Prevalence and predictors of frailty among older patients with lung cancer: a systematic review and meta-analysis

Abstract

Background

Older lung cancer patients with frailty are of higher risk of therapeutic side effects and mortality. Despite the fact that the estimated prevalence of frailty among older patients with lung cancer is widely reported, these results have not been synthesized. The aim of this review was to systematically assess the prevalence and related factors of frailty in older patients with lung cancer.

Methods

We searched PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Chinese Biomedical Literature Database (CBM), China National Knowledge Infrastructure (CNKI), and Wanfang databases for observational studies (published up to January 1, 2025) on the prevalence of frailty in older patients with lung cancer. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of included cohort or case-control studies, the Agency for Healthcare Research and Quality (AHRQ) tool was applied to assess the risk of bias in cross-sectional studies. Pooled estimates, subgroup analyses, meta-regression, and publication bias were conducted using Stata 17.0.

Results

In total, 44 articles comprising 61,587 patients were included in this study. The prevalence of frailty among older patients with lung cancer ranged from 5% to 91%, with an estimated prevalence of 46% [95% confidence interval (CI): 39–54%, I²=99.6%]. Moreover, Egger’s regression test suggested no publication bias (P=0.72). Subgroup analyses showed that frailty was more prevalent among female patients, and those older patients ≥70 years old, from developed countries, before radiotherapy, and assessed using the G8 tool.

Conclusions

Frailty is prevalent among older patients with lung cancer, and factors such as age, gender, country, treatment status, and frailty tool were associated with frailty. However, the findings should be interpreted with caution due to high heterogeneity and limited data from regions and subgroups. Early and routine frailty assessment with appropriate management may improve prognosis in this population.

Highlight box.

Key findings

• Following a systematic review and meta-analysis, it indicated that frailty was prevalent among older patients with lung cancer, and age, gender, country, treatment status, and frailty evaluation tool were associated with frailty. Early and routine assessment of frailty, coupled with appropriate management, may reduce the incidence of frailty and improve the prognosis of older patients with lung cancer.

What is known and what is new?

• The prevalence of frailty is 42% among older patients with various types of cancer, and 45% among lung cancer patients without an age limit.

• This meta-analysis provides up-to-data estimates of the pooled prevalence of frailty among older patients with lung cancer, which is 46%. Frailty is more prevalent among female patients, and those older patients ≥70 years old, from developed countries, before radiotherapy, and assessed using the G8 tool.

What is the implication, and what should change now?

• Routine assessment of frailty among older patients with lung cancer is very important.

• Researchers should develop a consensus definition of frailty and of standardizing measurement tool to assess the frailty among specific population.

Introduction

World Health Organization defines frailty as a progressive age-associated decline in physiological reserve and function affecting multiple organs, which confers extreme vulnerability to stressors and increases the risk of adverse health outcomes (1). Frailty can occur at any age, but older adults are more susceptible to frailty (2). The prevalence of frailty is 17.4% among the community-dwelling population aged 60 years and more, while up to 29.5% among those aged 85 years (3). Frailty is more prevalent among older patients with cancer. Han et al. found that frail older cancer survivors are at high risk of adverse symptoms and poor health-related quality of life after cancer treatment (4). Similarly, a systematic review of 2,916 older patients with cancer showed that frailty is associated with an increased risk of treatment-related complications and all-cause mortality (5). These findings suggest that older people with cancer are more susceptible to frailty and need more attention.

GLOBOCAN 2022 estimates indicated that lung cancer is the most frequently diagnosed cancer, with approximately 2.5 million new cases, which accounts for 12.4% of all cancer cases globally (6). The incidence of lung cancer rises rapidly among older patients. In the United States, 68% of new cases of lung cancer affect patients aged more than 65 years (7). Similarly, in Denmark, nearly half of patients with lung cancer are 70 years or older, and the incidence rate has been reported to double among those aged more than 80 years (8). Besides, several studies have shown that frailty is associated with a higher risk of treatment side effects and death among older patients with lung cancer (9-11).

