Preoperative Depressive Symptoms associated with Poor Functional Recovery after Surgery

Abstract

Background/Objectives

Depression screening and treatment for older adults are recommended in Age-Friendly Health Systems. Few studies have evaluated the association between depressive symptoms and post-operative functioning. We aim to determine the association between varying levels of depressive symptoms in the preoperative setting with post-operative functional recovery.

Design

Prospective cohort study

Setting

Two academic hospitals in Boston, MA.

Participants

Surgical patients aged ≥ 70 years (N=560).

Measurements

Participants were assessed preoperatively and 1 year post-operatively. Preoperative evaluation included the 15-item short-form Geriatric Depression Scale (GDS). Results were categorized: low (GDS 0–1), moderate (2–5), and high (6–15) symptom burden. Primary outcome was one-year instrumental activities of daily living (IADL) functional decline. Secondary outcomes included hospital stay >5 days, discharge to post-acute care(PAC) facility, readmission within 30 days.

Results

Mean participant age was 76.6±5 years, 58% were women, 81% underwent an orthopedic operation, 13% gastrointestinal, 6% vascular; 13% had functional decline at 1 year after their operation (by symptom burden: low – 5.5%, moderate – 14.8%, high – 38.6%). After adjusting for age, gender, and comorbidity, those with moderate or high depressive symptoms demonstrated greater odds of functional decline at 1 year compared to those with a low symptom burden (moderate: adjusted odds ratio [AOR] 2.7, 95%CI 1.3–5.3; high: AOR 9.3, CI 4.2–20.6), discharge to PAC facility (moderate: AOR 1.7, CI 1.2–2.6; high: AOR 2.7, CI 1.4–5.1), but demonstrated no significant association with 30-day readmission or hospital length of stay >5 days.

Conclusion

Greater burden of preoperative depressive symptoms is associated with increased likelihood of functional decline at 1 year after surgery and of discharge to PAC facility. Preoperative assessment of the burden of depressive symptoms in older adults undergoing elective surgery may be helpful in identifying patients at high risk of poor outcomes.

INTRODUCTION

In the U.S., adults 65 years and older undergo over 19 million operations a year¹ and represent over 50% of all surgical admissions.² Over 20% of older adults undergoing major surgery experience life-altering complications,³ which can precipitate loss of function and independence.^4–8 Almost half (43%) of older adults who undergo surgery experience an immediate postoperative functional decline.⁹ Most older adults report that they value function and quality of life over quantity.¹⁰ Post-operative functional outcome is an important patient-centered outcome that should be taken into account when counseling a patient on surgical decision-making; yet, few studies have evaluated important predictors of long-term functional outcomes of surgery.

Among older Americans, 71 years and older, 8–16% exhibit clinically significant depressive symptoms.¹¹ Cardiovascular studies support that patients with preoperative depression have a significantly increased likelihood of morbidity and mortality^12–14 in the months and years following surgery.^15,16 Notably, patients who were depressed prior to their coronary artery bypass graft (CABG) had poorer post-operative functional recovery.^15,17 Preoperative depression has also been associated with longer postoperative length of stay and unplanned hospital readmission in older adults who underwent CABG.^15,18 While the cardiology field is increasingly interested in assessing and treating depression in older CABG patients to improve post-operative outcomes, the assessment for and treatment of depression in older non-cardiac surgery patients to improve outcomes has largely been missed because evidence is lacking. In 2012, the American College of Surgeon’s (ACS) and American Geriatrics Society’s (AGS) published guidelines for Optimal Preoperative Assessment of the Older Adults supporting preoperative depression screening in older adults and depression screening is now emphasized again with the Aging-Friendly Healthcare Systems Initiative. Evidence for preoperative depression screenings is rooted in the association of depression with delirium with few supporting studies focused on the older, more vulnerable surgical population and the patient-centered outcome of a long-term post-operative functional recovery. The association between depression and post-operative functional recovery has been primarily featured in the orthopedic and cardiac surgery literature, conclusions differed^{19,20, 21}. Additional evidence justifying preoperative depression screening and treatment is needed.

