Abstract
Background
Understanding how patients’ frailty and the physiological stress of surgical procedures affect postoperative outcomes may inform risk stratification of older patients undergoing surgery. The objective of the study was to examine the association of peri-operative frailty with mortality, 30-day readmission and days at home after non-cardiac surgical procedures of different physiological stress.
Methods
This retrospective study used Medicare claims data from a 7.125% random sample of Medicare fee-for-service beneficiaries from 2015 to 2019 who were aged ≥ 65 years and underwent non-cardiac surgical procedure listed in the Operative Stress Score categories. The exposure of the study was claims-based frailty index (robust, < 0.15; pre-frail, 0.15 to < 0.25; mildly frail, 0.25 to < 0.35; and moderate-to-severely frail, ≥ 0.35) with Operative Stress Score categories being 1, very low stress to 5, very high stress. The primary outcome was all-cause mortality at 30 days and 365 days after the surgical procedure.
Results
In total, 1,019,938 patients (mean (SD) age of 76.1 (7.3) years; 52.3% female; 16.8% frail) were included. The cumulative incidence of mortality generally increased with Operative Stress Score category, ranging from 5.0% (Operative Stress Score 2) to 24.9% (Operative Stress Score 4) at 365 days. Within each category, increasing frailty was associated with mortality at 30 days (hazard ratio comparing moderate-to-severe frailty vs. robust ranged from 1.59–3.91) and at 365 days (hazard ratio 1.30–4.04). The variation in postoperative outcomes by patients’ frailty level was much greater than the variation by the operative stress category.
Conclusions
These results emphasise routine frailty screening before major and minor non-cardiac procedures and the need for greater clinician awareness of postoperative outcomes beyond 30 days in shared decision-making with older adults with frailty.
Keywords: peri-operative frailty, operative stress, mortality, readmission, home time
Introduction
An estimated 40% of all surgical procedures are performed on people aged ≥ 65 y in the USA [1,2]. Patients who are older have unique considerations putting them at increased peri-operative risk [3]. One such consideration is frailty, which is a state of decreased physiological reserve and increased vulnerability to stressors [4]. In a nationally representative cohort of community-dwelling Medicare beneficiaries, the 1-year mortality rate after major surgery increased from 6.0% for robust individuals to 27.8% for those with Fried frailty phenotype [3]. Patients who are older and frail also have a higher risk of complications, failure to rescue and functional decline after major surgery in multiple surgical specialties [3,5–11]. Consequently, improving surgery outcomes for this patient population is a priority for the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society [3,12,13].
A recent comprehensive analysis of the United States Veterans Administration Surgical Quality Improvement Program suggests that both pre-operative frailty, measured using the Risk Analysis Index (range: 0–81 points) [14], and the physiological stress of surgical procedures are associated strongly with mortality after non-cardiac procedures [15]. Notably, veterans with a Risk Analysis Index score of ≥ 40 points had higher mortality than those with an Risk Analysis Index score ≤ 20 points even after very low stress procedures. However, because most veterans were men aged < 65 y and long-term patient-centred outcomes were not examined, replication of the findings is needed in a general population of older adults, including older women. In addition, inclusion of age and sex in the Risk Analysis Index calculation – which can account for considerable variation (up to 14 points without cancer and up to 25 points with cancer) – limits the ability to distinguish the extent of poor outcomes attributed to frailty, independent of age and sex.
The objective of this retrospective nationwide cohort study was to examine the association of peri-operative frailty and physiological stress of procedures with mortality, readmission and long-term patient-centred outcomes in Medicare fee-for-service beneficiaries undergoing a wide range of non-cardiac surgery. This information can be useful for decision-making, peri-operative optimisation, and postoperative care in older adults before major and minor surgical procedures.
Methods
This retrospective study used Medicare claims data from a 7.125% random sample of Medicare fee-for-service beneficiaries from 2015 to 2019. Eligible patients were beneficiaries aged ≥ 65 y who had any of the 565 surgical procedure claims listed in the Operative Stress Score (OSS) categories (n = 3,665,970) in the Part B carrier claims [15]. The OSS categories were developed using a modified Delphi consensus method among surgeons and anaesthetists to categorise the physiological stress of surgical procedures, which ranges from 1 (very low stress) to 5 (very high stress); current procedural terminology codes are in online Supporting Information (Table S1) [15]. We did not study 226,576 procedure claims that were performed in nursing facilities, assisted living facilities or home and 389,428 claims that were not preceded by at least 12-month continuous enrolment in Medicare Parts A and B. For patients who received multiple procedures, we chose one procedure per patient that had the highest OSS category; if more than one procedure in the same OSS category was performed, we chose the first procedure, thus excluding 2,029,969 procedure claims. The final study population included 1,019,938 patients (Fig. 1).
Figure 1.
Study population selection flowchart
We measured peri-operative frailty using a claims-based frailty index [16–18] which estimates a deficit-accumulation frailty index (range: 0–1) with a higher number indicating greater frailty. The claims-based frailty index uses 93 variables defined by diagnosis, health services and durable medical equipment codes in the 365 days preceding the procedure date. It has been validated against physical performance measures (gait speed and grip strength) and clinical frailty measures (frailty phenotype and deficit-accumulation frailty index) [16–18]. Using previously published cut-points, we defined robust (< 0.15); pre-frail (0.15 to < 0.25); mildly frail (0.25 to < 0.35); and moderate-to-severely frail categories (≥ 0.35) [17,19].
