Visual Abstract
Keywords: hemodialysis
Abstract
Key Points
Of the 643 patients undergoing long-term hemodialysis enrolled in a clinical trial for chronic pain, 28% experienced at least one fall, an incidence of 0.68 per participant year.
Accidents were the most frequent cause of falls, and it was rare for them to be related to the hemodialysis procedure or occur at the hemodialysis unit.
In multivariable analyses, demographic characteristics, severity of pain symptoms, or medication use such as opioids were associated with the fall risk.
Background
Falls are believed to be common in patients undergoing maintenance hemodialysis, but little is known about their frequency or outcomes. In this prospective study, we sought to increase our knowledge regarding the incidence, timing, circumstances, and outcomes of falls in this population.
Methods
Between January 2021 and April 2023, adults undergoing maintenance hemodialysis from 103 US dialysis facilities were enrolled in the Heart Outcomes Prevention Evaluation Consortium trial, which randomized participants with moderate or severe chronic pain to a pain coping skills, cognitive behavioral therapy intervention, or usual care. Occurrence of falls was a prespecified trial outcome. The research team inquired about falls at each 4-week follow-up visit during the 36-week study. Multivariable regression was used to explore associations of demographic and clinical characteristics, including patient-reported symptoms, with fall risk.
Results
Of 643 trial participants, 178 (28%) experienced 293 falls over a cumulative follow-up period of 429 participant-years for an overall rate of 0.68 falls per participant-year (95% confidence interval, 0.61 to 0.76). Accidents were the most frequent cause of falls (38%). It was rare for falls to be related to the hemodialysis treatment or to occur in the hemodialysis unit. Of the 293 falls, 36 (12%) were evaluated in the emergency department without subsequent hospitalization, 41 (14%) resulted in a hospital admission, and 19 (7%) led to a fracture. In multivariable analyses, demographic characteristics, severity of pain symptoms, or medication use such as opioids at enrollment were not associated with the fall risk.
Conclusions
Falls were common in this cohort of maintenance hemodialysis patients with chronic pain, occurring in 28% of individuals during a planned follow-up of 36 weeks. Falls rarely occurred in the dialysis unit, with the vast majority occurring at participants' homes and due to accidental causes. There was no significant association between patient-reported symptoms or medication use and the risk of subsequent falls.
Clinical Trial registry name and registration number:
ClinicalTrials.gov, NCT04571619.
Introduction
Falls are common, particularly in older adults. More than one quarter of older adults fall each year, often causing major injury.1 Available data suggest that falls are also common in patients undergoing maintenance hemodialysis, with studies reporting a higher rate of falls compared with the general population.2,3 Furthermore, patients treated with maintenance dialysis who have previously experienced a fall are more than twice as likely to suffer new falls, compared with those who have not previously fallen.4
Notwithstanding the higher frequency of falls, little is known about the circumstances associated with falls among hemodialysis patients, including the timing of falls related to hemodialysis sessions,4 or other factors that may be unique to people on hemodialysis. For example, impaired balance is reported as a risk factor for falls, dynamic balance has been shown to worsen after hemodialysis treatment, and posthemodialysis impaired dynamic balance was found more frequently in patients who had falls.5 Patients undergoing maintenance dialysis also report a large number of symptoms, and the symptoms themselves or their treatments (such as opioids or benzodiazepines) may increase risk of falls. For example, among middle-aged and older adults, pain is associated with a higher risk of falls.6–10 It might be expected that commonly present symptoms or medication use would also be associated with a higher risk of falls among patients treated with dialysis. Yet, to the best of our knowledge, these associations have previously not been assessed.
The Heart Outcomes Prevention Evaluation (HOPE) Consortium Trial to Reduce Pain and Opioid Use in hemodialysis (HOPE Trial) evaluated the effectiveness of a cognitive behavioral therapy intervention called pain coping skills training to reduce pain interference in patients receiving maintenance hemodialysis.11 In this analysis, we explored the incidence, timing, circumstances, associations, and consequences of falls in patients undergoing maintenance hemodialysis.
Methods
Study Population
Details of the HOPE trial have been previously published.12 In brief, between January 2021 and April 2023, 643 individuals with kidney failure undergoing maintenance hemodialysis were enrolled from 103 hemodialysis facilities associated with 16 Clinical Centers nationwide. The Institutional Review Board (IRB) at the University of Pennsylvania served as the IRB of record for 11 non-US Department of Veterans Affairs (VA) enrolling sites, and the VA Central IRB served as the IRB of record for the five VA enrolling sites.
Eligibility criteria included 18 years or older, treatment with in-center maintenance hemodialysis for ≥90 days, moderate or severe chronic pain defined as self-report of pain on most days or every day during the past 3 months, and a score of ≥4 (out of a maximum of 10) on the Pain, Enjoyment of Life, and General Activity (PEG) scale.13 Individuals with current substance use disorder, suicidal intent, significant cognitive impairment, anticipated change in kidney replacement modality within 6 months, and life expectancy <6 months were ineligible for participation. Of the 643 participants, 319 were randomly assigned to the pain coping skills training and 324 to usual care, with follow-up for up to 36 weeks after randomization. Follow-up of the last randomized participant was completed in December 2023. Pain interference was measured by the Brief Pain Inventory Pain Interference (BPI) scale (BPI Interference) and Brief Pain Inventory Severity Severity Scale (BPI Severity) at baseline and then at weeks 12, 24, and 36.14
Ascertainment of Falls and Circumstances and Consequences of Falls
Participants were contacted by the research team every 4 weeks for the duration of their trial participation, either in person or by telephone, to ascertain adverse events and clinical outcomes, update medication records, and schedule calls to assess patient-reported outcomes. Occurrence of falls was a prespecified outcome in the trial, and the research team inquired about falls at each follow-up visit. Structured data elements and narrative descriptions for fall events submitted by site research teams during the trial were used by a subset of trial investigators after completion of the trial to extract information about the circumstances (cause, relationship to hemodialysis, location) and immediate consequences (visit to emergency department, hospital admission, fracture, medical care, and nature of medical care) of the falls. The data on every fall were audited and reconciled for consistent adjudication by two of the authors (D.M. Charytan and R. Mehrotra). Data on hospital admissions and death through 6 months of follow-up related to the fall were abstracted from serious adverse events forms filed by the clinical centers.
Covariate Data Sources
Baseline demographic and clinical data were obtained from case report forms completed by members of the research team at the time of enrollment of participants, with information obtained either by patient interview or review of medical records. Medications evaluated for their association with falls included benzodiazepines, opioids, soporifics, gabapentinoids, antidepressants, and antihistamines. Patient-reported data ascertained at baseline and every 12 weeks through week 36 included pain interference and severity, depression, anxiety, sleep disturbance, physical function and fatigue, and an overall assessment of symptom burden with the Dialysis Symptom Index (DSI). The collection schedule and instruments used to quantify other psychosocial factors and patient-reported data are detailed in the full protocol, which is available with the Supplemental Appendix.11
Statistical Analyses
Descriptive statistics (mean with SD, median with interquartile range [IQR], or frequency with percentage) were used to describe baseline demographic and clinical characteristics and patient-reported measures. Zero-inflated negative binomial regression with a log link and an offset for follow-up time was used to quantify associations with the number of falls. This approach uses a two-part model including a logistic regression model and a count outcome model. Variables for inclusion in the nested models were prespecified and on the basis of known or expected associations with fall risk and included: (1) model 1 with age, sex, race, body mass index, education, alcohol use, marijuana use, and randomized treatment assignment and (2) model 2 with covariates in model 1 and benzodiazepines, gabapentinoids, opioid, diabetes, and stroke. Given collinearity between patient-reported outcome instruments of interest quantified by variance inflation factors (a statistical metric used to detect multicollinearity in regression analysis, shown in Supplemental Table 1), BPI Interference, BPI Severity, Patient-Reported Outcomes Measurement Information System (PROMIS) Sleep, PROMIS Fatigue, and PROMIS Physical Function were tested for their association with falls in hierarchical models one variable at a time. In the zero-inflation logistic model component, we only adjusted for death. Specifically, the zeroinfl function of the R package pscl was used to fit the zero-inflated negative binomial regression. Given the low proportion of missing data, a complete case analysis was used. For all analyses, the overall level of significance was set to α=0.05. Data analyses were performed using R (version 4.3.2; https://cran.r-project.org/).
Results
Frequency of Falls
Demographic and clinical characteristics of trial participants are summarized according to baseline BPI score (Table 1) and by the number of falls (Table 2). Among the 643 participants, 178 (28%) experienced 293 falls over a cumulative follow-up period of 429 participant-years for an overall rate of 0.68 falls per participant-year (95% confidence interval, 0.61 to 0.76). The first fall occurred a median of 95 days (IQR, 44–155 days) from randomization.
Table 1.
