Modified 5‐Item Frailty Index and Postoperative Outcomes Following Parotidectomy for Malignancy

ABSTRACT

Objective

Frailty is a predictor of poor outcomes following surgery for malignancy. Our study investigates the association between the modified 5‐item frailty index (mFI‐5) and postoperative outcomes following parotidectomy for malignancy.

Study Design

Cross‐sectional database study.

Setting

National Surgical Quality Improvement Program (NSQIP).

Methods

The 2005–2018 NSQIP was retrospectively queried for patients undergoing parotidectomy for malignancy. Cases were stratified by the modified 5‐item frailty index (mFI‐5), which was calculated by the presence of 5 comorbidities: hypertension requiring medication, diabetes mellitus, congestive heart failure in the last 30 days, lung pathology, and functional health status. Univariate and multivariate analyses were performed to identify associations between mFI‐5 and demographic, comorbid, and complication variables.

Results

A total of 14,567 patients were identified and stratified by the mFI‐5. Univariate analysis showed an increasing incidence of American Society of Anesthesiologists (ASA) class, all comorbidities, and all complications with increasing mFI‐5 score (p < 0.001). Multivariable regressions showed increased odds of any surgical complication in patients with mFI‐5 ≥ 2 compared with mFI‐5 = 0 (aOR 2.56; 95% CI: 1.37–4.76, p = 0.011), as well as increased odds of unplanned readmission (aOR 4.02; 95% CI: 1.16–13.97, p = 0.032) and reoperation (aOR 6.61; 95% CI: 1.07–40.68, p = 0.042).

Conclusion

The mFI‐5 is associated with increased incidence of comorbidities and complications, as well as increased odds of surgical complications, unplanned readmission, and reoperation in patients undergoing parotidectomy for malignancy, independent of age and ASA class. The mFI‐5 can be utilized in the risk stratification of these patients.

1. Introduction

Parotidectomy is indicated in the management of parotid gland malignancies, benign tumors, and non‐neoplastic diseases [1]. Approximately 80% of salivary gland tumors occur in the parotid gland, of which 20% are malignant, accounting for 1%–5% of all head and neck malignancies [2, 3, 4, 5, 6]. Despite being an uncommon malignancy, the annual incidence of malignant parotid tumors ranges from 0.5 to 3.0 per 100,000 internationally [2, 4]. While parotidectomy remains the foundation of management for these malignancies, there is limited data on the prognostic factors influencing surgical outcomes, especially in the context of patient‐specific risk factors and comorbidities.

Frailty has emerged as a reliable indicator of outcomes in patients with head and neck cancer and a valuable prognostic indicator in assessing surgical risk. The modified frailty index 5 (mFI‐5) is a 5‐point score consisting of the presence of congestive heart failure, diabetes mellitus, chronic obstructive pulmonary disease or pneumonia, dependent functional health status at the time of surgery, and hypertension requiring medications [7]. Emerging literature has elucidated the relationship between frailty and poor outcomes in several domains of otolaryngology, such as head and neck cancer, free flap reconstruction, olfactory impairment, and facial fracture repair [8, 9, 10, 11, 12, 13, 14]. However, the application of frailty on salivary gland resections such as parotidectomy remains to be determined. In fact, studies of frailty in head and neck cancer surgery are often limited to glossectomy, laryngectomy, pharyngectomy, mandibulectomy, thyroidectomy, and esophagectomy, and exclude resections of salivary malignancy such as parotidectomy [14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47]. Understanding the impact of frailty on outcomes after parotidectomy and complications such as reoperations, unplanned readmissions, and overall morbidity is therefore critical for guiding perioperative management strategies and enhancing patient outcomes.

Benign salivary pathologies are often highly amenable to parotidectomy and have an uncomplicated postoperative course; frailty is therefore not a major consideration in the preoperative evaluation, because patients tend to be healthier and younger. However, frailty is particularly relevant in the setting of salivary malignancy, because malignancy may worsen baseline performance status, complicating decisions surrounding adjuvant therapy, which is indicated in patients with high‐risk pathologic features. To address this gap in the literature, we utilized the National Surgical Quality Improvement Program (NSQIP) database to identify patients undergoing parotidectomy for malignancy. By analyzing mFI‐5 scores, we aimed to elucidate the association between frailty and certain comorbidities. We hypothesized that higher (worse) mFI‐5 scores would correlate with increased odds of unplanned readmissions, reoperations, and surgical complications. To our knowledge, we report the first study that investigates the association between the mFI‐5 and outcomes in patients undergoing parotidectomy for malignancy.

