Falls in older patients treated for cancer in French outpatient departments: screening and prevention—a prospective observational study

Abstract

Purpose

International guidelines published in 2022 recommend annual fall risk (FR) screening in adults aged ≥ 65 years. Older cancer patients are at increased risk of falls due to cancer-related symptoms, treatment toxicities, and concomitant medications. Falls may negatively impact cancer care and outcomes.

Methods

This exploratory multicenter cross-sectional and prospective study included patients aged ≥ 65 years treated for cancer in ten oncology ambulatory care units. Patients were recruited during World Patient Safety Day in September 2024. FR screening and preventive recommendations followed international guidelines. A 3-month follow-up assessed implementation in patients at high FR. The primary objective was to describe the prevalence of fallers and the distribution in the three FR levels in older adults with cancer. The secondary objectives were to identify factors associated with being at high FR and assess the feasibility of fall prevention recommendations.

Results

Among 328 patients, 21% reported at least one fall in the previous year and 15% were classified as high FR. High FR was independently associated with older age (OR 1.33 per 5-year increase), taxane treatment (OR 3.74), and exposure to ≥ 2 fall-risk-increasing drugs (FRIDs) (OR 2.19). Polypharmacy was frequent (61%) but not associated with high FR. At 3 months, only 24% of high FR patients had fully implemented preventive recommendations, mainly biological assessments (60%), medication review (48%), and cardiovascular evaluation (43%).

Conclusion

High FR in older cancer patients is associated with potentially reversible pharmaceutical factors. Fall prevention recommendations remain poorly implemented, highlighting the need for structured multidisciplinary approaches.

Trial registration number

NCT06504628, registered on 2024-07-10.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00520-026-11108-9.

Introduction

A fall is an event which results in a person coming to rest inadvertently on the ground or floor or other lower level [1]. About 25% of older adults with cancer have experienced at least one fall in the past 6 months, according to two recent meta-analyses [2, 3]. Beyond their physical consequences, falls have a direct impact on the oncological care pathway. They may lead to therapeutic modifications, such as dose reductions, treatment interruptions, or even discontinuation [4]. Falls are also associated with increased hospitalizations, morbidity, and mortality in older cancer patients [5].

Adverse drug reactions (ADRs) of oncological treatments act as predisposing and precipitating factors for falls and fall-related injuries through peripheral and vestibular neuropathies, cognitive impairment, hand-foot syndrome, orthostatic hypotension, chemotherapy-induced osteoporosis, and musculoskeletal disorders [6, 7]. In France, in 2018, antineoplastic agents were the leading drug class responsible for hospitalizations due to ADRs (15.1%) [8]. Although taxanes and oxaliplatin are associated with neurotoxicity [9], there was no consensus on considering antineoplastic agents as fall risk-increasing drugs (FRIDs) in the STOPPFall list [10]. Polypharmacy, common in older adults treated for cancer, increases with addition of supportive medications related to chemotherapy. This increases the risk of ADRs, partly due to the risk of drug interaction and the frequent use of self-medication [11]. Polypharmacy is also independently associated with increased mortality [12]. Medication non-adherence is frequent and can, in part, be explained by complex dosing regimens often encountered in oncological care [13]. In addition to oncological and supportive treatments, patients may also be prescribed numerous non-specific FRIDs, such as psychotropics, cardiovascular drugs, medications for benign prostatic hyperplasia, incontinence, overactive bladder, and anticholinergics [14].

It is recommended that frailty be assessed and managed in older adults before initiating cancer treatment, starting at age 65, to better anticipate chemotherapy-induced toxicities and postoperative complications [15]. A randomized trial analyzed the impact of nurse-led geriatric assessment and management in oncology, demonstrating benefits not only in reducing grade 3–5 toxicities (50.7% vs 71.2%), likely linked to dose reductions, but also in decreasing the incidence of falls within the first 3 months after treatment initiation (11.7% vs 20.7%; RR 0.58, p < 0.01) [16].