We obtained age, sex, ethnicity (Asian or Pacific Islander, Black, Hispanic, White, and others) and dual eligibility from the Medicare beneficiary summary file. Chronic conditions were measured using the Chronic Condition Data Warehouse algorithms [20,21] and the comorbidity burden was quantified using the combined comorbidity score [22]. We categorised procedure setting (inpatient or outpatient) and admission type (elective or non-elective).
The primary outcome was all-cause mortality at 30 days and 365 days after the surgical procedure. The date of death was obtained by linking the Medicare data to Social Security files. Secondary outcomes were 30-day readmission and home time loss at 365 days. Readmission dates were obtained from inpatient claims and only measured for patients who received surgical procedures in the inpatient setting and were discharged alive. Home time is a patient-centred outcome calculated from administrative claims data that has been validated as a surrogate measure for functional status and quality of life [23–25]. The minimal clinically important difference in home time was 18.6 days over 365 days because this difference was associated with worsening functional status or death [24]. Home time loss was calculated by summating the number of inpatient days, skilled nursing facility days and days lost to death in the 365-day period from the day of discharge after an inpatient surgical procedure and from the day of an outpatient surgical procedure [24]. Patients were followed for 365 days until the earliest of the following events: the outcome of interest; death; disenrollment from the fee-for-service Medicare; and the end of the study period (31 December 2019).
Patient characteristics for subjects undergoing procedures in different OSS categories were compared using ANOVA or χ2 testing. For all-cause mortality, we estimated cumulative incidence at 30 days and 365 days by OSS and frailty category. Hazard ratios (95%CI) were estimated within each OSS category using Cox proportional hazards models that included: frailty category (robust (reference), pre-frail, mildly frail and moderate-to-severely frail categories); age (< 70 y (reference), 70 to < 75 y, 75 to < 80 y, 80 to < 85 y and ≥ 85 y); sex; ethnicity (Asian or Pacific Islander, Black, Hispanic, White (reference) and Others); combined comorbidity index (continuous); metastatic cancer; and admission type (elective (reference) or non-elective). For 30-day readmission, we estimated cumulative incidence by OSS and frailty category among those who received procedures in the inpatient setting and were discharged alive.
To account for competing risk by death, we estimated Fine-Gray subdistribution hazard ratios (95%CI) that included frailty category and the above-listed covariates. The 365-day home time loss was summarised by OSS and frailty category among those discharged alive. If a patient was censored due to disenrollment from Medicare fee-for-service or the end of the study period before the end of 365-day period we estimated the 365-day home time loss by multiplying 365 days with the proportion of home time loss during the observed period. Negative binomial regression was used to estimate incidence rate ratios of home time loss as a function of frailty category and the above-listed covariates, using the observation time as an offset.
We performed analyses using Stata version 17.0 (StataCorp LLC, College Station, TX, USA) and considered a two-sided p-value of < 0.05 to be statistically significant.
Results
Among 1,019,938 patients (mean (SD) age 76.1 (7.3) y; 52.3% female) who underwent non-cardiac surgical procedures in 2015–2019, most patients underwent very low stress (OSS 1 209,551 (20.5%)); low stress (OSS 2 488,952 (47.9%)); or moderate stress (OSS 3 288,121 (28.2%)) procedures. Only a small fraction underwent high stress (OSS 4 30,905 (3.0%)) or very high stress (OSS 5 2409 (0.2%)) procedures (Table 1). Common procedures under each OSS category were cystourethroscopy, insertion of central venous access device, neuroplasty on median nerve at carpal tunnel, and endovascular ablation therapy of incompetent extremity veins in OSS 1; Mohs micrographic surgery procedure, total knee arthroplasty, debridement on skin, repair inguinal hernia, and partial mastectomy in OSS 2; total hip arthroplasty, laparoscopic cholecystectomy, open treatment of femoral fracture or prosthetic replacement, and total shoulder arthroplasty in OSS 3; partial colectomy, thoracoscopy with lobectomy, and craniectomy for evacuation of hematoma in OSS 4; and Whipple-type procedures, direct repair of aneurysm, liver allotransplantation, and partial esophagectomy in OSS 5 (online Supporting Information Table S2).
Table 1.