Characteristics of participants enrolled in the Heart Outcomes Prevention Evaluation Consortium according to brief pain inventory category at baseline
| Characteristic | All (N=643) | BPI Interferencea | ||
|---|---|---|---|---|
| Low (1–3) (n=86) |
Medium (4–7) (n=400) |
High (≥8) (n=156) |
||
| Age at randomization (yr), mean (SD) | 60 (13) | 62 (12) | 60 (13) | 61 (11) |
| Race, n (%) | ||||
| American Indian/Alaskan Native | 22 (3) | 3 (3) | 14 (4) | 5 (3) |
| Asian/Asian American | 6 (1) | 0 (0) | 6 (2) | 0 (0) |
| Black | 307 (48) | 37 (43) | 190 (48) | 80 (51) |
| Multiple races | 11 (2) | 2 (2) | 7 (2) | 2 (1) |
| Native Hawaiian/other Pacific Islander | 6 (1) | 2 (2) | 2 (1) | 2 (1) |
| Race not reported | 80 (13) | 8 (9) | 47 (12) | 25 (16) |
| White | 210 (33) | 34 (40) | 134 (34) | 42 (27) |
| Hispanic ethnicity, n (%) | 119 (19) | 8 (9) | 73 (18) | 38 (24) |
| Male sex, n (%) | 350 (55) | 52 (60) | 226 (56) | 72 (46) |
| Education, n (%) | ||||
| Did not complete secondary school or high school | 138 (21) | 16 (19) | 73 (18) | 49 (31) |
| High school degree | 269 (42) | 33 (38) | 181 (45) | 55 (35) |
| Associate's or technical degree | 105 (16) | 16 (19) | 61 (15) | 28 (18) |
| College degree, doctoral, or postgraduate education | 130 (20) | 21 (24) | 85 (21) | 24 (15) |
| Current employment status, n (%) | ||||
| Full-time | 28 (4) | 7 (8) | 18 (5) | 3 (2) |
| Part-time | 25 (4) | 6 (7) | 15 (4) | 4 (3) |
| Disabled | 436 (68) | 53 (62) | 269 (68) | 114 (73) |
| Retired | 125 (20) | 19 (22) | 77 (19) | 29 (19) |
| Other | 26 (4) | 1 (1) | 19 (5) | 6 (4) |
| Married or with domestic partner, n (%) | 200 (31) | 34 (40) | 124 (31) | 42 (27) |
| Living alone, n (%) | 203 (32) | 19 (22) | 133 (33) | 51 (33) |
| Annual household Income, n (%) | ||||
| <$25,000 | 237 (37) | 26 (30) | 145 (36) | 66 (42) |
| $25,000–$49,999 | 101 (16) | 14 (16) | 62 (16) | 25 (16) |
| ≥$50,000 | 80 (12) | 6 (7) | 54 (14) | 20 (13) |
| Prefer not to answer | 224 (35) | 40 (47) | 139 (35) | 45 (29) |
| VA site participant, n (%) | 69 (11) | 9 (10) | 44 (11) | 16 (10) |
| Dialysis schedule, n (%) | ||||
| Monday-Wednesday-Friday | 367 (58) | 51 (59) | 231 (58) | 85 (54) |
| Tuesday-Thursday-Saturday | 242 (38) | 31 (36) | 150 (38) | 61 (39) |
| Tuesday-Thursday-Sunday | 14 (2) | 1 (1) | 8 (2) | 5 (3) |
| Other | 14 (2) | 3 (3) | 6 (2) | 5 (3) |
| Primary underlying kidney disease diagnosis, n (%) | ||||
| Diabetes | 239 (37) | 30 (35) | 151 (38) | 58 (37) |
| Hypertension | 118 (18) | 19 (22) | 70 (18) | 29 (19) |
| Glomerular disease | 23 (4) | 2 (2) | 14 (4) | 7 (4) |
| Other or unknown | 262 (41) | 35 (41) | 165 (41) | 62 (40) |
| Years since the initiation of maintenance hemodialysis, yr, n (%) | ||||
| <1 | 144 (22) | 21 (24) | 82 (20) | 41 (26) |
| 1–5 | 296 (46) | 41 (48) | 194 (48) | 61 (39) |
| >5 | 202 (31) | 24 (28) | 124 (31) | 54 (35) |
| Previous kidney transplant, n (%) | 37 (6) | 6 (7) | 24 (6) | 7 (4) |
| Cardiovascular disease, n (%) | 389 (63) | 48 (58) | 248 (64) | 93 (62) |
| Coronary artery disease, n (%) | 175 (28) | 23 (27) | 113 (29) | 39 (25) |
| Peripheral vascular disease, n (%) | 102 (16) | 13 (16) | 70 (18) | 19 (13) |
| Stroke or TIA, n (%) | 128 (20) | 15 (17) | 82 (21) | 31 (20) |
| Diabetes mellitus, n (%) | 380 (59) | 51 (59) | 224 (56) | 105 (67) |
| History of cancer, n (%) | 105 (16) | 16 (19) | 69 (17) | 20 (13) |
| Any psychologic diagnosis, n(%) | 259 (40) | 23 (27) | 154 (38) | 82 (53) |
| Neuropathic pain, n (%) | 424 (67) | 43 (50) | 258 (66) | 123 (79) |
| Chronic lung disease, n(%) | 75 (12) | 10 (12) | 41 (10) | 24 (15) |
| Chronic liver disease, n (%) | 23 (4) | 1 (1) | 18 (5) | 4 (3) |
| Autoimmune disorder, n (%) | 51 (8) | 6 (7) | 37 (9) | 8 (5) |
| Current smoking, n (%) | 98 (15) | 14 (16) | 52 (13) | 32 (21) |
| Any alcohol use, n (%) | 12 (2) | 1 (1) | 7 (2) | 4 (3) |
| Current cannabis use, n (%) | 102 (16) | 6 (7) | 71 (18) | 25 (16) |
| Current cocaine use, n (%) | 5 (0.8) | 1 (1) | 2 (0.5) | 2 (1) |
| History of heroin use, n (%) | 15 (2) | 3 (3) | 5 (1) | 7 (4) |
| BMI (kg/m2), mean (SD) | 31.3 (12.6) | 31.1 (8.7) | 31.7 (14.4) | 30.3 (9.1) |
| Predialysis systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, mean (SD) | 372 (58) | 50 (58) | 227 (57) | 95 (61) |
| Albumin (g/dl), mean (SD) | 3.9 (0.4) | 3.9 (0.4) | 3.9 (0.4) | 3.9 (0.4) |
| Hemoglobin (g/dl), mean (SD) | 10.9 (1.7) | 11.0 (1.6) | 10.9 (1.7) | 10.6 (1.6) |
| Medication use, n (%) | ||||
| Benzodiazepines | 34 (5) | 2 (2) | 22 (6) | 10 (6) |
| Opioids | 99 (15) | 10 (12) | 63 (16) | 26 (17) |
| Soporifics (sleeping aids) | 118 (18) | 14 (16) | 74 (18) | 30 (19) |
| Gabapentinoids | 182 (28) | 22 (26) | 112 (28) | 48 (31) |
| Antidepressants | 133 (21) | 16 (19) | 81 (20) | 36 (23) |
| Antihistamines | 134 (21) | 21 (24) | 77 (19) | 36 (23) |
Percentages are calculated using individuals with the number with available data as the denominator. One participant had missing brief pain inventory interference data. Employment status was missing in two participants in the medium brief pain inventory interference group. Dialysis schedule was missing in five participants in the medium brief pain inventory interference group. Cardiovascular disease history was missing in three participants in the low, ten in medium, and seven in the high brief pain inventory interference group. The history of coronary artery disease was missing two participants in the low, eight in the medium, and three in the high brief pain inventory interference group. Peripheral vascular disease history was missing in three participants in the low, nine in the medium, and eight in the high brief pain inventory interference group. Stroke or transient ischemic attack was missing in two participants in the medium and one in the high brief pain inventory interference group. Diabetes at baseline was missing in one participant in the medium brief pain inventory interference group. The history of cancer at baseline was missing in one participant in the medium brief pain inventory interference group. Neuropathic pain was missing in one participant in the medium brief pain inventory group. Chronic lung disease was missing in one participant in the low and three in the medium brief pain inventory interference group. Chronic liver disease was missing in one participant in the low, eight in the medium, and one in the high brief pain inventory interference group. The history of alcohol use was missing in ten participants in the low, 37 in the medium, and 15 in the high brief pain inventory interference group. Body mass index was missing in three participants in the medium and one in the high brief pain inventory interference group. Serum albumin was missing in 13 participants in the medium and two in the high brief pain inventory interference group. Hemoglobin was missing in seven participants in the medium brief pain inventory interference group. BMI, body mass index; BPI, brief pain inventory; TIA, transient ischemic attack; VA, Veterans Affairs.
BPI Interference, Brief Pain Inventory Pain Interference Scale score.
Table 2.
