS, Abernethy AP. Symptom burden and performance status among community-dwelling patients with serious illness. J Palliat Med. 2015;18(6):542–544. doi: 10.1089/jpm.2014.0381 [DOI] [PubMed] [Google Scholar]
- 7.Patel K v., Guralnik JM, Phelan EA, et al. Symptom Burden Among Community-Dwelling Older Adults in the United States. J Am Geriatr Soc. 2019;67(2):223–231. doi: 10.1111/jgs.15673 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Pandya C, Magnuson A, Flannery M, et al. Association Between Symptom Burden and Physical Function in Older Patients with Cancer. J Am Geriatr Soc. 2019;67(5):998–1004. doi: 10.1111/jgs.15864 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.McKee KY, Kelly A. Management of Grief, Depression, and Suicidal Thoughts in Serious Illness. Medical Clinics of North America. 2020;104(3):503–524. doi: 10.1016/j.mcna.2020.01.003 [DOI] [PubMed] [Google Scholar]
- 10.Murali KP, Yu G, Merriman JD, Vorderstrasse A, Kelley AS, Brody AA. Latent Class Analysis of Symptom Burden Among Seriously Ill Adults at the End of Life. Nurs Res. 2021;70(6):443–454. doi: 10.1097/NNR.0000000000000549 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Salanitro AH, Hovater M, Hearld KR, et al. Symptom burden predicts hospitalization independent of comorbidity in community-dwelling older adults. J Am Geriatr Soc. 2012;60(9):1632–1637. doi: 10.1111/j.1532-5415.2012.04121.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sheppard KD, Brown CJ, Hearld KR, et al. Symptom burden predicts nursing home admissions among older adults. J Pain Symptom Manage. 2013;46(4):591–597. doi: 10.1016/j.jpainsymman.2012.10.228 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wang L, Guyatt GH, Kennedy SA, et al. Predictors of persistent pain after breast cancer surgery: A systematic review and meta-analysis of observational studies. CMAJ. 2016;188(14):E352–E361. doi: 10.1503/cmaj.151276 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yang MMH, Hartley RL, Leung AA, et al. Preoperative predictors of poor acute postoperative pain control: A systematic review and meta-analysis. BMJ Open. 2019;9(4). doi: 10.1136/bmjopen-2018-025091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Oduyale OK, Eltahir AA, Stem M, et al. What Does a Diagnosis of Depression Mean for Patients Undergoing Colorectal Surgery? Journal of Surgical Research. 2021;260:454–461. doi: 10.1016/j.jss.2020.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chen J, Li JY, Tian GH, et al. A national snapshot of the impact of clinical depression on post-surgical pain and adverse outcomes after anterior cervical discectomy and fusion for cervical myelopathy and radiculopathy: 10-year results from the US Nationwide Inpatient Sample. PLoS One. 2021;16(10 October). doi: 10.1371/journal.pone.0258517 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Radbruch L, de Lima L, Knaul F, et al. Redefining Palliative Care—A New Consensus-Based Definition. J Pain Symptom Manage. 2020;60(4):754–764. doi: 10.1016/j.jpainsymman.2020.04.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Temel JS, Greer JA, Muzikansky A, et al. Early Palliative Care for Patients with Metastatic Non-Small-Cell Lung Cancer A Bs Tr Ac t. Vol 363.; 2010. [DOI] [PubMed] [Google Scholar]
- 19.Ferre AC, Demario BS, Ho VP. Narrative review of palliative care in trauma and emergency general surgery. Ann Palliat Med. 2022;11(2):936–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lee KC, Walling AM, Senglaub SS, Kelley AS, Cooper Z. Defining Serious Illness Among Adult Surgical Patients. J Pain Symptom Manage. 2019;58(5):844–850.e2. doi: 10.1016/j.jpainsymman.2019.08.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kelly MT, Sturgeon D, Harlow AF, Jarman M, Weissman JS, Cooper Z. Using Medicare Data to Identify Serious Illness in Older Surgical Patients. J Pain Symptom Manage. 2020;60(2):e101–e103. doi: 10.1016/j.jpainsymman.2020.04.002 [DOI] [PubMed] [Google Scholar]
- 22.Lee KC, Senglaub SS, Walling AM, Mosenthal AC, Cooper Z. Quality Measures in Surgical Palliative Care: Adapting Existing Palliative Care Measures to Improve Care for Seriously Ill Surgical Patients. Ann Surg. 2019;269(4):607–609. doi: 10.1097/SLA.0000000000003136 [DOI] [PubMed] [Google Scholar]
- 23.American College of Surgeons. Geriatric Surgery Verification Quality Improvement Program: Optimal Resources for Geriatric Surgery 2019 Standards.; 2019.