Characteristics of Medicare fee-for-service beneficiaries undergoing non-cardiac surgery and procedure in 2015–2019. Values are mean (SD) and number (proportion).
| Characteristics | Total n = 1,019,938 |
OSS 1 n = 209,551 |
OSS 2 n = 488,952 |
OSS 3 n = 288,121 |
OSS 4 n = 30,905 |
OSS 5 n = 2409 |
|---|---|---|---|---|---|---|
| Age; y | 76.1 (7.3) | 76.0 (7.2) | 75.9 (7.2) | 76.6 (7.6) | 76.2 (6.8) | 73.9 (5.7) |
| Sex; female | 533,254 (52%) | 100,221 (48%) | 246,557 (50%) | 169,830 (59%) | 15,719 (51%) | 927 (39%) |
| Ethnicity | ||||||
| Asian or Pacific Islander | 15,171 (2%) | 4338 (2%) | 5982 (1%) | 4258 (2%) | 546 (2%) | 47 (2%) |
| Black | 48,642 (5%) | 12,685 (6%) | 19,913 (4%) | 139,35 (5%) | 1987 (6%) | 122 (5%) |
| Hispanic | 37,105 (4%) | 9564 (5%) | 16,044 (3%) | 10,274 (4%) | 1131 (4%) | 92 (4%) |
| White | 894,822 (88%) | 177,826 (85%) | 435,179 (89%) | 253,200 (88%) | 26,544 (86%) | 2073 (86%) |
| Other | 24,198 (2%) | 5138 (3%) | 11834 (2%) | 6454 (2%) | 697 (2%) | 75 (3%) |
| Dual eligibility | 96,298 (9%) | 23,321 (11%) | 39,669 (8%) | 29,683 (10%) | 3438 (11%) | 187 (8%) |
| Frailty index | 0.19 (0.07) | 0.18 (0.07) | 0.18 (0.07) | 0.20 (0.07) | 0.20 (0.07) | 0.19 (0.06) |
| Frailty category | ||||||
| Robust (< 0.15) | 341,304 (34%) | 75,861 (36%) | 186,702 (38%) | 70,227 (24%) | 7852 (25%) | 662 (28%) |
| Pre-frail (0.15 to < 0.25) | 508,021 (50%) | 101,524 (49%) | 233,098 (48%) | 155,651 (54%) | 16,341 (53%) | 1407 (58%) |
| Mildly frail (0.25 to < 0.35) | 135,131 (13%) | 25,586 (12%) | 54,478 (11%) | 49,430 (17%) | 5332 (17%) | 305 (13%) |
| Moderate-to-severely frail (≥ 0.35) | 35,482 (4%) | 6580 (3%) | 14,674 (3%) | 12,813 (5%) | 1380 (5%) | 35 (2%) |
| Combined comorbidity index | 2.8 (3.3) | 3.0 (3.5) | 2.3 (3.1) | 3.3 (3.4) | 5.4 (3.8) | 6.2 (3.8) |
| Chronic conditions | ||||||
| Hypertension | 855,750 (84%) | 175,358 (84%) | 402,836 (82%) | 248,525 (86%) | 26,959 (87%) | 2072 (86%) |
| Diabetes | 394,223 (39%) | 84454 (40%) | 183,988 (38%) | 112,125 (39%) | 12,669 (41%) | 987 (41%) |
| Chronic kidney disease | 369,076 (36%) | 82820 (40%) | 161,875 (33%) | 109,250 (38%) | 14,111 (46%) | 1020 (42%) |
| Chronic obstructive pulmonary disease | 283,479 (28%) | 61,878 (30%) | 122,335 (25%) | 86,236 (30%) | 12,221 (40%) | 809 (34%) |
| Congestive heart failure | 234,227 (23%) | 46,977 (22%) | 99,855 (20%) | 76,113 (26%) | 10,557 (34%) | 725 (30%) |
| Atrial fibrillation | 192,911 (19%) | 40,933 (20%) | 89061 (18%) | 55,558 (19%) | 6941 (23%) | 418 (17%) |
| Myocardial infarction | 55,253 (5%) | 11,624 (6%) | 24,581 (5%) | 16,869 (6%) | 2062 (7%) | 119 (5%) |
| Dementia | 128,045 (13%) | 25861 (12%) | 52,829 (11%) | 44,902 (16%) | 4285 (14%) | 168 (7%) |
| Metastatic cancer | 51,191 (5%) | 15551 (7%) | 12,716 (3%) | 16112 (6%) | 6143 (20%) | 669 (28%) |
| Inpatient procedure | 382,545 (38%) | 16602 (8%) | 97,472 (20%) | 23,6015 (82%) | 30,047 (97%) | 2409 (100%) |
| Non-elective procedure | 157,001 (15%) | 14524 (7%) | 28,783 (6%) | 99,627 (35%) | 13481 (45%) | 496 (21%) |
OSS, Operative Stress Score.
Full details of patient characteristics are shown in Table 1. When grouped by OSS categories, the mean (SD) age of subjects ranged from 73.9 (5.7) y for OSS 5 to 76.6 (7.6) y for OSS 3. The proportion of females ranged from 38.5% for OSS 5 to 58.9% for OSS 3. There was an indication that patients who underwent an OSS 2 procedure were generally healthier, evidenced by the highest prevalence of robust state (38.2%) and the lowest mean combined comorbidity index (2.3 points). Patients who had an OSS 4 procedure were the most medically complex, based on the highest prevalence of mild frailty (17.3%) and moderate-to-severe frailty (4.5%) with a high mean combined comorbidity index (5.4 points). Patients who underwent an OSS 5 procedure had the lowest prevalence of moderate-to-severe frailty (1.5%) and the highest mean combined comorbidity index (6.2 points). The proportion of inpatient procedures increased with OSS category.