2. Methods

2.1. Study Design

The National Surgical Quality Improvement Program (NSQIP) is an annual database of surgical cases published by the American College of Surgeons (ACS) that consists of associated data on demographics, comorbid variables, and postoperative complications, including data from upward of 450 hospitals across the United States. Our study identified patients undergoing parotidectomy for primary malignancy that occurred between the years 2005 and 2018 from the 2005–2018 ACS‐NSQIP. We utilized CPT codes 42410, 42415, 42420, 42425, and 42426 to select for parotidectomy cases. We further selected for postoperative diagnosis of parotid malignancy by codes 142.0 and C07, for ICD‐9 and ICD‐10, respectively. Of note, this study did not require approval by the Rutgers New Jersey Medical School Institutional Review Board because the NSQIP consists entirely of de‐identified patient data. The ACS‐NSQIP has not verified the validity of the statistical analysis or conclusions derived herein.

For each case in our analysis, demographic data were collected, including age, sex, race, and American Society of Anesthesiologists Classification (ASA). Parotidectomy type was also included as an independent variable, detailing the parotidectomy for each case as either lateral (no facial nerve dissection), lateral (with facial nerve dissection and preservation), total (with facial nerve dissection and preservation), total (with facial nerve sacrifice), or total (with facial nerve sacrifice and unilateral ipsilateral neck dissection). Comorbidity data were also retrieved, including frequency data on variables such as obesity, smoking status, dyspnea, peripheral vascular disease, dialysis status, stroke history, disseminated cancer status, wound infection, steroid use, bleeding disorder, and systemic sepsis.

Cases were stratified by the mFI‐5. For each patient, the mFI‐5 was calculated by the presence of five comorbidities, each constituting one point in summation: (1) hypertension requiring medication, (2) diabetes mellitus, (3) congestive heart failure in the last 30 days, (4) lung pathology (chronic obstructive pulmonary disease or pneumonia), and (5) functional health status. Following this, we established mFI‐5 score cohorts as such, as supported by prior similar studies with mFI‐5 analyses: 0, 1, and ≥ 2 [7, 9, 12, 14, 17, 24, 31, 36, 39, 46]. Finally, we retrieved perioperative complications data occurring within 30 days of the primary procedure (parotidectomy), including surgical complications (surgical site infections, wound dehiscence, excessive bleeding, hematoma formation), medical complications, readmission, unplanned readmission, reoperation, and mortality.

2.2. Statistical Analysis

The statistical analysis for this study was performed via Version 28.0.1 of SPSS Statistics (IBM Corporation, Armonk, NY). Univariate analyses in the form of Fisher's exact test and Pearson chi‐square analysis were used to compare demographic and comorbid data among mFI‐5 score cohorts. Multivariable, binary logistic regression was used to assess the independent impact of mFI‐5 on adverse outcomes, controlling for statistically significant demographic and comorbid variables including age and ASA class. For all analyses, the alpha level was set at p < 0.050.

3. Results

A total of 14,567 parotidectomy patients were included as part of this retrospective cohort analysis and stratified by modified frailty index, with 7434 (51.03%) having mFI‐5 = 0, 5025 (34.50%) having mFI = 1, and 2108 (14.47%) having mFI ≥ 2. The most common type of parotidectomy in this cohort was lateral or superficial parotidectomy with dissection (Table 1).

TABLE 1.

Frequency and CPT code by parotidectomy type.

