Comparing Symptom Burden in Patients with Metastatic and Nonmetastatic Cancer

Abstract

Background: Symptoms of patients with advanced cancer have been extensively studied, but less attention has been paid to the symptomatology of patients with localized disease. We compared the symptom burden of outpatients with cancer, either metastatic or localized disease, seen in a palliative care comanagement clinic. Methods: We conducted a retrospective, cross-sectional study assessing patient symptoms with multiple patient surveys, including a modified Edmonton Symptom Assessment Survey (to assess pain, fatigue, anxiety, depression, and nausea), along with a spiritual well-being screen and questions about relationships and quality of life. Consecutive patients at a comprehensive cancer center completed the surveys prior to their first visit to an outpatient palliative care clinic providing palliative care concurrently with the patients' oncologic treatments. Patients were referred by their oncologist, radiation oncologist, oncologic surgeon, or primary care physician without respect to their stage or prognosis. Statistical analyses used were χ², Fisher's exact test, Student's t test, and multivariate linear regression. Significance was taken as p<0.05. Results: Two hundred and five patients were included in this study. One hundred and twenty-seven patients (62%) had metastatic cancer and 78 (38%) had nonmetastatic cancer. The characteristics of the two study groups were similar with regard to age, sex, ethnicity, language, religion, partnership status, and insurance status. Patients with metastatic disease and localized disease had similar mean symptom burden and mean individual symptom intensity for all symptoms evaluated except for nausea, which was worse in patients with metastatic disease (3.4 versus 2.5, p=0.0028). Nausea was also worse in younger patients. Conclusion: Among patients referred to a cancer center outpatient palliative care clinic, symptom burden is similar in patients with localized and metastatic disease. Even patients with relatively good prognoses may nevertheless have significant symptom burdens in physical, emotional, and existential domains. The palliative care needs of all patients must be assessed, regardless of stage or prognosis.

Introduction

One of the aims of palliative care is to improve quality of life in patients with cancer or other chronic disease.¹ Outpatient palliative care clinics are filling an important niche in the care of cancer patients, with a reported 22–59% of cancer centers surveyed in 2009 having outpatient palliative care services.² These clinics provide many services, including palliation of symptoms, continuity of care, and greater disease understanding for patients who have advanced disease.³

In assessing symptoms in patients with cancer, cancer stage is often used in the evaluation of symptoms. In 1994, Portenoy reported a study of 246 cancer patients. Those with metastatic cancer had higher symptom distress than those with early disease.⁴ However, it is not clear if outpatients with metastatic disease are more symptomatic than those with localized disease, or which symptoms are more burdensome to them. There is a growing literature on the assessment of symptom burden, spirituality, and quality of life in patients with advanced cancer, but there is little reported on patients with localized cancer.^5–9 One study including patients with localized disease was a large cross-sectional cohort study of symptoms in outpatients in a cancer registry.¹⁰ This study by Barbera reported the prevalence of a number of symptoms (pain, dyspnea, depressed mood, and nausea) similar to those found in a systematic review of palliative studies, but did not examine symptomatology based on stage.^8,10

Symptoms can arise from the disease process itself or from treatments of the disease. Given that some patients with localized disease undergo treatments with significant morbidity, and may endure the psychological distress of living with a diagnosis of cancer, we hypothesized that there are likely many patients with nonmetastatic cancer who nevertheless are burdened with clinically important symptoms, spiritual distress, and poor quality of life. By investigating symptomatology in patients with localized disease, providers may be better able to address the needs of a population subset who might be thought a priori to have a less significant symptom burden.

Methods

Study design

We conducted a retrospective cross-sectional analysis to compare symptom scores in patients with metastatic versus nonmetastatic cancer attending the Symptom Management Service (SMS) clinic at the University of California, San Francisco (UCSF) Helen Diller Family Comprehensive Cancer Center. The SMS clinic is a palliative care comanagement clinic at an academic cancer center with services provided by the following: attending physicians, palliative care fellows, nurse practitioners, nurses, chaplains, social workers, psychologists, nutritionists, and pharmacists. Referring providers to the SMS clinic include oncologists, radiation oncologists, oncologic surgeons, and primary care physicians. Study subjects were ambulatory adults who had their first visit in the SMS clinic between October 24, 2008 and December 31, 2009. Symptom score data were collected with an eight item self-administered survey in order to identify patients with heavy symptom burdens without respect to cancer diagnosis or stage. The survey is used by the SMS in routine clinical care and is based on validated surveys widely used to assess symptom burden in patients with cancer. Five of the items were from the previously validated Edmonton Symptom Assessment Survey (ESAS) and asked about pain, fatigue, anxiety, depression, and nausea.¹¹ A sixth item, for exploratory purposes, asked about relationship problems. These six questions had a response scale from 0 to 10 with 0=having no symptoms or relationship problems and 10=having the worst possible symptoms or relationship problems. The remaining two questions of the survey used a five point Likert scale and queried overall quality of life and spiritual well-being. The quality of life question was taken from the validated QUAL-E^® survey¹² and the spiritual well-being question was a one item validated screen, “Are you at peace?”¹³

