• Edmonton Symptom Assessment System (ESAS);
  • Palliative Performance Scale (PPS);
  • population cohort, palliative care;
  • health services research


  1. Top of page
  2. Abstract
  6. Acknowledgements


For ambulatory cancer patients, Ontario has standardized symptom and performance status assessment population-wide, using the Edmonton Symptom Assessment System (ESAS) and Palliative Performance Scale (PPS). In a broad cross-section of cancer outpatients, the authors describe the ESAS and PPS scores and their relation to patient characteristics.


This is a descriptive study using administrative healthcare data.


The cohort included 45,118 and 23,802 patients' first ESAS and PPS, respectively. Fatigue was most prevalent (75%), and nausea least prevalent (25%) in the cohort. More than half of patients reported pain or shortness of breath; about half of those reported moderate to severe scores. Seventy-eight percent had stable performance status scores. On multivariate analysis, worse ESAS outcomes were consistently seen for women, those with comorbidity, and those with shorter survivals from assessment. Lung cancer patients had the worst burden of symptoms.


This is the first study to report ESAS and PPS scores in a large, geographically based cohort with a full scope of cancer diagnoses, including patients seen earlier in the cancer trajectory (ie, treated for cure). In this ambulatory cancer population, the high prevalence of numerous symptoms parallels those reported in palliative populations and represents a target for improved clinical care. Differences in outcomes for subgroups require further investigation. This research sets the groundwork for future research on patient and provider outcomes using linked administrative healthcare data. Cancer 2010. © 2010 American Cancer Society.

Optimal symptom management is required to minimize suffering and maximize quality of life for patients with cancer throughout the course of their disease,1 not just at the end of life. Similarly, anticipation and management of transitions between the curative, palliative, and end-of-life phases of care are important to minimize distress, optimize care, and ensure adequate supports are available.2 Regrettably, reports suggest adequate symptom control, regardless of cancer phase, is not achieved for persons receiving cancer care.1, 3-5 The 2002 National Institute of Health's State-of-the-Science report on cancer symptom management on pain, fatigue, and depression1, 5 estimated cancer-related pain, depression, and fatigue to range from 14% to 100%, 1% to 42%, and 4% to 91%, respectively, including hundreds of studies of all cancer types and across the cancer continuum. Major limitations noted in the report, however, were the lack of large, population-based studies to provide more accurate estimates of the prevalence of cancer symptoms, the diverse definitions and tools used to measure cancer symptoms, the lack of standardized symptom-assessment processes, and the gap in knowledge of how patient characteristics affect specific cancer subgroups and symptoms.

Ontario, Canada's cancer system is uniquely positioned to address these research gaps. Since 2007, the provincial cancer authority in Ontario has systematically collected symptom and performance status scores in cancer outpatients by implementing 2 standardized tools in every cancer center: the Edmonton Symptom Assessment System (ESAS) and the Palliative Performance Scale (PPS), respectively. The ESAS6 is a well known and validated tool to screen for the presence and severity of symptoms. The identification of symptoms with this screening tool is meant to prompt a further detailed assessment, with possible intervention depending on the findings. The PPS7 is a well known and validated tool for assessing the performance status of a patient. It is known to be predictive of survival8 and, therefore, provides a measure to help in planning care. In essence, Ontario has implemented population-wide standardized symptom and performance status assessments for cancer outpatients in the cancer center or at home.

Therefore, Ontario possesses a unique and innovative data source that describes patient-reported symptoms and provider-reported performance status on a large scale with an unparalleled number of observations. This provides the opportunity to go beyond the existing literature to study symptoms and performance status at a population level for patients with a full scope of cancer diagnoses, broad range of ages, from ambulatory and/or home settings, and including patients seen much earlier in the course of their illness (ie, those treated for cure); this ambulatory cancer population is representative of what most oncologists encounter in daily clinical practice in a cancer center. Therefore, this study's purpose is to address the research gaps in cancer symptom management across the care continuum and describe the ESAS and PPS scores in a population-based cross-section of cancer patients and the relation of these scores to sociodemographic and disease factors.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Study Design

This is a cross-sectional study describing symptom scores and performance status outcomes for a population-based cohort of ambulatory cancer patients between 2007 and 2009. The study uses administrative healthcare data from across the province of Ontario, Canada. Data linkage was completed by using a unique encrypted patient identifier. The study was approved by the ethics committee of the Sunnybrook Health Sciences Centre and followed data confidentiality and privacy guidelines of the Institute for Clinical Evaluative Sciences.