Characteristics of participants enrolled in the Heart Outcomes Prevention Evaluation Consortium according to the presence of a fall during the course of the study
| Characteristic | All (N=643) | No Falls (n=465) | ≥1 Fall (n=178) | Two Falls (n=42) | >2 Falls (n=26) |
|---|---|---|---|---|---|
| Age at randomization (yr), mean (SD) | 60 (13) | 60 (13) | 61 (13) | 61 (13) | 62 (13) |
| Race, n (%) | |||||
| American Indian/Alaskan Native | 22 (3) | 17 (4) | 5 (3) | 2 (5) | 0 (0) |
| Asian/Asian American | 6 (1) | 4 (1) | 2 (1) | 1 (2) | 0 (0) |
| Black | 308 (48) | 228 (49) | 80 (45) | 15 (36) | 16 (62) |
| Multiple races | 11 (2) | 8 (2) | 3 (2) | 1 (2) | 0 (0) |
| Native Hawaiian/Other Pacific Islander | 6 (1) | 5 (1) | 1 (1) | 0 (0) | 0 (0) |
| Race not reported | 80 (12) | 56 (12) | 24 (13) | 8 (19) | 0 (0) |
| White | 210 (33) | 147 (32) | 63 (35) | 15 (36) | 10 (39) |
| Hispanic ethnicity, n (%) | 119 (19) | 84 (18) | 35 (20) | 12 (29) | 1 (4) |
| Male sex, n (%) | 350 (54) | 265 (57) | 85 (48) | 21 (50) | 13 (50) |
| Education, n (%) | |||||
| Did not complete secondary school or high school | 138 (21) | 100 (22) | 38 (21) | 11 (26) | 5 (19) |
| High school degree | 270 (42) | 203 (44) | 67 (38) | 11 (26) | 12 (46) |
| Associate's or technical degree | 105 (16) | 75 (16) | 30 (17) | 7 (17) | 1 (4) |
| College degree, doctoral, or postgraduate education | 130 (20) | 87 (19) | 43 (24) | 13 (31) | 8 (31) |
| Current employment status, n (%) | |||||
| Full-time | 28 (4) | 19 (4) | 9 (5) | 1 (2) | 2 (8) |
| Part-time | 25 (4) | 19 (4) | 6 (3) | 4 (10) | 0 (0) |
| Disabled | 437 (68) | 317 (68) | 120 (68) | 25 (60) | 17 (65) |
| Retired | 125 (19) | 89 (19) | 36 (20) | 8 (19) | 6 (23) |
| Other | 26 (4) | 20 (4) | 6 (3) | 4 (10) | 1 (4) |
| Married or with domestic partner, n (%) | 200 (31) | 152 (33) | 48 (27) | 13 (31) | 6 (23) |
| Living alone, n (%) | 203 (32) | 141 (30) | 62 (35) | 15 (36) | 8 (31) |
| Annual household income, n (%) | |||||
| <$25,000 | 237 (37) | 166 (36) | 71 (40) | 12 (29) | 14 (54) |
| $25,000–$49,999 | 101 (16) | 73 (16) | 28 (16) | 8 (19) | 3 (23) |
| ≥$50,000 | 81 (13) | 60 (13) | 21 (12) | 5 (12) | 5 (19) |
| Prefer not to answer | 224 (35) | 166 (36) | 58 (33) | 17 (40) | 4 (15) |
| VA site participant, n (%) | 574 (89) | 405 (87) | 169 (95) | 42 (100) | 23 (88) |
| Dialysis schedule, n (%) | |||||
| Monday-Wednesday-Friday | 367 (57) | 272 (59) | 95 (54) | 19 (45) | 11 (42) |
| Tuesday-Thursday-Saturday | 243 (38) | 171 (37) | 72 (41) | 20 (48) | 13 (50) |
| Tuesday-Thursday-Sunday | 14 (2) | 12 (3) | 2 (1) | 0 (0) | 0 (0) |
| Other | 14 (2) | 6 (1) | 8 (4) | 3 (7) | 2 (8) |
| Primary underlying kidney disease diagnosis, n (%) | |||||
| Diabetic kidney disease | 239 (37) | 171 (37) | 68 (38) | 17 (41) | 9 (35) |
| Hypertension | 118 (18) | 96 (21) | 22 (12) | 7 (17) | 5 (19) |
| Glomerular disease | 24 (4) | 17 (4) | 7 (4) | 1 (2) | 1 (4) |
| Other or unknown | 262 (41) | 181(39) | 81 (46) | 17 (40) | 11 (42) |
| Years since the initiation of maintenance hemodialysis, yr, n (%) | |||||
| <1 | 144 (22) | 99 (21) | 45 (25) | 10 (24) | 8 (31) |
| 1–5 | 296 (46) | 212 (46) | 84 (47) | 19 (45) | 11 (42) |
| >5 | 203 (32) | 154 (33) | 49 (28) | 13 (31) | 7 (27) |
| Previous kidney transplant, n (%) | 37 (6) | 26 (6) | 11 (6) | 1 (2) | 1 (4) |
| History of cardiovascular disease, n (%) | 389 (62) | 277 (61) | 112 (65) | 26 (63) | 16 (64) |
| Coronary artery disease, n (%) | 175 (28) | 126 (28) | 49 (28) | 11 (28) | 6 (23) |
| Peripheral vascular disease, n (%) | 102 (16) | 75 (17) | 27 (16) | 9 (22) | 2 (8) |
| Stroke or TIA, n (%) | 128 (20) | 90 (19) | 38 (21) | 12 (29) | 4 (15) |
| Diabetes mellitus, n (%) | 380 (59) | 264 (57) | 116 (66) | 28 (67) | 17 (65) |
| History of cancer, n (%) | 106 (17) | 74 (16) | 32 (18) | 8 (19) | 3 (12) |
| Any psychologic diagnosis, n (%) | 260 (40) | 175 (38) | 85 (48) | 24 (57) | 15 (58) |
| Neuropathic pain, n (%) | 425 (67) | 297 (65) | 128 (72) | 31 (74) | 20 (77) |
| Chronic lung disease, n (%) | 75 (12) | 51 (11) | 24 (13) | 6 (14) | 6 (23) |
| Chronic liver disease, n (%) | 23 (4) | 16 (3) | 7 (4) | 2 (5) | 1 (4) |
| Autoimmune disorder, n (%) | 51 (8) | 38 (8) | 13 (7) | 2 (5) | 1 (4) |
| Current smoking, n (%) | 99 (15) | 68 (15) | 31 (17) | 6 (14) | 6 (23) |
| Any alcohol use, n (%) | 12 (2) | 7 (2) | 5 (3) | 0 | 2 (8) |
| Current cannabis use, n (%) | 102 (16) | 70 (15) | 32 (18) | 6 (14) | 8 (31) |
| Current cocaine use, n (%) | 5 (1) | 4 (1) | 1 (1) | 1 (2) | 0 (0) |
| History of heroin use, n (%) | 15 (2) | 11 (2) | 4 (2) | 1 (2) | 0 (0) |
| BMI (kg/m2) | 31.3 (12.6) | 31.6 (14.0) | 30.6 (8.0) | 30.3 (7.2) | 31.1 (11.0) |
| Predialysis systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg, mean (SD) | 373 (58) | 270 (58) | 103 (58) | 21 (50) | 19 (73) |
| Albumin (g/dl), mean (SD) | 3.9 (0.4) | 3.9 (0.4) | 3.9 (0.4) | 4.0 (0.4) | 3.9 (0.4) |
| Hemoglobin (g/dl), mean (SD) | 10.9 (1.7) | 10.8 (1.6) | 10.9 (1.8) | 10.8 (1.3) | 10.7 (1.2) |
| Medication use, n (%) | |||||
| Benzodiazepines | 34 (5) | 20 (4) | 14 (8) | 3 (7) | 4 (15) |
| Opioids | 100 (16) | 71 (15) | 29 (16) | 4 (10) | 7 (27) |
| Soporifics (sleeping aids) | 118 (18) | 85 (18) | 33 (19) | 9 (21) | 9 (35) |
| Gabapentinoids | 182 (28) | 127 (27) | 55 (31) | 13 (31) | 10 (38) |
| Antidepressants | 133 (21) | 89 (19) | 44 (25) | 13 (31) | 10 (38) |
| Antihistamines | 135 (21) | 97 (21) | 38 (21) | 9 (21) | 10 (38) |
Percentages are calculated using individuals with the number with available data as the denominator. Employment status was missing in two participants: one without and one with history falls. Dialysis schedule was missing in five participants, four without falls, and one with falls. Cardiovascular disease history was missing 20 participants: 13 without falls and seven with falls. The history of coronary disease was missing in 13 participants: nine without falls and four with falls. Peripheral vascular disease history was missing in 20 participants: 11 with falls and nine without falls. Stroke or transient ischemic attack was missing three participants: three without falls and zero with falls. Diabetes at baseline was missing in one participant: zero without falls and one with falls. Cancer at baseline was missing one participant: one without history of falls and zero with falls. Neuropathic pain was missing in seven participants: six without history of falls and one with falls. Chronic lung disease was missing in four participants: four without history of falls and zero with falls. Chronic liver disease was missing in ten participants: seven without falls and three with falls. The history of alcohol use was missing in 52 participants: 36 without falls and 16 with falls. Body mass index was missing in four participants: four without falls and zero with falls. Serum albumin was missing in 15 participants: 14 without falls and one with falls. Hemoglobin was missing in seven participants: five without falls and two with falls. BMI, body mass index; TIA, transient ischemic attack; VA, Veterans Affairs.
The mean follow-up time was 0.7±0.2 years overall and did not differ according to the number of falls. Of the 178 participants who sustained at least one fall, 68 (38%) had more than one fall over the follow-up period with a median of two falls (IQR, 2–3; range, 2–6 falls). The second fall in these individuals occurred a median of 57 days (IQR, 22–110 days) after the first fall during the trial period.
Circumstances and Consequences of Falls
The circumstances and immediate consequences of falls are summarized in Table 3. Falls were most frequently attributed to accidents, and it was rare for the falls to be related to the hemodialysis treatment or for them to occur in the hemodialysis unit. Of the 293 falls, 30 (10%) were followed by a visit to the emergency department without hospitalization, 41 (14%) resulted in a hospital admission, and 19 (7%) led to a fracture. Overall, 84 (29%) falls required some type of medical attention, but it was rare for participants to receive a new pain medication (Table 2).