Mean (SD) follow-up time was 279 (121.1) days. The cumulative incidence of 30-day mortality for OSS 1–5 categories was 2.3%, 1.1%, 2.5%, 6.1% and 5.4%, respectively. Within each OSS category, greater frailty was associated with increased 30-day mortality after adjustment (Fig. 2). The cumulative incidence of 30-day mortality comparing patients in the robust category with patients in the moderate-to-severe frailty category was 0.7% vs. 7.2% (hazard ratio (HR) [95%CI] 1.59 [1.34–1.89]) after OSS 1 procedures; 0.1% vs. 5.0% (HR [95%CI] 3.46 [2.85–4.20]) after OSS 2 procedures; 0.8% vs. 10.8% (HR [95%CI] 1.98 [1.74–2.24]) after OSS 3 procedures; and 4.1% vs. 26.1% (HR [95%CI] 3.08 [2.53–3.75]) after OSS 4 procedures. The results for OSS 5 procedures were not reported due to small numbers.
Figure 2.
Frailty, Operative Stress Score (OSS), 30-day and 365-day all-cause mortality after non-cardiac surgery and procedure. Hazard ratios were estimated from Cox proportional hazards model, adjusting for age (< 70 y [reference], ≥ 70 to < 75 y, ≥ 75 to < 80 y, ≥ 80 to < 85 y and ≥ 85 y); sex; ethnicity (Asian or Pacific Islander, Black, Hispanic, White [reference] and others); combined comorbidity index (continuous); metastatic cancer; and admission type (elective [reference] or non-elective). The results for the OSS category 5 and moderate-to-severe frailty were not presented due to the small number of events (< 11). NR, not reported.
The cumulative incidence of 365-day mortality for OSS 1–5 categories was 9.9%, 5.0%, 10.8%, 24.9% and 22.1%, respectively. For OSS 1–4 categories, greater frailty was associated with increased 365-day mortality after adjustment (Fig. 2). The cumulative incidence comparing patients in the robust category vs. patients in the moderate-to-severe frailty category was 4.2% vs. 35.2% (HR [95%CI] 1.46 [1.36–1.58]) after OSS 1 procedures; 1.1% vs. 32.2% (HR [95%CI] 4.04 [3.75–4.34]) after OSS 2 procedures; 3.0% vs. 35.4% (HR [95%CI] 2.19 [2.04–2.34]) after OSS 3 procedures; and 10.2% vs. 53.7% (HR [95%CI] 2.27 [2.00–2.58]) after OSS 4 procedures. The results for OSS 5 procedures were not reported due to small numbers).
The cumulative incidence of 30-day readmission after OSS 1–5 categories inpatient procedures was 5.3%, 1.7%, 2.4%, 4.3% and 5.7%, respectively. For OSS 2–4 categories, greater frailty was associated with increased 30-day readmission after adjustment; for OSS 1 and 5 categories, the trends were not statistically significant (Fig. 3). The cumulative incidence comparing patients in the robust category vs. patients in the moderate-to-severe frailty category was 3.5% vs. 7.8% (adjusted subdistribution HR [95%CI] 1.22 [0.84–1.77]) after OSS 1 procedures; 0.4% vs. 6.8% (HR [95%CI] 3.26 [2.44–4.36]) after OSS 2 procedures; 0.9% vs. 5.8% (HR [95%CI] 2.32 [1.99–2.71]) after OSS 3 procedures; and 2.6% vs. 8.9% (HR [95%CI] 1.64 [1.20–2.25]) after OSS 4 procedures. The results for OSS 5 procedures were not reported due to small numbers).
Figure 3.
Frailty, Operative Stress Score (OSS) and 30-day readmission after non-cardiac surgery and procedure. The analysis only included patients who received procedures in the inpatient setting and were discharged alive. Subdistribution hazard ratios were estimated from Fine-Gray model to account for competing risk by death, adjusting for age (< 70 y [reference], ≥ 70 to < 75 y, ≥ 75 to <80 y, ≥ 80 to < 85 y and ≥ 85 y); sex; ethnicity (Asian or Pacific Islander, Black, Hispanic, White [reference] and others); combined comorbidity index (continuous); metastatic cancer; and admission type (elective [reference] or non-elective). The results for OSS 5 and moderate-to-severe frailty were not presented due to the small number of events (< 11). NR, not reported.
Mean (SD) home time loss during the 365-day follow-up for OSS 1–5 categories was 22 (71.6), 15 (53.5), 37 (81.7), 62 (105.1) and 57 (91.7) days, respectively. Mean (SD) home time loss comparing patients in the robust category and patients in the moderate-to-severe frailty category was 8 (44.1) days vs. 86 (123.4) days (adjusted rate ratio (RR) [95%CI] 1.93 [1.74–2.15]) after OSS 1 procedures; 3 (23.0) days vs. 85 (118.9) days (RR [95%CI] 4.20 [3.96–4.45]) after OSS 2 procedures; 12 (42.5) days vs. 113 (126.1) days (RR [95%CI] 2.71 [2.62–2.81]) after OSS 3 procedures; 31 (72.1) days vs. 149 (141.0) days (RR [95%CI] 2.04 [1.86–2.22]) after OSS 4 procedures; and 40 (75.1) days vs. 73 (102.3) days (RR [95%CI] 1.12 [0.70–1.80]) after OSS 5 procedures (Fig. 4). As the frailty level increased, home time loss was mainly due to an increase in the days spent in a skilled nursing facility and days lost to death.