Study Population

The cohort includes all patients at the time of their first assessment with ESAS or PPS. Patients were excluded when they had an invalid or missing provincial health insurance number (which precludes linkage) or when they did not link to the Ontario Cancer Registry. The assessment date had to occur between the date of diagnosis and date of death or last follow-up. If patients had more than 1 assessment on the same date, the complete assessment was chosen. If both were complete, the one with the higher score was chosen.

Data Sources and Collection

The main source of data for this paper is the Symptom Management Reporting Database held by Cancer Care Ontario, which contains information on symptom screening using the ESAS and performance status assessment using the PPS. The province-wide initiative to standardize symptom and performance status assessments was modeled on a successful pilot project,9, 10 which suggested that routine use of these tools decreased emergency department visits, hospital admissions, and in-hospital deaths.10 Details of the nature of the provincial initiative have been described elsewhere.9, 10 Briefly, the initiative's goal is to improve symptom management and collaborative palliative care planning through earlier identification, documentation, and communication of patients' symptoms and performance status. In 2008, about 45% of the lung cancer population and 15% of the remaining cancer patients in Ontario were screened,11 and patient capture is increasing.

The standardized assessments of ESAS and PPS are completed at each patient's cancer center or homecare visit. In the most common case, when patients visit a clinic at the cancer center, they have the option to electronically complete the ESAS via a web-based tool that allows patients to enter their own scores at a touch-screen kiosk or on paper, in which case the data are then entered into an electronic database. A printed summary of the symptom scores, including those from previous visits, is given to the patient for discussion with the primary clinical team on any potentially problematic symptoms. The clinician present during the visit will record the PPS score, which is also subsequently recorded electronically; all ESAS and PPS scores are reported to the Symptom Management Reporting Database.

As well, the study used other administrative databases: the Ontario Cancer Registry, which is a comprehensive population-based cancer registry created to capture all incident cases of cancer in the province12, 13; the Registered Persons Database, which contains sociodemographic information on all residents of Ontario who are eligible for the universal government funded healthcare plan14; and the Canadian Institute for Health Information's Discharge Abstract Database, which lists diagnostic and procedure codes from all inpatient and outpatient hospital admissions.15

Patients eligible for screening included those living in all regions of the province, with any cancer diagnoses, of any age, from ambulatory and/or home settings, and any treatment intent. Assessments happened on an opportunistic basis dependant on how each cancer center or home-care program implemented the standardized tools. Ideally, all patients were screened at every cancer clinic or home visit; however, those who were included were not systematically screened at regular intervals. The initiative initially focused on all lung cancer and patients seen in palliative care clinics. Participating centers were free to expand their target population as they wished.

Outcome Definitions

The ESAS6 is a widely used and validated tool to screen for the severity of symptoms. The 9 symptoms assessed on a scale of 0-10 (10 = worst) are anxiety, appetite, depression, drowsiness, nausea, pain, shortness of breath, tiredness, and well being. Its validity and reliability are well described.16-18 Previous research has categorized the severity of ESAS scores as none (0), mild (1-3), moderate (4-6), and severe (7-10).19 These scores were reported by the patient at the time of their visit to a cancer center or at the time of a visit from a home-care nurse.

The PPS7 is a validated tool for assessing the performance status of a patient, is known to be predictive of survival,8 and is considered a valid tool for assessing prognosis in a palliative population.20 It is 1 of 11 possible scores (0, 10, 20, ..., 100%, 100%=best). It describes a patient's level of ambulation, activity level, evidence of disease, ability to perform self-care, intake, and level of consciousness. The provincial initiative classified patients with PPS scores of 70% to 100% as stable, 40% to 60% as transitional, and 10% to 30% as end-of-life.9, 21 PPS was assessed by a physician or nurse at the time of a visit to the cancer center or at the time of a visit from a home-care nurse.