This study’s objectives were (1) to determine whether preoperative depressive symptoms were associated with increased risk of functional decline at one-year and (2) to evaluate the association between preoperative depressive symptoms and adverse postoperative outcomes (i.e., prolonged hospital stay, discharge to a post-acute care facility, 30-day readmission). We hypothesized that preoperative depressive symptoms would be associated with increased risk of 1-year functional decline and worse postoperative outcomes.

METHODS

Study population

The Successful Aging after Elective Surgery (SAGES) study is a prospective cohort study of older adults undergoing major elective surgery. The study design and methods have been previously described in detail.^22–24 In brief, eligible participants were at least 70 years of age and underwent an elective surgery with an anticipated length of stay of at least 3 days at one of two Harvard-affiliated academic medical centers. Elective surgery included: orthopedic, gastrointestinal, and vascular procedures (Supplementary Material S1). Exclusion criteria assessed during the screening process included evidence of dementia, delirium, hospitalization within 3 months, terminal condition, legal blindness, severe deafness, history of schizophrenia or psychosis, and history of alcohol abuse. A total of 560 patients met all eligibility criteria and were enrolled between June 8, 2010 and August 8, 2013. Written informed consent for study participation was obtained from all participants according to procedures approved by the institutional review boards of Beth Israel Deaconess Medical Center and Brigham and Women’s Hospital, the two study hospitals, and Hebrew SeniorLife, the study coordinating center, all located in Boston, Massachusetts. For this study, 558 patients who completed the Geriatric Depression Scale (GDS) prior to surgery were included in the final analyses. Two SAGES participants with an incomplete GDS were not included.

Data Source

Baseline comorbidities and discharge location were collected from patients’ medical records. Readmission was collected from interviews conducted in patients’ homes at 30 days post-surgery. Functional, cognitive, and psychosocial assessments were collected from patient interviews in their homes in the preoperative period and at one-year post-surgery.

Predictors

The main predictor of interest was the number of depressive symptoms, measured using the 15-item short-form GDS, a brief screening instrument for depression in older adults²⁵, and administered by trained interviewers at the baseline preoperative assessment. The GDS is designed to have little overlap with the more somatic symptoms of depression commonly associated with medical illness in older persons. The number of positive depressive symptoms was categorized into one of three categories reflecting the burden of depressive symptoms: 0 to 1 indicating low burden, 2 to 5 indicating moderate burden, and 6 to 15 indicating high burden. Based on previous studies, a GDS score of 6+ is considered as the threshold for clinically significant depression.²⁵

Outcomes

The primary outcome of interest was functional decline at one year after surgery, measured by comparing the preoperative score on instrumental activities of daily living (IADL) with the one-year postoperative score. IADLs were measured by the Lawton Scale, which assesses the ability to perform 7 complex activities (using the telephone, grocery shopping, using transportation, cooking, housekeeping, taking medications, and handling finances) with a total score of 0–14 (14 worst). One example of the questions is, “One month ago, did you go shopping for groceries… without help, with some help, or completely unable to do any shopping in a store”^26. The one-year time frame was chosen based on prior research using this interval^{27, 28} and prior work demonstrating that recovery of IADL function can take up to 6 months to 1 year after a major operation²⁹. IADLs were chosen as opposed to activities of daily living (ADLs) given that nearly all patients were independent in ADLs at baseline. Severe decline affecting ADLs was not observed.