Table 3.
Circumstances and immediate consequences of falls experienced by participants enrolled in the Heart Outcomes Prevention Evaluation consortium trial
| Circumstances of Fall, n (%) | All (N=293) |
|---|---|
| Primary cause | |
| Accident | 110 (38) |
| Gait instability/balance issues | 33 (11) |
| Hemodynamic instability (not related to hemodialysis) | 7 (2) |
| Balance/gait related to medications or substances | 2 (1) |
| Hemodynamic instability after hemodialysis | 2 (1) |
| Other causes | 103 (35) |
| Unable to assess | 42 (14) |
| Relatedness to hemodialysisa | |
| Related to hemodialysis b | 10 (3) |
| Unrelated | 271 (93) |
| Unable to assess | 10 (3) |
| Locationa | |
| Hemodialysis facilityc | 18 (6) |
| Home | 154 (53) |
| Insufficient data | 94 (32) |
| Other | 25 (9) |
| Resulted in | |
| ER visit without hospital admission | 30 (10) |
| Hospital admission | 41 (14) |
| Fracture | 19 (7) |
| Any medical care provided | 84 (29) |
| Type of medical care provided | |
| New pain medication (including OTC) | 10 (12) |
| Up-titration of existing pain meds | 0 |
| Casting/splinting | 3 (4) |
| Surgical fracture repair | 1 (1) |
| Care of laceration/soft tissue injury | 3 (4) |
| Surgery or radiologic intervention | 1 (1) |
| Other type of care | 66 (79) |
Missing in two participants.
Relatedness to hemodialysis was defined by falls occurring on the day of hemodialysis after the hemodialysis session.
Hemodialysis facility includes the waiting room and parking lot. Falls before the hemodialysis session were not considered related to hemodialysis regardless of location. ER, emergency room; OTC, over-the-counter
Although treatment of the fall itself infrequently required hospitalization, subsequent hospitalizations were common among patients with a fall. Overall, including hospitalizations related to the index fall, 83 (28%) and 135 (46%) fall events were followed by a hospital admission within 30 days and 6 months, respectively. There were no consistent associations between selected demographic or clinical factors or BPI interference or BPI severity with risk of sustaining a fall requiring hospital admission (Supplemental Table 2). In addition, 3 (1%) and 20 (7%) falls were followed by death within 30 days and 6 months, respectively.
Demographic and Clinical Characteristics, Symptom Burden, and Falls
Demographic and clinical characteristics were similar for the 178 participants who experienced at least one fall, compared with the 465 participants that did not have a fall (Table 2). Twenty-six (4%) participants had greater than two falls during follow-up. The use of medications with the potential to alter balance and gait stability was qualitatively higher in these participants compared with participants without falls. There were no meaningful differences in measures of pain severity or interference, depression, anxiety, sleep, fatigue, or physical function at the time of enrollment between those who experienced and those who did not experience a fall (Table 4). The pain coping skills treatment intervention was also not associated with falls risk. There were no clinically relevant differences in the prevalence of the three most frequently reported symptoms on the DSI between the two groups or between the distribution of responses for the most bothersome symptoms on the DSI (Table 4). In multivariable analyses, none of the included demographic or clinical characteristics were associated with the risk of falls in the study population (Table 5). In unadjusted models, BPI interference, BPI, Severity, and PROMIS Sleep Disturbance, Fatigue, and Physical Function scores were associated with risk of falls. Small albeit statistically significant associations between BPI interference, PROMIS Sleep, and PROMIS Fatigue remained after adjustment for demographics, comorbidities, and medications. BPI Interference was no longer associated with the risk of falls in the study population (Table 6) after additional adjustment for BPI severity, PROMIS Sleep, Fatigue, and Physical Function scores. BPI severity and PROMIS Sleep, Fatigue, and Physical Function scores were similarly not independently associated with falls in the model with simultaneous adjustment for all of these instruments (Supplemental Table 3).
Table 4.
Patient-reported outcomes at the time of enrollment in participants enrolled in the Heart Outcomes Prevention Evaluation Consortium trial, stratified by whether they experienced a fall during the course of the study
| Characteristic | All (N=643) | No Falls (N=465) | ≥1 Fall (n=178) |
|---|---|---|---|
| BPI interference, median (IQR) | 6.6 (5.1–7.9) | 6.4 (5.0–7.7) | 6.9 (5.6–8.1) |
| BPI severity, median (IQR) | 6.0 (4.5–7.5) | 6.0 (4.2–7.5) | 6.2 (4.5–7.5) |
| PHQ-9 (depression), median (IQR) | 8 (4–13) | 8 (4–12) | 9 (5–14) |
| GAD-7 (anxiety) score, median (IQR) | 5 (2–11) | 5 (2–11) | 7 (3–12) |
| PROMIS sleep disturbance T score, median (IQR) | 58.5 (52.3–65.0) | 57.3 (50.9–65.0) | 58.5 (53.6–65.0) |
| PROMIS fatigue T score, median (IQR) | 60.0 (53.7–66.4) | 58.8 (53.7–65.0) | 62.4 (55.1–67.8) |
| PROMIS physical function T score, median (IQR) | 33.2 (28.8–37.6) | 34.2 (28.8–38.5) | 32.3 (28.8–35.9) |
| DSI top three most frequent symptomsa, n (%) | |||
| Feeling tired | 567 (88) | 401 (86) | 166 (93) |
| Bone/joint pain | 492 (77) | 350 (75) | 142 (80) |
| Trouble staying asleep | 485 (75) | 349 (75) | 136 (76) |
| DSI top three most severe symptomsb | |||
| Feeling tired or lack of energy in the past week?, n (%) | |||
| Did not have symptom in the past week | 75 (12) | 63 (14) | 12 (7) |
| Not at all | 23 (4) | 18 (4) | 5 (3) |
| A little bit | 63 (10) | 49 (11) | 14 (8) |
| Somewhat | 125 (20) | 88 (19) | 37 (21) |
| Quite a bit | 138 (22) | 102 (22) | 36 (20) |
| Very much | 218 (34) | 144 (31) | 74 (42) |
| Missing | 1 | 1 | |
| Trouble falling asleep in the past week?, n (%) | |||
| Did not have symptom in the past week | 172 (27) | 131 (28) | 41 (23) |
| Not at all | 15 (2) | 12 (3) | 3 (2) |
| A little bit | 60 (9) | 43 (9) | 17 (10) |
| Somewhat | 102 (16) | 79 (17) | 23 (13) |
| Quite a bit | 106 (17) | 75 (16) | 31 (17) |
| Very much | 187 (29) | 124 (27) | 63 (35) |
| Missing | 1 | 1 | |
| Trouble staying asleep in the past week?, n (%) | |||
| Did not have symptom in the past week | 158 (25) | 116 (25) | 42 (24) |
| Not at all | 18 (3) | 16 (3) | 2 (1) |
| A little bit | 61 (10) | 48 (10) | 13 (7) |
| Somewhat | 112 (17) | 80 (17) | 32 (18) |
| Quite a bit | 114 (18) | 82 (18) | 32 (18) |
| Very much | 180 (28) | 123 (27) | 57 (32) |
For each of these scales, higher scores represent greater severity of the individual disorders. Brief pain inventory interference was missing in participant without falls. Brief pain inventory severity was missing in two participants without falls and one with falls. Patient health questionnaire-9 depression was missing in eight individuals without falls and six with falls. General anxiety disorder-7 anxiety was missing three participants without falls. Patient-reported outcome measurement information system sleep disturbance T score was missing five participants without falls and two with falls. Patient-reported outcome measurement information system fatigue T score was missing in six participants without falls and one with falls. Patient-reported outcome measurement information system physical function T score was missing in two participants without falls. Dialysis Symptom Index “tired” was missing in one participant without falls. BPI, brief pain inventory; DSI, dialysis symptom index; GAD-7, general anxiety disorder-7; IQR, interquartile range; PHQ-9, patient health questionnaire-9; PROMIS, patient-reported outcome measurement information system.
The default definition of most frequent symptoms is based on the most frequent “Yes” response among the entire study cohort (n=643).
The default definition of most severe symptoms is based on the most frequent “Very much” response among the entire study cohort (n=643).
Table 5.
