Abstract
Rationale & Objective:
Adults receiving maintenance hemodialysis (HD) frequently report pain, yet detailed descriptions of pain in this population are lacking. This study examines pain locations, characteristics, and associations with other symptoms in adults receiving HD.
Study Design:
Cross-sectional analysis.
Setting & Participants:
Adults with moderate to severe chronic pain receiving maintenance HD enrolled in the multicenter HOPE Consortium Trial from 2021 to 2023.
Exposures:
Sociodemographic, pain treatment, dialysis, medical comorbidity, and psychological and behavioral characteristics. Other patient-reported symptoms.
Outcome:
Pain interference and severity as assessed by the Brief Pain Inventory (BPI) Interference and Severity subscales (range 0–10).
Analytical Approach:
Multivariable regression with LASSO to examine associations between participant characteristics and pain interference/severity, and Spearman’s correlation to examine relationships between other symptoms and pain interference/severity at baseline.
Results:
Among 643 participants, the median (IQR) BPI interference was 6.6 (5.1–7.9) and severity was 6.0 (4.5–7.5). 84% of participants reported pain >1 year and 75% had daily pain. 89% and 66% of participants endorsed musculoskeletal and neuropathic pain, respectively. Of 32 body regions, the median (IQR) number of painful regions was 8 (4–14). Common regions in females were lower back (72%), knees (64%), legs (60%), and upper back (59%). A similar pattern existed for males. In LASSO analyses, cardiovascular disease and depression were associated with significantly higher pain interference whereas White race (ref: Black race) and non-Hispanic ethnicity were associated with significantly lower pain interference. Similar findings were noted for pain severity. Pain catastrophizing and symptoms of fatigue, depression, and anxiety were moderately correlated with pain interference (r>0.4).
Limitations:
Neither relationship directionality nor causality can be inferred.
Conclusions:
Among adults treated with HD who have chronic pain, pain locations were numerous and diverse, with substantial musculoskeletal and neuropathic characteristics. Factors associated with pain interference were predominantly sociodemographic and psychological rather than those related to comorbid diseases and dialysis.
Index Words: Pain, Dialysis, Symptoms, Depression, Trial
Plain-Language Summary
Adults receiving maintenance hemodialysis (HD) frequently experience chronic pain, but it remains poorly understood. We examined pain locations, characteristics, and relationships with other symptoms among 643 adults with moderate to severe chronic pain receiving maintenance HD enrolled in the multicenter HOPE Consortium Trial from 2021 to 2023. We found that pain locations were numerous and diverse, with substantial musculoskeletal and neuropathic pain characteristics. Factors associated with pain interference were predominantly sociodemographic and psychological rather than those related to comorbid diseases and dialysis. Given these findings, routine pain assessments and treatment plans tailored to the specific needs and symptoms of the individual patient seem warranted for this patient population.
Introduction
More than 50% of adults with chronic kidney failure undergoing treatment with maintenance hemodialysis (HD) experience pain, and many describe it as moderate or severe (1–6). Pain is also associated with a wide array of negative health consequences. Adults receiving maintenance HD reporting pain are more likely than those without pain to use emergency services, incur hospitalizations, and have worse survival (7–8). One potential explanation for these observations is that pain results in a greater likelihood of skipping or shortening HD treatments or not adhering to other aspects of dialysis care (9–10). Pain also contributes to additional disabling symptoms (e.g., insomnia, anxiety, and depression), and is a major factor diminishing quality of life and satisfaction (2,4,5,7,10,11). Unfortunately, pain is often underrecognized and undertreated among HD patients (1,12–14). Moreover, pain management in this population is challenging for many reasons, including lack of provider training in pain assessment, insufficient care coordination, and limited evidence-based safe effective pain treatment options. (1,10,15,16).
Despite its high prevalence and negative consequences, pain has been infrequently and insufficiently characterized among adults receiving maintenance HD (1,4,8,13,17,18). As noted in a systematic review, prior observations of pain in adults receiving maintenance HD are generally drawn from small homogenous cohorts from single centers, failed to use standardized, validated pain instruments, and did not define pain duration (e.g., distinguishing acute vs. chronic pain) (4,19). Additionally, most studies describing pain in adults receiving maintenance HD reported exclusively on pain severity without delineating other important pain characteristics, such as its impact on daily function (i.e., pain interference). Few examined the relationship of pain with other common symptoms reported by maintenance HD adults. Hence, generalizable inferences that could inform the management of chronic pain for the more than 450,000 Americans undergoing treatment with maintenance HD, the majority of whom have chronic pain, are currently lacking (20).
The HOPE Consortium Trial to Reduce Pain and Opioid Use in Hemodialysis (HOPE Trial) was a multicenter randomized trial to evaluate the effectiveness of a cognitive behavioral therapy-based intervention called Pain Coping Skills Training (PCST) for reducing pain interference and improving other pain-related outcomes among adults with ESKD and chronic pain receiving maintenance HD. (21). The HOPE Trial was designed to capture detailed information regarding pain interference, severity, and characteristics as well as other important patient symptoms to inform the inferences drawn about the study interventions. The objectives of this report were to examine pain locations, characteristics, and associations with other symptoms at baseline among HOPE Trial participants who represent a large, diverse, and multicenter cohort of adults with chronic pain receiving maintenance HD.
Methods
Study Sample and Design
Details of the design and methods and results of the HOPE Trial were previously published (21,22). For the purposes of this secondary data analysis, only baseline data were used. In brief, between January 22, 2021 and April 7, 2023, 643 adults with ESKD and chronic pain receiving maintenance HD from 103 dialysis facilities were randomized to either Pain Coping Skills Training, an intervention designed to increase self-efficacy for pain management, or usual care. The 103 outpatient dialysis facilities were affiliated with, or in geographic proximity to, 16 enrolling sites across the United States (Item S1). The major inclusion criteria were age ≥18 years, treatment with in-center maintenance hemodialysis for ≥90 days, English or Spanish fluency, self-report of moderate or severe chronic pain defined as 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, Activity, and Enjoyment of Life Scale (PEG) (23). The major exclusion criteria were current substance use disorder, suicidal intent, significant cognitive impairment, anticipated change in kidney replacement modality within 6 months, and life expectancy <6 months. The Institutional Review Board (IRB) at the University of Pennsylvania served as the IRB of record for the 11 non-Veterans Affairs (VA) sites (Protocol 843471), and the VA Central IRB served as the IRB of record for the 5 VA sites (CIRB Project 20–27). All participants provided written informed consent.
The trial was funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) through the National Institutes of Health Helping to End Addiction Long-term (HEAL) Initiative26. The trial was conducted under an investigational new drug (IND) application (15134) and is registered at clinicaltrials.gov (NCT04571619). An independent Data and Safety Monitoring Board (DSMB), appointed by the NIDDK, approved the protocol and reviewed safety, progress, and data quality. A patient advisory panel of individuals with end-stage kidney disease provided input throughout the planning and conduct of the trial. The trial was designed by the investigators with input from the NIDDK project scientist. Trial data were collected by the investigators and research coordinators at the enrolling centers and analyzed by the data coordinating center.
Variables and Data Sources
Baseline participant characteristics included those pertaining to demographic, pain treatment, kidney disease and dialysis, medical comorbidities, and psychological and behavioral features. These data, collected from participants at baseline (enrollment) using standardized case report forms by trained study personnel, included dialysis treatment variables (e.g., Kt/V, blood pressure) from the dialysis electronic medical records. Pain type was designated by study personnel according to descriptions on the case report form [e.g., musculoskeletal (bone, joint, muscle pain), neuropathic (e.g., tingling, numbness, shooting, nerve-like pain)], which was developed by the HOPE consortium with input from pain experts. Similarly, additional information regarding pain characteristics including site of pain and intensity by site were collected utilizing the CHOIR self-report body map, which reports pain separately by male and female sex (24). The Brief Pain Inventory (BPI) was used to assess pain interference and severity (25). The BPI is composed of the Interference subscale, which consists of seven pain interference items, and the Severity subscale, which consists of four pain intensity items. Both subscales are reported on a 0–10 scale, with a higher score indicating worse symptoms. Pain interference is the degree to which pain interferes with daily function and quality of life. The BPI has been validated in diverse populations (26). Baseline values for trial outcomes incorporated into these analyses included patient-reported indicators of pain and opioid use and several patient-reported symptoms important for adults receiving maintenance HD. As previously described (21) and summarized in Table S1 (27–39), all of these outcomes were ascertained in English or Spanish using computer-assisted telephone interviewing (CATI) administered by a centralized team, whose members were unaware of the participant’s treatment assignment.
