Undertreatment of Symptoms in Patients on Maintenance Hemodialysis

Article Outline

• Abstract
• Introduction
• Subjects and Methods
  • Patient Selection
  • Assessment of Symptoms
  • Assessment of Treatment of Potentially Treatable Symptoms
  • Statistical Analyses
• Results
  • Patient Characteristics
  • Prevalence and Treatment of Symptoms
• Discussion
• Acknowledgments
• References
• Copyright

Abstract 

Context

Hemodialysis patients suffer a large symptom burden, and little is known about how effectively symptoms are treated.

Objectives

To assess the management of treatable symptoms in hemodialysis patients, we administered a 30-item questionnaire on physical and emotional symptoms to patients receiving outpatient hemodialysis at the University of Virginia.

Methods

We asked patients whether they were prescribed therapy for potentially treatable symptoms and assessed who prescribed the therapy. By means of chart review, we also documented whether medications were prescribed for these symptoms.

Results

We approached 87 patients and enrolled 62 (71%). The most commonly reported, potentially treatable symptoms included bone/joint pain, insomnia, mood disturbance, sexual dysfunction, paresthesia, and nausea. Only 45% of patients with bone/joint pain reported receiving an analgesic medication. Twenty-three percent of patients with trouble falling asleep and 53% of patients with nausea reported receiving a medication to alleviate this symptom. Chart review revealed that 58% of patients who reported the presence of bone/joint pain were prescribed an analgesic, 23% of patients with trouble falling asleep were prescribed a sleep aid, and 42% of patients with nausea received an antiemetic. Primary care providers were more likely than nephrologists to provide for all symptoms except nausea and numbness or tingling in the feet, and this difference was significant for the treatment of worrying (3/3 vs. 0/3, P=0.05) and nervousness (4/5 vs. 0/5, P=0.02).

Conclusion

Potentially treatable symptoms in hemodialysis are undertreated. Pharmacologic therapy, particularly for emotional symptoms, was more commonly prescribed by primary care providers than nephrologists. Additional study of the barriers to symptom treatment and interventions that increase nephrologist and primary care provider symptom management are needed.

Key Words: Dialysis patients, quality of life, symptom burden, symptom management, symptom assessment

Back to Article Outline

Introduction 

The number of patients undergoing hemodialysis in the United States is growing.¹ Despite widespread effort, few interventions other than successful kidney transplantation have improved survival in this patient population. As a result, increased attention has been focused on improving the quality of life of patients dependent on chronic renal replacement therapy.², ³, ⁴ Recent studies demonstrate a high burden of physical and emotional symptoms in this patient population and reveal that symptoms are associated with impaired quality of life.⁵, ⁶, ⁷, ⁸, ⁹ Despite this, recent data suggest that nephrologists may not be aware of many of the symptoms that bother hemodialysis patients.¹⁰ Little is known about whether patients on hemodialysis receive appropriate treatment for their symptoms.

A few studies have shown that the treatment of depression and pain is suboptimal in this patient population⁷, ¹¹, ¹², ¹³ and that patients with chronic pain have higher rates of insomnia and depression than those without pain.¹⁴ Other symptoms that occur commonly in the hemodialysis population, such as insomnia, sexual dysfunction, and nausea, are potentially amenable to pharmacologic therapy, but the extent to which these symptoms are treated is not known. It is also unclear whether nephrologists or primary care providers are more likely to treat such symptoms in this population. To address these questions, we conducted a prospective observational study to assess whether pharmacologic therapy is prescribed for potentially treatable symptoms in hemodialysis patients and whether such therapy is more likely to be prescribed by nephrologists or primary care clinicians.

Back to Article Outline

Subjects and Methods 

Patient Selection 

This was a cross-sectional study of patients receiving thrice-weekly outpatient hemodialysis at the University of Virginia. We approached and enrolled patients over a 14-day period in May 2007. Inclusion criteria were age greater than 18 years, English language fluency as demonstrated by participant ability to discuss risks and benefits of study participation, and ability to successfully complete the CLOX test (a drawing of the face of a clock) as screening for cognitive function.¹⁵ Institutional review board (IRB) approval was obtained through the University of Virginia IRB, and each participant completed informed consent.

Assessment of Symptoms 

To assess the presence of symptoms, patients completed the Dialysis Symptom Index (DSI) either independently or with the aid of a researcher reading survey questions aloud.¹⁶ This 30-item survey asks patients to report the presence of physical and emotional symptoms over the prior week and to rate the symptom severity on a 4-point Likert scale from zero, which reflects that a symptom is not bothersome, to four, which reflects that a symptom is very bothersome. We added an item to the DSI by asking participants to report the presence or absence of “other pain” in addition to bone/joint pain, chest pain, and headache, which allowed us to identify any form of pain that may not have been captured on the original DSI.