However, the 2022 World Guidelines for Falls Prevention and Management recommend annual screening for adults aged 65 and older, either during routine medical follow-up or after a fall resulting in injury [17]. According to these guidelines, the fall risk (FR) assessment classifies patients into three categories, low, intermediate, or high, and recommends corresponding interventions for each level. For patients at high FR, recommendations include securing the environment and conducting cardiovascular, nutritional, laboratory, pharmacologic, physical, visual, auditory, emotional, and cognitive assessments. The feasibility of implementing all these recommendations in older adults receiving cancer treatment has not yet been evaluated. A French pilot study demonstrated that pharmacist-led interventions effectively reduced inappropriate prescriptions in older cancer patients [18].

Given the specific challenges associated with population aging in oncology, we conducted the CHUPACHOUP study as part of a 2023 regional call for innovative solutions launched by the Gérontopôle under the French National Fall-Prevention Plan. Its objectives were to describe the prevalence of fall history and the distribution of the three FR levels and to identify factors associated with being at high FR and assess the feasibility of fall prevention recommendations in this subgroup.

Methods

Study design, setting, and population

This was an exploratory, multicenter, cross-sectional and prospective study conducted in France, targeting patients aged 65 and older, treated for cancer at public or private centers and affiliated with French health insurance system. Ten of the 44 centers authorized to provide cancer care in Normandy participated in the study. Consecutive eligible patients were recruited during the week of the World Patient Safety Day in September 2024, in oncology and radiotherapy outpatient units [19]. Patients with dependance according to Activities of Daily Living and unable to consent were not included [20]. These non-inclusion criteria were specified in the call-for-projects policy document because dependent patients were to be referred directly to geriatric teams rather than undergo the screening step.

Ethics and regulations

This study was performed in line with the principles of the Declaration of Helsinki. The ethical approval was obtained on June 28, 2024, by the Comité de Protection des Personnes - Ile de France XI. The protocol was registered on ClinicalTrial.gov (NCT06504628). All investigators were certified with Good Clinical Practice for research. Informed consent was obtained from all individual participants included in the study.

Objectives and endpoints

The primary objective was to describe the prevalence of fall history over the past 12 months and the distribution of the three low, intermediate, and high FR subgroups according to the 2022 World Guidelines for Falls Prevention and Management [17].

The secondary objectives were to identify factors associated with high FR and evaluate the implementation of recommended interventions at 3 months in this subgroup.

Secondary endpoints included:

• Investigation of an association between the high FR subgroup and socio-demographic, medical and therapeutic characteristics of the study population

• Description of the proportion of patients in the high FR group who received the recommended interventions.

Data collection and procedure

Included patients underwent FR assessment using the algorithm in the 2022 World Guidelines for Falls Prevention and Management [17]. The three screening questions were: (1) Have you fallen in the past 12 months? (2) Do you feel unsteady when standing or walking? and (3) Are you worried about falling? (Supplementary Materials 2).

• Additional data regarding any previous assessment of FR were collected (Supplementary Materials 2). Oncological characteristics were retrieved from the medical records and the current medication prescriptions issued by general practitioners provided by the patients.

• A list of protocolized recommendations for each FR subgroup was provided to the patient and their general practitioners (Supplementary Materials 3 and 4). The three leaflets had been revised by the Users Committee of the Health Regional Agency to ensure the understanding of the messages. The prescription of physiotherapy was directly given by the investigators to the patients identified in both the intermediate and high FR subgroups. Those at low FR could be referred to the regional adapted physical activity program so-called IMAPAC, if not already active [21].

• At 3 months from inclusion, patients in the high FR subgroup were interviewed by phone call and data were also collected from medical records, to check the implementation of recommendations and, if not, the reason why (Supplementary Materials 5).

Coding and classifications of FRIDs using INN, ATC5, and STOPPFall list

Drug names entered in the electronic case report form were standardized and translated into International Nonproprietary Names (INN) to match the Anatomical Therapeutic Chemical classification level 5 (ATC5). ATC5 codes were extracted from the French National Health Data System via OpenMedic, and all collected drugs were mapped accordingly. The STOPPFall list was also translated into ATC5 codes to identify whether each drug belonged to the STOPPFall criteria (YES/NO) and to classify it as a FRID (Supplementary Materials 6, Supplementary Materials 7).