Statistical Analysis
BPI interference and severity were calculated as median and interquartile range (IQR) overall and per demographic, pain, kidney disease and dialysis, medical comorbidity, and psychological and behavioral characteristics of HOPE Trial participants at baseline. An illustrated pain map was constructed based on the proportion of participants indicating pain in a body part. Also, mean (SD) pain intensity (1–10 scale) was calculated per each body part among those participants reporting pain in that body part. To examine relationships between pain (i.e., BPI interference and BPI severity) and participant characteristics, the least absolute shrinkage and selection operator (LASSO) was used to create reduced multivariable models (Item S2) (40,41). Fully adjusted multivariable linear regression models also were used to evaluate the cross-sectional associations between participant characteristics and pain. Unadjusted and adjusted (42) Spearman’s correlations were used to examine the associations between patient-reported symptoms and the outcomes of BPI interference and severity. The proportion of missing data ranged from 0% to 4.98% except for Kt/V (10.26%) (Item S3). Multiple imputation was used to address missing observations and create 10 imputed datasets (43). Correlations and regression results were pooled across imputed datasets using Rubin’s rules (44). Data analyses were performed using R (currently version 4.3.2; https://cran.r-project.org/). For all analyses, the overall level of significance was set to α = 0.05. Confidence intervals rather than p-values were reported since we did not adjust for multiple comparisons as no specific a priori hypotheses informed the analyses.
Results
Participant Characteristics at Baseline
Among the final analytic cohort of 643 participants, the overall median (IQR) brief pain inventory (BPI) interference score was 6.6 (5.1–7.9) and severity score was 6.0 (4.5–7.5) (Table 1, Figure S1). In terms of sociodemographic characteristics, participants who were Black, Hispanic, or female had higher BPI interference and severity scores compared to their White, Non-Hispanic, and male counterparts, respectively. Similarly, participants who spoke primarily Spanish, were unemployed, and had less than a high school education had higher BPI interference and severity scores compared to others in their respective categories. In the context of kidney disease and dialysis characteristics, participants with a prior kidney transplant had lower pain interference and severity scores compared to participants without prior kidney transplant. In contrast, participants with a Kt/V < 1.2 and hemoglobin < 10 g/dL had the highest BPI pain interference and severity scores. Among medical comorbidities, participants with diabetes and chronic lung disease had higher BPI interference and severity scores while less consistent small differences in were observed across other conditions. Among psychological and behavioral characteristics, participants with depression, anxiety, and bipolar disorder or schizophrenia had higher levels of BPI interference and severity than participants without.
Table 1:
Pain Interference and Severity by Participant Characteristics at Baseline
| Characteristic* | N (%)** | BPI Interference Median (IQR) | BPI Severity Median (IQR) |
|---|---|---|---|
| Overall | 643 | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Demographic | |||
| Age | |||
| [18,55) | 206 (32%) | 6.5 (5.3–7.7) | 6.0 (4.2–7.0) |
| [55,70) | 300 (47%) | 6.6 (5.0–7.9) | 6.2 (4.8–7.8) |
| [70,Inf) | 137 (21%) | 6.6 (5.0–8.0) | 5.8 (4.2–7.5) |
| Race | |||
| Black | 308 (48%) | 6.9 (5.3–8.0) | 6.5 (4.8–7.8) |
| White | 210 (33%) | 6.3 (4.7–7.4) | 5.2 (4.0–7.0) |
| Other races | 125 (19%) | 6.7 (5.4–8.0) | 6.2 (4.8–7.5) |
| Sex at Birth | |||
| Male | 350 (54%) | 6.3 (5.0–7.7) | 5.8 (4.2–7.2) |
| Female | 293 (46%) | 6.9 (5.4–8.1) | 6.2 (4.8–7.8) |
| Hispanic Ethnicity | |||
| Hispanic or Latino | 119 (19%) | 6.9 (5.6–8.4) | 6.5 (5.0–7.8) |
| Not Hispanic or Latino | 517 (80%) | 6.4 (5.0–7.7) | 6.0 (4.2–7.5) |
| Unknown | 1 (0.16%) | 5.9 (5.9–5.9) | 4.8 (4.8–4.8) |
| Not reported | 6 (1%) | 7.3 (6.3–8.0) | 5.4 (4.3–6.6) |
| Preferred Language | |||
| English | 584 (91%) | 6.4 (5.0–7.9) | 6.0 (4.5–7.5) |
| Spanish | 59 (9%) | 7.3 (6.1–8.6) | 6.5 (5.0–7.8) |
| Education | |||
| Did not complete secondary school or less than high school | 47 (7%) | 7.4 (6.1–8.9) | 6.8 (5.5–8.2) |
| Some secondary school or high school education | 91 (14%) | 7.0 (5.3–8.4) | 6.8 (5.0–7.8) |
| High school or secondary school degree complete | 270 (42%) | 6.6 (5.3–7.7) | 6.0 (4.5–7.5) |
| Associate’s or technical degree complete | 105 (16%) | 6.6 (5.0–8.0) | 5.5 (4.0–7.2) |
| College or baccalaureate degree complete | 106 (16%) | 6.2 (4.6–7.7) | 5.8 (4.2–7.0) |
| Doctoral or postgraduate education | 24 (4%) | 6.4 (5.4–7.0) | 5.4 (4.4–8.2) |
| Employment | |||
| Full-time | 28 (4%) | 6.2 (4.1–7.0) | 5.0 (3.8–6.6) |
| Part-time | 25 (4%) | 5.7 (4.0–7.0) | 5.5 (4.2–7.0) |
| Not employed | 590 (92%) | 6.6 (5.2–8.0) | 6.0 (4.5–7.5) |
| Relationship Status | |||
| Married/domestic partner | 200 (31%) | 6.4 (4.7–7.6) | 6.0 (4.5–7.5) |
| Widowed/divorced/separated | 236 (37%) | 6.6 (5.4–8.1) | 6.0 (4.8–7.8) |
| Never married | 207 (32%) | 6.7 (5.2–7.8) | 6.2 (4.2–7.5) |
| VA Site Participant | |||
| No | 574 (89%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 69 (11%) | 6.6 (5.4–7.9) | 6.2 (4.8–7.5) |
| Participant Residence | |||
| Northeast | 273 (43%) | 6.4 (5.1–7.7) | 6.0 (4.5–7.5) |
| Midwest | 153 (24%) | 7.1 (5.8–8.4) | 6.2 (4.5–7.5) |
| South | 116 (18%) | 6.3 (4.2–7.7) | 6.2 (4.2–7.5) |
| West | 87 (14%) | 6.4 (5.5–7.6) | 5.5 (4.2–7.1) |
| Pain Treatment | |||
| Any Non-pharmacologic Treatment to Manage Chronic Pain | |||
| No | 78 (12%) | 6.9 (5.0–8.3) | 6.9 (4.6–7.8) |
| Yes | 560 (88%) | 6.4 (5.1–7.9) | 6.0 (4.5–7.5) |
| Any Opioid Medication to Treat Pain | |||
| No | 505 (79%) | 6.4 (5.0–7.9) | 6.0 (4.2–7.5) |
| Yes | 138 (21%) | 6.9 (5.4–8.3) | 6.0 (4.8–7.5) |
| Any Non-Opioid Medication to Treat Pain | |||
| No | 200 (31%) | 6.4 (4.4–7.7) | 5.8 (4.0–7.5) |
| Yes | 443 (69%) | 6.6 (5.4–8.0) | 6.2 (4.5–7.5) |
| Kidney Disease and Dialysis | |||
| Years on Dialysis | |||
| <1 year | 144 (22%) | 6.6 (5.0–8.0) | 5.9 (4.2–7.5) |
| 1–5 years | 296 (46%) | 6.4 (5.1–7.7) | 6.0 (4.2–7.5) |