Assessment of Treatment of Potentially Treatable Symptoms 

A priori, study investigators identified symptoms on the DSI for which pharmacologic therapy was potentially available. These “potentially treatable” symptoms included nausea, numbness or tingling, bone or joint pain, other pain, worry, nervousness, sadness, anxiety, and insomnia. Patients with one or more of these symptoms were asked if they received a medication to treat the symptom(s), and if so, whether it was prescribed by their nephrologist, primary care provider, or another provider.

To augment our assessment of the use of pharmacologic therapy as reported by the patient and to minimize the impact of patient recall bias, we conducted chart reviews of each patient's dialysis record to abstract prescribed medications. Medications were recorded if they belonged to one of four groups: analgesics, antidepressants or anxiolytics, antiemetics, and sleep aids. Oxycodone, fentanyl, tramadol, gabapentin, acetaminophen, cyclobenzaprine, and nonsteroidal anti-inflammatory medications were recorded as analgesics. Antiemetics included promethazine, ondansetron, prochlorperazine, and metoclopramide. Benzodiazepines, selective serotonin reuptake inhibitors, serotonin-norepinephrine reuptake inhibitors, tricyclic antidepressants, quetiapine, and bupropion were recorded as antidepressants or anxiolytics. Zolpidem, trazodone, mirtazapine, and diphenhydramine were recorded as sleep aids.

Statistical Analyses 

Our analyses were principally descriptive. We report the prevalence and severity of symptoms using proportions, means, or medians, as appropriate. Symptom treatment is reported as the proportion of patients who reported a symptom for whom pharmacologic therapy was prescribed. Differences in symptom treatment (nephrologist vs. primary care provider) were assessed using Fisher's exact test and the Chi-squared statistics as appropriate. A P value less than 0.05 was considered significant. All analyses were conducted using NCSS^® (Number Cruncher Statistical System; J. Hintze, Kaysville, UT).

Back to Article Outline

Results 

Patient Characteristics 

Eighty-seven patients were randomly approached; 17 refused to participate, one did not speak English, and seven were excluded for cognitive impairment. This resulted in a study population of 62 patients who completed the DSI. More than half (56%) of the participants were African American, 55% were men, and 39% were diabetic. Participants had been on dialysis at a mean of four years (range 0–24.5 years) and experienced an average of two hospitalizations in the previous year (Table 1).

Table 1. Participant Demographics (n=62)

SD=standard deviation; HD=hemodialysis.

Values are percent (number) unless otherwise indicated.

Prevalence and Treatment of Symptoms 

Table 2 depicts the most prevalent and severe symptoms and their treatment. Most patients reported receiving no treatment for their symptoms. Forty-five percent of patients with bone or joint pain reported a medication prescribed for the symptom, whereas just less than 25% were treated for sadness and worrying. Half (51%) of the patients who reported either bone or joint pain or other pain had a medication documented in the chart for that symptom, compared with 42% for nausea and 23% for sleep disturbance. Interestingly, for the symptoms of trouble falling asleep and nausea, more patients report having a medication prescribed for the symptom than was recorded in the patient's medical record. The opposite is true for bone or joint pain—fewer patients report a medication prescribed for pain than was recorded in the chart.

Table 2. Symptom Prevalence, Mean Severity, and Treatment

— Refers to data not gathered by this study.

Table 3 shows the prescription patterns for treatment of patient-reported symptoms, as recorded by chart review. Opioids and gabapentin were most frequently prescribed for pain (32% for both). Acetaminophen, tramadol, and nonsteroidal anti-inflammatory medications also were prescribed. Benzodiazepines and selective serotonin reuptake inhibitors or serotonin-norepinephrine reuptake inhibitors were most frequently prescribed for depression and anxiety (40% for both). Zolpidem was most frequently prescribed for insomnia (57%). Promethazine was most frequently prescribed for nausea (75%). Primary care physicians were more likely than nephrologists to prescribe medications for all symptoms except nausea and numbness or tingling in the feet, but the only statistically significant differences in prescribing for symptoms were for worry and nervousness (Table 4). The most prevalent symptoms reported by patients were more likely to be treated by primary care physicians. However, few significant differences were seen, in part because of the small number of patients for whom medications were prescribed (Table 4). Patient demographic and clinical characteristics had no impact on the treatment of symptoms (data not shown).