All drugs belonging to taxanes (ATC4 code L01CD) and platinum salts (L01XA) were grouped together. These 2 classes were analyzed separately to independently assess their association with FR level.

Statistics

Analyses were performed at the Biostatistics and Clinical Research Unit of the University Hospital using SAS V9.4 (or later) and/or R software version 3.6.2 (or later). Due to the exploratory nature of the study, a sample size calculation was not performed.

Qualitative variables were described by frequencies, percentages, and 95% confidence interval (CI), while quantitative ones were presented as means (standard deviation) or medians (interquartile range, IQR) depending on their distribution. Univariate analysis of factors associated with high FR was performed using chi-square test or Fisher’s exact test for qualitative variables and the unpaired t-test or Mann-Whitney test for quantitative variables, depending on their distribution. A multivariate logistic regression was performed for all variables with a p-value less than 0.20 in univariate analysis. A p-value of < 0.05 was considered statistically significant.

Results

Among 416 older patients treated in ambulatory care units (ACUs), 386 were eligible. One patient was excluded because of missing data, 33 declined to participate, and 24 could not be invited to participate because investigators lacked sufficient time. In total, 328 patients were included in the analysis.

It was found that 73% of patients (n = 236) were classified as having a low FR, 12% (n = 40) as intermediate FR and 15% (n = 48) as high FR. The 3 months follow-up was performed in 42 of the 48 patients in the high FR group (Fig. 1).

Fig. 1.

Flow chart

During the FR assessment (Table 1), 21% (CI 95%, 16–26) of patients reported having experienced at least one fall in the past 12 months. Instability while standing or walking was the most observed assessment criterion, reported by 39% of patients across all FR subgroups. The subgroup most frequently reporting a fear of or concern about falling was the intermediate FR (53% of patients).

Table 1.

Description of fall-risk assessment criteria

Fall-risk assessment criteria n (%)	All (n = 328)	Low FR (n = 236)	Intermediate FR (n = 40)	High FR (n = 48)
1. Fall in the last 12 months	68 (21)	11 (5)	7 (17)	48 (100)
2. Feel instability while standing or walking	127 (39)	66 (28)	32 (80)	27 (56)
3. Fear of falling	83 (25)	40 (17)	21 (53)	21 (44)
Prevalence, % (CI 95%)			12 (9–16)	15 (11–19)

The baseline characteristics of the included patients are presented in Table 2. Lung, digestive, gynecological, and urological cancers were the most represented cancer types. Digestive and gynecological cancers were the most frequent in the high FR subgroup. An equal number of patients aged 65 to 74 and those aged 75 and older were observed in the high FR subgroup. Across all FR subgroups, most patients (94%, n = 269) were receiving chemotherapy, of whom 25% (n = 81) platinum salts and 16% (n = 54) taxanes. A higher proportion of patients were identified in the high FR subgroup when receiving taxanes (p = 0.032). Polypharmacy was observed in 61% (n = 172) of the overall sample, raising up to 72% (n = 28) in the high FR subgroup. The number of FRIDs was significantly associated with higher FR level (p = 0.023).

Table 2.

Comparison of baseline characteristics according to fall-risk level (N = 328)