Covariate Definitions

Age, sex, vital status, and neighborhood income quintile (based on postal code linkage22) were taken from the Registered Persons Database. Cancer type, diagnosis, and diagnosis date were taken from Ontario Cancer Registry. For patients with multiple primary tumors, we chose the first diagnosis date that preceded the first assessment date. Comorbidity was calculated by using the Deyo23 modification of the Charlson score based on diagnoses coded in Discharge Abstract Database in the 12 months before first assessment with scores for primary and metastatic cancer subtracted. The Symptom Management Reporting Database provided assessment location and treatment intent (derived from 2 related variables to increase completeness).

Statistical Analysis

Analyses are primarily descriptive. Mean and median scores are reported. Because of the large sample size, essentially all Student t tests and chi-squared tests were significant at the .001 level. For this reason, no statistical results are presented, and results are interpreted based on clinically meaningful differences. A difference of 2 in the median ESAS score24 was considered important.

A series of univariate and multivariate logistic regression models were performed for a few commonly studied cancer symptoms. Multivariate models included age, sex, Charlson score, income quintile, cancer type, and survival from time of assessment as input variables (intent was not included because it was missing for two-thirds of the cohort). On the basis of previous research, individual symptom scores were dichotomized into scores of 0-3 (none to mild) versus ≥4 (moderate to severe).19, 25 We reran the models and compared our results using a cutoff of ≥7 (severe). Modeling results were similar with either cutpoint. We present data with a cutoff of ≥4, which was our a priori preference.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Our study cohort included 45,118 (ESAS) and 23,802 (PPS) unique patients whose first assessment contributes to this study. We excluded 4606 patients (ESAS) and 2159 patients (PPS) when they had an invalid unique identifier, when their the first assessment was before their diagnosis date or after their death date, when we were not able to link a patient to the cancer registry, or when a patient was less than 18 years of age. Although the aim of the Ontario Cancer Symptom Management Collaborative initiative was to have ESAS and PPS scores documented at every visit, many more ESAS scores were collected than PPS scores, which is likely a result of provider compliance.

Table 1 presents the characteristics of the patients in each cohort. The average age was 63 years old; 56% of the cohort were women. About 15% of each cohort had comorbidities beyond a diagnosis of cancer. The 4 most common cancers (ie, lung, breast, gastrointestinal, and genitourinary) represented about 75% of cases in both cohorts. The treatment intent information was missing for the majority of cases. Thirty-two percent of the ESAS cohort and 43% of the PPS cohort had died by the time of analysis. The vast majority of assessments were completed in the clinic, rather than at home. Uptake of the screening program was highly variable among cancer centers and regional home care programs (data not shown).

Table 1. Patient Characteristics of ESAS and PPS Cohort
 First ESASFirst PPS
  1. ESAS indicates Edmonton Symptom Assessment System; PPS, Palliative Performance Score.

Age at diagnosis median62.9 y62.9 y
Age at first assessment median66.0 y66.0 y
Charlson comorbidity
Income quintile
 Low Q117.4784217.24090
 High Q521.2953921.95199
 Unknown or missing70.03157266.515821
Cancer type
 Head and neck3.716642.7630
 Central nervous system1.15111.4321
 Primary unknown0.72920.8178
Vital status at time of analysis
Assessment location

Table 2 demonstrates the frequency distribution of each ESAS symptom score. The symptoms are presented in order of most-commonly to least-commonly reported. Feeling tired was reported most frequently (75% of patients reported a score of at least 1 for feeling tired) and nausea least frequently (25% of patients reported a score of at least 1 for feeling nausea). Approximately half of patients reported problems with pain or shortness of breath, and about half of those had a pain or shortness of breath score greater than or equal to 4, suggesting moderate to severe intensity.