In addition, we evaluated three secondary outcomes. Two outcomes were evaluated at discharge: length of stay of more than 5 days and discharge to a post-acute care facility. A length of stay longer than 5 days was chosen as 5 days was the median for SAGES; additionally, this outcome has been used in previously published SAGES studies.³⁰ The third outcome, readmission within 30 days of discharge, was assessed at the 1-month follow-up interview, with the following question: “Since we spoke to you the last time on (last interview date), have you been to a hospital at least overnight”. A 30-day readmission outcome was chosen because the Centers for Medicare and Medicaid Services use it as a hospital quality of care indicator.³¹

Other study variables

Demographic and health data were obtained from patient interviews and medical record review. The following characteristics were assessed as potential confounders: age, gender, and Charlson Comorbidity Index (CCI). CCI was calculated based on diagnoses abstracted from medical records.³² The Modified Mini-Mental State (3MS) Examination, a widely used screening for dementia, scaled from 0–100, was used as a descriptive variable.³³ A score of ≤84 indicates cognitive impairment, and 84–100 is considered cognitively intact. The Activities of Daily Living (ADL), measured using the Katz Scale by the following activities: bathing, grooming, dressing, feeding, transfers (bed to chair), toileting, and walking, was also used as a descriptive variable.³⁴

Analysis

We compared patients across the three categories of symptom burden (low, moderate, and high) as determined by GDS score. Bivariate differences between the participant characteristics and depressive symptom groups were assessed using Chi square test for binary variables, and Analysis of Variance (ANOVA) test for numeric variables. To control for potential confounders, we estimated multivariable logistic regression models for each of the four outcomes of interest, controlling for age (continuous), gender, and CCI (continuous) as covariates. We had no missing data for the covariates. Further, since 81% of the cohort underwent orthopedic procedures, a restricted analysis of orthopedic surgeries was conducted for comparison with the analysis of the entire cohort.

We excluded those patients with a missing outcome from the respective analysis: a total of 11 patients did not complete the 1-month interview, 7 who dropped out, 3 who refused or were unobtainable, and 1 who died, and were therefore excluded from the 30-day readmission analysis, and 47 patients did not complete the 12 month interview, 17 who dropped out, 25 who refused or unobtainable, 5 who died, and were therefore excluded from the 1-year IADL decline outcome analysis. Additionally, we conducted multivariable regression models in sensitivity analyses to assess the impact of delirium, type of surgery, and of deaths on our results (see results section for details). First, because delirium has been shown to be associated with delayed functional recovery, we fit logistic regression models which adjusted for delirium in addition to age, gender and CCI. Second, because the underlying disease leading to the type of surgery may affect the presence of preoperative depression and the likelihood one would have post-operative functional decline, we performed logistic regression models which adjusted for type of surgery, as well. Third, our primary outcome (functional decline) and one secondary outcome (30-day readmission) were assessed following discharge (12 months and 1 month, respectively), resulting in some patients being lost to follow up, and we have no information on their functional status or readmission. One possible reason for the loss to follow up is death. Excluding patients that died before the follow up interview could possibly lead to biased results, as death itself can be indicative of the outcome. Our second sensitivity analysis included logistic regressions with composite outcomes (IADL and/or death, and 30-day readmission and/or death).

All analyses were performed using SAS Version 9.4 and Stata/MP 14.2 software. A value of p<0.05 was considered the threshold for statistical significance.

RESULTS

Description of the cohort

In our final cohort of 558 older surgical patients, mean age at the time of the surgery 77±5 years, 58% of the patients were female, 93% were white, 59% were married, and 30% lived alone (Table 1). 46% of the subjects had a CCI of 0. Almost all (93%) had a 3MS score of 85–100, indicating normal cognition. 81% of the patients underwent an orthopedic procedure, with the remainder undergoing a gastrointestinal procedure (13%) or a vascular procedure (6%). Most patients reported being able to independently (i.e., without needing help) complete ADLs (93%) and IADLs (73%) at baseline. There were 256 (46%) patients who reported low, 233 (42%) patients who reported moderate, and 69 (12%) patients who reported a high burden of depressive symptoms.

Table 1.