Parotidectomy type	CPT code	% (N)
Lateral (no dissection)	42,410	15.6 (2275)
Lateral (dissection)	42,415	56.4 (8212)
Total (dissection)	42,420	21.5 (3127)
Total (sacrifice)	42,425	2.2 (315)
Total (unilateral dissection)	42,426	4.4 (639)

3.1. Univariate Associations Between mFI‐5 and Demographics/Comorbidities

Older patients tended to correlate more with a higher mFI‐5 score; for example, patients aged 60–81 represented a higher portion of those with mFI‐5 ≥ 2 as compared to mFI‐5 = 1 or mFI‐5 = 0 (62.4% vs. 54.2% vs. 27.6%, p < 0.001). Additionally, males were more likely to have mFI‐5 ≥ 2 as compared to females (57.9% vs. 42.1%, p = 0.010). Black patients were more likely to have a score of mFI‐5 ≥ 2 as compared to mFI‐5 = 1 or mFI‐5 = 0 (9.7% vs. 8.8% vs. 6.7%, p < 0.001). For comorbidities, patients with mFI‐5 ≥ 2 were more likely than patients with mFI‐5 = 0 to be obese (60.9% vs. 31.9%, p < 0.001), to be a smoker (28.5% vs. 23.1%, p < 0.001), to have dyspnea (14.9% vs. 1.9%, p < 0.001), to have a history of stroke (4.3% vs. 0.6%, p < 0.001), to have current disseminated cancer (3.7% vs. 1.5%, p < 0.001), to have a history of steroid use (4.2% vs. 1.6%, p < 0.001), or to have a bleeding disorder (3.9% vs. 0.7%, p < 0.001) (Table 2).

TABLE 2.

Cohort demographics and comorbidities.

	mFI‐5 score 0 (n = 7434)	mFI‐5 score 1 (n = 5025)	mFI‐5 score 2+ (n = 2108)	Total (n = 14,567)	p‐value
Age cohorts in years, % (N)					< 0.001
18–40	25.3 (1878)	2.9 (148)	1.6 (31)	14.1 (2057)
41–60	44.4 (3298)	31.4 (1580)	24.8 (523)	37.1 (5401)
61–80	27.6 (2053)	54.2 (2724)	62.4 (1316)	41.8 (6093)
≥ 81	2.8 (205)	11.4 (573)	11.3 (238)	7 (1016)
Sex, % (N)					0.010
Female	53.8 (3997)	44.1 (2218)	42.1 (887)	48.8 (7102)
Male	46.2 (3432)	55.8 (2805)	57.9 (1219)	51.2 (7456)
Parotidectomy type, % (N)					< 0.001
Lateral (no dissection)	16.5 (1227)	14.4 (722)	15.5 (326)	15.6 (2275)
Lateral (dissection)	56.8 (4226)	56.5 (2839)	54.4 (1147)	56.4 (8212)
Total (dissection)	21.4 (1591)	21.1 (1058)	22.7 (478)	21.5 (3127)
Total (sacrifice)	1.8 (136)	2.4 (121)	2.8 (58)	2.2 (315)
Total (unilateral dissection)	3.4 (255)	5.7 (285)	4.7 (99)	4.4 (639)
Race cohorts, % (N)					< 0.001
White	77.1 (4616)	79.4 (3451)	77.5 (1432)	78 (9499)
Black	6.7 (402)	8.8 (384)	9.7 (180)	7.9 (966)
Other	16.2 (967)	11.8 (511)	12.7 (235)	14.1 (1713)
ASA class, % (N)					< 0.001
I	17.7 (1314)	0.6 (28)	0.1 (2)	9.2 (1344)
II	61.8 (4585)	46.8 (2348)	21.6 (445)	50.8 (7388)
III	19.9 (1478)	49.9 (2506)	70.6 (1486)	37.6 (5470)
IV	0.6 (47)	2.7 (136)	7.7 (161)	2.4 (344)
Comorbidities, % (N)
Obesity	31.9 (2350)	45.7 (2277)	60.9 (1273)	40.9 (5900)	< 0.001
Smoking	23.1 (1715)	24.3 (1220)	28.5 (601)	24.3 (3536)	< 0.001
Dyspnea	1.9 (143)	6.1 (305)	14.9 (315)	5.2 (763)	< 0.001
History of stroke	0.6 (9)	2.1 (23)	4.3 (14)	1.6 (46)	< 0.001
Disseminated cancer	1.5 (111)	3.1 (154)	3.7 (78)	2.4 (343)	< 0.001
Steroid use	1.6 (117)	3.9 (194)	4.2 (89)	2.7 (400)	< 0.001
Bleeding disorder	0.7 (53)	2.8 (142)	3.9 (83)	1.9 (278)	< 0.001

Note: Totals used in calculating percentages excluded missing cases for each variable.

Abbreviations: ASA = American Society of Anesthesiologists; PVD = peripheral vascular disease.