Table 2. Percentage of Patients Reporting Each ESAS Symptom
ScoreTiredWell BeingAppetiteAnxiousPainDrowsyShortness of BreathDepressionNausea
  1. ESAS indicates Edmonton Symptom Assessment System.


Table 3 shows the frequency distribution of PPS scores. In this cross-section of ambulatory cancer patients, seventy-eight percent of the patients had stable scores (70% to 100%), 21% were transitional (40% to 60%), and 1% would be considered at the end of life (10% to 30%).

Table 3. Percentage of Patients Reporting Each PPS Score
PPS Score% of Patients
  1. PPS indicates Palliative Performance Score.


Table 4 provides the mean and median score of each symptom by covariate, presented in order of most-commonly to least-commonly reported symptom. The highest mean symptom scores were for tiredness and issues with well being. Changes of 2 in the median score were seen for certain symptoms by the covariates age, comorbidity, treatment intent, and cancer type. Lung cancer patients generally had the highest symptom burden, followed closely by those with an unknown primary cancer.

Table 4. Mean and Median Individual ESAS Symptom Scores
  TiredWell BeingAppetiteAnxiousPainDrowsyShortness of BreathDepressionNausea
No.Mean (Med)Mean (Med)Mean (Med)Mean (Med)Mean (Med)Mean (Med)Mean (Med)Mean (Med)Mean (Med)
  1. ESAS indicates Edmonton Symptom Assessment System; Med, median.