Figure 4.
Frailty, Operative Stress Score (OSS) and home time loss in 365 days after non-cardiac surgery and procedure. The analysis only included patients who were discharged alive after the index surgical procedure. Home time loss is presented in the mean by summating the number of inpatient days, skilled nursing facility (SNF) days and days lost to death in the 365-day period from the day of discharge after an inpatient surgical procedure and from the day of an outpatient surgical procedure. Incident rate ratio (IRR) was estimated from negative binomial regression adjusting for age (< 70 y [reference], ≥ 70 to < 75 y, ≥ 75 y to < 80 y, ≥ 80 to < 85 y and ≥ 85 y); sex; ethnicity (Asian or Pacific Islander, Black, Hispanic, White [reference] and others); combined comorbidity index (continuous); metastatic cancer; and admission type (elective [reference] or non-elective).
Discussion
In this nationwide cohort of Medicare fee-for-service beneficiaries undergoing non-cardiac surgical procedures of all operative stress levels, patients with greater frailty had higher rates of mortality at 30 days and 365 days, higher readmission rates at 30 days and more home time loss over 365 days compared with patients who were robust. Notably, the variation in postoperative outcomes by patient frailty level was much greater than the variation by the OSS category. Incremental home time loss associated with frailty was mainly due to more days spent in a skilled nursing facility and days lost to death. These results support the utility of peri-operative frailty assessment for surgical risk stratification, informed decision-making and peri-operative care planning in older adults.
Our study results are largely consistent with the findings of a previous large study using the United States Veterans Administration Surgical Quality Improvement Program. Shinall et al. found that veterans (mean age 61.0 y, 7.7% female) who were frail had higher postoperative mortality up to 180 days across all levels of operative stress [15]. Similar to our study, veterans who underwent OSS 2 procedures were healthier and had lower mortality than those having OSS 1 procedures. Our study expands on this by including a more representative sample (mean age 76.1 y, 52.3% female), by following patients for 365 days after surgery and adjusting for age, sex, ethnicity, comorbidities and admission type.
Whether or not a procedure is high risk (defined as having ≥ 1% inpatient mortality [26]) depends on the physiological stress of the surgery and frailty level of the patient. While previous studies have highlighted the importance of surgical frailty on surgical outcomes [27–29], our study featuring an extensive dataset and more contemporary evidence, shows a more notable impact on postoperative outcomes based on frailty levels compared with OSS category. We observed a more substantial variation in the 30-day mortality by frailty level (for OSS 3 procedures, 0.8% in robust patients to 10.8% in patients with moderate-to-severe frailty) than the operative stress level (1.1% after OSS 2 procedures to 6.1% after OSS 4 procedures). Patients with frailty had > 1% 30-day mortality after OSS 1 and 2 procedures, which suggests that in patients with frailty, there is no such thing as a ‘low-risk’ procedure. Moreover, patients having an OSS 5 procedure had the lowest prevalence of moderate-to-severe frailty, yet the highest mean combined comorbidity index, which indicates selection of less frail patients with a single life-threatening condition. This selection may explain why frailty was not as strongly associated with 365-day mortality after OSS 5 procedures.
For patients who are older, mortality is not the only outcome of interest; recovery time, quality of life and functional status after surgery are also important [30,31]. Our results on home time loss due to inpatient stay, skilled nursing facility stay and death by OSS and frailty category can be useful to counsel patients and their family about the expected postoperative course. Home time is correlated closely with poor self-rated health; mobility impairment; depression; limited social activity; and difficulty in self-care [24]. The difference in the mean home time loss between patients with moderate-to-severe frailty and patients who were robust in our study far exceeded the minimal clinically important difference of 18.6 days [24]. For example, patients undergoing OSS 4 procedures, who were medically most complex and frail, had a mean home time loss of 62 days, ranging from 31 days in patients who were robust and 149 days in patients with moderate-to-severe frailty. Furthermore, increased home time loss was primarily driven by more skilled nursing facility days and premature death as frailty increased. We acknowledge that it is not possible to attribute all home time loss to surgery in the absence of a no surgery comparison group. Nonetheless, these results suggest that patients who are older and frail have a higher risk of death; even if they survive the procedure, they exhibit lack of resilience (i.e. inability to recover from the stress of surgery) and spend a considerable part of the year in an institutional setting after surgery.
In patients who are older and being considered for minor and major surgery, decision-making should be tailored to each individual based on the expected benefits of the surgery; alternative treatment options; immediate postoperative risk; and long-term recovery trajectory. Our study shows a disparity between the expected surgical outcomes and the reality in patients who are older and frail. Although procedures such as colectomy, prostatectomy or nephrectomy are traditionally fast-tracked with enhanced recovery after surgery protocols, with durations of stay ranging from 3–8 days in the literature [32,33], we found that, on average, patients with frailty lost over 100 days of home time in the year following such surgeries.