Participant Characteristics at Baseline

Characteristic	Full Sample (N=558)	GDS 0–1 (N=256)	GDS 2–5 (N=233)	GDS 6–15 (N=69)	p-valuea
Age - mean (SD)	76.6 (5.0)	76 (4.7)	77 (5.1)	77.8 (5.1)	0.0105
Female – n (%)	325 (58.2)	136 (53.1)	146 (62.7)	43 (62.3)	0.0781
Non-white – n (%)	42 (7.5)	12 (4.7)	22 (9.4)	8 (11.6)	0.0541
Education - mean years (SD)	15 (2.9)	15.5 (2.8)	14.5 (2.9)	14.5 (3.4)	< 0.001
Married – n (%)	331 (59.3)	160 (62.5)	135 (57.9)	36 (52.2)	0.2571
Lives Alone – n (%)	167 (29.9)	73 (28.5)	66 (28.3)	28 (40.6)	0.1187
3MS < 84 – n (%)b	36 (6.5)	12 (4.7)	14 (6.0)	10 (14.5)	0.0108
Charlson Comorbidity Index – n (%)
0	256 (45.9)	131 (51.2)	104 (44.6)	21 (30.4)	0.0295
1	138 (24.7)	58 (22.7)	61 (26.2)	19 (27.5)
2+	164 (29.4)	67 (26.2)	68 (29.2)	29 (42.0)
Dependence in Activities of Daily Living (ADL)c - n (%)
ADL Independent	516 (92.5)	248 (96.9)	210 (90.1)	58 (84.1)	< 0.001
Dependence in Instrumental Activities of Daily Living (IADL)d - n (%)
IADL Independent	407 (72.9)	219 (85.5)	151 (64.8)	37 (53.6)	< 0.001
Surgery Type – n (%)
Orthopedice	453 (81.2)	206 (80.5)	188 (80.7)	59 (85.5)	0.8571
Gastrointestinalf	71 (12.7)	35 (13.7)	30 (12.9)	6 (8.7)
Vascularg	34 (6.1)	15 (5.9)	15 (6.4)	4 (5.8)

^aThe p-values are for ANOVA tests for continuous variables and Chi-square statistical tests for categorical variables. For binary variables, these values signify statistical differences between GDS groups for the given row variable. For variables with more than two categories (Charlson Comorbidity Index and Surgery types), the p-value is testing the distribution of GDS scores across all three rows.

^b1 missing. The Modified Mini-Mental State Test(3MS) was designed to assess general cognitive ability in the elderly. (71–84 = Cognitively Impaired, 85–100 = Cognitively Intact)²⁹

^cADL (Activities of Daily Living) are measured the Katz Scale by the following activities: bathing, grooming, dressing, feeding, transfers (bed to chair), toileting, and walking.³⁰ “ADL Independent” is defined by not needing help in completing any of the ADL activities. “ADL Dependent” is defined by needing help in one or more of the ADL activities.

^dIADL (Instrumental Activities of Daily Living) are measured by the Lawton Scale by the following activities: shopping, preparing meals, managing medications, using the telephone, housework, laundry, driving or using public transportation, and managing finances.²³ “

^eTotal hip or knee replacement; lumbar, cervical or sacral laminectomy

^fOpen colectomy; laparoscopic colectomy

^gLower extremity arterial bypass surgery; open abdominal aortic aneurysm repair; lower extremity amputation

Baseline patient characteristics by number of depressive symptoms

Subjects with a greater number of symptoms tend to be older, have fewer years of education, and have multimorbidity, as indicated by a CCI score of 2 or higher. In terms of cognitive performance, patients with greater symptom burden tended to score in the lower range (i.e.,71–84) of the 3MS test. Additionally, they tended to have higher levels of physical impairment at baseline, measured both by ADL and IADL dependence. Across the categories of symptom burden, no significant differences were seen in terms of gender, race, marital status, whether patient lives alone, or type of surgery.

Primary Outcome: 1-year Functional Decline

Thirteen percent of the cohort did not return to their previous preoperative function by 1-year following their surgery. Table 2 describes the rates of 1-year functional decline in IADL across the categories of depressive symptom burden. The rate of functional decline increased with greater symptom burden (5.5% vs. 14.8% vs. 38.6%, p <0.001). After adjusting for age, gender, and CCI, greater symptom burden was significantly associated with a higher risk of not returning to their preoperative level of functioning (moderate: adjusted odds ratio (AOR) 2.7, 95% CI: 1.3–5.3; high: 9.3, 95% CI 4.2–20.6), compared to individuals with low symptom burden. Figure 1 presents the adjusted predicted probabilities of 1-year functional decline by categories of symptom burden.