| >5 years | 203 (32%) | 6.7 (5.4–8.0) | 6.0 (4.8–7.5) |
| Prior Kidney Transplant | |||
| No | 606 (94%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 37 (6%) | 5.9 (5.1–7.3) | 5.5 (4.5–6.5) |
| Predialysis Blood Pressure | |||
| SBP >= 140 mmHg or DBP >= 90 mmHg | 373 (58%) | 6.6 (5.1–8.0) | 6.0 (4.5–7.5) |
| SBP < 140 mmHg and DBP < 90 mmHg | 270 (42%) | 6.6 (5.1–7.7) | 6.0 (4.2–7.2) |
| Kt/V | |||
| <1.2 | 50 (9%) | 7.0 (5.5–8.6) | 6.9 (4.6–7.8) |
| >=1.2 | 527 (91%) | 6.6 (5.1–7.8) | 5.8 (4.2–7.2) |
| Albumin | |||
| >=4 g/dL | 314 (50%) | 6.4 (5.1–7.9) | 6.0 (4.2–7.5) |
| <4 g/dL | 314 (50%) | 6.6 (5.0–8.0) | 6.1 (4.6–7.5) |
| Hemoglobin | |||
| >=10 g/dL | 490 (77%) | 6.4 (5.1–7.9) | 6.0 (4.5–7.5) |
| <10 g/dL | 146 (23%) | 6.9 (5.4–8.1) | 6.5 (4.8–7.7) |
| Medical Comorbidities | |||
| Body Mass Index (BMI) | |||
| <25 | 170 (27%) | 6.6 (5.4–8.0) | 6.0 (4.5–7.5) |
| 25–30 | 169 (26%) | 6.6 (4.9–7.9) | 6.2 (4.8–7.8) |
| >30 | 300 (47%) | 6.6 (5.0–7.7) | 6.0 (4.2–7.2) |
| Coronary Artery Disease (CAD) | |||
| No | 455 (72%) | 6.6 (5.0–8.0) | 6.0 (4.5–7.5) |
| Yes | 175 (28%) | 6.6 (5.4–7.7) | 6.0 (4.6–7.5) |
| Heart Failure | |||
| No | 466 (73%) | 6.4 (5.0–7.9) | 6.0 (4.5–7.5) |
| Yes | 169 (27%) | 6.6 (5.3–7.6) | 6.0 (4.4–7.5) |
| Stroke | |||
| No | 512 (80%) | 6.6 (5.0–7.9) | 6.0 (4.5–7.5) |
| Yes | 128 (20%) | 6.6 (5.4–7.9) | 6.2 (4.4–7.8) |
| Atrial Fibrillation/Flutter | |||
| No | 498 (82%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 113 (18%) | 6.4 (4.7–7.6) | 5.8 (4.5–7.2) |
| Peripheral Vascular Disease (PVD) | |||
| No | 521 (84%) | 6.6 (5.2–7.9) | 6.0 (4.5–7.5) |
| Yes | 102 (16%) | 6.3 (4.9–7.7) | 5.5 (4.5–7.5) |
| Diabetes Mellitus (DM) | |||
| No | 262 (41%) | 6.4 (5.0–7.6) | 5.9 (4.5–7.2) |
| Yes | 380 (59%) | 6.8 (5.3–8.0) | 6.2 (4.5–7.8) |
| Cancer | |||
| No | 536 (83%) | 6.6 (5.1–8.0) | 6.1 (4.5–7.5) |
| Yes | 106 (17%) | 6.4 (5.0–7.6) | 5.5 (4.0–7.0) |
| Chronic lung disease | |||
| No | 564 (88%) | 6.4 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 75 (12%) | 7.0 (5.6–8.3) | 6.5 (4.8–7.6) |
| Don’t know | 4 (1%) | 6.4 (4.8–6.8) | 5.2 (3.4–5.4) |
| Chronic liver disease | |||
| No | 610 (95%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 23 (4%) | 6.6 (5.2–7.5) | 6.0 (4.1–7.0) |
| Don’t know | 10 (2%) | 6.6 (5.4–6.9) | 6.0 (4.8–8.5) |
| Arthritis | |||
| No | 281 (44%) | 6.4 (5.0–7.7) | 6.0 (4.2–7.2) |
| Yes | 347 (54%) | 6.6 (5.3–8.0) | 6.1 (4.5–7.5) |
| Don’t know | 15 (2%) | 6.6 (5.1–7.3) | 6.4 (5.1–7.0) |
| Autoimmune Disorder (SLE, Vasculitis, IBD) | |||
| No | 592 (92%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 51 (8%) | 6.6 (5.1–7.4) | 5.8 (4.5–7.0) |
| Sickle Cell Anemia | |||
| No | 626 (97%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 11 (2%) | 6.7 (6.2–8.9) | 7.0 (6.6–8.5) |
| Don’t know | 6 (1%) | 6.6 (5.7–6.9) | 4.8 (3.8–6.2) |
| Psychological and Behavioral | |||
| Depression | |||
| No | 435 (68%) | 6.4 (5.0–7.6) | 5.8 (4.2–7.5) |
| Yes | 208 (32%) | 6.9 (5.4–8.4) | 6.2 (4.8–7.5) |
| Anxiety | |||
| No | 509 (79%) | 6.4 (5.0–7.7) | 6.0 (4.2–7.5) |
| Yes | 134 (21%) | 7.0 (5.7–8.4) | 6.4 (5.0–7.8) |
| Bipolar Disorder or Schizophrenia | |||
| No | 604 (94%) | 6.4 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 39 (6%) | 7.6 (6.0–8.3) | 6.5 (5.5–7.2) |
| Smoking Status | |||
| Current smoker | 99 (15%) | 7.0 (4.9–8.4) | 6.8 (4.6–7.8) |
| Former smoker | 222 (35%) | 6.6 (5.4–7.8) | 5.9 (4.8–7.5) |
| Never smoker | 322 (50%) | 6.4 (5.0–7.7) | 6.0 (4.2–7.5) |
| Alcohol Use | |||
| No | 579 (90%) | 6.6 (5.1–8.0) | 6.0 (4.5–7.5) |
| Yes | 64 (10%) | 6.3 (4.9–7.4) | 5.5 (3.8–6.8) |
| Illicit Drug Use | |||
| No | 540 (84%) | 6.6 (5.0–7.9) | 6.0 (4.2–7.5) |
| Yes | 103 (16%) | 6.7 (5.6–7.9) | 6.2 (4.8–7.5) |
Note that percentages are calculated using all nonmissing values for each variable. Missingness ranged from 0% to 10.26%.
VA: Veterans Affairs; SBP: Systolic Blood Pressure; DBP: Diastolic Blood Pressure; SLE: Systemic Lupus Erythematosus; IBD: Inflammatory Bowel Disease
Pain Distribution among Participants at Baseline by Sex
Overall, out of a total of 32 body regions, the median (IQR) number of painful body regions was 8 (4,14) (Figure S2). The proportion of female participants experiencing pain by any given body region was higher than that among male participants except for the pelvis (Figure 1, Tables S2–S3. Among female participants, the body regions (collapsed) where more than one-half of participants experienced pain were lower back (72%), knees (64%), legs (60%), upper back (59%), feet (58%), arms (57%), hands (54%), shoulders (54%), and hips (53%). Among male participants, the body regions where more than one-half of participants experienced pain were the lower back (67%), legs (56%), knees (53%), and feet (50%). Among males and females, the three body regions with the highest pain intensity (among those reporting pain in the body region) were lower back, chest, and feet (Tables S2–S3).
Figure 1:
Proportion of Participants Experiencing Pain by Body Region at Baseline
Pain Interference and Severity by Pain Characteristics at Baseline
Eighty-four percent of participants reported pain for more than 1 year and 75% experienced pain daily (Table 2). Eighty-nine percent of participants endorsed musculoskeletal pain while 66% endorsed neuropathic pain. Common reported locations of pain included hemodialysis vascular access (27%), visceral (25%), and chronic headache (24%). While a trend was observed in which patients with pain of longer duration had higher BPI interference scores, there was no clear pattern between duration of pain and pain severity. Participants reporting pain on a daily basis had higher pain interference and severity [interference 6.7 (5.4–8.0) and severity 6.2 (4.8–7.8)] compared with those reporting lower frequencies of pain. Among categories of types of pain, patients reporting musculoskeletal [6.6 (5.1–8.0) vs. 6.3 (5.0–7.4)], neuropathic [6.6 (5.3–8.0) vs. 6.3 (4.9–7.7)], vascular access [6.9 (5.4–8.1) vs. 6.4 (5.0–7.9)], chronic headache [7.0 (5.6–8.3) vs. 6.4 (5.0–7.8)], and visceral [6.9 (5.7–8.1) vs. 6.4 (5.0–7.7)] pain had higher pain interference compared with those who did not. Participants with vascular pain had higher BPI interference and severity compared with those without [6.9 (5.8–7.9) vs 6.4 (5.0–7.9); 6.4 (4.8–7.5) vs. 6.0 (4.2–7.5), respectively].