Table 3. Medication Use^a

Type of Medication	Prevalenceb %
Pain
Opioids	32
Gabapentin	32
Acetaminophen	26
Tramadol	11
Nonsteroidal anti-inflammatories	16
Mood symptoms
Benzodiazapines	40
Selective serotonin reuptake inhibitors/serotonin-norepinephrine reuptake inhibitors	40
Tricyclics	20
Bupropion	30
Sleep
Zolpidem	57
Trazodone	43
Mirtazapine	14
Diphenhydramine	14
Nausea
Promethazine	75
Metoclopramide	25

Table 4. Symptom Treatment by Provider

PCP=primary care physician.

Back to Article Outline

Discussion 

We found that symptoms that are potentially amenable to pharmacologic therapy in hemodialysis patients are prevalent but commonly undertreated. Primary care physicians seem to be more likely than nephrologists to treat dialysis patients' symptoms, particularly emotional symptoms.

In this study, we attempted to focus on potentially treatable symptoms. As pain is reported by 50% or more of dialysis patients and is commonly multifactorial, we attempted to assess the presence of and provision of therapy for neuropathic, somatic, and visceral pain. Although numbness and tingling are difficult to manage medically, our goal was to use these symptoms as a marker of neuropathic pain, which may be amenable to treatment.

Pruritus has been shown to negatively impact dialysis patient's quality of life and is associated with a 17% higher mortality risk.¹⁷ However, we did not include this symptom in our investigation for several reasons. One, there are a multitude of pharmacologic treatments--gabapentin, tacrolimus, and the k-opiod receptor antagonist nalfurafine--that can be used to treat this symptom but are also used for other indications. Two, there is a lack of consensus on the optimal treatment of pruritus. Three, the negative impact of pruritus on quality of life and mortality may be mediated through disturbances in sleep.¹⁷ Nonetheless, given the frequency with which patients on dialysis experience pruritus and the implications of this symptom, future studies should examine whether providers are implementing available treatment for this symptom.

Our findings on undertreatment of pain in the hemodialysis population are consistent with prior estimates.⁷, ¹¹, ¹² Davison reported that approximately 50% of hemodialysis patients with chronic pain received treatment, and 75% of those patients reported ineffective management.⁷ We found very similar results. Similarly, Watnick et al. described that only 16% of hemodialysis patients with depression were receiving psychiatric treatment (either counseling or pharmacologic therapy).¹³ Although we did not directly assess the presence of depression in our patients, our estimate of the use of antidepressants in patients who report “sadness” is similar to that of Watnick et al.¹³ The observation that depression is undertreated in hemodialysis patients is increasingly important, with recent findings demonstrating that depression is associated with increased mortality in this population.¹⁸, ¹⁹, ²⁰

We expand on prior findings by investigating the treatment of other symptoms, including nausea, sleep disturbance, and anxiety-related symptoms. Dialysis patients frequently experience nausea because of a variety of factors, including decreased gastrointestinal motility and medication effects. Chronic nausea may contribute to anorexia and poor nutrition, which have been linked with increased mortality, making treatment critical.²¹, ²², ²³ Our findings suggest that despite the high prevalence of nausea in hemodialysis patients, treatment is infrequently implemented.

Sleep disturbance is also common in dialysis patients²⁴, ²⁵ and is also undertreated. Our study used the categorization of “sleep aids” to describe medications with sedative effects intended to improve the sleep of patients. Thus, these medications represent various pharmacologic classes from antidepressants to anxiolytics to antihistamines. Cautious utilization of sleep aids for dialysis patients with sleep disturbance could potentially decrease daytime fatigue and serve to improve overall quality of life.

Although our small study population and low rates of pharmacologic treatment of symptoms limit the ability to detect significant differences among providers caring for hemodialysis patients, we found that primary care physicians are more likely to treat symptoms than nephrologists. Additional study with larger numbers of patients who are receiving medications for symptoms is needed to determine if primary care providers are more likely to prescribe medications for symptomatic relief. This issue has significant impact on the management of dialysis patients. The fact that many hemodialysis patients consider their nephrologist to be their primary care physician²⁶, ²⁷ may explain the undertreatment of certain symptoms. Previous studies have demonstrated that nephrologists commonly underrecognize patients' symptom burden.¹⁰ Without recognition of symptoms, dialysis providers are unlikely to implement treatment. As nephrologists increasingly focus on issues such as dialysis access and the management of bone metabolism and anemia, less time may be devoted to symptom management. Primary care physicians who partner with nephrologists in providing care for dialysis patients may find it easier to focus on symptoms, as they are not required to concentrate on dialysis-related parameters. Primary care physicians also may be more comfortable and familiar with strategies to manage certain symptoms. It is important to note that we did not distinguish between nonphysician dialysis care providers (e.g., nurse practitioners) and nephrologists in our survey, because advanced practice nurses and physician assistants may provide much of the primary care for hemodialysis patients.²⁸ Future studies will need to differentiate among renal providers to determine who may be more likely to provide treatment for symptoms. Nonetheless, our findings highlight the importance of a multidisciplinary approach to symptom management in hemodialysis patients and illustrate the opportunity for collaboration among providers to treat bothersome symptoms in this patient population.