- 24.Centers for Medicare and Medicaid Research Data Access Center. Health and Retirement Survey - Medicare Linked Data. https://resdac.org/cms-data/files/hrs-medicare.
- 25.Agency for Healthcare Research and Quality. HCUP Central Distributor SASD Description of Data Elements - All States. Healthcare Cost and Utilization Project (HCUP).
- 26.Qian Y, Chen X, Tang D, Kelley AS, Li J. Prevalence of Memory-Related Diagnoses among U.S. Older Adults with Early Symptoms of Cognitive Impairment. Journals of Gerontology - Series A Biological Sciences and Medical Sciences. 2021;76(10):1846–1853. doi: 10.1093/gerona/glab043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Centers for Medicare and Medicaid Research Data Assistance Center. CMS Cell Size Suppression Policy. https://resdac.org/articles/cms-cell-size-suppression-policy.
- 28.Schane RE, Walter LC, Dinno A, Covinsky KE, Woodruff PG. Prevalence and risk factors for depressive symptoms in persons with chronic obstructive pulmonary disease. J Gen Intern Med. 2008;23(11):1757–1762. doi: 10.1007/s11606-008-0749-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ornstein KA, Aldridge MD, Garrido MM, Gorges R, Meier DE, Kelley AS. Association between hospice use and depressive symptoms in surviving spouses. JAMA Intern Med. 2015;175(7):1138–1146. doi: 10.1001/jamainternmed.2015.1722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Cooper Z, Mitchell SL, Gorges RJ, Rosenthal RA, Lipsitz SR, Kelley AS. Predictors of Mortality Up to 1 Year after Emergency Major Abdominal Surgery in Older Adults. J Am Geriatr Soc. 2015;63(12):2572–2579. doi: 10.1111/jgs.13785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ferrante LE, Pisani MA, Murphy TE, Gahbauer EA, Leo-Summers LS, Gill TM. Functional trajectories among older persons before and after critical illness. JAMA Intern Med. 2015;175(4):523–529. doi: 10.1001/jamainternmed.2014.7889 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Tsuda Y, Yasunaga H, Horiguchi H, Ogawa S, Kawano H, Tanaka S. Association between dementia and postoperative complications after hip fracture surgery in the elderly: analysis of 87,654 patients using a national administrative database. Arch Orthop Trauma Surg. 2015;135(11):1511–1517. doi: 10.1007/s00402-015-2321-8 [DOI] [PubMed] [Google Scholar]
- 33.Cooper Z, Lilley EJ, Bollens-Lund E, et al. High Burden of Palliative Care Needs of Older Adults During Emergency Major Abdominal Surgery. J Am Geriatr Soc. 2018;66(11):2072–2078. doi: 10.1111/jgs.15516 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kugelman DN, Haglin JM, Carlock KD, Konda SR, Egol KA. The association between patient education level and economic status on outcomes following surgical management of (fracture) non-union. Injury. 2019;50(2):344–350. doi: 10.1016/j.injury.2018.12.013 [DOI] [PubMed] [Google Scholar]
- 35.Herrera-Escobar JP, Seshadri AJ, Rivero R, et al. Lower education and income predict worse long-term outcomes after injury. Journal of Trauma and Acute Care Surgery. 2019;87(1):104–110. doi: 10.1097/TA.0000000000002329 [DOI] [PubMed] [Google Scholar]