Table 2:
Pain Interference and Severity by Pain Characteristics at Baseline
| Characteristic | N (%)* | BPI Interference median (IQR) | BPI Severity median (IQR) |
|---|---|---|---|
| How long has participant had pain? | |||
| Less than 6 months | 46 (7%) | 6.1 (3.6–7.1) | 5.8 (4.2–6.9) |
| 6 months to 1 year | 57 (9%) | 6.4 (5.0–8.3) | 6.2 (4.0–7.8) |
| More than 1 year | 535 (84%) | 6.6 (5.3–8.0) | 6.0 (4.5–7.5) |
| Applied for or received disability insurance for pain condition | |||
| No | 509 (79%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 134 (21%) | 6.6 (5.4–7.8) | 6.2 (4.5–7.5) |
| How often does participant experience pain? | |||
| Once per week | 9 (1%) | 5.1 (3.4–5.6) | 4.2 (4.0–4.5) |
| Several days per week | 140 (22%) | 6.4 (4.4–7.3) | 5.2 (3.8–6.8) |
| A few times per month | 13 (2%) | 5.0 (4.6–7.1) | 4.8 (2.2–6.8) |
| Daily | 476 (75%) | 6.7 (5.4–8.0) | 6.2 (4.8–7.8) |
| Experience musculoskeletal pain | |||
| No | 71 (11%) | 6.3 (5.0–7.4) | 6.0 (4.5–7.4) |
| Yes | 567 (89%) | 6.6 (5.1–8.0) | 6.0 (4.5–7.5) |
| Experience neuropathic pain | |||
| No | 215 (34%) | 6.3 (4.9–7.7) | 5.9 (4.2–7.2) |
| Yes | 423 (66%) | 6.6 (5.3–8.0) | 6.0 (4.6–7.5) |
| Experience hemodialysis vascular access pain | |||
| No | 465 (73%) | 6.4 (5.0–7.9) | 6.0 (4.5–7.5) |
| Yes | 173 (27%) | 6.9 (5.4–8.1) | 6.2 (4.8–7.8) |
| Experience other vascular pain | |||
| No | 544 (85%) | 6.4 (5.0–7.9) | 6.0 (4.2–7.5) |
| Yes | 94 (15%) | 6.9 (5.8–7.9) | 6.4 (4.8–7.5) |
| Experience chronic headache | |||
| No | 487 (76%) | 6.4 (5.0–7.8) | 6.0 (4.2–7.5) |
| Yes | 151 (24%) | 7.0 (5.6–8.3) | 6.2 (4.8–7.8) |
| Experience cancer pain | |||
| No | 630 (99%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 8 (1%) | 5.9 (5.2–7.0) | 5.5 (4.9–7.1) |
| Experience visceral pain | |||
| No | 480 (75%) | 6.4 (5.0–7.7) | 6.0 (4.5–7.5) |
| Yes | 158 (25%) | 6.9 (5.7–8.1) | 6.1 (4.5–7.6) |
| Arm pain most disturbing to you | |||
| No | 577 (91%) | 6.6 (5.1–7.9) | 6.0 (4.5–7.5) |
| Yes | 59 (9%) | 6.6 (4.6–8.2) | 6.2 (4.4–7.8) |
| Leg pain most disturbing to you | |||
| No | 477 (75%) | 6.4 (5.0–7.9) | 6.0 (4.5–7.5) |
| Yes | 159 (25%) | 6.9 (5.4–8.1) | 6.2 (4.5–7.8) |
Note that percentages are calculated using all nonmissing values for each variable. Missingness ranged from 0% to 1.09%.
Association between Participant Characteristics and Pain Interference at Baseline
Using LASSO, a subset of variables (race, sex, ethnicity, education, employment, geographic location, opioid medication to treat pain, non-opioid medication to treat pain, any cardiovascular disease, diabetes, depression, and anxiety) was selected for the final reduced multivariable adjusted model for pain interference (Table 3). In this model, White race [mean difference −0.76 (95% CI: −1.13, −0.38), ref: Black race] and non-Hispanic ethnicity [mean difference −0.79 (95% CI: −1.29, −0.29), ref: Hispanic ethnicity] were independently associated with lower BPI interference scores while the presence of any cardiovascular disease [mean difference 0.33 (95% CI: 0.01, 0.64)], depression [mean difference 0.40 (95% CI: 0.04, 0.76)], and Midwest participant residence [mean difference 0.45 (0.04, 0.86), ref: Northeast residence] were independently associated with higher BPI interference scores. These results were similar in sensitivity analyses that did not include pain treatment variables (any non-pharmacologic treatment to manage chronic pain, any opioid medication to treat pain, and any non-opioid medication to treat pain) (Table S4).
Table 3:
Association between Participant Characteristics, Pain Interference, and Pain Severity at Baseline
| Characteristic* | BPI Interference | BPI Severity | ||
|---|---|---|---|---|
| Unadjusted Mean Diff. (95% CI) | LASSO Mean Diff. (95% CI) | Unadjusted Mean Diff. (95% CI) | LASSO Mean Diff. (95% CI) | |
| Demographic | ||||
| Age | ||||
| [18,55) | Ref. | --- | Ref. | Ref. |
| [55,70) | −0.12 (−0.49, 0.25) | --- | 0.35 (−0.02, 0.73) | 0.47 (0.10, 0.83) |
| [70,Inf) | −0.22 (−0.67, 0.23) | --- | −0.06 (−0.51, 0.39) | 0.06 (−0.41, 0.53) |
| Race | ||||
| Black | Ref. | Ref. | Ref. | Ref. |
| White | −0.53 (−0.90, −0.17) | −0.76 (−1.13, −0.38) | −0.87 (−1.23, −0.50) | −0.91 (−1.27, −0.55) |
| Other races | 0.03 (−0.40, 0.46) | −0.63 (−1.14, −0.11) | −0.22 (−0.65, 0.21) | −0.33 (−0.77, 0.11) |
| Sex at Birth | ||||
| Male | Ref. | Ref. | Ref. | Ref. |
| Female | 0.42 (0.09, 0.74) | 0.27 (−0.05, 0.58) | 0.46 (0.13, 0.78) | 0.43 (0.11, 0.75) |
| Hispanic Ethnicity | ||||
| Hispanic or Latino | Ref. | Ref. | Ref. | --- |
| Not Hispanic or Latino | −0.68 (−1.09, −0.27) | −0.79 (−1.29, −0.29) | −0.40 (−0.82, 0.02) | --- |
| Ethnicity unknown or not reported | −0.44 (−2.01, 1.13) | −0.63 (−2.15, 0.90) | −1.10 (−2.69, 0.50) | --- |
| Preferred Language | ||||
| English | Ref. | --- | Ref. | --- |
| Spanish | 0.73 (0.18, 1.28) | --- | 0.33 (−0.24, 0.89) | --- |
| Education | ||||
| Less than High School | Ref. | Ref. | Ref. | Ref. |
| High School | −0.39 (−0.82, 0.03) | −0.22 (−0.64, 0.21) | −0.49 (−0.92, −0.07) | −0.40 (−0.82, 0.01) |
| Above High School | −0.66 (−1.09, −0.22) | −0.44 (−0.88, 0.00) | −0.80 (−1.24, −0.37) | −0.65 (−1.09, −0.22) |
| Employment | ||||
| Employed | Ref. | Ref. | Ref. | --- |
| Not employed | 0.76 (0.18, 1.34) | 0.56 (−0.01, 1.13) | 0.56 (−0.02, 1.15) | --- |
| Relationship Status | ||||
| Married/domestic partner | Ref. | --- | Ref. | --- |
| Widowed/divorced/separated | 0.41 (0.02, 0.81) | --- | 0.32 (−0.07, 0.71) | --- |
| Never married | 0.41 (0.01, 0.82) | --- | 0.17 (−0.23, 0.58) | --- |
| VA Site Participant | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | −0.02 (−0.54, 0.50) | --- | 0.18 (−0.34, 0.71) | --- |
| Participant Residence | ||||
| Northeast | Ref. | Ref. | Ref. | --- |
| Midwest | 0.62 (0.21, 1.03) | 0.45 (0.04, 0.86) | 0.09 (−0.33, 0.50) | --- |
| South | −0.39 (−0.83, 0.05) | −0.28 (−0.71, 0.16) | −0.20 (−0.64, 0.25) | --- |
| West | 0.05 (−0.45, 0.54) | 0.26 (−0.25, 0.77) | −0.38 (−0.88, 0.12) | --- |
| Pain Treatment | ||||
| Any Non-pharmacologic Treatment to Manage Chronic Pain | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | −0.14 (−0.63, 0.35) | --- | −0.22 (−0.72, 0.27) | --- |
| Any Opioid Medication to Treat Pain | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.44 (0.05, 0.83) | 0.36 (−0.04, 0.75) | 0.32 (−0.08, 0.71) | 0.33 (−0.06, 0.72) |
| Any Non-Opioid Medication to Treat Pain | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.50 (0.16, 0.85) | 0.30 (−0.04, 0.65) | 0.39 (0.04, 0.74) | 0.32 (−0.03, 0.67) |
| Kidney Disease and Dialysis | ||||
| Years on Dialysis | ||||
| <1 year | Ref. | --- | Ref. | --- |
| 1–5 years | −0.10 (−0.51, 0.32) | --- | 0.09 (−0.32, 0.51) | --- |
| >5 years | 0.15 (−0.29, 0.60) | --- | 0.30 (−0.15, 0.74) | --- |
| Prior Kidney Transplant | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | −0.41 (−1.10, 0.28) | --- | −0.59 (−1.28, 0.11) | --- |