To date, only one systematic review investigated the rate at which frailty is observed in people suffering from lung cancer (12), which included all ages and mixed study designs, limiting its applicability to older adults, among whom age-related physiological decline amplifies frailty’s impact. This systematic review addresses this gap by focusing exclusively on patients ≥60 years and observational studies to provide targeted evidence for frailty management in this vulnerable population and nurses management for older patients with lung cancer. We present this article in accordance with the guidelines of Meta-analysis and Systematic Reviews of Observational Studies (13) and the PRISMA reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1747/rc) (14).

Methods

Search strategy

A comprehensive search was conducted across a range of data sources, such as PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Chinese Biomedical Literature Database (CBM), China National Knowledge Infrastructure (CNKI), and Wanfang databases, from the inception to January 1, 2025. Both MeSH terms and unrestricted keywords focusing on older patients with lung cancer and frailty were applied in the search. Besides, the reference list of related review articles was reviewed to identify additional studies. The search strategy of each database is presented in Appendix 1.

Study selection

The eligibility conditions included: (I) non-interventional research presenting findings on how frequently the condition occurs or risk factors of frailty among patients with primary lung cancer; (II) all participants aged 60 years or more; (III) frailty was diagnosed based on established frailty models or validated frailty tools, such as Fried frailty phenotype, frailty index (FI), modified FI (m-FI), comprehensive geriatric assessments (CGA), vulnerable elderly survey-13 (VES-13), and G8 score; and (IV) articles published in Chinese and English. Studies were excluded if they met the following criteria: (I) incomplete data; (II) sample size was less than 50; and (III) low-quality methodology.

Data collection

Two reviewers independently screened eligible studies and extracted relevant data using predefined forms. Disagreements were resolved via discussion. The following data were extracted from each study: the first author’s name, publication year, country of study, study setting, cancer type and stage, age, treatment status, frailty evaluation tool, the prevalence of frailty, and gender prevalence. If any variable was missing or unclear, attempts were made to contact study authors; otherwise, the data were recorded as not reported.

Quality assessment

Quality evaluation of the included studies was conducted independently by two reviewers. For studies with cohort or case-control designs, the Newcastle-Ottawa Scale (NOS) (15) was utilized to assess domains such as selection, comparability, exposure, and outcomes. Scores ranged from 0 to 9, with a score of 0–3 suggesting a high risk, a score of 4–6 suggesting a moderate risk, and a score of 7–9 suggesting a low risk. The Agency for Healthcare Research and Quality (AHRQ) tool (16) was applied to assess the quality of cross-sectional studies. The tool consists of 11-item questions with answers “yes”, “no”, or “unclear”. “Yes” receives 1 point and “no” or “unclear” receives 0 point. A score of 0–3 suggests low quality, a score of 4–7 suggests medium quality, and a score of 8–11 suggests high quality. Considering frailty as exposure, not endpoint in the cohort or case-control studies, the AHRQ tool was employed to assess the quality of the included studies. Disagreements were resolved by consulting with a third reviewer.

Statistical analysis

Data were analyzed using Stata 17.0 software (Stata Corporation, College Station, TX, USA). The latent heterogeneity among the included studies was assessed using Cochran’s Q and I² statistic. Heterogeneity was considered significant when P value <0.10 or I²>50% (17). Pooled prevalence and 95% confidence intervals (CIs) of frailty among older patients with lung cancer were calculated using a random-effects (DerSimonian and Laird) model when heterogeneity was significant; otherwise, a fixed-effects (Mantel-Haenszel) model was employed. Subgroup analyses and meta-regression were conducted to explore the potential sources of heterogeneity. Potential publication bias was assessed through visual evaluation of funnel plot symmetry (under fixed-effect assumptions) and further examined using Egger’s regression test (P<0.10 is indicated as significant).

Results

Study selection

In total, 1,536 records were identified via databases. Furthermore, 8 pertinent studies were uncovered from manual reviews of the reference lists. Following the elimination of duplicates and the evaluation of titles, abstracts, and complete texts, a total of 44 full-text articles met the inclusion standards for both narrative and statistical evaluation. A comprehensive depiction of the study selection flow is shown in Figure 1.

Figure 1.