	All patients n (%)	Fall risk subgroups		p-value
		Low or Intermediate n (%)	Highn (%)
Number of patients	328 (100)	280 (85)	48 (15)
Sex ♦
Male	174 (53)	149 (53)	25 (52)	0.87
Female	153 (47)	130 (47)	23 (48)
Age (years)
Mean ± standard deviation	73 ± 6	73 ± 6	75 ± 7	0.06
65–74	201 (61)	177 (63)	24 (50)	0.09
≥ 75	127 (39)	103 (37)	24 (50)
County living places
At-risk counties (Manche (50) and Orne (61))	99 (30)	80 (29)	19 (40)	0.13
Other counties	229 (70)	200 (71)	29 (60)
Center type
University hospitals and Comprehensive cancer centers	172 (52)	155 (55)	17 (35)	0.039*
Community hospitals	71 (22)	57 (20)	14 (29)
Private hospitals	85 (26)	68 (24)	17 (35)
Cancer’s localization
Hematologic	27 (8)	24 (9)	3 (6)
Upper aerodigestive tract	21 (6)	16 (6)	5 (10)
Lung	75 (23)	67 (24)	8 (17)
Digestive	73 (22)	61 (22)	12 (25)
Gynecological and breast	61 (19)	50 (18)	11 (23)
Urologic	48 (15)	42 (15)	6 (12)
Skin	15 (5)	12 (4)	3 (6)
Sarcoma	0 (0)
Others	8 (2)	8 (3)	0 (0)
Fall risk screening
Previous assessment by a healthcare professional	22 (7)	19 (7)	3 (6)	> 0.99
Recommendations issued	17 (77)	15 (79)	2 (67)
Performed in year preceding inclusion	12 (55)	11 (58)	1 (33)
Oncological treatment	287 (88)	247 (88)	40 (83)	0.35
Chemotherapy agents or immunotherapy	269 (94)	231 (94)	38 (95)	> 0.99
Hormonotherapy	23 (9)	21 (10)	2 (5)	0.55
Targeted therapy	50 (19)	41 (18)	9 (23)	0.49
Radiotherapy ♦	76 (28)	64 (28)	12 (28)	0.99
Total dose received at inclusion (in Grays), median (IQR)	32.0 (14.0–54.0)	33.5 (14.5–54.0)	19.0 (10.0–38.0)
Number of fractions received, median (IQR)	10.0 (5.0–20.0)	10.5 (5.0–21.0)	7.0 (4.0–18.0)
Current non-oncological medications ♦	288 (93)	247 (93)	41 (91)	0.76
< 5 drugs	111 (39)	100 (41)	11 (28)	0.13
≥ 5 drugs	172 (61)	144 (59)	28 (72)
Median number of oncological treatments (IQR)	1 (1–2)	1 (1–2)	1 (1–2)	0.65
Oncological treatment ATC2
L01 Antineoplastic agents (%raw)(%column)	264 (80)	226 (86)(81)	38 (14)(79)	0.81
L02 Endocrine therapy	17 (5)	16 (94)(6)	1 (6)(2)	0.49
L04 Immunosuppressants	8 (2)	7 (87.5)(2.5)	1 (12.5)(2)	> 0.99
Taxanes	54 (16)	41 (76)(15)	13 (24)(27)	0.032*
Platinum salts	81 (25)	71 (88)(25)	10 (12)(21)	0.51
Median number of FRIDs (IQR) ♦	1 (0–2)	1 (0–2)	1 (0–2)	0.034*
Patients with 0 FRID	133 (41)	117 (88)(42)	16 (12)(33)	0.023*
Patients with 1 FRID	92 (28)	83 (90)(30)	9 (10)(19)
Patients with ≥ 2 FRIDs	102 (31)	79 (77)(28)	23 (23)(48)

n number, FRID fall-risk-increasing drugs, IQR interquartile range

^*Significant p-value

^♦Missing data

In the multivariate analysis (Table 3), age was significantly associated with FR, with a 30% increase in risk per 5-year age increment (95% CI, 1.03–1.73, p = 0.028). Receiving taxanes alone was associated with a 3.74-fold higher risk of being in the high FR subgroup (95% CI, 2.12–6.60, p < 0.01). Finally, being prescribed at least 2 FRIDs doubled the risk of being at high FR (OR = 2.19, 95% CI, 1.23–3.92, p = 0.008).

Table 3.