Interestingly, chart documentation of treatment did not consistently match patient-reported treatment. This may reflect that the severity of a symptom affects patients' perceptions of prescribing practices. Patients may be less likely to report an ineffective medication prescribed for a symptom than an effective medication. A potential explanation of the finding that pain medications are more likely to be documented in the medical record than being reported by patients may be physician documentation practices. Given that opioids comprise one-third of pain medications prescribed for patients in this population, physicians may be more likely to carefully document scheduled medications rather than medications that are perceived as less potent.

There are important limitations to our study. First, we focused on a small population at a single center, which limits the generalizability of our findings. Future studies will need to validate our findings in a larger patient population and examine whether demographic and/or clinical factors impact the provision of symptom-alleviating therapy. Second, some of the symptoms we believe to be potentially treatable may not be viewed as treatable by providers. Third, estimation of the frequency of pharmacologic therapy for symptoms is inherently difficult. Despite concerns that recall bias would cause patients to minimize the pharmacologic treatment of symptoms, we found that patients overreported medications prescribed for most of the symptoms when compared with the rates of documented prescriptions. Also, lower rates of documentation of pharmacologic treatment of symptoms found in chart review compared with patient report may reflect the fact that medications are less likely to be documented in the dialysis medical record if prescribed by a primary care physician. We attempted to address these concerns by conducting both a chart review to identify medications that were prescribed for patients and using patient report. However, it is possible that we were unable to identify all medications prescribed to patients for bothersome symptoms using the methods we used. Nonetheless, by using a comprehensive approach of patient recall and chart review, our study was more likely to capture symptom treatment than studies that have used a single method. Fourth, we cannot be certain that medications were used for the symptom we investigated. Specifically, medications that we labeled as “sleep aids” may have been prescribed for reasons other than sleep disturbance, such as anxiety or depression. Fifth, as this study records prescribing practices based on patient recall, there are inherent biases in patient memory of which physician prescribed a medication. Also, a medication may be initially prescribed by a nephrologist and subsequent refills written by a primary care physician. Thus, it is difficult to draw concrete conclusions from these data on which provider is responsible for prescribing treatment. Lastly, we cannot be sure that the absence of treatment did not relate to patients' lack of interest in treatment or that pharmacologic therapy had not previously been implemented and either failed or led to untoward side effects, leading to their discontinuation.

In conclusion, this study builds on the foundation of previous research by elucidating opportunities to improve the management of certain symptoms in hemodialysis patients. Hemodialysis patients suffer a large symptom burden, and potentially treatable symptoms appear to be undertreated. Improving the implementation of symptom-alleviating therapies and making the approach to symptom management in hemodialysis patients a multidisciplinary process that involves both renal and primary providers may result in improvements in patients' quality of life.

Back to Article Outline

Acknowledgments 

The authors thank Robert M. Arnold for contributing to the editing and revision process of this article.