Table 2.

Association between 1-year Functional Decline, Hospital Length of Stay, Discharge to Post-Acute Care Facility, and 30-day Readmission with Preoperative Depressive Symptoms

	% outcome	Unadjusted Odds Ratio (95% CI)	Adjusted Odds Ratio (95% CI)a	p-value
IADLb Decline at 1 yearc
Total Cohort	67/511 (13.1%)
GDSd 0–1	13/238 (5.5%)	REF	REF
GDS 2–5	32/216 (14.8%)	3.0 (1.5,5.9)	2.7 (1.3,5.3)	0.005
GDS 6–15	22/57 (38.6%)	10.9 (5.0,23.6)	9.3 (4.2,20.6)	<0.001
Hospital length of stay > 5 days
Total Cohort	147/558 (26.3%)
GDS 0–1	57/256 (22.3%)	REF	REF
GDS 2–5	64/233 (27.5%)	1.3 (0.9,2.0)	1.2 (0.8,1.8)	0.44
GDS 6–15	26/69 (37.7%)	2.1 (1.2,3.7)	1.7 (0.9,3.0)	0.10
Discharge to post-acute care facility
Total Cohort	318/558 (57%)
GDS 0–1	121/256 (47.3%)	REF	REF
GDS 2–5	147/233 (63.1%)	1.9 (1.3,2.7)	1.7 (1.2,2.6)	0.006
GDS 6–15	50/69 (72.5%)	2.9 (1.6,5.3)	2.7 (1.4,5.1)	0.002
30 day readmissione
Total Cohort	67/547 (12.2%)
GDS 0–1	23/252 (9.1%)	REF	REF
GDS 2–5	34/230 (14.8%)	1.7 (1.0,3.0)	1.7 (0.9,3.0)	0.08
GDS 6–15	10/65 (15.4%)	1.8 (0.8,4.0)	1.7 (0.7,3.8)	0.23

^aAdjusted for age, gender, and Charlson Comorbidity Index

^bIADL (Instrumental Activities of Daily Living) are measured by the Lawton Scale by the following activities: shopping, preparing meals, managing medications, using the telephone, housework, laundry, driving or using public transportation, and managing finances.²³

^c47 subjects did not have a 12 month interview and were not included in this analysis

^d15-item short-form Geriatric Depression Scale (GDS) measures depressive symptoms. Results are categorized: low (GDS 0–1), moderate (2–5), and high (6–15) symptom burden.

^e11 subjects did not have a 1 month interview and were not included in this analysis

Figure 1.

displays patient outcomes organized by how patients scored on the Geriatric Depression Scale (GDS). The patients are divided in to three groups: low (GDS 0–1), moderate (2–5), and high (6–15) symptom burden. For each of these groups, the percentage reporting four different outcomes are displayed. These outcomes are functional decline at 1 year after surgery, hospital length of stay >5 days, discharge to post-acute care facility, and 30 day readmission. For all four outcomes, statistical significance of moderate and high symptom burden compared to low burden is indicated by “*” (p-value < 0.05), “**” (p-value < 0.01) and “***” (p-value < 0.005).

Other Adverse Post-Operative Outcomes

Category of symptom burden was not significantly associated with either hospitalization longer than 5 days (moderate: AOR 1.2, 95% CI: 0.8–1.8; high: 1.7, 95% CI 0.9–3.0), or readmission within 30 days (moderate: adjusted odds ratio (AOR) 1.7, 95% CI: 0.9–3.0; high: 1.7, 95% CI 0.7–3.8) in the adjusted analyses. Even after adjustment for age, gender, and CCI, greater symptom burden was associated with higher rates of discharge to a post-acute care facility (moderate: AOR: 1.7, 95% CI: 1.2–2.6; high: 2.7 95% CI: 1.4–5.1) compared to those with low symptom burden.