Unadjusted and adjusted associations of demographic and clinical characteristics with falls in participants enrolled in Heart Outcomes Prevention Evaluation Consortium clinical trial
| Covariate | Level | Unadjusted | Model 1a | Model 2b | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Rate Ratio (95% CI) | P Value | Overall P Value | Rate Ratio (95% CI) | P Value | Overall P Value | Rate Ratio (95% CI) | P Value | Overall P Value | ||
| Age at randomization (yr) | 1.004 (0.99 to 1.02) | 0.55 | 1 (0.99 to 1.01) | 1.00 | 0.999 (0.98 to 1.01) | 0.84 | ||||
| Sex at birth | Male | Ref. | Ref. | Ref. | ||||||
| Female | 1.25 (0.91 to 1.72) | 0.17 | 1.22 (0.86 to 1.72) | 0.27 | 1.33 (0.92 to 1.90) | 0.13 | ||||
| Race | American Indian/Alaskan Native | Ref. | 0.80 | Ref. | 0.70 | Ref. | 0.67 | |||
| Asian/Asian American | 1.35 (0.23 to 7.95) | 0.74 | 1.14 (0.19 to 6.76) | 0.89 | 1.37 (0.22 to 8.49) | 0.74 | ||||
| Black | 1.25 (0.47 to 3.31) | 0.66 | 1.15 (0.43 to 3.11) | 0.78 | 1.25 (0.46 to 3.38) | 0.66 | ||||
| Multiple races | 0.92 (0.19 to 4.38) | 0.92 | 1.10 (0.18 to 6.73) | 0.92 | 1.13 (0.19 to 6.85) | 0.90 | ||||
| Native Hawaiian/Other Pacific Islander | 0.39 (0.04 to 4.25) | 0.44 | 0 (0 to Inf) | 0.99 | 0 (0 to Inf) | 0.99 | ||||
| Race not reported | 1.02 (0.36 to 2.92) | 0.96 | 1.10 (0.38 to 3.18) | 0.86 | 1.22 (0.42 to 3.57) | 0.72 | ||||
| White | 1.38 (0.52 to 3.80) | 0.52 | 1.21 (0.44 to 3.30) | 0.71 | 1.28 (0.47 to 3.50) | 0.63 | ||||
| BMI (kg/m2) | 0.998 (0.98 to 1.01) | 0.82 | 1 (0.98 to 1.02) | 0.96 | 0.998 (0.98 to 1.01) | 0.77 | ||||
| Education | Did not complete secondary school or high school | Ref. | 0.13 | Ref. | 0.16 | Ref. | 0.09 | |||
| High school degree | 0.99 (0.66 to 1.50) | 0.98 | 1.07 (0.69 to 1.66) | 0.78 | 1.02 (0.65 to 1.60) | 0.93 | ||||
| Associate's or technical degree | 0.89 (0.53 to 1.49) | 0.66 | 0.97 (0.56 to 1.68) | 0.91 | 0.99 (0.57 to 1.73) | 0.98 | ||||
| College degree, doctoral, or postgraduate education | 1.48 (0.94 to 2.33) | 0.09 | 1.60 (0.98 to 2.62) | 0.06 | 1.69 (1.03 to 2.79) | 0.04 | ||||
| Alcohol use | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.66 (0.64 to 4.35) | 0.30 | 1.39 (0.52 to 3.67) | 0.51 | 1.54 (0.55 to 4.27) | 0.41 | ||||
| Cannabis use status | Never | Ref. | 0.34 | Ref. | 0.37 | Ref. | 0.32 | |||
| Current | 1.28 (0.86 to 1.93) | 0.23 | 1.32 (0.85 to 2.06) | 0.22 | 1.40 (0.88 to 2.22) | 0.16 | ||||
| Former | 0.90 (0.61 to 1.34) | 0.61 | 0.91 (0.58 to 1.41) | 0.67 | 0.94 (0.60 to 1.48) | 0.80 | ||||
| Indicator for assignment to PCST arm | No | Ref. | Ref. | Ref. | ||||||
| Yes | 0.91 (0.67 to 1.24) | 0.56 | 0.99 (0.71 to 1.37) | 0.93 | 0.99 (0.71 to 1.39) | 0.96 | ||||
| Baseline medication use: benzodiazepines | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.81 (1.00 to 3.28) | 0.05 | 1.51 (0.77 to 2.98) | 0.23 | ||||||
| Baseline medication use: gabapentinoids | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.25 (0.90 to 1.73) | 0.19 | 1.2 (0.82 to 1.75) | 0.34 | ||||||
| Current opioid use | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.16 (0.81 to 1.67) | 0.42 | 1.02 (0.67 to 1.56) | 0.91 | ||||||
| History of diabetes mellitus | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.30 (0.94 to 1.80) | 0.11 | 1.46 (1.01 to 2.12) | 0.05 | ||||||
| History of stroke or TIA | No | Ref. | Ref. | Ref. | ||||||
| Yes | 1.01 (0.69 to 1.48) | 0.96 | 0.97 (0.64 to 1.47) | 0.89 | ||||||
BMI, body mass index; CI, confidence interval; Inf, infinity; PCST, pain coping skills treatment; TIA, transient ischemic attack.
Model 1 adjusts for age, sex, race, body mass index, education, alcohol, marijuana, and treatment assignment.
Model 2 adjusts for the covariates in model 1, benzodiazepines, gabapentinoids, opioid, diabetes, and stroke or transient ischemic attack.
Table 6.
Association of selected patient-reported outcomes with risk of falls in participants enrolled in Heart Outcomes Prevention Evaluation Consortium clinical trial
| Patient-Reported Outcome | Crude | Model 1a | Model 2b | |||
|---|---|---|---|---|---|---|
| Rate Ratio (95% CI) | P Value | Rate Ratio (95% CI) | P Value | Rate Ratio (95% CI) | P Value | |
| BPI interference | 1.11 (1.02 to 1.20) | 0.01 | 1.11 (1.02 to 1.21) | 0.02 | 1.10 (1.01 to 1.20) | 0.03 |
| BPI severity | 1.08 (1.002 to 1.17) | 0.04 | 1.10 (1.01 to 1.19) | 0.04 | 1.08 (0.99 to 1.18) | 0.07 |
| PROMIS sleep disturbance T score | 1.03 (1.01 to 1.04) | 0.002 | 1.03 (1.01 to 1.04) | 0.001 | 1.02 (1.01 to 1.04) | 0.01 |
| PROMIS fatigue T score | 1.03 (1.01 to 1.04) | 0.002 | 1.02 (1.01 to 1.04) | 0.01 | 1.02 (1.01 to 1.04) | 0.01 |
| PROMIS physical function T score | 0.97 (0.94 to 0.99) | 0.003 | 0.97 (0.95 to 0.99) | 0.01 | 0.98 (0.95 to 1.001) | 0.06 |
| BPI Interferencec | 1.03 (0.92 to 1.15) | 0.63 | 1.03 (0.92 to 1.16) | 0.62 | 1.03 (0.92 to 1.16) | 0.58 |
BPI, brief pain inventory; CI, confidence interval; PROMIS, patient-reported outcome measurement information system.
Model 1 adjusts for age, sex, race, body mass index, education, alcohol, marijuana, and treatment assignment.
Model 2 adjusts for the covariates in model 1, benzodiazepines, gabapentinoids, opioid, diabetes, and stroke/transient ischemic attack.
All models additionally adjust for brief pain inventory severity, patient-reported outcome measurement information system sleep disturbance, patient-reported outcome measurement information system fatigue, and patient-reported outcome measurement information system physical function.
Discussion
In this multicenter clinical trial of individuals with chronic pain and kidney failure requiring maintenance hemodialysis, falls were common, occurring in 28% of individuals during a planned follow-up of 36 weeks. Falls were only rarely attributed to hemodynamic instability or occurred in the dialysis unit, with the vast majority of those that could be characterized occurring at participants' homes and due to accidental causes. Although most falls were not severe enough to require dedicated medical evaluation, 14% resulted in hospitalization and 7% resulted in a fracture. Associations between the severity of any symptom such as chronic pain or sleep disturbances with the risk of subsequent falls were weak and were not robust to statistical adjustment, while medication use such as benzodiazepines or opioids use at baseline was not independently associations with subsequent falls.
Previous studies have consistently identified a high incidence of falls in the maintenance hemodialysis population. For example, a longitudinal study of 95 individuals receiving hemodialysis by McAdams-DeMarco et al. showed that 28.3% experienced a fall during a median follow-up of 6.7 months.15 Another 8-week study, including 308 individuals from 7 dialysis units in Belgium, recorded falls in 12.7% of participants, with an incidence rate of approximately 1.2 falls per patient-year. A larger study from the United States including 762 prevalent patients from 14 dialysis units in Atlanta and San Francisco reported that 28.4% had experienced a fall in the previous 12 months. Fifty-seven percent of those with at least one fall experienced additional falls during follow-up.16 A much larger study using administrative data from 81,653 older individuals starting hemodialysis between 2010 and 2012 found that 7.6% had a fall resulting in serious injury during the first year of dialysis.17 Finally, in a cross-sectional cohort of older, central European maintenance hemodialysis patients, 46% of patients had experienced a fall in the previous year—a rate 4.7-fold higher than in a concurrent control group without CKD.18
Our study is consistent with these findings in demonstrating a high incidence of falls in a prospectively studied cohort of individuals receiving maintenance hemodialysis. The 28% incidence we observed during 36 weeks of follow-up provides, to the best of our knowledge, the first estimate of the frequency of falls among hemodialysis patients who experience chronic pain. Although our observed incidence was modestly lower, it is qualitatively consistent with the high fall rate observed in these prior cohorts of patients treated with dialysis with minor differences likely attributable to the differences in inclusion criteria, such as the restriction to incident or elderly patients in several previous studies and the younger age and requirement for chronic pain in this study. The high rate of 28% over 36 weeks we observed in this population of patients undergoing maintenance hemodialysis is also notable for being similar to the estimated reported in a recent study of 2822 community living older adults in the United States (35% in the prior year), despite a mean age nearly 20 years lower in the current study population (60 versus 79 years).19
Our analysis extends on prior work in a number of ways. In contrast to most prior reports, we prospectively enrolled a large number of patients and recruited from multiple centers across a broad swath of the United States, including dialysis units in both urban and rural areas. In addition, rather than using administrative data or identifying falls retrospectively,16,17 we prospectively identified falls regardless of severity or outcome, and we had access to descriptive narratives including data on location, circumstances, and outcomes related to the fall. This rich data source allowed us to better define the circumstance of falls, revealing that most falls occur at home and only rarely occur at the dialysis unit or in association with dialysis. This finding provides robust confirmation of prior observations from two small studies (N=49)20,21 and of a larger, 8-week study.22 Furthermore, only 3.4% of falls were deemed related to dialysis (occurring after dialysis on a dialysis day), with even fewer (0.7%) attributed to hemodynamic instability after dialysis and only a small proportion (2.4%) related to other hemodynamic issues.