Independent risk factors identified using a generalized linear model (binomial distribution, logit link), accounting for center effect

Independent risk factors identified	aOR	95% CI	p-value
Age (units = 5)	1,33	1.03–1.73	0.028
At-risk counties	1.56	0.79–3.04	0.20
Taxanes alone	3.74	2.12–6.60	< 0.01
Number of medications included in STOPPFall list
0	Ref
1	0.7	0.21–1.37	0.29
≥ 2	2.19	1.23–3.92	0.008

c-index = 0.67; Hosmer and Lemeshow goodness-of-fit, p = 0.71, aOR adjusted odds ratio, CI confidence interval

At the 3-month follow-up, preventive recommended interventions for patients at high FR (n = 42) were fully implemented in only 24% (n = 10) of patients and partially implemented in 71% (n = 30) (Fig. 2). The three most frequently carried out interventions were the biological dosage (60%, n = 25), medication review (48%, n = 20), and cardiovascular evaluation (43%, n = 18). Home environment safety interventions were the least frequently implemented (31%, n = 13) followed by cognitive assessment (26%, n = 11), and visual and auditory acuity evaluation (29%, n = 12).

Fig. 2.

Distribution and reasons for non-implementation of preventive recommendations at 3 months

Medication reviews (n = 20) could have been performed by the general practitioner (n = 6), the oncologist (n = 1), in a geriatric ACU (n = 5) or by a mobile geriatric team (n = 7).

Discussion

This multicenter, prospective, real-life study is, to our knowledge, the first to implement the 2022 World Guidelines for Falls Prevention and Management in older adults undergoing active cancer treatment [17]. It provides new evidence that systematic FR screening can be integrated into routine oncology care and documents the feasibility of implementing the recommended preventive measures.

Prevalence and determinants of FR

The observed prevalence of self-reported falls (21%) is consistent with recent meta-analyses in older adults with cancer (~ 20–30%) [2, 3]. These rates are slightly lower than those reported in community-dwelling older adults without cancer (30–40%) [5], likely reflecting a potential selection bias of individuals eligible for active cancer treatment, frailer than the general aged population. Nonetheless, falls remain a major safety issue in oncology, given their potential to interrupt treatment due to unplanned admissions for fall-related injuries, and worsen prognosis [4].

A significant center effect was observed: FR was higher in general and private hospitals than in university hospitals and comprehensive cancer centers. In France, marked disparities in medical density exist. Patients followed in general hospitals tended to live in rural, medically underserved areas. The higher FR observed in private hospitals may reflect differences in referral patterns, case mix, or access to multidisciplinary geriatric support; however, these factors were not collected, and this interpretation remains exploratory. Overall, the center effect suggests that structural and organizational inequalities may amplify the vulnerability of older adults treated for cancer.

Age, taxane-based chemotherapy, and exposure to two or more FRIDs were independently associated with high FR. However, the mean age between the high FR and the other combined subgroup was not statistically different. Especially our results underline the importance of systematic FR screening from the age of 65, as recommended by the WHO, the 2022 World Guidelines for Falls Prevention and Management [17] and the 2023 ASCO Guideline on the assessment of vulnerabilities in older adults with cancer [15]. The other two independently associated factors (taxane-based chemotherapy and exposure to at least two FRIDs) are pharmacological and potentially modifiable.

The strong association between taxane exposure and high FR (aOR 3.74) deserves particular attention. Neurotoxicity induced by taxanes is well documented and may induce gait instability, proprioceptive deficits, and sensory ataxia [22–25]. A previous study reported that older adults receiving two neurotoxic chemotherapy regimens had 9.5 falls per 1000 person-months versus 7.76 for those receiving one [26]. Another study in patients with chemotherapy-induced peripheral neuropathy showed a higher risk with taxanes than with platinum salts (p = 0.022), with no significant influence of age [23]. However, to our knowledge, none has demonstrated a direct association with FR. Taxanes multi-weekly regimens have been reported to increase neurotoxicity risk, but dosing schedules have not been collected in our study to look into that [27]. Platinum salts are also neurotoxic and ototoxic [28]. The ototoxicity of cisplatin is particularly well-established and may cause irreversible cochlear damage leading to sensorineural hearing loss, tinnitus, and balance disorders, resulting from cumulative injury to inner and outer hair cells [29]. However, the association between platinum salts exposure and high FR was not significant in our results, possibly due to frequent systematic dose reductions in older adults. These findings support the systematic integration of FR assessment into chemotherapy planning and follow-up, with targeted corrective interventions such as rehabilitation or adapted physical activity when indicated.