Demographic and clinical data were obtained from the SMS clinic database (sex, age, and stage of cancer), and the UCSF computerized medical record system (insurance status, ethnicity, partnership status, and language).

Those excluded were patients who did not complete a survey, had no medical record number, or did not have any demographic data in the UCSF database.

The study protocol was approved by the UCSF institutional review board.

Measures

The symptom burden score was a cumulative score of the six questions with the 0–10 scale, with a total possible score of 60 points (range 0–60). Higher scores indicated increased reported symptomatology.

For analysis of age, three subgroups were used: ≤54 years, 55–64 years, and ≥65 years. These subgroups were created to divide the study population into tertiles.

Language was categorized into four categories: English, Spanish, Chinese, and other.

Statistical analysis

The primary analysis compared the total symptom burden score in patients with metastatic cancer with those with nonmetastatic cancer. A secondary analysis compared individual symptom scores between the two study groups, as well as symptom scores in different population subsets (prostate cancer and breast cancer) and as predicted by sex, age, ethnicity, and language.

Analyses were performed using SAS software, version 9.1. For the analysis of demographic data, χ² and Fisher's exact tests were performed to assess for any differences between the two study groups for categorical variables. Student's t test was performed for a continuous variable (age). Bivariate analyses of symptom burden scores and individual symptom scores as related to predictor variables of status of cancer (metastatic versus nonmetastatic cancer), type of cancer (prostate and breast), gender, age, ethnicity, and language were performed using Student's t test. Multivariate linear regressions were then performed for all these predictor variables to see which were significant. Significance was taken as p≤0.05.

Results

A total of 205 patients were included in the study. Two hundred and twenty-nine patients were seen between October 24, 2008 and December 31, 2009. Two patients were excluded for having a non-cancer diagnosis, one for lack of computerized medical record demographic data, and 21 for not completing the survey at their first visit. One hundred and twenty-seven patients (62%) had metastatic cancer and 78 (38%) had nonmetastatic cancer. The characteristics of the two study groups were similar with regard to age, gender, ethnicity, language, partnership status, and insurance status.

The average age of patients with metastatic cancer was similar to that of patients with nonmetastatic cancer (58.9 versus 56.9, respectively, p=0.30). Forty-five percent of the overall study population was male, with similar representation of men and women in each of the two study groups (χ²=0.64). The majority (66%) of the study population was white, non-Hispanic, with similar representation of ethnic groups across the two study groups (Fisher's exact test=0.21). Eighty-nine percent of the study population was English speaking, with no significant difference in language between the two groups (Fisher's exact test=0.70). Fifty-three percent of the study population was married or in a partnership. Sixty-four percent of the population had some coverage by Medicare, MediCal (Medicaid), or both.

The cancer types most represented in this study population are breast and prostate (Table 1).

Table 1.

Study Cancer Types ^a

	All patients (n=205)	Metastatic Cancer (n=78)	Nonmetastatic Cancer (n=127)	p-value
Type of cancer,^b n (%)
Breast	44 (22)	22 (11)	22 (11)	0.011 (χ)
Prostate	35 (17)	21 (10)	14 (7)
GI	24 (12)	16 (8)	8 (4)
GU	16 (8)	10 (5)	6 (3)
Gynecologic	18 (9)	16 (8)	2 (1)
Head and neck	18 (9)	9 (4)	13 (6)
Lung	14 (7)	13 (6)	1 (0)
Other	36 (18)	20 (10)	12 (6)

Patients with leukemia included in group with metastatic cancer. Percentages might not add up to 100% because of rounding.

Gastrointestinal (GI) cancer: liver, anal, rectal, gastric, esophageal, and colon. Genitourinary (GU): bladder, renal cell, other GU (excluding prostate). Other: melanoma, leukemia, lymphoma, thyroid, sarcoma, and carcinoid.