| Predialysis Blood Pressure | ||||
| SBP >= 140 mmHg or DBP >= 90 mmHg | Ref. | --- | Ref. | --- |
| SBP < 140 mmHg and DBP < 90 mmHg | 0.09 (−0.24, 0.41) | --- | −0.13 (−0.46, 0.20) | --- |
| Kt/V | ||||
| <1.2 | Ref. | --- | Ref. | --- |
| >=1.2 | −0.31 (−0.89, 0.27) | --- | −0.50 (−1.12, 0.11) | --- |
| Albumin | ||||
| >=4 g/dL | Ref. | --- | Ref. | --- |
| <4 g/dL | 0.03 (−0.29, 0.36) | --- | 0.15 (−0.18, 0.49) | --- |
| Hemoglobin | ||||
| >=10 g/dL | Ref. | --- | Ref. | --- |
| <10 g/dL | 0.09 (−0.29, 0.48) | --- | 0.17 (−0.21, 0.56) | --- |
| Medical Comorbidities | ||||
| Body Mass Index (BMI) | ||||
| <25 | Ref. | --- | Ref. | Ref. |
| 25–30 | −0.18 (−0.62, 0.26) | --- | 0.13 (−0.32, 0.57) | 0.15 (−0.28, 0.58) |
| >30 | −0.18 (−0.57, 0.21) | --- | −0.25 (−0.65, 0.14) | −0.27 (−0.66, 0.12) |
| Stroke | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 0.16 (−0.24, 0.56) | --- | 0.30 (−0.11, 0.70) | --- |
| Any Cardiovascular Disease (CVD) | ||||
| No | Ref. | Ref. | Ref. | --- |
| Yes | 0.24 (−0.09, 0.56) | 0.33 (0.01, 0.64) | 0.08 (−0.24, 0.41) | --- |
| Peripheral Vascular Disease (PVD) | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | −0.21 (−0.66, 0.24) | --- | 0.03 (−0.43, 0.48) | --- |
| Diabetes Mellitus (DM) | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.28 (−0.05, 0.60) | 0.13 (−0.19, 0.45) | 0.31 (−0.02, 0.64) | 0.23 (−0.10, 0.55) |
| Cancer | ||||
| No | Ref. | --- | Ref. | Ref. |
| Yes | −0.20 (−0.64, 0.23) | --- | −0.55 (−0.98, −0.11) | −0.39 (−0.83, 0.05) |
| Chronic Lung or Liver Disease | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 0.27 (−0.19, 0.72) | --- | 0.21 (−0.25, 0.67) | --- |
| Don’t know | −0.41 (−1.71, 0.89) | --- | 0.03 (−1.32, 1.38) | --- |
| Arthritis | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 0.17 (−0.16, 0.50) | --- | 0.31 (−0.02, 0.64) | --- |
| Don’t know | −0.10 (−1.18, 0.98) | --- | 0.60 (−0.51, 1.71) | --- |
| Autoimmune Disorder (SLE, Vasculitis, IBD) | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | −0.18 (−0.78, 0.41) | --- | −0.16 (−0.76, 0.44) | --- |
| Sickle Cell Anemia | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 1.04 (−0.20, 2.28) | --- | 1.45 (0.21, 2.70) | --- |
| Don’t know | −0.66 (−2.33, 1.01) | --- | −1.05 (−2.81, 0.72) | --- |
| Psychological and Behavioral | ||||
| Depression | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.56 (0.22, 0.90) | 0.40 (0.04, 0.76) | 0.41 (0.06, 0.76) | 0.33 (−0.03, 0.69) |
| Anxiety | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.71 (0.31, 1.10) | 0.38 (−0.05, 0.80) | 0.48 (0.08, 0.88) | 0.19 (−0.23, 0.62) |
| Bipolar Disorder or Schizophrenia | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 0.73 (0.05, 1.40) | --- | 0.32 (−0.36, 1.00) | --- |
| Smoking Status | ||||
| Current smoker | Ref. | --- | Ref. | --- |
| Former smoker | −0.12 (−0.61, 0.38) | --- | −0.15 (−0.65, 0.35) | --- |
| Never smoker | −0.31 (−0.78, 0.16) | --- | −0.37 (−0.85, 0.10) | --- |
| Alcohol Use | ||||
| No | Ref. | --- | Ref. | Ref. |
| Yes | −0.46 (−0.99, 0.08) | --- | −0.76 (−1.29, −0.22) | −0.58 (−1.11, −0.05) |
| Illicit Drug Use | ||||
| No | Ref. | --- | Ref. | --- |
| Yes | 0.39 (−0.04, 0.83) | --- | 0.31 (−0.13, 0.75) | --- |
The unadjusted models each have 1 covariate, the reduced model for BPI Interference has 12 covariates, and the reduced model for BPI Severity has 12 covariates. See the statistical details section for more information about the group lasso procedure for the reduced models.
Any CVD included coronary artery disease (CAD), heart failure, or atrial fibrillation/flutter
VA: Veterans Affairs; SBP: Systolic Blood Pressure; DBP: Diastolic Blood Pressure; SLE: Systemic Lupus Erythematosus; IBD: Inflammatory Bowel Disease
Association between Participant Characteristics and Pain Severity at Baseline
Using LASSO, a subset of variables (age, race, sex, education, opioid medication to treat pain, non-opioid medication to treat pain, body mass index, diabetes, cancer, depression, anxiety, and alcohol use) was selected for the final reduced multivariable adjusted model for pain severity (Table 3). In this model, White race [mean difference −0.91 (95% CI: −1.27, −0.55), ref: Black race], above high school education [mean difference −0.65 (95% CI: −1.09, −0.22), ref: less than high school education], and alcohol use [mean difference −0.58 (95% CI: −1.11, −0.05) were independently associated with lower BPI severity scores while age 55 to < 70 years [mean difference 0.47 (95% CI: 0.10, 0.83), ref: 18 to < 55 years] and female sex [mean difference 0.43 (95% CI: 0.11, 0.75)] were independently associated with higher BPI severity scores. These results were similar in sensitivity analyses that did not include pain treatment variables (any non-pharmacologic treatment to manage chronic pain, any opioid medication to treat pain, and any non-opioid medication to treat pain) (Table S4).
Similar findings were observed in fully adjusted multivariable linear regression models for pain interference and pain severity, including sensitivity analyses without pain treatment variables (Table S4–S5).
Correlations between Pain Interference, Pain Severity, and Patient Reported Symptoms at Baseline
Unadjusted moderate positive correlations (r >= 0.4) were found between BPI interference and pain catastrophizing [r = 0.48 (0.41, 0.53)], dialysis symptom index severity [r=0.45 (95% CI: 0.39, 0.51)], fatigue [r=0.44 (95% CI: 0.38, 0.50)], depressive symptoms (PHQ-9) [r= 0.43 95% CI: (0.37, 0.5)], catastrophizing (CSQ-24) [r= 0.43 (95% CI: 0.36, 0.49)], and anxiety symptoms (GAD-7) [r= 0.43 (95% CI: 0.36, 0.49)] (Table S6). A moderate positive correlation (r >= 0.4) was found between pain catastrophizing and BPI severity [0.48 (95% CI: 0.42, 0.54)]. When adjusted for age and sex, correlation values were similar to the unadjusted results (Table S6).
Discussion
Among a large diverse cohort of adults with chronic pain who were receiving maintenance hemodialysis across the United States, we report several findings important to their care. First, while musculoskeletal pain is the most common source of chronic pain in this patient population, there is tremendous heterogeneity in pain sources. Therefore, routine comprehensive pain assessments are necessary and treatment plans should be tailored to the specific needs and symptoms of the individual. Second, sociodemographic and psychosocial factors were more likely than kidney disease, dialysis, and other comorbid conditions to be related to pain, particularly pain interference. These factors should be considered by clinicians in their approach to pain assessments of adults receiving maintenance hemodialysis. Third, we observed a prominent concurrence of pain with pain catastrophizing as well as with symptoms of anxiety, depression, and fatigue, which underscores the importance of including assessments for psychosocial and neurobehavioral conditions in pain management.