PRISMA flow diagram for study selection. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Study characteristics

Table 1 presents the characteristics of the included studies. In total, 61,587 lung cancer cases in older adults were reported across 44 included studies. Twenty-one studies included patients with mean or median age ≥70 years old, and thirteen studies included patients with mean or median age <70 years old. Twenty-five studies included participants from developing countries, nineteen studies from developed countries. Seven articles focused on outpatient settings, while fifteen articles focused on inpatient settings. Groningen Frailty Indicator (GFI, n=8), CGA (n=6), FRAIL scale (n=6), and G8 (n=6) were the main tools used to diagnose frailty. However, only fifteen studies reported data on the type of lung cancer, and sixteen studies specified lung cancer stage.

Table 1. Characteristics of the included studies.

Study (year)	Country	Setting	Study design	Lung cancer type and stage	Age (years) mean/median	Treatment status	Frailty evaluation tool	Sample size	Frail population	Gender prevalence, n/N (%)
										Male	Female
Gironés et al. 2012 (18)	Spain	Outpatient	Cohort study	Different types all stages	77	N/A	Comprehensive Geriatric Assessment	83	60	N/A	N/A
Decoster et al. 2017 (19)	Belgium	N/A	Cohort study	Different types all stages	76	Before specific treatment	G8	245	204	N/A	N/A
Kirkhus et al. 2017 (20)	Norway	Outpatient	Cohort study	Different types all stages	76.9	Before specific treatment	Modified Geriatric Assessments	59	35	N/A	N/A
Schulkes et al. 2017 (21)	The Netherlands	N/A	Cohort study	Different types all stages	77	Before specific treatment	G8	142	108	63/88 (71.6)	45/54 (83.3)
Antonio et al. 2018 (22)	Spain	Inpatient	Cohort study	NSCLC stage IIA–IIIB	≥75	Before chemotherapy	Comprehensive Geriatric Assessment	85	54	51/76 (67.1)	4/9 (44.4)
Franco et al. 2018 (23)	USA	Inpatient	Cohort study	NSCLC stage I/II	74	Before radiotherapy	Modified Frailty Index	139	101	54/67 (80.6)	47/72 (65.3)
Gironés et al. 2018 (24)	Spain	Outpatient	Cohort study	NSCLC stage IV	77	Before chemotherapy	Comprehensive Geriatric Assessment	93	23	21/84 (25.0)	2/9 (22.2)
Ørum et al. 2018 (25)	Denmark	Outpatient	Cohort study	Different types all stages	≥70	Before specific treatment	Comprehensive Geriatric Assessment	93	59	N/A	N/A
Raghavan et al. 2018 (26)	USA	N/A	Cohort study	NSCLC stage I/II	>70	Before radiotherapy	Modified Frailty Index	102	43	N/A	N/A
Agemi et al. 2019 (27)	Japan	Inpatient	Cohort study	Different types all stages	79	Before specific treatment	G8	101	83	64/81 (79.0)	19/20 (95.0)
Liu et al. 2019 (28)	China	Inpatient	Cohort study	Different types all stages	76.4	Before specific treatment	FRAIL scale	324	91	51/172 (29.7)	40/152 (26.3)
Tsubata et al. 2019 (29)	Japan	N/A	Cohort study	Different types all stages	75	Before chemotherapy	Comprehensive Geriatric Assessment	68	15	10/57 (17.5)	5/11 (45.5)
Wang et al. 2019 (30)	China	N/A	Cohort study	Different types all stages	65	Before chemotherapy	Frailty Index-LAB	1,020	50	36/728 (4.9)	14/292 (4.8)
Chen et al. 2021 (31)	China	Inpatient	Cohort study	NSCLC stage I–III	≥65	Before surgery	FRAIL scale	423	117	72/253 (28.5)	45/170 (26.5)
Couderc et al. 2021 (32)	France	Outpatient	Cohort study	Different types all stages	78.7	Before specific treatment	G8	228	208	N/A	N/A
Kaneda et al. 2021 (33)	Japan	Outpatient	Cohort study	NSCLC stage I–IIIc	70.4	Before surgery	Frailty index	193	13	N/A	N/A
Zheng et al. 2021 (34)	China	Inpatient	Cross-sectional study	NSCLC all stages	68.4	Before chemotherapy	Tilburg Frailty Indicator	253	139	N/A	N/A