The association between two or more FRIDs and high FR (aOR 2.19) is consistent with recent studies [10, 30]. Cumulative exposure to high-risk medications appears to predict falls more accurately than total prescription count, supporting deprescribing initiatives prioritizing psychotropics, opioids, and anticholinergic agents [10, 14]. Interestingly, polypharmacy itself, although frequent, was not independently associated with FR. Yet several recent studies have reported high polypharmacy prevalence in older adults with cancer: between 45.2% and 90.8%. Polypharmacy is often associated with an increased risk of fall-related hospitalizations, particularly in cases of excessive polypharmacy (≥ 10 medications) (OR = 3.19; 95% CI 1.61–6.32) [34]. This suggests that prescription quality matters more than quantity. The polypharmacy rates observed here are consistent with recent studies in older adults with cancer [11, 12]. Structured medication reviews, ideally pharmacist- or geriatrician-led, should therefore be systematically integrated into oncology care [18, 35].

Implementation of fall-prevention recommendations

At 3 months, only 24% of high-risk patients had fully benefited from the implementation of the recommended preventive interventions, while 71% received only partial interventions. Laboratory tests, medication reviews, and cardiovascular assessments were the most frequently completed measures, reflecting actions easily integrated into routine practice. On the contrary, home-safety checks, cognitive assessments, and sensory evaluations were rarely performed. The main motives were patients’ refusal of some recommended measures (psycho-cognitive assessments, home-safety evaluations, nutritional assessments). Since not all patients presented deficits in these domains, this underlines the importance of individualized approaches and patient-centered care. Similar limitations have been described in general geriatric cohorts, where adherence to multidimensional fall-prevention programs rarely exceeds 30% [36]. Moreover, the feasibility of these programs is limited by both individual and organizational constraints. Key barriers are multifactorial and include time constraints, limited staffing, competing oncological priorities, and lack of coordination among specialists [36, 37]. A 3-month period is probably too short to schedule specialist referrals. Our findings highlight the gap between fall-prevention recommendations and their real-world implementation feasibility across healthcare settings. Even simple interventions, such as vitamin D supplementation or referral to a physical therapist, were applied inconsistently, suggesting organizational rather than clinical barriers. The results highlight an urgent need for structured, team-based pathways rather than relying solely on practitioner vigilance.

Organizational and clinical implications

Our results emphasize the clinical and public health relevance of systematically integrating FR screening into cancer care pathways of older patients. Fall prevention in oncology requires integrated interdisciplinary processes involving oncologists, general practitioners, pharmacists, physiotherapists, and geriatricians [35, 37, 38]. Implementation could be optimized through co-management models similar to those recommended by the International Society of Geriatric Oncology (SIOG) and the ESMO/SIOG Cancer in the Elderly Working Group [39, 40]. Collaborative models integrating oncologists, geriatricians, and pharmacists have demonstrated 20–40% reductions in medication-related problems and improved adherence to preventive interventions [35, 38, 41]. The OPTIMEDOC program (optimization of medications in the care pathway of older adults), funded by the regional Health Agency, offers multidisciplinary combined hospital-community care meetings in which geriatricians, clinical pharmacists, general practitioners, and community pharmacists collaborate to revise medication use in general population aged ≥ 75 with polypharmacy. This program could be a pragmatic model transferable to oncology setting in adults aged ≥ 65 to facilitate the deprescription of STOPPFall drugs [10, 30]. Beyond individual management, such models could strengthen coordination between hospital and community-based providers.