Except for a few small recent studies (45,46), pain interference, which indicates the degree to which pain negatively impacts a person’s daily function and activity, has been inadequately described in adults receiving HD. Despite most HOPE participants having received non-pharmacologic and/or pharmacologic pain treatments, their average pain interference and severity remained moderate to severe, which underscores the significant functional disability incurred by pain. Moreover, these findings corroborate that current pain management strategies are not sufficiently effective for many adults with chronic pain on HD and underscores the need to improve clinical management of pain (1,10,12). Consideration should be given to cross-cutting and multimodal pain interventions that can treat pain and related symptoms of a variety of etiologies and locations, such as those evaluated in the HOPE Consortium Trial (21).
We found that musculoskeletal pain was the most common type of pain, and that low back and lower limb pain were the leading locations of pain, similar to prior studies (1,4,19). However, the multitude of painful body regions and pain locations (e.g., vascular access, visceral, headache) as well as the preponderance of musculoskeletal and neuropathic pain substantially improves the characterization of pain in this population and underscores the numerous and diverse phenotypes of pain. To address this observed burden and heterogeneity of pain, a comprehensive care plan tailored to the needs and symptoms of the individual patient by the multidisciplinary dialysis team is needed on a regular basis, consistent with KDIGO recommendations (47). However, ameliorating the unsatisfactory state of pain management among HD adults is not solely a call to action for the dialysis care team. Pain did not appear to be related to the hemodialysis procedure in terms of dialysis adequacy and other laboratory-based parameters. Importantly, we did not collect data regarding hemodialysis complications (e.g., cramping) or adherence. Nonetheless, we believe it is imperative that all medical professionals (e.g., primary care providers, pain specialists, palliative care providers, social workers) who care for adults receiving maintenance HD be involved in their pain evaluation. Models of dialysis patient care with engagement of PCPs are being explored and some have already been shown to improve health-related quality of life including symptoms related to kidney disease (48). Therefore, integrating and coordinating care with PCPs and other medical professionals for adults with ESKD could greatly increase adoption of symptom-based screening assessment and pain management.
Prior studies examining individual patient factors associated with pain in adults receiving HD had inconsistent findings, which in part may be driven by small sample sizes and limited racial and ethnic diversity (1,46,49). Notably, we found several sociodemographic, psychological, and behavioral factors associated with pain interference and severity. Participants who identified as Black reported a higher degree of pain interference and severity, while those who identified as Hispanic ethnicity and female sex had higher degrees of pain interference and severity, respectively. Studies of general populations with pain suggest that the pain experience may differ among patients of different sexes and races (50–53), which may be due to biological and socioeconomic factors, environmental stressors, psychological responses, and disparities in pain assessment and treatment by clinicians (3, 16, 50, 52,54, 56, 57). We now report these disparities among adults receiving HD. This has potentially significant implications because of the overrepresentation of minoritized populations among those needing maintenance HD (55), but warrants further detailed investigation.
The general lack of substantial associations observed between kidney disease, dialysis factors, and other medical comorbidities with pain interference and severity suggests pain in maintenance HD patients may have a variety of etiologies and not always have discrete attributions. In particular, laboratory-based dialysis parameters (e.g., Kt/V, albumin), which have been utilized traditionally by the dialysis care team to assess quality of dialysis care, did not appear useful in differentiating the burden of pain across patients. One unintended consequence of an emphasis on easily quantifiable measures is less focus on symptom assessment and management. As noted by Weisbord et al, adults on HD are infrequently asked about pain by the dialysis team (14). The results in this report underscore the need to routinely assess meaningful patient symptoms such as pain.
We observed moderate correlations between pain interference and symptoms of depression, anxiety, and fatigue. A diagnosis of depression was also significantly associated with pain interference. Adults receiving maintenance HD have a significant and diverse burden of symptoms with pain, depression, anxiety, and fatigue noted to be especially common and co-occurring (10,17,58,59). For example, similar to our observations, others have noted significant associations between depression, anxiety, and pain frequency and severity (2,5). In the past few years, a greater appreciation has developed for the interrelationships between these symptoms through symptom cluster analyses among HD patients (59,60). In fact, the proportion of patients receiving HD with a high probability of fatigue, moderate-severe pain, and moderate-severe depressive symptoms is similar to that among adults with cancer (60). It has been previously posited that perception of pain can be intensified by depression and anxiety, and in turn pain can worsen depression and anxiety (2,59). The psychological and biological mechanisms linking these debilitating symptoms are not understood (59). Nonetheless, awareness of these relationships bears relevance to the clinical care of dialysis patients with chronic pain. Others have noted that it is unlikely that pain will be successfully treated without addressing psychological and social concerns (1). It is recommended that pain management programs should be based on a biopsychosocial model of pain that includes assessments for psychosocial and neurobehavioral conditions to be effective in treating pain in adults receiving HD (61,62).
Our study has limitations. The HOPE Trial Consortium cohort was composed of adults receiving maintenance hemodialysis who had moderate to severe chronic pain as defined by this study’s inclusion criteria (21) and may have resulted in a more homogenous cohort for some of the other characteristics. Therefore, these findings do not necessarily extend to all dialysis patients with pain. However, the cohort was drawn from 16 centers and 103 dialysis facilities in multiple geographic regions and in urban and rural settings. It was sociodemographically diverse with both Spanish- and English-speaking participants, and the age, sex, race, and Hispanic ethnicity composition was generally similar to that of the overall US dialysis patient population (55). Moreover, the high pain interference and severity scores among HOPE participants confirm the tremendous pain burden in this cohort and its appropriateness for studies of pain interventions. Second, the design of these analyses was cross-sectional. Therefore, neither directionality nor causality of observed relationships can be inferred.
In summary, among a large, diverse, and national sample of adults undergoing maintenance HD with chronic pain, the burden of pain is substantial despite pain treatments, and its characteristics are heterogeneous. Based on these findings, routine pain assessments and treatment plans tailored to the specific needs and symptoms of the individual patient seem warranted for this patient population, especially targeting high risk groups such as adults identifying as Black or Hispanic, and female, as well as those with depression, anxiety, and fatigue. Further investigation into the biologic, psychological, and social mechanisms of pain is needed to advance our understanding of its diversity and disparities in patients on maintenance HD and to develop effective treatment strategies.
Supplementary Material
Figure S1: Distribution of BPI Interference and Severity Scores among Participants
Figure S2: Distribution of Total Number of Body Regions Reported by Participants as Painful
Item S1: HOPE Consortium, Cores, and Data and Safety Monitoring Board Members
Item S2: The Development of the Reduced Models
Item S3: Variable Missingness
Table S1: Summary Data for Proportion of Participants Experiencing Pain by Body Region at Baseline
Table S2: Summary Data for Proportion of Participants Experiencing Pain by Body Region at Baseline (collapsed)
Table S3: Multivariable Linear Regression Results for Association between Participant Characteristics, Pain Interference, and Pain Severity at Baseline
Table S4: Spearman Correlations between Pain Interference, Pain Severity, and Patient Symptoms at Baseline
Table S5: Association between Participant Characteristics, Pain Interference, and Pain Severity at Baseline without Pain Treatment Variables
Acknowledgements:
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 North America; Northwest Kidney Centers; Puget Sound Kidney Centers; The Rogosin Institute; University of Illinois Hospital Dialysis Unit, Chicago, IL; and Associates in Nephrology, S.C. Chicago, IL.