Several studies have identified age or frailty as important risk factors for falls in the hemodialysis population.15,16,22,23 Despite enrolling a relatively large number of patients, analyzing nearly 300 falls, and capturing detailed data on comorbidities and psychosocial factors, we were unable to confirm independent associations of age or sex with the risk of falls in this population. Although we did not directly measure frailty in this trial, measures of fatigue and sleep, items that may partly overlap with frailty, had weak associations with falls.
Interestingly, we also did not find consistent associations of pain with the likelihood of falls. Pain is likely, at least partly, to reflect the degree of comorbidity, arise from musculoskeletal injuries, and limit mobility. We therefore expected that measures of pain would be strong predictors of the risk of falls. To the best of our knowledge, we are the first to assess the relationship of pain with the risk of falls in patients treated with maintenance hemodialysis. Although BPI interference scores were marginally higher in patients with falls compared with those without falls, none of these factors—the degree of pain interference, the severity of pain, or the use of opioids—was robustly associated with the likelihood of falls. Similarly, we did not find any association with other symptoms experienced commonly by patients undergoing long-term dialysis with the risk of falls.
Our findings have several important implications. They suggest that at the current time, peridialytic orthostasis or weakness induced by hemodynamic and electrolyte shifts during dialysis are relatively infrequent causes of falls. This may reflect attention to falls as a preventable safety event in hemodialysis units and vigilance by staff regarding these potential causes of falls. Despite prior research suggesting that frailty and comorbidity are the primary drivers of falls, our negative findings suggest that more complicated processes may be at play and that deriving comorbidity-based predictive scores is unlikely to be straightforward. More detailed assessments of balance and mobility may be needed to accurately identify those at risk. Finally, the low proportion of falls occurring at dialysis in contrast to the high proportion of accidental falls and falls occurring at home suggests that the dialysis-specific interventions are likely to be low yield in contrast to home safety assessments or strength, balance, and mobility training as means of preventing falls in this population.
Limitations of this study should be acknowledged. This was a clinical trial population and enrollment required chronic pain according to study-specific definitions. The use of opiates and antidepressant medications or other characteristics such as the age of participants in this clinical trial population may also differ from the general US dialysis population, thereby further limiting generalizability. In addition, although we were able to assess association of falls with qualitative measures of pain, we could not assess whether the presence of pain compared with its absence affects fall risk, and our findings should be extrapolated cautiously to individuals without pain. In addition, despite prospective collection of narrative reports of all falls, data elements regarding the timing and nature of falls were not standardized and had to be extrapolated from those narratives. We did not assess frailty using a validated scale. Based on previous studies and our findings,15,24,25 it is likely that frailty is a stronger risk factor for falls than pain among patients with kidney diseases. Finally, because data on the dialysis prescription and mineral and bone parameters were not collected, we are unable to analyze associations of these parameters with the risk of falls.
In summary, falls are a common event among patients treated with maintenance in-center hemodialysis, with a minority of individuals who fall requiring subsequent medical evaluation and intervention. As has been recommended in the nephrology literature,26,27 intervening to prevent falls with a limited geriatric assessment could provide important opportunities to improve quality of life and outcomes in this population.28 The Centers for Disease Control and Prevention has recommended one such approach.29 Our results highlight the importance of factors outside the dialysis unit in determining fall risk and demonstrate that better treatment of pain is unlikely to lessen the burden of falls in the hemodialysis population. Finally, they highlight the need for additional studies to identify patients at risk and the optimal strategies for intervention.
Supplementary Material
Acknowledgments
The authors are grateful to the trial participants, personnel at the participating dialysis facilities, and the following dialysis provider organizations or nephrology practices: DaVita; Dialysis Clinic, Inc.; Fresenius Medical Care; Northwest Kidney Centers; Puget Sound Kidney Centers; The Rogosin Institute; University of Illinois Hospital Dialysis Unit, Chicago, Illinois; and Associates in Nephrology, S.C., Chicago, Illinois. The authors acknowledge the important contributions of the patient advisors, research coordinators, pain coping skills treatment coaches, computer-assisted telephone interviewing team, and members of the Data and Safety Monitoring Board. Project officers from the National Institute of Diabetes and Digestive and Kidney Diseases worked collaboratively with the investigators in designing the study, monitoring the study performance, interpreting data, and preparing the manuscript. The opinions expressed herein do not necessarily reflect those of the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institutes of Health, the Department of Health and Human Services, or the Government of the United States. The authors thank the participants, research staff, and dialysis personnel in the HOPE trial for their contributions to the success of the HOPE trial.
Footnotes
The HOPE Consortium members include: HOPE Consortium, Cores, and Data and Safety Monitoring Board Members: Hennepin Healthcare Clinical Center and Enrolling Site: Kirsten L. Johansen (PI), Gavin Bart, Erin E. Krebs, James B. Wetmore, Maria Pacheco-Hernandez, Ursula Munet, Rudy Qamhiyeh, Mike Wambua; Massachusetts General Hospital Clinical Center and Enrolling Site: Sahir Kalim (PI), Sagar U. Nigwekar, Daniel E. Weiner, Kome Ekor, Beza Mengesha, Shananssa Percy; New York University Langone Health Clinical Center and Enrolling Site: David M. Charytan (PI), Keith S. Goldfeld, Joshua D. Lee, Nobuyuki Miyawaki, Jennifer S. Scherer, Amanda J. Shallcross, Miri Cazes, Sobaata Chaudry, Paula Dutka, Yasmine Flores, Daniela Fraticelli Ortiz, Candace Grant, Colin Keane, Pragna Krishnamurthy, Ashley Macina, Angela McCarthy, Kathleen Rice, Grace Robinson, Dalila Varela, Javaughn Ways; University of Illinois Chicago Clinical Center and Enrolling Site: Michael J. Fischer (PI), Ardith Z. Doorenbos, Christopher Holden, James P. Lash, Mark B. Lockwood, Alana D. Steffen, Cheryl Gilmartin, Amanda Goldstein, Monya Meinel, Kimberly Silva, Guillermo Zamora; University of Pittsburgh Clinical Center and Enrolling Site: Manisha Jhamb (PI), Hailey W. Bulls, Megan E. Hamm, Sanjana Kamat, Jane M. Liebschutz, Jennifer L. Steel, Jonathan G. Yabes, Precious Lacey, Donna Olejniczak, Mary Schopp, Melissa Weimer, Vincent Wood; University of Pennsylvania Enrolling Site: Nwamaka D. Eneanya (PI), Sarah J. Schrauben (PI), Stephany Almonte-Then, Chigozie Amonu, Nicholas Bishop, Diane Park, Taylor Stallings; University of Washington Clinical Center and Enrolling Site: Rajnish Mehrotra (PI), Nisha Bansal, Elenore P. Bhatraju, Steven D. Weisbord, Lisa Anderson, Sydney Johnson, Kaeleb Laszlo, Lori Linke; University of New Mexico Enrolling Site: Mark L. Unruh (PI), Christos P. Argyropoulos, Shane Pankratz, Davin Quinn, George Garcia, Monica Bajana Meza, Monica Cardona, Tammy Seaman Weidner, Greg Trejo; Rogosin Institute Enrolling Site: Daniel Cukor (PI), Nathaniel Berman, Nelson Chen, Ines Chicos, Stephanie Donahue, Anna Gong; Vanderbilt University Medical Center C Clinical Center and Enrolling Site: Kerri L. Cavanaugh (PI), Carrie E. Brintz, David A. Edwards, T. Alp Ikizler, Puneet Mishra, Thomas Stewart, DeVitra Berry, Don Merrimon, Samuel Opeke, Sarah Pleasant, Hadassah Pegues, Christopher Roach, Sonya Williams; West Virginia University Enrolling Site: Bethany Pellegrino (PI), Daniel W. McNeil, Alvin H. Moss, Rebecca J. Schmidt, Cheryl Dalton, Kristy O'Connell, Maryanne Wilkinson; Yale School of Medicine/VA Connecticut Healthcare System Clinical Center and Enrolling Site: William C. Becker (PI), Justin M. Belcher, Susan T. Crowley, Denise Esserman, Caroline G. Falker, Alicia A. Heapy, Svetlana Vassilieva, Samara Zuniga; Durham VA Health Care System Enrolling Site: Patrick H. Pun (PI), Wissam M. Kourany, James Lefler, Teresa Purdy, Jenika Hammond, Khristian Harris, Sara Hoffman, Jeanette Rutledge; VA Portland Health Care System: Benjamin J. Morasco (PI), Christopher K. Blazes, Christopher S. Stauffer, Melissa Adams, Deza’Rae Collins, Richard Torres; Dallas VA Medical Center Enrolling Site: Jeffrey Penfield (PI), Monica Barbosa, Levi Beeks, Erik Guajardo, Sindi Sanchez; VA New York Harbor Health Care Enrolling Site: Mansi Mehta (PI), Adrian Cosmin, David S. Goldfarb, Sabrina Felson, Brian Sands, Frank Modersitzki; Scientific Data and Research Center: University of Pennsylvania: Laura M. Dember (PI), Martin D. Cheatle, John T. Farrar, Jesse Y. Hsu, Steve Joffe, Kyle M. Kampman, Francis J. Keefe, J Richard Landis, Ted Barrell, Leah Bernardo, Natalie Kuzla, Jonah Joffe, Joanna Walsh; Patient Advisors: Dawn P. Edwards, Robert E. Grindstaff*, Andre Hoover, Stephen Lerner*, Roger Mims, Nina Quintana, Steven Sousa, Darlene Villareal, David M. White, Caroline Wilkie, Joel Williams; Pain Coping Skills Training Coaches: Veronica Dyer, New York University Langone Health; Eshika Kalam, Rogosin Institute; Blanca Contreras, University of Illinois Chicago; Carlyn Clark, University of Washington; Heather Howell and Kevin Payne, West Haven VA Medical Center; Pain Coping Skills Training Fidelity Monitors: Andrew M. Busch and Sarah Cameron, Hennepin Healthcare; Computer-Assisted Telephone Interviewing (CATI) Centers: University of New Mexico - Adamaris Arteaga Leon, Elena Ashley, Gabriela Chacon Palma, Katy Chalamidas, Oluoma Edeh, Lindsay Gear, Cameron Guy, Grace Kimura, Alexander Leon Cupe, Henry Luo, Katherine McDaniels, Sofia McLaren, Valeria Mejia, Claire Mullins, Uchechukwu Okereke, Gabriel Rudow, Giselle Rodriguez Sosa, Lily Sullivan, John Torres, Hugo Vilchis; University of Pittsburgh - University of Pittsburgh: Scott Beach, Amber Barton, Victoria Casilli, Lynda Connelly, Jane Dirks, Luke Farkas, Ana Geibel, Olivia Kirsch, Paula Kubrick, Patricia Lietz, Brandon Self, Susan String Fellow, Olivia Wilson; National Institute of Diabetes and Digestive and Kidney Diseases: Kevin C. Abbott, Paul L. Kimmel, Jenna M. Norton, Tracy L. Rankin; Data and Safety Monitoring Board: Bruce Barton (Chair), Jane Atkinson, Melissa Bensouda, Roger B. Fillingim, Michael Freeman, Andrew Garland, Jennifer J. Gassman, Allen R. Nissenson, Maile Robb*, Friedhelm Sandbrink, Brigitte Schiller, Eric Storch, David Thomas, Roger Weiss. *Deceased.