Back to Article Outline

References 

1. US Renal Data System. USRDS 2007 annual data report. Bethesda, MD: The National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2007;
   • View In Article
2. Evans RW, Manninen DL, Garrison LP, et al. The quality of life in patients with end-stage renal disease. N Engl J Med. 1985;312:553–559
   • View In Article
   • MEDLINE
   • CrossRef
3. Kimmel PL, Peterson RA, Weihs KL, et al. Aspects of quality of life in hemodialysis patients. J Am Soc Nephrol. 1995;6:1418–1426
   • View In Article
4. Unruh ML, Weisbord SD, Kimmel PL. Health related quality of life in nephrology research and clinical practice. Semin Dial. 2005;18:82–90
   • View In Article
   • MEDLINE
   • CrossRef
5. Weisbord SD, Carmody SS, Bruns FJ, et al. Symptom burden, quality of life, advance care planning and the potential value of palliative care in severely ill hemodialysis patients. Nephrol Dial Transplant. 2003;18:1345–1352
   • View In Article
   • MEDLINE
   • CrossRef
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;42:713–721
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (897 KB)
   • CrossRef
7. Davison SN. Pain in hemodialysis patients: prevalence, cause, severity and management. Am J Kidney Dis. 2003;42:1239–1247
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (64 KB)
   • CrossRef
8. 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;16:2487–2494
   • View In Article
9. Murtagh FE, Addington-Hall J, Higginson IJ. The prevalence of symptoms in end-stage renal disease: a systematic review. Adv Chronic Kidney Dis. 2007;1:82–89
   • View In Article
10. Weisbord SD, Fried LF, Mor MK, et al. Renal provider recognition of symptoms in patients on maintenance hemodialysis. Clin J Am Soc Nephrol. 2007;2:960–967
    • View In Article
    • CrossRef
11. Bailie GR, Mason NA, Bragg-Gresham JL, Gillespie BW, Young EW. Analgesic prescription patterns among hemodialysis patients in the DOPPS: potential for underprescription. Kidney Int. 2004;65:2419–2425
    • View In Article
    • MEDLINE
    • CrossRef
12. Carreon M, Fried LF, Palevsky PM, et al. Clinical correlates and treatment of bone/joint pain and difficulty with sexual arousal in patients on maintenance hemodialysis. Hemodial Int. 2008;12:268–274
    • View In Article
    • CrossRef
13. Watnick S, Kirwin P, Mahnensmith R, Concato J. The prevalence and treatment of depression among patients starting dialysis. Am J Kidney Dis. 2003;41:105–110
    • View In Article
    • Abstract
    • Full-Text PDF (51 KB)
    • CrossRef
14. Davison SN, Jhangri GS. The impact of chronic pain on depression, sleep, and the desire to withdraw dialysis in hemodialysis patients. J Pain Symptom Manage. 2005;30:465–473
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (110 KB)
    • CrossRef
15. Royall DR, Cordes JA, Polk M. CLOX: an executive drawing task. J Neurol Neurosurg Psychiatry. 1998;64:588–594
    • View In Article
    • MEDLINE
    • CrossRef
16. 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;27:226–240
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1597 KB)
    • CrossRef
17. Pisoni RL, Wikström B, Elder SJ, et al. Pruritus in haemodialysis patients: international results from the Dialysis and Practice Patterns Study (DOPPS). Nephrol Dial Transplant. 2006;21:3495–3505
    • View In Article
    • MEDLINE
    • CrossRef
18. Peterson RA, Kimmel PL, Sacks CR, et al. Depression, perception of illness and mortality in patients with end-stage renal disease. Int J Psychiatry Med. 1991;21:343–354
    • View In Article
    • MEDLINE
    • CrossRef
19. Hedayati SS, Bosworth HB, Briley LP, et al. Death or hospitalization of patients on chronic hemodialysis is associated with a physician-based diagnosis of depression. Kidney Int. 2008;74:930–936
    • View In Article
    • CrossRef
20. Kimmel PL, Peterson RA, Weihs KL, et al. Multiple measurements of depression predict mortality in a longitudinal study of chronic hemodialysis outpatients. Kidney Int. 2000;57:2093–2098
    • View In Article
    • MEDLINE
    • CrossRef
21. Qureshi AR, Alvestrand JC, Divino-Filho JC, et al. Inflammation, malnutrition, and cardiac disease as predictors of mortality in hemodialysis patients. J Am Soc Nephrol. 2002;13:S28–S36
    • View In Article
22. Iseki K, Kawazoe N, Fukiyama K. Serum albumin is a strong predictor of death in chronic dialysis patients. Kidney Int. 1993;44:115–119
    • View In Article
    • MEDLINE
    • CrossRef
23. Acchiardo SR, Moore LW, Latour PA. Malnutrition as the main factor in morbidity and mortality of hemodialysis patients. Kidney Int. 1983;16:S199–S203
    • View In Article
24. Kimmel P. Sleep disorders in chronic renal disease. J Nephrol. 1989;1:59–65
    • View In Article
25. Parker K. Sleep disturbances in dialysis patients. Sleep Med Rev. 2003;7:131–143
    • View In Article
    • Abstract
    • Full-Text PDF (199 KB)
    • CrossRef
26. Shah N, Dahl NV, Kapoian T, Sherman RA, Walker JA. The nephrologist as a primary care provider for the hemodialysis patient. Int Urol Nephrol. 2005;37:113–117
    • View In Article
    • MEDLINE
    • CrossRef
27. Bender FH, Holley JL. Most nephrologists are primary care providers for chronic dialysis patients: results of a national survey. Am J Kidney Dis. 1996;28:67–71
    • View In Article
    • Abstract
    • Full-Text PDF (470 KB)
    • CrossRef
28. Bolton WK. Nephrology nurse practitioners in a collaborative care model. Am J Kidney Dis. 1998;31:786–793
    • View In Article
    • Abstract
    • Full-Text PDF (149 KB)
    • CrossRef