We conducted two series of sensitivity analyses: (1) adjusting for delirium in addition to age, gender, and CCI and (2) including death as a composite outcome in the logistic regression models. These yielded similar results for all four outcomes (Supplementary Table S1). We conducted an additional analysis, restricted to orthopedic surgeries which made up 81% of the cohort. This yielded similar results on all four outcomes (Supplementary Table S2). An additional analysis to adjust for surgery type in addition to delirium, age, gender, and CCI yielded similar results on all four outcomes (Supplementary Table S1a).

DISCUSSION

As rates of major surgery continue to increase in older adults, the improvement of post-operative outcomes has assumed increasing importance. This study found a high prevalence of depressive symptoms in older adults scheduled for major elective surgery; 54% self-reported moderate or high burden of depressive symptoms, with 12% reported a high burden (i.e., GDS score 6+). Those with moderate burden had a nearly 3-fold increased risk of IADL decline at one year, and those with high burden had a 9-fold increased risk, with an exposure-response gradient across the increasing levels of depressive symptom burden (Figure 1). In addition, this gradient was also strongly demonstrated in risk of discharge to a post-acute care facility. Both of these findings stress the strong clinical relevance of depressive symptoms for postoperative outcomes. The lack of full functional recovery following surgery holds tremendous significance, since these patients remain dependent and require long-term help to accomplish IADLs, either by loved ones or facilities.

Preoperative depressive symptoms may be associated with worse functional postoperative outcomes due to several possible mechanisms. First, depressive symptoms may decrease motivation for physical and/or social activities that are important for maintaining or regaining function. Patients with depression are more challenging to engage in physical therapy, which is crucial for functional recovery.³⁵ Second, depressive symptoms can also affect adherence to prescribed medical regimens³⁶ and may affect functioning through important biological (e.g., neuroendocrine and inflammatory) mechanisms.³⁷ Further, depression can negatively influence a patient’s perception of their self-efficacy; therefore, limiting their ability to participate fully or effectively in medical care, rehabilitation programs,³⁸ or social activities, all of which are important to enhance functional recovery. Patient-reported outcomes, regardless of type of surgery or health care profession that is reporting the outcome, need to be patient-centered (i.e., focused on what is most important to the patient). Our study therefore chose to focus on evaluating whether an older person would be able to return to being able to complete their instrumental activities of daily living (IADLs).

While existing guidelines recommend preoperative depression assessment and management, our findings are novel as they provide evidence that even moderate preoperative depressive symptoms are associated with long-term functional recovery in the older adult population spanning various types of major elective surgery. Additionally, our findings highlight the high prevalence of moderate to high burden of depressive symptoms in an elective surgical cohort. The relationship between depression and function has been published in the orthopedic and cardiac surgery literature but have found mixed results. In a study evaluating delirium and cognitive impairment in older adults admitted for hip fracture, Givens et al. found 28% of their cohort had significant depressive symptoms (i.e., GDS>5) and that depressive symptoms were associated with greater odds of ADL decline, decline in ambulation, and nursing home placement or death at 1-month follow-up; however, associations were no longer present at 6-month follow-up.¹⁹ Wilcox et al. found that older adult patients with persistent depressive symptoms were at greater risk for worse functional status 30 months following coronary interventions (i.e., CABG, cardiac catheterization, coronary angiography).²⁰ Beyond the orthopedic and cardiac literature, Kwon et al. found that among older adults who underwent a non-emergent abdominal or thoracic surgery, depression was associated with functional decline in the univariate analysis; however, when controlled for confounders, the effect was no longer present.²¹ Additionally, one study of older adults who underwent a major abdominal surgery, those with depression in the preoperative setting were more likely to suffer from sustained post-operative functional decline.²⁹