Contributor Information
Collaborators: Kirsten L. Johansen, Gavin Bart, Erin E. Krebs, James B. Wetmore, Maria Pacheco-Hernandez, Ursula Munet, Rudy Qamhiyeh, Mike Wambua, Sahir Kalim, Sagar U. Nigwekar, Daniel E. Weiner, Kome Ekor, Beza Mengesha, Shananssa Percy, David M. Charytan, Keith S. Goldfeld, Joshua D. Lee, Nobuyuki Miyawaki, Jennifer S. Scherer, Amanda J. Shallcross, Miri Cazes, Sobaata Chaudry, Paula Dutka, Yasmine Flores, Daniela Fraticelli Ortiz, Candace Grant, Colin Keane, Pragna Krishnamurthy, Ashley Macina, Angela McCarthy, Kathleen Rice, Grace Robinson, Dalila Varela, Javaughn Ways, Michael J. Fischer, Ardith Z. Doorenbos, Christopher Holden, James P. Lash, Mark B. Lockwood, Alana D. Steffen, Cheryl Gilmartin, Amanda Goldstein, Monya Meinel, Kimberly Silva, Guillermo Zamora, Manisha Jhamb, Hailey W. Bulls, Megan E. Hamm, Sanjana Kamat, Jane M. Liebschutz, Jennifer L. Steel, Jonathan G. Yabes, Precious Lacey, Donna Olejniczak, Mary Schopp, Melissa Weimer, Vincent Wood, Nwamaka D. Eneanya, Sarah J. Schrauben, Stephany Almonte-Then, Chigozie Amonu, Nicholas Bishop, Diane Park, Taylor Stallings, Rajnish Mehrotra, Nisha Bansal, Elenore P. Bhatraju, Steven D. Weisbord, Lisa Anderson, Sydney Johnson, Kaeleb Laszlo, Lori Linke, Mark L. Unruh, Christos P. Argyropoulos, Shane Pankratz, Davin Quinn, George Garcia, Monica Bajana Meza, Monica Cardona, Tammy Seaman Weidner, Greg Trejo, Daniel Cukor, Nathaniel Berman, Nelson Chen, Ines Chicos, Stephanie Donahue, Anna Gong, Kerri L. Cavanaugh, Carrie E. Brintz, David A. Edwards, T. Alp Ikizler, Puneet Mishra, Thomas Stewart, DeVitra Berry, Don Merrimon, Samuel Opeke, Sarah Pleasant, Hadassah Pegues, Christopher Roach, Sonya Williams, Bethany Pellegrino, Daniel W. McNeil, Alvin H. Moss, Rebecca J. Schmidt, Cheryl Dalton, Kristy O'Connell, Maryanne Wilkinson, William C. Becker, Justin M. Belcher, Susan T. Crowley, Denise Esserman, Caroline G. Falker, Alicia A. Heapy, Svetlana Vassilieva, Samara Zuniga, Patrick H. Pun, Wissam M. Kourany, James Lefler, Teresa Purdy, Jenika Hammond, Khristian Harris, Sara Hoffman, Jeanette Rutledge, Benjamin J. Morasco, Christopher K. Blazes, Christopher S. Stauffer, Melissa Adams, Deza’Rae Collins, Richard Torres, Jeffrey Penfield, Monica Barbosa, Levi Beeks, Erik Guajardo, Sindi Sanchez, Mansi Mehta, Adrian Cosmin, David S. Goldfarb, Sabrina Felson, Brian Sands, Frank Modersitzki, Laura M. Dember, Martin D. Cheatle, John T. Farrar, Jesse Y. Hsu, Steve Joffe, Kyle M. Kampman, Francis J. Keefe, J Richard Landis, Ted Barrell, Leah Bernardo, Natalie Kuzla, Jonah Joffe, Joanna Walsh, Dawn P. Edwards, Robert E. Grindstaff, Andre Hoover, Stephen Lerner, Roger Mims, Nina Quintana, Steven Sousa, Darlene Villareal, David M. White, Caroline Wilkie, Joel Williams, Veronica Dyer, New York University Langone Health, Eshika Kalam, Rogosin Institute, Blanca Contreras, University of Illinois Chicago, Carlyn Clark, University of Washington, Heather Howell, Kevin Payne, Andrew M. Busch, Sarah Cameron, Hennepin Healthcare, Adamaris Arteaga Leon, Elena Ashley, Gabriela Chacon Palma, Katy Chalamidas, Oluoma Edeh, Lindsay Gear, Cameron Guy, Grace Kimura, Alexander Leon Cupe, Henry Luo, Katherine McDaniels, Sofia McLaren, Valeria Mejia, Claire Mullins, Uchechukwu Okereke, Gabriel Rudow, Giselle Rodriguez Sosa, Lily Sullivan, John Torres, Hugo Vilchis, Scott Beach, Amber Barton, Victoria Casilli, Lynda Connelly, Jane Dirks, Luke Farkas, Ana Geibel, Olivia Kirsch, Paula Kubrick, Patricia Lietz, Brandon Self, Susan String Fellow, Olivia Wilson, Kevin C. Abbott, Paul L. Kimmel, Jenna M. Norton, Tracy L. Rankin, Bruce Barton, Jane Atkinson, Melissa Bensouda, Roger B. Fillingim, Michael Freeman, Andrew Garland, Jennifer J. Gassman, Allen R. Nissenson, Maile Robb, Friedhelm Sandbrink, Brigitte Schiller, Eric Storch, David Thomas, and Roger Weiss
Disclosures
Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/CJN/C343.
Author Contributions
Conceptualization: David M. Charytan, Paul L. Kimmel, Rajnish Mehrotra.
Data curation: David M. Charytan, Rajnish Mehrotra, Wenbo Wu.
Formal analysis: Jesse Y. Hsu, Wenbo Wu.
Funding acquisition: David M. Charytan, Laura M. Dember, Sahir Kalim, Rajnish Mehrotra, Beth Pellegrino.
Investigation: David M. Charytan, Rajnish Mehrotra.
Methodology: David M. Charytan, Laura M. Dember, Sahir Kalim, Rajnish Mehrotra, Alvin H. Moss.
Project administration: David M. Charytan, Laura M. Dember, Sahir Kalim, Paul L. Kimmel, Natalie Kuzla, Rajnish Mehrotra, Alvin H. Moss, Beth Pellegrino, Sarah Schrauben.
Resources: Laura M. Dember, Jesse Y. Hsu, Beth Pellegrino.
Software: Jesse Y. Hsu, Wenbo Wu.
Supervision: David M. Charytan, Laura M. Dember, Paul L. Kimmel, Rajnish Mehrotra.
Writing – original draft: David M. Charytan, Rajnish Mehrotra, Alvin H. Moss, Manar Shalak.