Mean symptom burden scores in patients with metastatic cancer and nonmetastatic cancer were similar (25.2 versus 22.2, p=0.099). When individual symptom scores were analyzed, there were no differences in the report of relationship problems, spiritual well-being, or quality of life between the two groups. Patients with metastatic cancer had greater fatigue than did patients with nonmetastatic cancer (mean 6.1 versus 5.4, p=0.043) (Table 2). They also had worse nausea (3.4 versus 2.5, p=0.028). However, on multiple linear regression, only nausea remained statistically significant.

Table 2.

Individual Symptom Scores in Patients with Metastatic vs. Nonmetastatic Cancer

Symptom, mean	Metastatic cancer	Nonmetastatic Cancer	p-value
Pain	4.7	4.2	0.25
Fatigue	6.1	5.4	0.043
Anxiety	4.5	4.1	0.35
Depression	4.2	3.8	0.31
Relationship problems	2.1	2.2	0.83
Nausea	3.4	2.5	0.028
Peace	2.7	3.0	0.10
Quality of Life	3.0	3.2	0.12

Symptoms for patients with metastatic versus localized cancer were analyzed among patients with the two most common cancer diagnoses: prostate and breast cancer. Patients with metastatic prostate cancer had similar symptoms as those with localized prostate cancer, except that those with metastatic prostate cancer had greater fatigue (5.5 versus 3.4, p=0.022). Patients with nonmetastatic breast cancer had a slightly better quality of life than did patients with metastatic breast cancer (3.5 versus 2.8, p=0.042).

In bivariate analysis of age as a predictor of symptom score, lower age was associated with worse nausea. In a stratified comparison of three different age groups using Student's t test, patients ≤54 years of age had worse nausea than did patients ≥65 years of age (3.8 versus 2.5, p=0.012). Age was also significant in the multiple linear regression analysis in which age was run as a continuous variable (p=0.027). Younger patients (≤54 years old) also had greater severity of pain and anxiety than did older patients (>54 years old), but this was not statistically significant.

In bivariate analysis, looking at sex as a predictor of symptom intensity, pain, and fatigue appeared to be statistically significant, with women reporting greater symptom burden than men, although only pain was statistically significant in multiple linear regression (Table 3).

Table 3.

Individual Symptom Scores in Men versus Women

Symptom, mean	Men	Women	p-value
Pain	3.5	5.3	<0.0001
Fatigue	5.3	6.3	0.0032
Anxiety	4.0	4.7	0.053
Depression	3.9	4.3	0.35
Relationship problems	1.9	2.4	0.26
Nausea	2.7	3.4	0.080
Peace	2.8	2.8	0.81
Quality of life	2.9	3.2	0.075

Nonwhite patients reported worse pain than did white patients (5.5 versus 4.1, p=0.0035). This finding from the bivariate analysis was statistically significant in early regression models, but became less significant after age was accounted for. Quality of life was reported to be better by English-speaking patients than by non-English speaking patients (3.1 versus 2.3, p=0.0009).

Discussion

This study examined symptom burden as well as individual symptom intensity among patients with metastatic versus those with nonmetastatic cancer attending an outpatient palliative care clinic. In this palliative care clinic's population, symptom burden as well as symptom intensity (physical symptoms, spiritual well-being, and quality of life) were similar between the two groups. Nausea was the only symptom that was worse in patients with metastatic disease.

The mean scores found in this study for fatigue, depression, pain, and anxiety were similar to that of patients previously described in studies of patients with advanced cancer.^5,14–19 Nausea was slightly worse in our study population than in most of those other studies, but similar to that reported in two prior studies.^19,20 In a study of outpatients with metastatic cancer referred for palliative radiotherapy, mean nausea level on the ESAS survey was 1.41, versus 3.4 for patients with metastatic disease in our population.¹⁵ Studies of ambulatory palliative care patients found that the highest median ESAS score was for fatigue; fatigue had the highest mean ESAS score in our study as well. ^10,21

Multiple secondary analyses were performed to see what other patient characteristics were associated with increased symptom intensity. Age has previously been studied in relation to symptoms, although the results have been inconsistent, with some studies showing older patients exhibiting or complaining less of pain and others showing no difference in pain in relation to age.^10,22,23 Barbera reported worse moderate-to-severe pain, anxiety, depression, and nausea scores in patients ages 30–59.¹⁰ Although our age stratifications were coarser than those of the study by Barbera because of study sample size, our study similarly found worse nausea in younger patients (≤54 years of age) than in older patients (≥65 years of age). In our study, patients ≤54 years of age also had greater pain and anxiety (but not depression) severity, although this was not statistically significant.