All451183.57 (3)3.08 (3)2.63 (1)2.27 (1)2.17 (1)2.05 (0)2.01 (0)1.68 (0)0.85 (0)
Age, y
 18-294283.21 (3)2.46 (2)2.12 (1)1.86 (0)1.69 (0)1.97 (1)1.1 (0)1.4 (0)1.01 (0)
 30-3912053.48 (3)2.91 (2)2.4 (1)2.42 (2)2.23 (1)1.96 (1)1.28 (0)1.72 (0)1 (0)
 40-4940713.61 (3)3.13 (3)2.48 (1)2.45 (2)2.38 (1)2.04 (1)1.49 (0)1.86 (0)0.99 (0)
 50-5987933.69 (3)3.2 (3)2.61 (2)2.5 (2)2.42 (1)2.09 (1)1.86 (0)1.86 (0)0.99 (0)
 60-69126393.38 (3)2.97 (2)2.52 (1)2.25 (1)2.11 (1)1.91 (0)2.01 (0)1.62 (0)0.83 (0)
 70-79121713.52 (3)3.02 (3)2.67 (1)2.1 (1)2.03 (0)2.02 (0)2.24 (1)1.57 (0)0.77 (0)
 80-8954033.9 (4)3.27 (3)3.01 (2)2.16 (1)2.11 (0)2.32 (1)2.35 (1)1.62 (0)0.71 (0)
 90-1004084.49 (5)3.53 (3)3.46 (3)2.34 (1)2.2 (1)2.87 (2)2.22 (1)1.67 (0)0.83 (0)
 Women252733.7 (3)3.12 (3)2.63 (2)2.45 (1)2.22 (1)2.01 (0)1.88 (0)1.75 (0)0.89 (0)
 Men198453.41 (3)3.03 (3)2.63 (1)2.04 (1)2.11 (1)2.09 (1)2.18 (1)1.58 (0)0.8 (0)
 0388343.44 (3)2.97 (2)2.51 (1)2.2 (1)2.1 (1)1.94 (0)1.89 (0)1.61 (0)0.81 (0)
 ≥162844.36 (5)3.76 (4)3.4 (3)2.66 (2)2.6 (2)2.7 (2)2.77 (2)2.08 (1)1.08 (0)
Income quintile
 178423.87 (4)3.32 (3)2.93 (2)2.48 (1)2.45 (1)2.26 (1)2.32 (1)1.97 (0)0.99 (0)
 288493.71 (3)3.14 (3)2.75 (2)2.32 (1)2.29 (1)2.13 (1)2.12 (1)1.75 (0)0.87 (0)
 391043.56 (3)3.1 (3)2.62 (1)2.28 (1)2.18 (1)2.07 (0)2.02 (0)1.69 (0)0.89 (0)
 496703.43 (3)2.96 (2)2.47 (1)2.16 (1)2.07 (1)1.93 (0)1.9 (0)1.54 (0)0.79 (0)
 595393.35 (3)2.94 (2)2.46 (1)2.14 (1)1.94 (0)1.89 (0)1.76 (0)1.5 (0)0.76 (0)
 Curative70543.08 (3)2.48 (2)2.03 (1)1.91 (1)1.74 (0)1.75 (0)1.79 (0)1.38 (0)0.65 (0)
 Palliative64924.74 (5)4.19 (4)3.79 (3)2.94 (2)3.39 (3)3.04 (2)2.66 (2)2.38 (1)1.48 (0)
 Unknown or missing315723.44 (3)2.99 (3)2.53 (1)2.21 (1)2.02 (1)1.91 (0)1.92 (0)1.6 (0)0.77 (0)
Cancer type
 Breast97213.24 (3)2.67 (2)2.03 (1)2.14 (1)1.99 (1)1.67 (0)1.46 (0)1.54 (0)0.64 (0)
 Central nervous system5114.42 (5)3.35 (3)2.31 (1)2.37 (1)2.14 (1)3.11 (2)1.27 (0)2.03 (1)0.88 (0)
 Gastrointestinal84163.68 (3)3.22 (3)3.01 (2)2.25 (1)2.18 (1)2.19 (1)1.6 (0)1.69 (0)1.07 (0)
 Genitourinary64312.93 (2)2.6 (2)2.09 (1)1.79 (0)1.85 (0)1.77 (0)1.65 (0)1.36 (0)0.65 (0)
 Gynecologic39853.55 (3)3.08 (3)2.62 (2)2.43 (1)2.11 (1)1.82 (0)1.49 (0)1.67 (0)0.88 (0)
 Hematology37313.69 (3)2.95 (2)2.33 (1)2.02 (1)2.09 (1)2.13 (1)1.87 (0)1.54 (0)0.76 (0)
 Head and Neck16643.27 (3)3.18 (3)3.14 (2)2.19 (1)2.51 (1)2.01 (0)1.71 (0)1.78 (0)0.78 (0)
 Lung86354.25 (4)3.77 (4)3.43 (3)2.84 (2)2.58 (2)2.52 (1)3.75 (3)2.08 (1)1.07 (0)
 Other6803.87 (4)3.19 (3)2.48 (1)2.29 (1)2.27 (1)2.05 (0)1.83 (0)1.63 (0)0.85 (0)
 Primary unknown2924.54 (4)3.8 (4)3.61 (3)2.79 (2)2.82 (2)2.88 (2)2.37 (1)1.94 (0)1.13 (0)
 Sarcoma2673.62 (3)3.38 (3)2.64 (2)2.26 (1)2.81 (2)2.22 (1)1.83 (0)1.66 (0)0.79 (0)
 Skin7853.21 (2)2.89 (2)2.25 (1)2.06 (1)1.93 (0)2 (0)1.48 (0)1.56 (0)0.76 (0)

Focusing on a few symptoms, Table 5 summarizes the log-odds model for reporting a symptom score ≥4 (moderate to severe) controlling for specific covariates. Women consistently reported higher scores than men (except shortness of breath, which was not statistically different). Those with comorbidity had more than one-third higher odds of reporting moderate to severe scores. As well, those who died within 90 days of first assessment had more than twice the odds of reporting scores ≥4. Income showed a modest association with symptom score: the lower the income, the higher the mean symptom score. Generally, lung cancer patients had the worst burden of symptoms.

Table 5. Results of Multivariate Logistic Regression for Individual Symptom
 Odds of Symptom Score ≥4 (95% CI)
PainShortness of BreathAnxiousDepressionTired
  1. CI indicates confidence interval.