Bandidwattan-awong et al. 2022 (35)	Thailand	N/A	Cohort study	Different types advanced stages	74	Before chemotherapy	G8	73	59	N/A	N/A
Bensken et al. 2022 (36)	USA	N/A	Cohort study	Different types all stages	>65	N/A	Claims Frailty Index	8,817	2,119	N/A	N/A
Chen FF et al. 2022 (37)	China	Inpatient	Cohort study	NSCLC stage I–III	67.7	Before surgery	Groningen Frailty Indicator	176	58	N/A	N/A
Chen J et al. 2022 (38)	China	Inpatient	Cross-sectional study	NSCLC stage I–III	≥60	Before chemotherapy	Tilburg Frailty Indicator	362	232	N/A	N/A
Cheng et al. 2022 (39)	USA	N/A	Cohort study	NSCLC all stages	74.1	N/A	Veterans Affairs Frailty Index	42,204	23,443	23,055/41,579 (55.4)	388/625 (62.1)
Cooper et al. 2022 (40)	USA	Inpatient	Cohort study	Different types all stages	76.7	Before surgery	Comprehensive Geriatric Assessment	73	28	16/36 (44.4)	12/37 (32.4)
Jeon et al. 2022 (41)	Korea	Outpatient	Cross-sectional study	Different types all stages	>65	Before chemotherapy	FRAIL scale	53	10	N/A	N/A
Shen et al. 2022 (42)	China	Inpatient	Cross-sectional study	NSCLC all stages	66.5	N/A	Groningen Frailty Indicator	167	117	N/A	N/A
Wu et al. 2022 (43)	China	Inpatient	Cross-sectional study	NSCLC stage I–IIIa	≥60	Before surgery	Groningen Frailty Indicator	227	77	N/A	N/A
Goldsmith et al. 2023 (44)	UK	N/A	Cohort study	N/A	71.8	Before specific treatment	CSHA Frailty Index	314	137	N/A	N/A
Guo et al. 2023 (45)	China	Inpatient	Cohort study	Different types stage I–III	68	Before surgery	28-item Frailty Index	334	118	78/228 (34.2)	40/106 (37.7)
Hou et al. 2023 (46)	China	N/A	Cross-sectional study	N/A	<70	Before specific treatment	Frailty Index	538	190	N/A	N/A
Long et al. 2023 (47)	China	Inpatient	Cross-sectional study	Different types all stages	≥60	Before chemotherapy	FRAIL scale	108	49	23/53 (43.4)	26/55 (47.3)
Lou 2023 (48)	China	Inpatient	Cross-sectional study	N/A	>60	Before specific treatment	AGILE scale	215	24	15/122 (12.3)	9/93 (9.7)
Tian et al. 2023 (49)	China	N/A	Cohort study	N/A	≥65	Before surgery	Modified Frailty Index	1,372	388	275/781 (35.2)	113/591 (19.1)
Wang et al. 2023 (50)	China	Inpatient	Cross-sectional study	Different types all stages	≥60	N/A	Groningen Frailty Indicator	106	40	18/44 (40.9)	48/62 (77.4)
Jing et al. 2024 (51)	China	Inpatient	Cross-sectional study	N/A	67.3	After surgery	Tilburg Frailty Indicator	260	84	N/A	N/A
Lee et al. 2025 (52)	USA	N/A	Cohort study	NSCLC stage III–IV	73	Before specific treatment	46-item Frailty Index	155	28	13/70 (18.6)	15/85 (17.6)
Liu et al. 2024 (53)	China	Inpatient	Cohort study	Different types stage II–III	69.4	Before chemotherapy	Groningen Frailty Indicator	204	85	30/88 (34.1)	55/116 (47.4)
Sha et al. 2024 (54)	China	Inpatient	Cross-sectional study	Different types stage I–III	67.4	Before surgery	Groningen Frailty Indicator	150	72	N/A	N/A
Sun et al. 2024 (55)	China	Inpatient	Cross-sectional study	NSCLC all stages	≥60	Before surgery	Frailty Index	328	130	N/A	N/A
Tao et al. 2024 (56)	China	Inpatient	Cohort study	Different types stage IIIB–IV	68.4	N/A	FRAIL scale	234	100	65/160 (40.6)	35/74 (47.3)
Wang et al. 2024 (57)	China	Inpatient	Cross-sectional study	N/A	67.1	N/A	Groningen Frailty Indicator	404	316	N/A	N/A
Xiang 2024 (58)	China	Inpatient	Cross-sectional study	Different types stage IIIB–IV	67	N/A	Groningen Frailty Indicator	376	226	N/A	N/A
Xu et al. 2024 (59)	China	Inpatient	Cross-sectional study	Different types all stages	68.2	After surgery	Frailty Index	350	123	77/229 (33.6)	46/121 (38.0)
Yan et al. 2024 (60)	China	Inpatient	Cross-sectional study	NSCLC stage III–IV	75.0	After targeted therapy	G8	238	197	131/154 (85.1)	66/84 (78.6)
Zhang et al. 2024 (61)	China	Inpatient	Cross-sectional study	Different types all stages	72.7	N/A	FRAIL scale	98	36	N/A	N/A