Financial Disclosure:
Dr. Scherer is on the clinical advisory board of Monogram Health, has speaking fees from Vifor Pharmaceuticals and Cara Therapeutics, and royalties from Uptodate. Dr. Nigwekar received compensation from the American Society of Nephrology for his role as the Deputy Editor of the Clinical Journal of Society of Nephrology and from the National Kidney Foundation for his role as Associate Editor of the Advances in Kidney Disease and Health, consulting fees and/or funding from CSL Behring, Fresenius Medical Care North America, Inozyme Pharma, Epizon Pharma, Vera Therapeutics, Alexion, RubiconMD, and Sanofi, and royalty payments from UpToDate. Dr. Charytan reports consulting fees from Eli Lilly/Boehringer Ingelheim, Astra ZenecaAllena Pharmaceuticals (DSMB), Gilead, Novo Nordisk, GSK, Medtronic, Merck, CSL Behring, Zogenix, Renalytix, LG Chemical, Alentis Therapeutics, Fresenius/NxStage; Research funding from Medtronic; Clinical trial support from Gilead, NovoNordisk, Amgen, Boehringer Ingelheim/Eli Lilly, Astra Zeneca; Patents or royalties: UpToDate.com. Fees related to editorial work at CJASN; and expert witness fees related to proton pump inhibitors and anti-depressants. Dr. Jhamb has research support from NIH, Dialysis Clinic, Inc., Pfizer, Bayer and CKD Leaders Network; personal consultancy fees from Boehringer-Ingelheim, Eli-Lilly, Xcenda LLC and CKD Leaders Network. Dr. Mehrotra is the Editor-in-Chief of the Journal of the American Society of Nephrology and the Senior Editor-in-Chief of ASN Portfolio of Journals. Dr. Keefe has travel support from ENTRUST-PE ENhancing TRUSTworthiness in Pain Evidence (ENTRUST-PE). Dr. Dember received compensation from the National Kidney Foundation for her role as Deputy Editor of the American Journal of Kidney Diseases, honoraria from the American Society of Nephrology, consulting fees from AstraZeneca, Merck, and Alucent Biomedical, and compensation for serving on Data and Safety Monitoring Boards for the National Institute of Diabetes and Digestive and Kidney Diseases, CSL Behring, and Vertex Pharmaceuticals. Dr. Johansen has done consulting for GSK, Akebia, and Vifor in the last 3 years. The remaining authors declare that they have no relevant financial interests.
Support:
This work was funded by the following grants from the National Institute of Diabetes and Digestive and Kidney Diseases: U01DK123786, U01DK123787, U01DK123812, U01DK123813, U01DK123814, U01DK123816, U01DK123817, U01DK123818, U01DK123821, K26DK138374. 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.
Footnotes
Disclaimer: 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.
References
- 1.Davison SN. Pain in hemodialysis patients: prevalence, cause, severity, and management. Am J Kidney Dis 2003;42(6):1239–1247. doi: 10.1053/j.ajkd.2003.08.025 [DOI] [PubMed] [Google Scholar]
- 2.Davison SN, Jhangri GS. The impact of chronic pain on depression, sleep, and the desire to withdraw from dialysis in hemodialysis patients. J Pain Symptom Manage 2005;30(5):465–473. doi: 10.1016/j.jpainsymman.2005.05.013 [DOI] [PubMed] [Google Scholar]
- 3.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. Published 2021 Feb 19. doi: 10.1177/2054358121993995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Brkovic T, Burilovic E, Puljak L. Prevalence and severity of pain in adult end-stage renal disease patients on chronic intermittent hemodialysis: a systematic review. Patient Prefer Adherence. 2016;10:1131–1150. Published 2016 Jun 23. doi: 10.2147/PPA.S103927 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cohen SD, Patel SS, Khetpal P, Peterson RA, Kimmel PL. Pain, sleep disturbance, and quality of life in patients with chronic kidney disease. Clin J Am Soc Nephrol 2007;2(5):919–925. doi: 10.2215/CJN.00820207 [DOI] [PubMed] [Google Scholar]
- 6.Kimmel PL, Emont SL, Newmann JM, Danko H, Moss AH. ESRD patient quality of life: symptoms, spiritual beliefs, psychosocial factors, and ethnicity. Am J Kidney Dis 2003;42(4):713–721. doi: 10.1016/s0272-6386(03)00907-7 [DOI] [PubMed] [Google Scholar]
- 7.Harris TJ, Nazir R, Khetpal P, et al. Pain, sleep disturbance and survival in hemodialysis patients. Nephrol Dial Transplant 2012;27(2):758–765. doi: 10.1093/ndt/gfr355 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Davison SN. The prevalence and management of chronic pain in end-stage renal disease. J Palliat Med 2007;10(6):1277–1287. doi: 10.1089/jpm.2007.0142 [DOI] [PubMed] [Google Scholar]
- 9.Weisbord SD, Mor MK, Sevick MA, et al. Associations of depressive symptoms and pain with dialysis adherence, health resource utilization, and mortality in patients receiving chronic hemodialysis. Clin J Am Soc Nephrol 2014;9(9):1594–1602. doi: 10.2215/CJN.00220114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Weisbord SD. Patient-Centered Dialysis Care: Depression, Pain, and Quality of Life. Semin Dial 2016;29(2):158–164. doi: 10.1111/sdi.12464 [DOI] [PubMed] [Google Scholar]
- 11.Davison SN, Jhangri GS. Impact of pain and symptom burden on the health-related quality of life of hemodialysis patients. J Pain Symptom Manage 2010;39(3):477–485. doi: 10.1016/j.jpainsymman.2009.08.008 [DOI] [PubMed] [Google Scholar]
- 12.Claxton RN, Blackhall L, Weisbord SD, Holley JL. Undertreatment of symptoms in patients on maintenance hemodialysis. J Pain Symptom Manage 2010;39(2):211–218. doi: 10.1016/j.jpainsymman.2009.07.003 [DOI] [PubMed] [Google Scholar]
- 13.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]
- 14.Weisbord SD, Fried LF, Mor MK, et al. Renal provider recognition of symptoms in patients on maintenance hemodialysis. Clin J Am Soc Nephrol 2007;2(5):960–967. doi: 10.2215/CJN.00990207 [DOI] [PubMed] [Google Scholar]
- 15.Roy PJ, Weltman M, Dember LM, Liebschutz J, Jhamb M; HOPE Consortium. Pain management in patients with chronic kidney disease and end-stage kidney disease. Curr Opin Nephrol Hypertens 2020;29(6):671–680. doi: 10.1097/MNH.0000000000000646 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tobin DG, Lockwood MB, Kimmel PL, et al. Opioids for chronic pain management in patients with dialysis-dependent kidney failure. Nat Rev Nephrol 2022;18(2):113–128. doi: 10.1038/s41581-021-00484-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Abdel-Kader K, Unruh ML, Weisbord SD. Symptom burden, depression, and quality of life in chronic and end-stage kidney disease. Clin J Am Soc Nephrol 2009;4(6):1057–1064. doi: 10.2215/CJN.00430109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Murtagh FE, Addington-Hall J, Higginson IJ. The prevalence of symptoms in end-stage renal disease: a systematic review. Adv Chronic Kidney Dis 2007;14(1):82–99. doi: 10.1053/j.ackd.2006.10.001 [DOI] [PubMed] [Google Scholar]
- 19.Lambourg E, Colvin L, Guthrie G, et al. The prevalence of pain among patients with chronic kidney disease using systematic review and meta-analysis. Kidney Int 2021;100(3):636–649. doi: 10.1016/j.kint.2021.03.041 [DOI] [PubMed] [Google Scholar]
- 20.Flythe JE, Hilliard T, Castillo G, et al. Symptom Prioritization among Adults Receiving In-Center Hemodialysis: A Mixed Methods Study. Clin J Am Soc Nephrol 2018;13(5):735–745. doi: 10.2215/CJN.10850917 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dember LM, Hsu JY, Bernardo L, et al. 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]
- 22.Dember LM, Hsu JY, Mehrotra R, et al. 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]
- 23.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]
- 24.Scherrer KH, Ziadni MS, Kong JT, et al. Development and validation of the Collaborative Health Outcomes Information Registry body map. Pain Rep 2021;6(1):e880. Published 2021 Jan 15. doi: 10.1097/PR9.0000000000000880 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singap 1994;23(2):129–138. [PubMed] [Google Scholar]
- 26.Keller S, Bann CM, Dodd SL, Schein J, Mendoza TR, Cleeland CS. Validity of the brief pain inventory for use in documenting the outcomes of patients with noncancer pain. Clin J Pain. 2004;20(5):309–318. doi: 10.1097/00002508-200409000-00005 [DOI] [PubMed] [Google Scholar]
- 27.Sullivan MJL BS, Pivik J. The Pain Catastrophizing Scale: Development and validation. Psychological Assessment 1995;7:524–32. [Google Scholar]
- 28.Cohen SR, Sawatzky R, Russell LB, Shahidi J, Heyland DK, Gadermann AM. Measuring the quality of life of people at the end of life: The McGill Quality of Life Questionnaire-Revised. Palliat Med 2017;31:120–9. [DOI] [PubMed] [Google Scholar]
- 29.Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 1999;282:1737–44. [DOI] [PubMed] [Google Scholar]