Such long-term postoperative outcomes based on frailty level can inform discussions for decision-making and postoperative care planning. In fact, if a treatment leads to significant functional impairment, individuals often opt to forgo such interventions [30]. In the era of shared decision-making, it is the responsibility of the healthcare team to convey postoperative outcomes relevant to patients. Peri-operative shared decision-making in patients who are older poses unique challenges given factors such as the inherent knowledge disparity in the patient-physician relationship; heightened emotional contexts; and the absence of established frameworks and systems [34]. It is, therefore, crucial to involve a multidisciplinary team, comprising anaesthetists, surgeons and geriatricians, to offer diverse expertise tailored to the patient’s goals to help make an informed decision [35–37]. Once the decision has been made to proceed with surgery, patients who are older and frail should be referred to appropriate consultants (geriatrics, peri-operative medicine, anaesthesia) and allied health services (physiotherapy, dietetics) for peri-operative optimisation and closer postoperative monitoring beyond 30 days after surgery.
The peri-operative team could use our results in conjunction with the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator to paint a realistic picture of expected 30-day and 1-year postoperative outcomes based on patient frailty and the procedure [38]. The clinical frailty index threshold for frailty of 0.25 in our study approximates to clinical frailty scale stage 5 [39], FRAIL scale 2 points [40], Fried frailty phenotype 2 points [41] and a comprehensive geriatric assessment-based frailty index 0.26 [42, 43]. Implementation of routine pre-operative frailty screening has been shown to increase pre-operative evaluation and improve postoperative mortality [44].
Our study included Medicare fee-for-service beneficiaries who underwent 565 procedures defined by the OSS. The results from our study may not apply to patients who have Medicare Advantage Plan or those who undergo procedures not included in the OSS, such as cardiac surgery. Although we had a large sample size, the number of patients with moderate-to-severe frailty who underwent OSS 5 procedures was small, likely due to patient selection, limiting our ability to fully elucidate the association of frailty and postoperative outcomes after OSS 5 procedures. Though we meticulously adjusted for all known confounders present in the Medicare data, the possibility of unmeasured confounders cannot be entirely dismissed and may have introduced a degree of bias. Selection bias is a consideration, particularly regarding patients who ultimately underwent surgery, influenced perhaps by factors like financial and social support, potentially leading to a selective study population. In addition, the data in this study spans 2015–2019, raising concerns about potential obsolescence considering advancements in patient optimisation, peri-operative medicine and surgical techniques. However, the persistently high mortality and morbidity among surgical patients who are older, coupled with the ageing population, underscores the importance of our study. Lastly, we did not have patients with similar clinical characteristics who did not undergo any procedure. Therefore, it was not possible to determine whether poor postoperative outcomes experienced by patients with frailty were due to surgery itself.
In summary, this nationwide retrospective cohort study of patients aged ≥ 65 years revealed that those with frailty experience worse short- and long-term postoperative outcomes regardless of the level of operative stress. This underscores the significance of frailty level over procedural stress in identifying high-risk patients for death and loss of home-time. These results emphasise routine frailty screening before major and minor non-cardiac procedures and the need for greater surgeon awareness of postoperative outcomes beyond 30 days in shared decision-making with older adults with frailty.
Supplementary Material
Table S1. List of Current Procedural Terminology codes of procedures in the Operative Stress Score
Table S2. Five most common procedures by Operative Stress Score
Acknowledgments
CMP and JJL contributed equally. DK received personal fee from Alosa Health and VillageMD for unrelated work. This study was supported by the National Institute on Aging of the National Institutes of Health (R01AG071809 and K24AG073527). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. All data are from Medicare claims data which has strict data security regulations in place that prohibit data sharing. Request for access to data from the US Centers for Medicare and Medicaid Services can be made through https://www.resdac.org. A signed data use agreement is required before data can be accessed. No statistical code is available. No competing interests declared.