3.2. Univariate Associations Between mFI‐5 and Complication Rates

On univariable analysis of postoperative outcomes, patients with mFI‐5 ≥ 2 were more likely than patients with mFI‐5 = 0 to have higher rates of overall complications (6.3% vs. 3.1%, p < 0.001), medical complications (2.6% vs. 0.8%, p < 0.001), surgical complications (4.2% vs. 2.5%, p < 0.001), unplanned readmission (2.8% vs. 1.0%, p < 0.001), reoperation (2.7% vs. 1.0%, p < 0.001), and mortality (0.8% vs. 0.1%, p < 0.001) (Table 3).

TABLE 3.

Incidence of perioperative complications.

	mFI‐5 score 0 (n = 7434)	mFI‐5 score 1 (n = 5025)	mFI‐5 score 2+ (n = 2108)	Total (n = 14,567)	p‐value
Any surgical complication, % (N)	2.5 (187)	3.1 (156)	4.2 (88)	3.0 (431)	< 0.001
Any medical complication, % (N)	0.8 (61)	1.3 (64)	2.6 (54)	1.2 (179)	< 0.001
Any complication, % (N)	3.1 (234)	4.1 (208)	6.3 (132)	3.9 (574)	< 0.001
Unplanned readmission, % (N)	1.0 (76)	1.9 (95)	2.8 (58)	1.6 (229)	< 0.001
Reoperation, % (N)	1.0 (76)	2.0 (102)	2.7 (57)	1.6 (235)	< 0.001
Mortality, % (N)	0.1 (7)	0.2 (12)	0.8 (17)	0.2 (36)	< 0.001

Note: Totals used in calculating percentages excluded missing cases for each variable.

3.3. Multivariable Associations Between mFI‐5 and Complication Rates

On multivariable binary logistic regression analysis also adjusting for age and ASA class, patients with mFI‐5 score ≥ 2 as compared to patients with mFI‐5 = 0 were significantly more likely to have any surgical complication (aOR 2.56; 95% CI: 1.37–4.76, p = 0.011), unplanned readmission (aOR 4.02; 95% CI: 1.16–13.97, p = =0.032), or reoperation (aOR 6.61; 95% CI: 1.07–40.68). However, patients with mFI ≥ 2 as compared to mFI = 0 were not more likely to have any medical complication. Finally, of note, neither increasing age nor ASA class was independently associated with a higher likelihood of complications, after adjusting for mFI‐5 (Table 4).

TABLE 4.

Factors associated with development of perioperative complications in multivariate analysis.

Dependent variable	Independent covariates	Significance	OR	95% CI
Any complication	mFI‐5 score = 0 (reference)
	mFI‐5 score = 1	0.042	0.56	0.32–0.98
	mFI‐5 score = 2+	0.564	1.23	0.61–2.48
	Age = 18–40 (reference)
	Age = 41–60	0.770	1.13	0.50–2.55
	Age = 61–80	0.590	1.26	0.54–2.92
	Age ≥ 81	0.700	1.23	0.43–3.53
	ASA class I (reference)
	ASA class II	0.394	1.63	0.531–4.98
	ASA class III	0.169	2.29	0.70–7.44
	ASA class IV	0.053	4.87	0.98–24.27
Any surgical complication	mFI‐5 score = 0 (reference)
	mFI‐5 score = 1	0.041	0.51	0.27–0.97
	mFI‐5 score = 2+	0.011	2.56	1.37–4.76
	Age = 18–40 (reference)
	Age = 41–60	0.993	1.00	0.42–2.36
	Age = 61–80	0.918	0.95	0.39–2.34
	Age ≥ 81	0.791	0.86	0.272–2.70
	ASA class I (reference)
	ASA class II	0.146	3.04	0.68–13.57
	ASA class III	0.077	4.09	0.86–19.42
	ASA class IV	0.050	7.67	0.99–58.98
Any medical complication	mFI‐5 score = 0 (reference)
	mFI‐5 score = 1	0.307	0.61	0.23–1.58
	mFI‐5 score = 2+	0.635	0.74	0.22–2.52
	Age = 18–40 (reference)
	Age = 41–60	0.647	0.71	0.16–3.14
	Age = 61–80	0.497	1.66	0.38–7.20
	Age ≥ 81	0.671	1.47	0.25–8.57
	ASA class I (reference)
	ASA class II	0.491	0.55	0.10–3.05
	ASA class III	0.928	1.09	0.18–6.39
	ASA class IV	0.308	3.36	0.33–34.59
Unplanned readmission	mFI‐5 score = 0 (reference)
	mFI‐5 score = 1	0.322	1.77	0.57–5.45
	mFI‐5 score = 2+	0.032	4.02	1.16–13.97
	Age = 18–40 (reference)
	Age = 41–60	0.595	0.62	0.11–3.64
	Age = 61–80	0.893	0.89	0.17–4.73
	Age ≥ 81	0.881	1.16	0.16–8.19
	ASA class I (reference)
	ASA class II	n.a.	n.a.	n.a.
	ASA class III	n.a.	n.a.	n.a.
ASA class IV	n.a.	n.a.	n.a.
Reoperation	mFI‐5 score = 0 (reference)
	mFI‐5 score = 1	0.189	3.01	0.58–15.59
	mFI‐5 score = 2+	0.042	6.61	1.07–40.68
	Age = 18–40 (reference)
	Age = 41–60	n.a.	n.a.	n.a.
	Age = 61–80	n.a.	n.a.	n.a.
	Age ≥ 81	n.a.	n.a.	n.a.
	ASA class I (reference)
	ASA class II	n.a.	n.a.	n.a.
	ASA class III	n.a.	n.a.	n.a.
	ASA class IV	n.a.	n.a.	n.a.