Future perspectives

These findings open perspectives for future research: interventional trials combining digital monitoring of falls and neuropathic symptoms, geriatric assessment, and pharmacist-led deprescribing [37, 38, 38–41]. Randomized studies integrating these approaches could assess cost-effectiveness and efficacy on fall occurrence reduction. To start with, the subsequent research project will be a feasibility pilot randomized trial in adults aged 65 years or older referred to taxane-based chemotherapy. Patients with baseline intermediate or high FR will be randomized 1:1 to usual care (as in CHUPACHOUP study) or to an advanced practice nurse-coordinated personalized multimodal fall-prevention intervention. This intervention would encompass a comprehensive geriatric assessment and management, a medication optimization, remotely supervised strength and balance exercises, and a home assessment by an occupational therapist. Falls will be recorded monthly over the phone and FR reassessed by advanced nurses every 3 months up to 12 months. The primary objective on feasibility would be to confirm the implementation (i.e., intervention delivery and acceptability) of the multimodal intervention. Patients with baseline low FR will be followed up in a parallel observational cohort but could be offered the pilot trial if their FR worsen during cancer care pathway. This approach directly aligns with WHO and 2022 World Guidelines for Falls Prevention and Management recommendations to develop context-adapted and scalable fall-prevention frameworks [17, 39].

Strengths and limitations

The strengths of this study rely on its multicenter design and the integration of the STOPPFall classification mapped to ATC codes. This is the first study to provide specific data on older patients treated for cancer, whereas the literature has so far been limited to the general geriatric population. The real-world setting also improves the generalizability of the results. However, several limitations must be acknowledged. The cross-sectional design precludes causal inference. Because funding did not permit follow-up of the entire cohort, follow-up was limited to the high FR subgroup. Although the 2022 World Guidelines recommend reassessment at 12 months for the low and intermediate FR subgroups, such follow-up would likely have increased attrition and would have been difficult to interpret because patients’ clinical status and cancer treatment could change substantially during that interval. Selection bias is possible, as we recruited in ACU and not out consultations; patients under oral cancer therapies for breast or prostate cancer were barely represented in our sample. Self-reported falls may lead to underestimation if patients had cognitive recall impairments; this was not assessed. Apart from the gait speed to separate intermediate from low subgroups, no other functional test nor neurological examination were performed to better understand their potential contribution in fall prevalence. Finally, despite adjusting for center effect, unmeasured confounding factors, such as socioeconomic status or rural living place may have influenced the results.

Conclusion

This study showed that, although the risk of falls in older adults undergoing cancer treatment is a major concern, the implementation of fall-prevention guidelines into practice remains limited. Taxane exposure and prescription of ≥ 2 FRIDs represent key, modifiable risk factors that should systematically trigger FR assessment and multidisciplinary management. Medication reviews and coordinated patient-centered specific care, at treatment initiation and in the follow-up, are potentially relevant tracks to reduce FR. Embedding structured fall-prevention protocols into oncology care could therefore strengthen patient safety and treatment adherence, contributing to the broader implementation of the 2022 World Guidelines in complex care populations.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

Study conception and design: Bérangère BEAUPLET and Anaïs BRIANT. Material preparation: Bérangère BEAUPLET, Poppy EVENDEN, François FOURNEL. Data collection: Romain LEVARD, Océane MESNILGRANTE, Priscille LE BON, Jean-Matthieu L’ORPHELIN, Soazig LEBAUBE, Maud VILLEMIN, Elise DELARUE, Angélique DUMONT, Samira BASTI, Marianne HAMON, Laure KALUZINSKI, Lucie POTTIER, Delphine DUMONT, Eddy FOURNIER, Patricia BERNARDIN, Tony CALACHE, Guillaume BOUDIN, Helen HOMOKOS, Cyril MAGNAN and Alexandre MEURANT. Formal analysis: Anaïs BRIANT. The first draft of the manuscript was written by Romain LEVARD and Océane MESNILGRANTE and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The study was funded by the Direction Générale de l’Offre de Soins, through an annual grant of the Normandy Interregional Oncogeriatric Coordination Unit at OncoNormandie.

Data availability

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

Declarations

Conflict of interests

The authors declare no competing interests.