- 30.Spitzer RL, Kroenke K, Williams JB, Lowe B. A brief measure for assessing generalized Anxiety disorder: the GAD-7. Arch Intern Med 2006;166:1092–7. [DOI] [PubMed] [Google Scholar]
- 31.Harland N, Georgieff K,. Development of the Coping Strategies Questionnaire 24, a Clinically Utilitarian Version of the Coping Strategies Questionnaire. Rehabilitation Psychology 2003;48:4. [Google Scholar]
- 32.Gruber-Baldini AL, Velozo C, Romero S, Shulman LM. Validation of the PROMIS((R)) measures of self-efficacy for managing chronic conditions. Qual Life Res 2017;26:1915–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Cella D, Riley W, Stone A, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. J Clin Epidemiol 2010;63:1179–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Buysse DJ, Yu L, Moul DE, et al. Development and validation of patient-reported outcome measures for sleep disturbance and sleep-related impairments. Sleep 2010;33:781–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Schneider S, Choi SW, Junghaenel DU, Schwartz JE, Stone AA. Psychometric characteristics Of daily diaries for the Patient-Reported Outcomes Measurement Information System (PROMIS(R)): a preliminary investigation. Qual Life Res 2013;22:1859–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Rose M, Bjorner JB, Gandek B, Bruce B, Fries JF, Ware JE Jr. The PROMIS Physical Function item bank was calibrated to a standardized metric and shown to improve measurement efficiency. J Clin Epidemiol 2014;67:516–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Zimet GD DN, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. Journal of Personality Assessment 1988;52:30–41. [Google Scholar]
- 38.Weisbord SD, Fried LF, Arnold RM, et al. Development of a symptom assessment instrument for chronic hemodialysis patients: the Dialysis Symptom Index. J Pain Symptom Manage 2004;27:226–40. [DOI] [PubMed] [Google Scholar]
- 39.Hahn EA, Devellis RF, Bode RK, et al. Measuring social health in the patient-reported Outcomes measurement information system (PROMIS): item bank development and testing. Qual Life Res 2010;19:1035–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Chen Q, Wang S. Variable selection for multiply-imputed data with application to dioxin exposure study. Stat Med 2013;32(21):3646–3659. doi: 10.1002/sim.5783 [DOI] [PubMed] [Google Scholar]
- 41.Marino M, Buxton OM, Li Y. Covariate Selection for Multilevel Models with Missing Data. Stat (Int Stat Inst). 2017;6(1):31–46. doi: 10.1002/sta4.133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Liu Q, Li C, Wanga V, Shepherd BE. Covariate-adjusted Spearman’s rank correlation with probability-scale residuals. Biometrics. 2018;74(2):595–605. doi: 10.1111/biom.12812 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Van Buuren S, Brand JPL, Groothuis-Oudshoorn CGM, Rubin DB. Fully conditional specification in multivariate imputation. J Statistical Computation and Simulation. 2006;76(12):1049–64. [Google Scholar]
- 44.Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York: John Wiley and Sons, 1987. [Google Scholar]
- 45.Dos Santos PR, Mendonça CR, Hernandes JC, et al. Pain in Patients With Chronic Kidney Disease Undergoing Hemodialysis: A Systematic Review. Pain Manag Nurs 2021;22(5):605–615. doi: 10.1016/j.pmn.2021.05.009 [DOI] [PubMed] [Google Scholar]
- 46.Marzouq MK, Samoudi AF, Samara A, Zyoud SH, Al-Jabi SW. Exploring factors associated with pain in hemodialysis patients: a multicenter cross-sectional study from Palestine. BMC Nephrol 2021;22(1):96. Published 2021 Mar 17. doi: 10.1186/s12882-021-02305-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Davison SN, Levin A, Moss AH, et al. Executive summary of the KDIGO Controversies Conference on Supportive Care in Chronic Kidney Disease: developing a roadmap to improving quality care. Kidney Int 2015;88(3):447–459. doi: 10.1038/ki.2015.110 [DOI] [PubMed] [Google Scholar]
- 48.Hynes DM, Fischer M, Fitzgibbon M, et al. Integrating a Medical Home in an Outpatient Dialysis Setting: Effects on Health-Related Quality of Life. J Gen Intern Med 2019;34(10):2130–2140. doi: 10.1007/s11606-019-05154-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Brkovic T, Burilovic E, Puljak L. Risk Factors Associated with Pain on Chronic Intermittent Hemodialysis: A Systematic Review. Pain Pract 2018;18(2):247–268. doi: 10.1111/papr.12594 [DOI] [PubMed] [Google Scholar]
- 50.Campbell CM, Edwards RR, Fillingim RB. Ethnic differences in responses to multiple experimental pain stimuli. Pain. 2005;113(1–2):20–26. doi: 10.1016/j.pain.2004.08.013 [DOI] [PubMed] [Google Scholar]
- 51.Hastie BA, Riley JL, Fillingim RB. Ethnic differences and responses to pain in healthy young adults. Pain Med 2005;6(1):61–71. doi: 10.1111/j.1526-4637.2005.05009.x [DOI] [PubMed] [Google Scholar]
- 52.Fillingim RB, Hastie BA, Ness TJ, Glover TL, Campbell CM, Staud R. Sex-related psychological predictors of baseline pain perception and analgesic responses to pentazocine. Biol Psychol 2005;69(1):97–112. doi: 10.1016/j.biopsycho.2004.11.008 [DOI] [PubMed] [Google Scholar]
- 53.Paller CJ, Campbell CM, Edwards RR, Dobs AS. Sex-based differences in pain perception and treatment. Pain Med 2009;10(2):289–299. doi: 10.1111/j.1526-4637.2008.00558.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Meints SM, Cortes A, Morais CA, Edwards RR. Racial and ethnic differences in the experience and treatment of noncancer pain. Pain Manag 2019;9(3):317–334. doi: 10.2217/pmt-2018-0030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Johansen KL, Chertow GM, Foley RN, et al. US Renal Data System 2020 Annual Data Report: epidemiology of kidney disease in the United States. Am J Kidney Dis 2021;77(4)(suppl 1):A7–A8. doi: 10.1053/j.ajkd.2021.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.FitzGerald C, Hurst S. Implicit bias in healthcare professionals: a systematic review. BMC Med Ethics. 2017;18(1):19. Published 2017 Mar 1. doi: 10.1186/s12910-017-0179-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Drwecki BB. Education to identify and combat racial bias in pain treatment. AMA J Ethics. 2015;17(3):221–228. Published 2015 Mar 1. doi: 10.1001/journalofethics.2015.17.3.medu1-1503 [DOI] [PubMed] [Google Scholar]
- 58.Devaraj SM, Roumelioti ME, Yabes JG, et al. Correlates of Rates and Treatment Readiness for Depressive Symptoms, Pain, and Fatigue in Hemodialysis Patients: Results from the TĀCcare Study. Kidney360. 2023;4(9):e1265–e1275. doi: 10.34067/KID.0000000000000213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Lockwood MB, Steel JL, Doorenbos AZ, Contreras BN, Fischer MJ. Emerging Patient-Centered Concepts in Pain Among Adults With Chronic Kidney Disease, Maintenance Dialysis, and Kidney Transplant. Semin Nephrol 2021;41(6):550–562. doi: 10.1016/j.semnephrol.2021.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Jhamb M, Abdel-Kader K, Yabes J, et al. Comparison of Fatigue, Pain, and Depression in Patients With Advanced Kidney Disease and Cancer-Symptom Burden and Clusters. J Pain Symptom Manage 2019;57(3):566–575.e3. doi: 10.1016/j.jpainsymman.2018.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Cheatle MD. Biopsychosocial Approach to Assessing and Managing Patients with Chronic Pain. Med Clin North Am 2016;100(1):43–53. doi: 10.1016/j.mcna.2015.08.007 [DOI] [PubMed] [Google Scholar]
- 62.Nair D, Cukor D, Taylor WD, Cavanaugh KL. Applying A Biopsychosocial Framework to Achieve Durable Behavior Change in Kidney Disease. Semin Nephrol 2021;41(6):487–504. doi: 10.1016/j.semnephrol.2021.10.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Distribution of BPI Interference and Severity Scores among Participants
Figure S2: Distribution of Total Number of Body Regions Reported by Participants as Painful
Item S1: HOPE Consortium, Cores, and Data and Safety Monitoring Board Members
Item S2: The Development of the Reduced Models
Item S3: Variable Missingness
Table S1: Summary Data for Proportion of Participants Experiencing Pain by Body Region at Baseline
Table S2: Summary Data for Proportion of Participants Experiencing Pain by Body Region at Baseline (collapsed)
Table S3: Multivariable Linear Regression Results for Association between Participant Characteristics, Pain Interference, and Pain Severity at Baseline
Table S4: Spearman Correlations between Pain Interference, Pain Severity, and Patient Symptoms at Baseline
Table S5: Association between Participant Characteristics, Pain Interference, and Pain Severity at Baseline without Pain Treatment Variables