References
- 1.Etzioni DA, Liu JH, O’Connell JB, Maggard MA, Ko CY. Elderly patients in surgical workloads: a population-based analysis. Am Surg 2003; 69: 961–5. 10.1177/0003134803069011 [DOI] [PubMed] [Google Scholar]
- 2.Berian JR, Rosenthal RA, Baker TL, et al. Hospital Standards to Promote Optimal Surgical Care of the Older Adult: A Report From the Coalition for Quality in Geriatric Surgery. Ann Surg 2018; 267: 280–90. 10.1097/SLA.0000000000002185 [DOI] [PubMed] [Google Scholar]
- 3.Gill TM, Vander Wyk B, Leo-Summers L, Murphy TE, Becher RD. Population-Based Estimates of 1-Year Mortality After Major Surgery Among Community-Living Older US Adults. JAMA Surg 2022; 157: e225155. 10.1001/jamasurg.2022.5155 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Clegg A, Young J, Iliffe S, Rikkert MO, Rockwood K. Frailty in elderly people. Lancet 2013; 381: 752–62. 10.1016/S0140-6736(12)62167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Adams P, Ghanem T, Stachler R, Hall F, Velanovich V, Rubinfeld I. Frailty as a predictor of morbidity and mortality in inpatient head and neck surgery. JAMA Otolaryngol Head Neck Surg 2013; 139: 783–9. 10.1001/jamaoto.2013.3969 [DOI] [PubMed] [Google Scholar]
- 6.Cooper Z, Rogers SO Jr., Ngo L, et al. Comparison of Frailty Measures as Predictors of Outcomes After Orthopedic Surgery. J Am Geriatr Soc 2016; 64: 2464–71. 10.1111/jgs.14387 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.George EM, Burke WM, Hou JY, et al. Measurement and validation of frailty as a predictor of outcomes in women undergoing major gynaecological surgery. BJOG 2016; 123: 455–61. 10.1111/1471-0528.13598 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McIsaac DI, Bryson GL, van Walraven C. Association of Frailty and 1-Year Postoperative Mortality Following Major Elective Noncardiac Surgery: A Population-Based Cohort Study. JAMA Surg 2016; 151: 538–45. 10.1001/jamasurg.2015.5085 [DOI] [PubMed] [Google Scholar]
- 9.McIsaac DI, Taljaard M, Bryson GL, et al. Frailty as a Predictor of Death or New Disability After Surgery: A Prospective Cohort Study. Ann Surg 2020; 271: 283–9. 10.1097/SLA.0000000000002967 [DOI] [PubMed] [Google Scholar]
- 10.Shah R, Attwood K, Arya S, et al. Association of Frailty With Failure to Rescue After Low-Risk and High-Risk Inpatient Surgery. JAMA Surg 2018; 153: e180214. 10.1001/jamasurg.2018.0214 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Suskind AM, Walter LC, Jin C, et al. Impact of frailty on complications in patients undergoing common urological procedures: a study from the American College of Surgeons National Surgical Quality Improvement database. BJU Int 2016; 117: 836–42. 10.1111/bju.13399 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chow WB, Rosenthal RA, Merkow RP, et al. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. J Am Coll Surg 2012; 215: 453–66. 10.1016/j.jamcollsurg.2012.06.017 [DOI] [PubMed] [Google Scholar]
- 13.Mohanty S, Rosenthal RA, Russell MM, Neuman MD, Ko CY, Esnaola NF. Optimal Perioperative Management of the Geriatric Patient: A Best Practices Guideline from the American College of Surgeons NSQIP and the American Geriatrics Society. J Am Coll Surg 2016; 222: 930–47. 10.1016/j.jamcollsurg.2015.12.026 [DOI] [PubMed] [Google Scholar]
- 14.Hall DE, Arya S, Schmid KK, et al. Development and Initial Validation of the Risk Analysis Index for Measuring Frailty in Surgical Populations. JAMA Surg 2017; 152: 175–82. 10.1001/jamasurg.2016.4202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shinall MC Jr., Arya S, Youk A, et al. Association of Preoperative Patient Frailty and Operative Stress With Postoperative Mortality. JAMA Surg 2020; 155: e194620. 10.1001/jamasurg.2019.4620 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kim DH, Schneeweiss S, Glynn RJ, Lipsitz LA, Rockwood K, Avorn J. Measuring Frailty in Medicare Data: Development and Validation of a Claims-Based Frailty Index. J Gerontol A Biol Sci Med Sci 2018; 73: 980–7. 10.1093/gerona/glx229 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kim DH, Glynn RJ, Avorn J, et al. Validation of a Claims-Based Frailty Index Against Physical Performance and Adverse Health Outcomes in the Health and Retirement Study. J Gerontol A Biol Sci Med Sci 2019; 74: 1271–6. 10.1093/gerona/gly197 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim DH, Patorno E, Pawar A, Lee H, Schneeweiss S, Glynn RJ. Measuring Frailty in Administrative Claims Data: Comparative Performance of Four Claims-Based Frailty Measures in the U.S. Medicare Data. J Gerontol A Biol Sci Med Sci 2020; 75: 1120–5. 10.1093/gerona/glz224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shi SM, Steinberg N, Oh G, et al. Change in a Claims-Based Frailty Index, Mortality, and Healthcare Costs in Medicare Beneficiaries. J Gerontol A Biol Sci Med Sci 2023; 78: 1198–203. 10.1093/gerona/glad010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gorina Y, Kramarow EA. Identifying chronic conditions in Medicare claims data: evaluating the Chronic Condition Data Warehouse algorithm. Health Serv Res 2011; 46: 1610–27. 10.1111/j.1475-6773.2011.01277.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Weeks WB, Huynh G, Cao SY, Smith J, Weinstein JN. Assessment of Year-to-Year Patient-Specific Comorbid Conditions Reported in the Medicare Chronic Conditions Data Warehouse. JAMA Netw Open 2020; 3: e2018176. 