Age, y
 30-391.67 (1.26, 2.21)1.31 (0.94, 1.83)1.49 (1.14, 1.93)1.15 (0.86, 1.54)1.11 (0.88, 1.4)
 40-491.8 (1.39, 2.32)1.42 (1.05, 1.93)1.29 (1.01, 1.64)1.26 (0.97, 1.64)1.13 (0.92, 1.39)
 50-591.81 (1.41, 2.33)1.79 (1.33, 2.41)1.38 (1.09, 1.75)1.2 (0.93, 1.56)1.17 (0.96, 1.44)
 60-691.39 (1.08, 1.78)1.75 (1.3, 2.35)1.12 (0.88, 1.41)1 (0.77, 1.29)0.94 (0.77, 1.15)
 70-791.29 (1, 1.66)2.01 (1.5, 2.71)1.01 (0.8, 1.28)0.92 (0.71, 1.19)0.97 (0.8, 1.19)
 80-891.32 (1.02, 1.7)2.27 (1.68, 3.07)1.04 (0.82, 1.33)0.95 (0.73, 1.24)1.19 (0.97, 1.46)
 90-1001.29 (0.92, 1.8)2.15 (1.48, 3.12)1.07 (0.78, 1.48)0.97 (0.68, 1.39)1.69 (1.27, 2.26)
 Women1.22 (1.15, 1.29)0.96 (0.91, 1.01)1.47 (1.39, 1.55)1.27 (1.19, 1.35)1.37 (1.31, 1.44)
 ≥11.3 (1.23, 1.39)1.38 (1.3, 1.47)1.3 (1.22, 1.38)1.34 (1.25, 1.43)1.54 (1.45, 1.63)
Income quintile
 11.41 (1.32, 1.51)1.35 (1.25, 1.45)1.2 (1.12, 1.29)1.42 (1.32, 1.53)1.26 (1.18, 1.34)
 21.27 (1.19, 1.36)1.18 (1.1, 1.27)1.1 (1.03, 1.17)1.21 (1.12, 1.3)1.16 (1.09, 1.23)
 31.21 (1.13, 1.3)1.15 (1.07, 1.23)1.12 (1.05, 1.2)1.17 (1.09, 1.27)1.1 (1.04, 1.17)
 41.13 (1.06, 1.21)1.06 (0.98, 1.14)1.01 (0.94, 1.08)1.02 (0.95, 1.1)1.02 (0.97, 1.09)
Cancer type
 Breast0.72 (0.67, 0.77)0.28 (0.26, 0.3)0.54 (0.51, 0.58)0.67 (0.62, 0.73)0.59 (0.55, 0.63)
 Central nervous system0.73 (0.59, 0.9)0.22 (0.17, 0.28)0.72 (0.59, 0.88)0.93 (0.75, 1.16)1.12 (0.93, 1.36)
 Gastrointestinal0.84 (0.79, 0.9)0.25 (0.23, 0.27)0.7 (0.66, 0.75)0.79 (0.73, 0.85)0.79 (0.74, 0.85)
 Genitourinary0.84 (0.77, 0.91)0.28 (0.26, 0.3)0.64 (0.59, 0.69)0.71 (0.65, 0.78)0.62 (0.58, 0.67)
 Gynecologic0.75 (0.68, 0.82)0.27 (0.25, 0.3)0.67 (0.61, 0.73)0.7 (0.63, 0.77)0.69 (0.64, 0.75)
 Hematology0.89 (0.82, 0.98)0.36 (0.33, 0.39)0.62 (0.57, 0.68)0.75 (0.68, 0.83)0.87 (0.8, 0.94)
 Head and neck1.08 (0.96, 1.22)0.29 (0.26, 0.33)0.71 (0.63, 0.8)0.9 (0.79, 1.03)0.66 (0.59, 0.74)
 Other0.94 (0.78, 1.12)0.37 (0.31, 0.45)0.71 (0.6, 0.85)0.74 (0.61, 0.91)0.93 (0.79, 1.1)
 Primary unknown1.12 (0.87, 1.45)0.43 (0.33, 0.56)0.94 (0.73, 1.21)0.82 (0.62, 1.09)1.04 (0.81, 1.33)
 Sarcoma1.23 (0.94, 1.6)0.35 (0.26, 0.47)0.7 (0.53, 0.93)0.79 (0.58, 1.07)0.81 (0.62, 1.04)
 Skin0.69 (0.58, 0.83)0.23 (0.19, 0.28)0.61 (0.52, 0.73)0.74 (0.61, 0.89)0.56 (0.48, 0.65)
Survival from assessment date
 ≥90 days11111
 <90 days3.21 (3.02, 3.42)2.39 (2.24, 2.55)2.11 (1.98, 2.25)2.48 (2.33, 2.65)4.61 (4.28, 4.96)


