Prospective study of falls and risk factors for falls in adults with advanced cancer.
Stone, Carol A Lawlor, Peter G Savva, George M Bennett, Kathleen Kenny, Rose Anne
Stone, Carol A
Lawlor, Peter G
Savva, George M
Bennett, Kathleen
Kenny, Rose Anne
An error occurred retrieving the object's statistics
Advisors
Editors
Other Contributors
Departments
Our Lady's Hospice and Care Services, Harold's Cross, Ireland. carolstone130@yahoo.co.uk
Date
2012-06-10
Date Submitted
Keywords
Other Subjects
Subject Mesh
Accidental Falls
Aged
Aging
Female
Humans
Incidence
Male
Middle Aged
Neoplasms
Palliative Care
Proportional Hazards Models
Prospective Studies
Regression Analysis
Risk
Risk Factors
Time Factors
Aged
Aging
Female
Humans
Incidence
Male
Middle Aged
Neoplasms
Palliative Care
Proportional Hazards Models
Prospective Studies
Regression Analysis
Risk
Risk Factors
Time Factors
Planned Date
Start Date
Collaborators
Principal Investigators
Alternative Titles
Publisher
Abstract
Retrospective studies of inpatients with cancer suggest that a cancer diagnosis confers a high risk of falls. In adults with advanced cancer, we aimed to prospectively document the incidence of falls, identify the risk factors, and determine if falls in this population occur predominantly in older patients.
Patients admitted consecutively to community and inpatient palliative care services with metastatic or locoregionally advanced cancer who were mobile without assistance were recruited. Risk-factor assessment was conducted on initial encounter. Patients underwent follow-up via weekly telephone contact for 6 months or until time of fall or death. Relationship between covariates and time to fall was examined using hazard ratios (HRs) derived from univariate and multivariate Cox proportional hazards models.
Of 185 participants (52.4% men; mean age 68 ± standard deviation of 12.6 years), 50.3% fell; 35 (53%) of 66 participants age < 65 years and 58 (48.7%) of 119 age ≥ 65 years fell; 61.3% of falls occurred in the community; 42% resulted in injury. Median time to fall was 96 days (95% CI, 64.66 to 127.34). Primary brain tumor or brain metastasis (HR 2.5; P = .002), number of falls in the preceding 3 months (HR, 1.27; P = .005), severity of depression (HR, 1.12; P = .012), benzodiazepine dose (HR, 1.05; P = .004), and cancer-related pain (HR, 1.96; P = .024) were independently associated with time to fall in multivariate analysis.
Fifty percent of adults with advanced cancer, regardless of age, will experience a fall associated with high risk of physical injury. There is a compelling need to assess the efficacy of assessment and management of modifiable fall risk factors in patients with advanced cancer.
Patients admitted consecutively to community and inpatient palliative care services with metastatic or locoregionally advanced cancer who were mobile without assistance were recruited. Risk-factor assessment was conducted on initial encounter. Patients underwent follow-up via weekly telephone contact for 6 months or until time of fall or death. Relationship between covariates and time to fall was examined using hazard ratios (HRs) derived from univariate and multivariate Cox proportional hazards models.
Of 185 participants (52.4% men; mean age 68 ± standard deviation of 12.6 years), 50.3% fell; 35 (53%) of 66 participants age < 65 years and 58 (48.7%) of 119 age ≥ 65 years fell; 61.3% of falls occurred in the community; 42% resulted in injury. Median time to fall was 96 days (95% CI, 64.66 to 127.34). Primary brain tumor or brain metastasis (HR 2.5; P = .002), number of falls in the preceding 3 months (HR, 1.27; P = .005), severity of depression (HR, 1.12; P = .012), benzodiazepine dose (HR, 1.05; P = .004), and cancer-related pain (HR, 1.96; P = .024) were independently associated with time to fall in multivariate analysis.
Fifty percent of adults with advanced cancer, regardless of age, will experience a fall associated with high risk of physical injury. There is a compelling need to assess the efficacy of assessment and management of modifiable fall risk factors in patients with advanced cancer.
Language
en
ISSN
1527-7755
eISSN
ISBN
DOI
10.1200/JCO.2011.40.7791
PMID
22585687