Before specific treatment, patients may receive surgery, chemotherapy, radiotherapy, chemoradiotherapy, targeted therapy or immunotherapy based on their condition. CSHA Frailty Index, the Canadian Study of Health and Aging Frailty Index; N/A, not available; NSCLC, non-small cell lung cancer.

Quality of studies

All included studies were assessed using the AHRQ tool. the methodological rigor of the included research ranged from moderate to strong, with AHRQ scores between 7 and 9 out of a maximum of 11. The results of the assessment based on the AHQR tool are presented in Table S1.

Prevalence of frailty among older patients with lung cancer

Among the 44 included studies, the prevalence of frailty ranged from 5% to 91%. The estimated prevalence of frailty among older patients with lung cancer was 46% (95% CI: 39–54%) (Figure 2). The random-effects model was adopted because of high heterogeneity (I²=99.6%, P<0.001).

Figure 2.

Forest plot of the estimated prevalence of frailty among older patients with lung cancer. CI, confidence interval.

Meta regression analyses

Meta-regression indicated that age (β=0.160, P=0.20), country of study (β=−0.070, P=0.32), setting (β=0.004, P=0.97), gender (β=−0.040, P=0.61), treatment status (β=−0.007, P=0.82), and study design (β=−0.004, P=0.96) were not significant moderators of the overall heterogeneity (Table 2).

Table 2. Meta-regression analyses.

Risk factor	No. of studies	Coefficient	95% CI of coefficient	Standard error	t value	P value
Age	34	0.106	−0.061, 0.273	0.082	1.30	0.20
Country	44	−0.070	−0.211, 0.070	0.070	−1.01	0.32
Setting	32	0.004	−0.191, 0.199	0.095	0.04	0.97
Gender	21	−0.040	−0.194, 0.115	0.076	−0.52	0.61
Treatment status	34	−0.007	−0.070, 0.056	0.031	−0.23	0.82
Study design	44	−0.004	−0.149, 0.141	0.072	−0.06	0.96
Frailty evaluation tool	44	0.025	0.006, 0.045	0.010	2.61	0.01

CI, confidence interval.

Subgroup analyses of the frailty occurrence in older patients suffering from lung cancer

Subgroup evaluation yielded an estimate of frailty frequency in older adults with lung cancer aged ≥70 and <70 years old was 55% and 44% respectively. The prevalence of frailty differed between countries of varying economic development levels, with 43% in developing countries and 50% in developed countries. The prevalence of frailty was comparable across different settings and study designs, with 48% in both outpatient and inpatient settings, 47% in cohort studies, and 46% in cross-sectional studies. Frailty occurrence rate was 39% and 43% among male and female participants, respectively. Subgroup analysis based on treatment status indicated that the rate of frailty detected in older patients diagnosed with lung cancer before surgery, after surgery, before chemotherapy, before specific treatment, and before radiotherapy was 32%, 34%, 42%, 54%, and 58%, respectively. The prevalence of frailty varied widely in subgroup analysis based on frailty evaluation tool: G8: 83%; modified Geriatric Assessments (m-GA): 59%; Veterans Affairs Frailty Index (VAFI): 56%; GFI: 51%; Tilburg Frailty Indicator (TFI): 51%; CGA: 48%; m-FI: 48%; Canadian Study of Health and Aging Frailty Index (CSHA FI):44%; 28-item FI: 35%; FRAIL scale:33%; FI: 29%; Claims Frailty Index (CFI): 24%; 46-item FI: 18%; AGILE scale: 11%; and Frailty Index-Laboratory (FI-LAB): 5%. Comprehensive subgroup findings are presented in Table 3 and Figures S1-S7.