Limitations of this study must be considered: first, the majority of subjects are orthopedic patients; therefore, our findings may not be translatable to other major operations. Replication of our findings will be important since the degree of functional decline after an abdominal operation may be different from an orthopedic operation. Second, the SAGES study includes patients from a circumscribed geographic area, with high educational level, and low racial diversity. These factors limit the generalizability of our findings which will need to be confirmed in more diverse cohorts. Third, because the SAGES cohort only enrolled older adults who had an anticipated post-operative length of stay of 3 or more days, our findings may not be generalizable to patients who will be discharged the same day or within 2 days after surgery. Our cohort may be highly selected, and the rate of depression and outcomes may be skewed by a population who otherwise cannot tolerate same day or short-admission surgeries. Fourth, due to the limitation of our dataset, we could not evaluate the potential moderating effect of level of rehabilitation engagement ³⁵ or acute post-operative pain²¹. Fifth, while certain orthopedic operations may disproportionately affect specific IADL functions (e.g., finances, telephone use), we were unable to assess IADLs individually. Lastly, this study may have had limited power to examine some of the outcomes, such as length of stay and readmissions.

We have shown that moderate and high levels of depressive symptom burden in the preoperative setting is a strong predictor of poor functional outcomes after elective surgery. The concept of screening for depression prior to surgery in older adults is not new, as it has been recommended by NSQIP/AGS Guidelines for Optimal Preoperative Assessment of the Older Adult since 2012 and is now promoted as part of the Age-Friendly Health System. Guidelines for geriatric surgery continually recommend preoperative screening for depression as it is a risk factor for delirium^{39, 40}. We would recommend extending this to include the presence of even moderate levels of depressive symptoms which may not meet the level of a depression diagnosis.

Despite extensive recommendations, there remains a discordance in translating these guidelines to effective assessment and management of depression in the preoperative period. Depression screenings are not a routine part of the preoperative assessment. As surgical decision-making evolves to focus more on patient-centered outcomes, like long-term functional recovery, our approach to identifying patients at risk of decline will need to include preoperative screening for depressive symptoms⁴⁰. An acceptable preoperative screening tool is the 2-item Patient Health Questionnaire-2 (PHQ-2)³⁹, as it is easy to use and has been validated in older adults⁴¹.

In addition, interventions found effective in addressing depressive symptom burden and safe to use in frail, older population will need to be tailored to the surgical setting. An example of such intervention would be a tailoring of the Problem Solving Treatment (PST), which is a cognitive behavioral treatment that has shown to successfully treat depressive symptoms in homebound and hospitalized older adults, and has been delivered by laypersons.^{42, 43} Older adults who have undergone the PST have shown positive results within 4 sessions. Current research is being conducted to study the adaptation of PST in a surgical geriatric population. In addition to developing interventions targeting preoperative depressive symptoms, helping patients and families prepare for post-operative outcomes, such as a delayed functional recovery or a sustained functional decline, is necessary to proactively plan for increased need for long-term care by family members or hired caregivers. Depending on the timeframe for treatment, assessing for preoperative depressive symptoms and intervening upon positive cases may require delaying surgery. Further research is needed to better define effective approaches for preoperative optimization for moderate and severe depressive symptom burden.

CONCLUSION

Our findings lend strong support for depressive symptom burden screening prior to surgery, and particularly highlight the impact of even a moderate burden of depressive symptoms on surgical outcomes. Preoperative screening for depressive symptoms can help in several ways: (1) to identify those that may be at risk for functional decline after surgery and in whom measures can be taken to prepare for the functional decline (i.e., additional home health resources, social support resources, setting realistic expectations for patients and families) and (2) to identify those in which a preoperative intervention may be useful, such as (a) treating depressive symptoms through medications or counseling and/or (b) proactively preventing functional decline after surgery with pre- and post-operative physical therapy. Because depressive symptoms may be amenable to treatment which could prevent functional decline, screening for depressive symptoms in the preoperative settings may be helpful to enhance post-operative outcomes in older populations.

This study lays the groundwork for future efforts to implement screens for depressive symptoms in routine clinical care, and for intervention development to reduce depressive symptoms and prevent its associated functional decline in older adults in the postoperative setting.