Note: Values for which there were not enough cases in multivariable analysis to calculate were labeled as “n.a.” Any complication—consists of the following variables: surgical complication, medical complication. Any surgical complication—consists of excessive bleeding, wound dehiscence, and surgical infection. Excessive bleeding—excessive bleeding requiring transfusion within 72 h postop. Surgical infection—consists of superficial infection, deep infection, and organ space infection. Any medical complication—consists of medical infection, hypercoagulation event, and ventilation/intubation. Medical infection—consists of urinary tract infection, sepsis, septic shock, and pneumonia. Hypercoagulation event—consists of pulmonary embolism, deep vein thrombosis, and myocardial infarction. Ventilation/intubation—ventilator use and/or intubation within 48 h postoperatively.

4. Discussion

The current literature discusses several prognostic factors of parotid cancers. One retrospective review of 224 patients undergoing surgery for primary parotid malignancies found that advanced age is a poor prognostic factor for survival, locoregional failure, and distant metastasis [48]. Several studies showed that high‐grade tumors were seen more often in patients over 45 years old and are also considered to be of prognostic significance, with high‐grade tumors having worse survival outcomes compared to low‐grade tumors [6, 48, 49]. This study is the first to examine the association between the mFI‐5 and outcomes following parotidectomy for malignancy. Our analysis revealed that higher mFI‐5 scores correlated with increased incidences of comorbidities and complications. Specifically, patients with mFI‐5 scores ≥ 2 had significantly higher odds of surgical complications, unplanned readmissions, and reoperations.

Our study demonstrated a significant association between higher mFI‐5 scores and increased rates of surgical complications, with patients scoring mFI‐5 ≥ 2 having over twice the odds of experiencing surgical complications compared to patients with mFI‐5 = 0 (aOR 2.56; 95% CI: 1.37–4.76, p = 0.011). These findings are consistent with prior research on frailty in head and neck surgery. Abt et al. highlighted that the modified frailty index strongly predicts Clavien‐Dindo IV complications, surgical morbidity, and mortality in patients undergoing flap reconstructions, with moderate correlations observed for reoperation and readmission [32]. Their study underscored the particular relevance of frailty indices in oncology patients, where functional health impairments are often not captured by traditional comorbidity indices but are integral to predicting and managing surgical risk. Our results also align with findings from skull base surgeries, where higher mFI scores were independently associated with increased odds of surgical complications after adjusting for confounders such as age, ASA class, and surgical site [50]. These studies collectively highlight the role of frailty as a robust and independent predictor of adverse surgical outcomes, emphasizing the need for frailty‐based preoperative assessments to better anticipate complications and tailor perioperative care.