Table 3. Subgroup analyses of the included studies.

Subgroups	Number of studies included	Frailty
		Prevalence (%)	95% CI (%)	I2 (%)	P value
Age
≥70 years	21	55	45, 64	99.0	<0.001
<70 years	13	44	28, 60	99.4	<0.001
Country
Developed	19	50	40, 61	97.7	<0.001
Developing	25	43	33, 54	99.2	<0.001
Setting
Inpatient	25	48	39, 56	98.1	<0.001
Outpatient	7	48	14, 82	99.5	<0.001
Gender
Female	21	43	31, 56	98.5	<0.001
Male	21	39	27, 52	99.6	<0.001
Treatment status
Before radiotherapy	2	58	28, 87	95.9	<0.001
Before specific therapy	11	54	35, 73	99.3	<0.001
Before chemotherapy	10	42	21, 63	99.2	<0.001
After surgery	2	34	30, 38	0.0	0.46
Before surgery	9	32	24, 40	95.9	<0.001
Study design
Cohort study	27	47	37, 56	99.7	<0.001
Cross-sectional study	17	46	35, 57	98.4	<0.001
Frailty evaluation tool
G8	6	83	79, 88	75.0	0.001
m-GA	1	59	47, 72	–	–
VAFI	1	56	55, 56	–	–
GFI	8	51	37, 64	97.3	<0.001
TFI	3	51	32, 70	97.1	<0.001
CGA	6	48	30, 65	94.8	<0.001
m-FI	3	48	19, 76	98.4	<0.001
CSHA FI	1	44	38, 49	–	–
28-item FI	1	35	30, 41	–	–
FRAIL scale	6	33	26, 40	84.4	<0.001
FI	4	29	13, 46	98.3	<0.001
CFI	1	24	23, 25	–	–
46-item FI	1	18	12, 24	–	–
AGILE scale	1	11	7, 15	–	–
FI-LAB	1	5	4, 6	–	–

CFI, Claims Frailty Index; CGA, Comprehensive Geriatric Assessment; CI, confidence interval; CSHA FI, Canadian Study of Health and Aging Frailty Index; FI, Frailty Index; FI-LAB, Frailty Index-Laboratory; GFI, Groningen Frailty Indicator; m-FI, modified Frailty Index; m-GA, modified Geriatric Assessment; TFI, Tilburg Frailty Indicator; VAFI, Veterans Affairs Frailty Index.

Publication bias

Visual inspection of funnel plots exhibited no significant publication bias (Figure S8). Furthermore, Egger’s regression test indicated no publication bias among the included studies (P=0.72).

Discussion

This meta-analysis identified a frailty prevalence of 46% (95% CI: 39–54%) among older patients diagnosed with lung cancer, which was higher than that (42%) among older patients with various types of cancer. In addition, the prevalence of frailty in this study was higher than that among lung cancer patients without an age limit (45%) (5,12). Lung cancer is a disease of older adults with a median age of 71 years at diagnosis (62) and older adults are more likely to develop frailty due to age-related cumulative decline in physiological systems (63). Therefore, implementing routine frailty assessments for older lung cancer patients should be prioritized, as this will facilitate the establishment of evidence-based management protocols tailored to this vulnerable population.

Subgroup analysis based on gender showed that the prevalence of frailty among older patients with lung cancer was higher in females (43%) than in males (39%), which was consistent with that reported by other studies (64-66). Women have lower muscle strength and lean body mass, which makes them more vulnerable to sarcopenia (67). The relationship between sarcopenia and frailty has been confirmed by previous studies (68). In addition, women have a longer life expectancy (69). Both reasons can contribute to the higher risk of frailty among female patients with lung cancer.