Barbera also reported worse symptoms in women, which is consistent with some prior studies.¹⁰ Zimmerman also reported that women in their study had worse symptom burden and worse depression, anxiety, loss of appetite, drowsiness, and fatigue.²⁰ The data are inconclusive on whether women experience worse pain than men. In our study, there was a statistically significant difference with women having worse pain than men. The study by Barbera reported women also having worse pain, although it is not clear if this was statistically significant, and Zimmerman did not notice a significant difference.^10,20

Regarding ethnicity, there was a weak association between ethnicity and pain in this study. Nonwhite patients had worse pain than white patients in bivariate analysis, a relationship that held up in early regression models. There is conflicting information in the literature on symptoms as related to ethnicity, and previously mostly studied in specific populations. Fu reported in a study of women with breast cancer that Hispanic women were more likely to report symptoms than white women.²⁴ Yoon reported that nonwhite women with breast cancer were less likely than white women to report severe symptoms.²⁵ Rabow studied a more diverse group of patients seen for outpatient palliative care consultations and included patients with advanced heart failure, chronic obstructive pulmonary disease, and cancer, and found that nonwhite patients reported greater pain than did white patients.²⁶

Study Limitations

Although our findings are largely consistent with prior literature, there are a number of potential limitations. The SMS study includes only patients who were referred to the clinic, generating the potential for referral bias. Clearly, many patients with low symptom burdens are not being referred to the clinic and not included in our analysis. However, our data do suggest that there is a subpopulation of patients with localized cancer who nonetheless have significant symptoms and palliative care needs. Our results may be generalizable, in particular to other cancer center palliative care clinics, and are a key reminder that patients not at the end of life may still suffer in physical, emotional, and existential domains.

The study was not controlled for treatment being received by patients. The morbidity of various oncologic treatments, such as emetogenic chemotherapy, might explain some of the findings demonstrated. Other study limitations include sample size and the use of some nonvalidated survey questions. The study was powered to answer the primary question of looking at the difference in symptom burden between patients with metastatic and those with nonmetastatic cancer. However, the study might not be adequately powered to detect differences between other groups of the secondary analyses (type of cancer, gender, age, ethnicity, and language). Therefore, interesting differences in certain symptom intensities based on those patient characteristics could have been missed in this study. Finally, one of the survey questions (relationship problems) was not previously validated, and the associated results will need to be assessed with validated surveys. Although the ESAS and QUAL-E are validated surveys, our survey was a modified ESAS. Our survey included only five of the nine ESAS questions and only one question from the QUAL-E.

Conclusion

Despite these potential limitations, the findings of our study suggest that clinicians should, and may already in certain settings, recognize that patients with limited disease can be as symptomatic as those with metastatic disease. In our study population, the similar symptom burden between the two groups may be the result of a nonstandardized yet internally consistent referral pattern of clinicians at one particular cancer center. However, despite the possible overrepresentation of symptomatic patients with localized cancer in this study, it does ultimately show that patients with localized cancer may have great, and potentially unaddressed, palliative care needs. Although many outpatient palliative care clinics care for patients with advanced cancer, increased awareness of the needs of patients with localized disease will, one hopes, encourage both referrals as well as expansion of outpatient palliative care services to adequately address these needs. Future studies should be performed looking at symptoms of outpatients with localized disease who are referred, compared with those who are not.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

WHO definition of palliative care. www.who.int/cancer/palliative/definition/en/ (Last accessed on April 11, 2011).

Hui

, Elsayem

, De la Cruz

, Berger

, Zhukovsky

, Palla

, Evans

, Fadul

, Palmer

, Bruera

: Availability and integration of palliative care at US cancer centers. JAMA, 2010; 303:1054–1061.

Meier

, Beresford

: Outpatient clinics are a new frontier for palliative care. J Palliat Med, 2008; 11:823–828.

Portenoy

, Thaler

, Kornblith

, Lepore

, Friedlander-Klar

, Kiyasu

, Sobel

, Coyle

, Kemeny

, Norton

, Scher

: The Memorial Symptom Assessment Scale: an instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer, 1994; 30A:1326–1336.

Yennurajalingam

, Urbauer

, Casper

, Reyes-Gibby

, Chacko

, Poulter

, Bruera

: Impact of a palliative care consultation team on cancer-related symptoms in advanced cancer patients referred to an outpatient supportive care clinic. J Pain Symptom Manage, 2011; 41:49–56.