Writing – review & editing: David M. Charytan, Laura M. Dember, Denise Esserman, Jesse Y. Hsu, Sahir Kalim, Paul L. Kimmel, Natalie Kuzla, Mark B. Lockwood, Rajnish Mehrotra, Nobuyuki Miyawaki, Alvin H. Moss, Beth Pellegrino, Patrick H. Pun, Rudy Qamhiyeh, Jennifer Scherer, Sarah Schrauben, Manar Shalak, Daniel E. Weiner.
Funding
R. Mehrotra: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123786). M.B. Lockwood: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123787). L.M. Dember, J.Y. Hsu, and N. Kuzla: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123813). D.M. Charytan, W. Wu, N. Miyawaki, and J. Scherer: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123814). R. Qamhiyeh: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123816). D. Esserman and P.H. Pun: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123817). S. Kalim and D.E. Weiner: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123818). S. Schrauben: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123812). A.H. Moss, M. Shalak, and B. Pellegrino: National Institute of Diabetes and Digestive and Kidney Diseases (U01DK123821).
Data Availability Statements
Original data created for the study are or will be available in a persistent repository upon publication. Clinical Trial Data. NIDDK Repository. This is pending at this time.
Supplemental Material
This article contains the following supplemental material online at http://links.lww.com/CJN/C344.
Supplemental Table 1. Correlation matrix for selected patient-reported outcomes.
Supplemental Table 2. Crude and adjusted associations of covariates with injurious fall event rate among the randomized participants in the HOPE trial.
Supplemental Table 3. Fully adjusted model for associations of BPI severity, BPI interference, and PROMIS instruments with falls.
References
- 1.Bergen G, Stevens MR, Burns ER. Falls and fall injuries among adults aged ≥65 years - United States, 2014. MMWR Morb Mortal Wkly Rep. 2016;65(37):993–998. doi: 10.15585/mmwr.mm6537a2 [DOI] [PubMed] [Google Scholar]
- 2.Papakonstantinopoulou K, Sofianos I. Risk of falls in chronic kidney disease. J Frailty Sarcopenia Falls. 2017;2(2):33–38. doi: 10.22540/jfsf-02-033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.López-Soto PJ De Giorgi A Senno E, et al. Renal disease and accidental falls: a review of published evidence. BMC Nephrol. 2015;16:176. doi: 10.1186/s12882-015-0173-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Perez-Gurbindo I, María Álvarez-Méndez A, Pérez-García R, Cobo PA, Carrere MTA. Factors associated with falls in hemodialysis patients: a case-control study. Rev Lat Am Enfermagem. 2021;29:e3505. doi: 10.1590/1518-8345.5300.3505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Shirai N, Yamamoto S, Osawa Y, Tsubaki A, Morishita S, Narita I. Dynamic and static balance functions in hemodialysis patients and non-dialysis dependent CKD patients. Ther Apher Dial. 2023;27(3):412–418. doi: 10.1111/1744-9987.13931 [DOI] [PubMed] [Google Scholar]
- 6.Welsh VK, Clarson LE, Mallen CD, McBeth J. Multisite pain and self-reported falls in older people: systematic review and meta-analysis. Arthritis Res Ther. 2019;21(1):67. doi: 10.1186/s13075-019-1847-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cai Y, Leveille SG, Shi L, Chen P, You T. Chronic pain and risk of injurious falls in community-dwelling older adults. J Gerontol A Biol Sci Med Sci. 2021;76(9):e179–e186. doi: 10.1093/gerona/glaa249 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ogliari G Ryg J Andersen-Ranberg K, et al. Association of pain and risk of falls in community-dwelling adults: a prospective study in the Survey of Health, Ageing and Retirement in Europe (SHARE). Eur Geriatr Med. 2022;13(6):1441–1454. doi: 10.1007/s41999-022-00699-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Davison SN, Rathwell S, Ghosh S, George C, Pfister T, Dennett L. The prevalence and severity of chronic pain in patients with chronic kidney disease: a systematic review and meta-analysis. Can J Kidney Health Dis. 2021;8:2054358121993995. doi: 10.1177/2054358121993995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Weisbord SD Fried LF Arnold RM, et al. Prevalence, severity, and importance of physical and emotional symptoms in chronic hemodialysis patients. J Am Soc Nephrol. 2005;16(8):2487–2494. doi: 10.1681/ASN.2005020157 [DOI] [PubMed] [Google Scholar]
- 11.Dember LM Hsu JY Bernardo L, et al.; HOPE Consortium. The design and baseline characteristics for the HOPE consortium trial to reduce pain and opioid use in hemodialysis. Contemp Clin Trials. 2024;136:107409. doi: 10.1016/j.cct.2023.107409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Dember LM Hsu JY Mehrotra R, et al.; HOPE Consortium. Pain coping skills training for patients receiving hemodialysis: the HOPE consortium randomized clinical trial. JAMA Intern Med. 2025;185(2):197–207. doi: 10.1001/jamainternmed.2024.7140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Krebs EE Lorenz KA Bair MJ, et al. Development and initial validation of the PEG, a three-item scale assessing pain intensity and interference. J Gen Intern Med. 2009;24(6):733–738. doi: 10.1007/s11606-009-0981-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cleeland CS, Ryan KM. Pain assessment: global use of the brief pain inventory. Ann Acad Med Singap. 1994;23(2):129–138. PMID: 8080219 [PubMed] [Google Scholar]
- 15.McAdams-DeMarco MA Suresh S Law A, et al. Frailty and falls among adult patients undergoing chronic hemodialysis: a prospective cohort study. BMC Nephrol. 2013;14:224. doi: 10.1186/1471-2369-14-224 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kutner NG, Zhang R, Huang Y, Wasse H. Falls among hemodialysis patients: potential opportunities for prevention? Clin Kidney J. 2014;7(3):257–263. doi: 10.1093/ckj/sfu034 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bowling CB, Hall RK, Khakharia A, Franch HA, Plantinga LC. Serious fall injury history and adverse health outcomes after initiating hemodialysis among older U.S. adults. J Gerontol A Biol Sci Med Sci. 2018;73(9):1216–1221. doi: 10.1093/gerona/glx260 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Račić M Petković N Bogićević K, et al. Comprehensive geriatric assessment: comparison of elderly hemodialysis patients and primary care patients. Ren Fail. 2015;37(7):1126–1131. doi: 10.3109/0886022X.2015.1057459 [DOI] [PubMed] [Google Scholar]
- 19.Thomas J, Almidani L, Ramulu P, Varadaraj V. Falls and multiple falls among United States older adults with vision impairment. Am J Ophthalmol. 2025;271:166–174. doi: 10.1016/j.ajo.2024.11.012 [DOI] [PubMed] [Google Scholar]
- 20.Shirai N, Yamamoto S, Osawa Y, Tsubaki A, Morishita S, Narita I. Dysfunction in dynamic, but not static balance is associated with risk of accidental falls in hemodialysis patients: a prospective cohort study. BMC Nephrol. 2022;23(1):237. doi: 10.1186/s12882-022-02877-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Abdel-Rahman EM, Yan G, Turgut F, Balogun RA. Long-term morbidity and mortality related to falls in hemodialysis patients: role of age and gender - a pilot study. Nephron Clin Pract. 2011;118(3):c278–c284. doi: 10.1159/000322275 [DOI] [PubMed] [Google Scholar]
- 22.Desmet C, Beguin C, Swine C, Jadoul M.; Université Catholique de Louvain Collaborative Group. Falls in hemodialysis patients: prospective study of incidence, risk factors, and complications. Am J Kidney Dis. 2005;45(1):148–153. doi: 10.1053/j.ajkd.2004.09.027 [DOI] [PubMed] [Google Scholar]
- 23.Cook WL Tomlinson G Donaldson M, et al. Falls and fall-related injuries in older dialysis patients. Clin J Am Soc Nephrol. 2006;1(6):1197–1204. doi: 10.2215/CJN.01650506 [DOI] [PubMed] [Google Scholar]
- 24.Delgado C Shieh S Grimes B, et al. Association of self-reported frailty with falls and fractures among patients new to dialysis. Am J Nephrol. 2015;42(2):134–140. doi: 10.1159/000439000 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.van Loon IN, Joosten H, Iyasere O, Johansson L, Hamaker ME, Brown EA. The prevalence and impact of falls in elderly dialysis patients: frail elderly patient outcomes on dialysis (FEPOD) study. Arch Gerontol Geriatr. 2019;83:285–291. doi: 10.1016/j.archger.2019.05.015 [DOI] [PubMed] [Google Scholar]
- 26.Morley JE. Chapter 33: Falls in Elderly Patients with Kidney Disease, 2009. Accessed August 29, 2024. https://www.asn-online.org/education/distancelearning/curricula/geriatrics/ [Google Scholar]
- 27.Farrington K Covic A Nistor I, et al. Clinical practice guideline on management of older patients with chronic kidney disease stage 3b or higher (eGFR <45 mL/min/1.73 m2): a summary document from the European Renal Best Practice Group. Nephrol Dial Transplant. 2017;32(1):9–16. doi: 10.1093/ndt/gfw411 [DOI] [PubMed] [Google Scholar]
- 28.Brown EA, Farrington K. Geriatric assessment in advanced kidney disease. Clin J Am Soc Nephrol. 2019;14(7):1091–1093. doi: 10.2215/CJN.14771218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Algorithm for Fall Risk Screening, Assessment, and Intervention. Center for Diseases Control and Prevention, 2019. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Original data created for the study are or will be available in a persistent repository upon publication. Clinical Trial Data. NIDDK Repository. This is pending at this time.