The stratified analysis based on country indicated that the prevalence of frailty in developing countries (43%) was lower than that in developed countries (50%), which is consistent with that observed in patients with breast cancer (70). Research on frailty among older patients with lung cancer began in developed countries from 2012 and extended to developing economies, notably China [2019] and Thailand [2021]. Therefore, frailty can be better identified and managed in developed countries, which may partly explain the disparity in the prevalence of frailty among different regions.

Our findings also indicated that older adults with lung cancer in inpatient (48%) and outpatient settings (48%) exhibited a high rate of frailty. Given that frailty is associated with healthcare utilization, including hospitalization visits and outpatient healthcare (71-73). A cross-sectional study of 7,987 adults aged ≥50 years from three European countries showed that older adults with frailty needed 3.1 times higher frequency of hospitalization and a higher frequency of outpatient visits compared to those without frailty (71).

Subgroup analysis based on treatment phase indicated that frailty rates was the lowest among older patients with lung cancer assessed prior to surgery (32%), followed by those assessed after surgery (34%), prior to chemotherapy (42%), prior to specific treatment (54%), and prior to radiotherapy (58%). Studies on frailty among older patients with lung cancer before radiotherapy primarily target individuals aged ≥70 years (23,26). In contrast, preoperative frailty research in this population focuses on those aged ≥60 years (31,37,43). This younger preoperative cohort aligns with established evidence suggesting that younger age protects against frailty.

Fifteen distinct frailty evaluation tools were employed in this systematic review and meta-analysis of 44 studies, and GFI, CGA, FRAIL scale, and G8 score were the top 4 most frequently used tools. The estimated prevalence of frailty in older patients with lung cancer varied hugely from 5% in studies using FI-LAB to 83% in studies using the G8 score, which was consistent with that reported by other studies (12,70). The G8 score has been introduced as a highly sensitive frailty-screening tool for older adults with cancer; however, its limited specificity may contribute to the high prevalence of frailty in clinical assessments (74). This variation also aligns with our meta-regression results, which identified the frailty evaluation tool as a significant moderator contributing to between-study heterogeneity. The absence of an internationally recognized gold standard persists in frailty assessment. More than 70 measurement tools have been developed, embodying heterogeneous conceptual models and culturally specific adaptations in diverse populations (75). This heterogeneity arises from several factors: (I) all existing frailty instruments exhibit inherent trade-offs between sensitivity and specificity; (II) the absence of a universally accepted operational definition hinders precise case identification; (III) tool characteristics differ based on target users (clinicians, researchers, or epidemiologists) and primary functions (screening, predictive, or outcome assessment) (76). Therefore, it is important to develop a consensus on the definition of frailty and standardize the measurement tools used to assess frailty in a specific population.

In addition, there are some limitations in this study. First, although high heterogeneity is frequently observed among observational studies, subgroup analyses conducted in this meta-analysis failed to explain the source of heterogeneity. Second, despite a comprehensive database search, studies were predominantly from Europe, the USA, and China, limiting generalizability to other regions. Third, subgroup analyses based on lung cancer type and stage were precluded by the limited data available from the included studies, an aspect that warrants further studies. Forth, the very high heterogeneity observed (I²=99.6%) was only partially explained by the variables examined in subgroup and meta-regression analyses. Important unmeasured or inconsistently reported factors very likely contributed, including lung cancer stage such as early vs. locally advanced vs. metastatic, pathological subtype like adenocarcinoma vs squamous cell carcinoma, burden of comorbidities, and specific treatments received. These clinical characteristics are known to influence physiological reserve and thus frailty prevalence. This limitation underscores the need for future primary studies to systematically report cancer stage, histology, and comorbidity burden when assessing frailty.

Conclusions

This review estimated that 46% of older patients with lung cancer exhibited frailty. Besides, subgroup analyses revealed that age, gender, country, treatment status, and the frailty assessment tool were associated with the prevalence of frailty among older patients with lung cancer. Early and routine assessment of frailty, coupled with appropriate management, may reduce the incidence of frailty and improve the prognosis of older patients with lung cancer. Besides, it is necessary to reach a consensus on the definition of frailty and standardize the measurement tools used to assess frailty among specific populations.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.