Interestingly, our study found no significant association between increased mFI‐5 scores and medical or overall complications, which contrasts with prior research in other head and neck surgical populations. Abt et al. demonstrated that frailty is strongly correlated with Clavien‐Dindo IV complications and overall morbidity in flap reconstruction cases [32]. Similarly, in intradural skull base surgeries, higher mFI scores were associated with significantly increased odds of medical complications and prolonged hospital stays [50]. The absence of a significant association in our cohort may be partially explained by the mFI‐5 and newer versions of the ACS‐NSQIP missing key frailty dimensions, such as unintentional weight loss, subjective exhaustion, and reduced mobility, which are critical in other frailty phenotypes and may play a greater role in predicting medical complications. Abt et al. used the 11‐point modified frailty index, which, while including more data points, cannot be used for ACS‐NSQIP versions beyond 2014 because these versions do not include all of the 11 required variables [32, 51]. Another potential explanation is that the impact of frailty on medical complications may be mediated by preoperative optimization in this patient population. For instance, multidisciplinary prehabilitation programs have been shown to reduce frailty and improve postoperative outcomes, particularly in older or oncology patients [52, 53]. These interventions could mitigate the risk of medical complications in frail patients undergoing parotidectomy, thereby blunting the association observed.

Our study found that neither increasing age nor higher ASA class was independently associated with a higher likelihood of complications following parotidectomy, after adjusting for mFI‐5. This finding aligns with previous research suggesting that frailty, rather than chronological age or ASA classification, serves as a more accurate predictor of postoperative outcomes. For instance, a study by Adams et al. demonstrated that the mFI was significantly associated with morbidity and mortality in patients undergoing inpatient head and neck surgery, whereas age and ASA class were less predictive. ⁴⁴ Similarly, in vestibular schwannoma surgeries, Dicpinigaitis et al. found that increasing frailty, not increasing age, resulted in extended hospital stays [54]. These findings underscore the limitations of relying solely on age or ASA class for preoperative risk assessment.

With the average life expectancy increasing in developed countries, many studies have found frailty to be considered an essential factor to consider in preoperative patients [14, 55]. It is important to note that frailty and age, while related, are also independent factors, and there is strong evidence that frailty is a better prognostic indicator for postoperative outcomes than age [14]. Studies have shown that age alone should not be used as a factor by which to exclude patients from aggressive surgical treatment of esophageal, thyroid, and laryngeal cancers [14, 56, 57, 58]. While there are many tools to assess frailty, the mFI‐5 allows for a condensed and simplified tool that has effectively predicted postoperative complications and mortality across surgical specialties [59]. Importantly, its condensed nature allows for it to be a potential tool that can be used in day‐to‐day clinical practice [14].

This study has several limitations. As with many studies relying on the ACS‐NSQIP database, our analysis is restricted to short‐term outcomes within 30 days postoperatively, limiting our ability to evaluate long‐term complications, functional recovery, or survival [32, 50]. Moreover, the ACS‐NSQIP database does not report complications specific to parotidectomy procedures, such as facial nerve injury, Frey's syndrome, or salivary fistula. Data specific to tumor presentation such as histology, group stage, clinical tumor‐nodal‐metastasis classification, adjuvant therapy, locoregional recurrence, and survival are also not reported. Another limitation lies in the retrospective nature of the study and the use of a deidentified national database, which introduces selection bias and heterogeneity in surgical and perioperative practices across participating institutions [32, 50]. For example, it is possible that patients with higher frailty scores were selectively excluded from surgery or underwent less aggressive interventions, skewing the observed outcomes. Future studies incorporating institutional datasets and long‐term outcomes are necessary to validate our findings and expand upon these limitations.

5. Conclusion

The mFI‐5 is associated with increased incidence of comorbidities and complications, as well as increased odds of surgical complications, unplanned readmission, and reoperation in patients undergoing parotidectomy for malignancy, independent of age and ASA class. Clinically, the mFI‐5 can be utilized in the risk stratification of patients undergoing parotidectomy. Ultimately, frailty is highly multifactorial, encompasses many risk factors, and should be considered holistically in preoperative and postoperative evaluations.

Conflicts of Interest

The authors declare no conflicts of interest.

Shaari A. L., Kumar K., Saad A. M., et al., “Modified 5‐Item Frailty Index and Postoperative Outcomes Following Parotidectomy for Malignancy,” Laryngoscope Investigative Otolaryngology 10, no. 5 (2025): e70220, 10.1002/lio2.70220.

Data Availability Statement

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