Zimmermann

, Burman

, Swami

, Krzyzanowska

, Leighl

, Moore

, Rodin

, Tannock

: Determinants of quality of life in patients with advanced cancer. Support Care Cancer. 2011; 19:621–629.

Winkelman

, Lauderdale

, Balboni

, Phelps

, Peteet

, Block

, Kachnic

, VanderWeele

, Balboni

: The relationship of spiritual concerns to the quality of life of advanced cancer patients: preliminary findings. J Palliat Med, 2011; 14:1022–1028.

Teunissen

, Wesker

, Kruitwagen

, de Haes

, Voest

, de Graeff

: Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manage, 2007; 34:94–104.

Whitmer

, Pruemer

, Nahleh

, Jazieh

: Symptom management needs of oncology outpatients. J Palliat Med, 2006; 9:628–630.

10.

Barbera

, Seow

, Howell

, Sutradhar

, Earle

, Liu

, Stitt

, Husain

, Sussman

, Dudgeon

: Symptom burden and performance status in a population-based cohort of ambulatory cancer patients. Cancer, 2010; 116:5767–5776.

11.

Nekolaichuk

, Watanabe

, Beaumont

: The Edmonton Symptom Assessment System: a 15-year retrospective review of validation studies (1991–2006). Palliat Med, 2008; 22:111–122.

12.

Steinhauser

, Clipp

, Bosworth

, McNeilly

, Christakis

, Voils

, Tulsky

: Measuring quality of life at the end of life: Validation of the QUAL-E. Palliat Support Care, 2004, 2:3–14.

13.

Steinhauser

, Voils

, Clipp

, Bosworth

, Christakis

, Tulsky

: “Are you at peace?”: one item to probe spiritual concerns at the end of life. Arch Intern Med, 2006; 166:101–105.

14.

Cheung

, Le

, Zimmermann

: Symptom clusters in patients with advanced cancers. Support Care Cancer, 2009; 17:1223–1230.

15.

Bradley

, Davis

, Chow

. Symptom distress in patients attending an outpatient palliative radiotherapy clinic. J Pain Symptom Manage, 2005; 30:123–131.

16.

Fadul

, El Osta

, Dalal

, Poulter

, Bruera

: Comparison of symptom burden among patients referred to palliative care with hematologic malignancies versus those with solid tumors. J Palliat Med, 2008; 11:422–427.

17.

Fan

, Hadi

, Chow

: Symptom clusters in patients with advanced-stage cancer referred for palliative radiation therapy in an outpatient setting. Support Cancer Ther, 2007; 4:157–162.

18.

Bruera

, Michaud

, Vigano

, Neumann

, Watanabe

, Hanson

: Multidisciplinary symptom control clinic in a cancer center: a retrospective study. Support Care Cancer, 2001; 9:162–168.

19.

Follwell

, Burman

, Le

, Wakimoto

, Seccareccia

, Bryson

, Rodin

, Zimmermann

: Phase II study of an outpatient palliative care intervention in patients with metastatic cancer. J Clin Oncol, 2009; 27:206–213.

20.

Zimmermann

, Burman

, Follwell

, Wakimoto

, Seccareccia

, Bryson

, Le

, Rodin

: Predictors of symptom severity and response in patients with metastatic cancer. Am J Hosp Palliat Care, 2010; 27:175–181.

21.

Riechelmann

, Krzyzanowska

, O'Carroll

, Zimmermann

: Symptom and medication profiles among cancer patients attending a palliative care clinic. Support Care Cancer, 2007; 15:1407–1412.

22.

Vigano

, Bruera

, Suarez–Almazor

: Age, pain intensity, and opioid dose in patients with advanced cancer. Cancer, 1998; 83:1244–1250.

23.

Stein

, Miech

: Cancer pain in the elderly hospice patient. J Pain Symptom Manage, 1993; 8:474–482.

24.

, Crew

, Jacobson

, Greenlee

, Yu

, Campbell

, Ortiz

, Hershman

: Ethnicity and persistent symptom burden in breast cancer survivors. J Cancer Surviv, 2009; 3:241–250.

25.

Yoon

, Malin

, Tao

, Tisnado

, Adams

, Timmer

, Ganz

, Kahn

: Symptoms after breast cancer treatment: are they influenced by patient characteristics?. Breast Cancer Res Treat, 2008; 108:153–165.

26.

Rabow

, Dibble

: Ethnic differences in pain among outpatients with terminal and end-stage chronic illness. Pain Med, 2005; 6:235–241.