- 36.Tang VL, Jing B, Boscardin J, et al. Association of functional, cognitive, and psychological measures with 1-year mortality in patients undergoing major surgery. JAMA Surg. 2020;155(5):412–418. doi: 10.1001/jamasurg.2020.0091 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Chen L, Au E, Saripella A, et al. Postoperative outcomes in older surgical patients with preoperative cognitive impairment: A systematic review and meta-analysis. J Clin Anesth. 2022;80:110883. doi: 10.1016/j.jclinane.2022.110883 [DOI] [PubMed] [Google Scholar]
- 38.Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch Intern Med. 2003;163(20):2433–2445. doi: 10.1001/archinte.163.20.2433 [DOI] [PubMed] [Google Scholar]
- 39.Etminan M, Brophy JM, Collins G, Nazemipour M, Mansournia MA. To Adjust or Not to Adjust: The Role of Different Covariates in Cardiovascular Observational Studies. Am Heart J. 2021;237:62–67. doi: 10.1016/j.ahj.2021.03.008 [DOI] [PubMed] [Google Scholar]
- 40.Lilley EJ, Khan KT, Johnston FM, et al. Palliative care interventions for surgical patients a systematic review. JAMA Surg. 2016;151(2):172–183. doi: 10.1001/jamasurg.2015.3625 [DOI] [PubMed] [Google Scholar]
- 41.Yefimova M, Aslakson RA, Yang L, et al. Palliative Care and End-of-Life Outcomes Following High-risk Surgery. In: JAMA Surgery. Vol 155. American Medical Association; 2020:138–146. doi: 10.1001/jamasurg.2019.5083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Lee KC, Walling AM, Senglaub SS, et al. Improving Serious Illness Care for Surgical Patients: Quality Indicators for Surgical Palliative Care. Ann Surg. Published online June 3, 2020:1–7. doi: 10.1097/SLA.0000000000003894 [DOI] [PubMed] [Google Scholar]
- 43.Bischoff KE, O’Riordan DL, Fazzalaro K, Kinderman A, Pantilat SZ. Identifying Opportunities to Improve Pain Among Patients With Serious Illness. J Pain Symptom Manage. 2018;55(3):881–889. doi: 10.1016/j.jpainsymman.2017.09.025 [DOI] [PubMed] [Google Scholar]
- 44.Elsamadicy AA, Adogwa O, Cheng J, Bagley C. Pretreatment of Depression before Cervical Spine Surgery Improves Patients’ Perception of Postoperative Health Status: A Retrospective, Single Institutional Experience. World Neurosurg. 2016;87:214–219. doi: 10.1016/j.wneu.2015.11.067 [DOI] [PubMed] [Google Scholar]
- 45.Kelley AS, Bollens-Lund E. Identifying the Population with Serious Illness: The “Denominator” Challenge. J Palliat Med. 2018;21(S2):S7–S16. doi: 10.1089/jpm.2017.0548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.AARP, National Alliance for Caregiving. Caregiving in the U.S. 2020.; 2020. www.greenwaldresearch.com
- 47.AARP. Caregiving Out-of-Pocket Costs Study.; 2021.
- 48.Bell JF, Whitney RL, Young HM. Family Caregiving in Serious Illness in the United States: Recommendations to Support an Invisible Workforce. J Am Geriatr Soc. 2019;67(S2):S451–S456. doi: 10.1111/jgs.15820 [DOI] [PubMed] [Google Scholar]
- 49.Kavalieratos D, Corbelli J, Zhang D, et al. Association between palliative care and patient and caregiver outcomes: A systematic review and meta-analysis. JAMA - Journal of the American Medical Association. 2016;316(20):2104–2114. doi: 10.1001/jama.2016.16840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Library of Congress. H.R.3759 – 115th Congress (2017–2018): RAISE Family Caregivers Act. Congress.gov.
- 51.Kelley AS, Hanson LC, Ast K, et al. The Serious Illness Population: Ascertainment via Electronic Health Record or Claims Data. J Pain Symptom Manage. 2021;62(3):e148–e155. doi: 10.1016/j.jpainsymman.2021.04.012 [DOI] [PMC free article] [PubMed] [Google Scholar]