10.1001/jamanetworkopen.2020.18176 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gagne JJ, Glynn RJ, Avorn J, Levin R, Schneeweiss S. A combined comorbidity score predicted mortality in elderly patients better than existing scores. Journal of clinical epidemiology 2011; 64: 749–59. 10.1016/j.jclinepi.2010.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Groff AC, Colla CH, Lee TH. Days Spent at Home - A Patient-Centered Goal and Outcome. N Engl J Med 2016; 375: 1610–2. 10.1056/NEJMp1607206 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lee H, Shi SM, Kim DH. Home Time as a Patient-Centered Outcome in Administrative Claims Data. J Am Geriatr Soc 2019; 67: 347–51. 10.1111/jgs.15705 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mishra NK, Shuaib A, Lyden P, et al. Home time is extended in patients with ischemic stroke who receive thrombolytic therapy: a validation study of home time as an outcome measure. Stroke 2011; 42: 1046–50. 10.1161/STROKEAHA.110.601302 [DOI] [PubMed] [Google Scholar]
- 26.Schwarze ML, Barnato AE, Rathouz PJ, et al. Development of a list of high-risk operations for patients 65 years and older. JAMA Surg 2015; 150: 325–31. 10.1001/jamasurg.2014.1819 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mrdutt MM, Papaconstantinou HT, Robinson BD, Bird ET, Isbell CL. Preoperative Frailty and Surgical Outcomes Across Diverse Surgical Subspecialties in a Large Health Care System. J Am Coll Surg 2019; 228: 482–90. 10.1016/j.jamcollsurg.2018.12.036 [DOI] [PubMed] [Google Scholar]
- 28.Panayi AC, Orkaby AR, Sakthivel D, et al. Impact of frailty on outcomes in surgical patients: A systematic review and meta-analysis. Am J Surg 2019; 218: 393–400. 10.1016/j.amjsurg.2018.11.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tjeertes EKM, van Fessem JMK, Mattace-Raso FUS, Hoofwijk AGM, Stolker RJ, Hoeks SE. Influence of Frailty on Outcome in Older Patients Undergoing Non-Cardiac Surgery - A Systematic Review and Meta-Analysis. Aging Dis 2020; 11: 1276–90. 10.14336/AD.2019.1024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fried TR, Bradley EH, Towle VR, Allore H. Understanding the treatment preferences of seriously ill patients. N Engl J Med 2002; 346: 1061–6. 10.1056/NEJMsa012528 [DOI] [PubMed] [Google Scholar]
- 31.Seghers P, Wiersma A, Festen S, et al. Patient Preferences for Treatment Outcomes in Oncology with a Focus on the Older Patient-A Systematic Review. Cancers (Basel) 2022; 14: 1147. 10.3390/cancers14051147 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chen S, He Z, Yao S, et al. Enhanced Recovery After Surgery Protocol Optimizes Results and Cost of Laparoscopic Radical Nephrectomy. Front Oncol 2022; 12: 840363. 10.3389/fonc.2022.840363 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ljungqvist O, Scott M, Fearon KC. Enhanced Recovery After Surgery: A Review. JAMA Surg 2017; 152: 292–8. 10.1001/jamasurg.2016.4952 [DOI] [PubMed] [Google Scholar]
- 34.Vogel A, Guinemer C, Furstenau D. Patients’ and healthcare professionals’ perceived facilitators and barriers for shared decision-making for frail and elderly patients in perioperative care: a scoping review. BMC Health Serv Res 2023; 23: 197. 10.1186/s12913-023-09120-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bougeard AM, Nair A, Santhirapala R. Perioperative shared decision making - How do we train clinicians? PEC Innov 2023; 3: 100181. 10.1016/j.pecinn.2023.100181 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chan SP, Ip KY, Irwin MG. Peri-operative optimisation of elderly and frail patients: a narrative review. Anaesthesia 2019; 74 (Suppl 1): 80–9. 10.1111/anae.14512 [DOI] [PubMed] [Google Scholar]
- 37.Zietlow KE, Wong S, Heflin MT, et al. Geriatric Preoperative Optimization: A Review. Am J Med 2022; 135: 39–48. 10.1016/j.amjmed.2021.07.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bilimoria KY, Liu Y, Paruch JL, et al. Development and evaluation of the universal ACS NSQIP surgical risk calculator: a decision aid and informed consent tool for patients and surgeons. J Am Coll Surg 2013; 217: 833–42 e1–3. 10.1016/j.jamcollsurg.2013.07.385 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in elderly people. CMAJ 2005; 173: 489–95. 10.1503/cmaj.050051 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Morley JE, Malmstrom TK, Miller DK. A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J Nutr Health Aging 2012; 16: 601–8. 10.1007/s12603-012-0084-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56: M146–56. 10.1093/gerona/56.3.m146 [DOI] [PubMed] [Google Scholar]
- 42.Sison SDM, Shi SM, Oh G, Jeong S, McCarthy EP, Kim DH. Claims-Based Frailty Index and Its Relationship with Commonly Used Clinical Frailty Measures. J Gerontol A Biol Sci Med Sci 2024. Epub 1 April. 10.1093/gerona/glae094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Rockwood K, Mitnitski A. Frailty in relation to the accumulation of deficits. J Gerontol A Biol Sci Med Sci 2007; 62: 722–7. 10.1093/gerona/62.7.722 [DOI] [PubMed] [Google Scholar]
- 44.Varley PR, Buchanan D, Bilderback A, et al. Association of Routine Preoperative Frailty Assessment With 1-Year Postoperative Mortality. JAMA Surg 2023; 158: 475–83. 10.1001/jamasurg.2022.8341 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. List of Current Procedural Terminology codes of procedures in the Operative Stress Score
Table S2. Five most common procedures by Operative Stress Score