  1. Top of page
  2. Abstract
  6. Acknowledgements

This study describes the ESAS and PPS scores in a cross section of over 45,000 and 23,000 patients, respectively, making this the largest population-based study to examine cancer symptom and performance outcomes to our knowledge. This population is also unique because it includes a full scope of cancer diagnoses, broad range of ages, exclusively ambulatory outpatients, and patients seen much earlier in the course of their illness (ie, those treated for cure).

This study addresses knowledge gaps identified in the 2002 National Institute of Health's State-of-the-Science report on cancer symptom management on pain, fatigue, and depression.1, 5 That report estimated cancer-related pain, depression, and fatigue to range from 14% to 100%, 1% to 42% and 4% to 91%, respectively, including studies of all cancer types and across the cancer continuum; however, the report's findings were limited because the range of prevalence rates varied widely and because the reviewed studies used an assortment of tools to define symptoms. In our ambulatory, population-based, cancer cohort, the prevalence of having mild to severe symptoms was 53% for pain, 44% for depression, and 50% (drowsy) to 75% (tired) for fatigue, generally falling within the report's ranges. Moreover, the prevalence of fatigue and depression in our cohort is consistent with other reports.26-30 The high prevalence of these symptoms represents areas for continued research and treatment to improve symptom management. Furthermore these results are important to practitioners because they focus on a defined ambulatory population, are population-based, and originate from validated symptom assessment tools.

Cancer symptom management has also been studied widely in palliative and end-of-life care populations; yet compared with a large systematic review of symptom burden in incurable cancer patients, our study's outpatient cohort also has a significant burden of symptoms. The systematic review30 of 46 studies included approximately 25,000 patients from a variety of palliative cancer subpopulations using several different instruments to assess symptoms. In the review, fatigue was the most common symptom, reported in 75% of patients (same as current study); pain was reported by 71% of patients (53% in current study), dyspnea by 35% of patients (49% in current study), depressed mood by 39% (44% in current study), and nausea in 31% (25% in current study). Therefore, in our cohort of cancer outpatients, where only 32% to 44% had died by the time of data analysis, symptom prevalence was similar, in some cases, to those found in the systematic review of palliative studies.

Although our results clearly show that patients closer to death (ie, those dying within 90 days of first assessment) have twice to quadruple the odds of reporting moderate to severe symptom scores, the high symptom prevalence generally in this ambulatory population is still noteworthy. Comparing our results to studies that specifically report ESAS scores from a palliative care unit or clinic,26, 27, 31-38 the symptom prevalence and symptom scores from these palliative patients are consistently higher than in our study. For example, in a study of 1296 patients attending a palliative radiation oncology outpatient clinic,26 the median symptom score was 5 for fatigue, 2 for dyspnea, 3 for pain, and 2 for depression compared with our results of 3, 0, 1, and 0, respectively. Nonetheless, despite our cohort of “healthier” patients, namely with PPS scores between 70% and 100% (stable), approximately 1 in 2 patients reported the presence of pain or shortness of breath. One in 4 patients reported pain or shortness of breath of moderate to severe intensity (≥4). Management of these symptoms represents an important target for quality improvement because shortness of breath or pain commonly prompt visits to the emergency department or admission to hospital at the end of life.39-41

Our results also describe how patient characteristics are associated with particular symptom scores. Not surprisingly, comorbidity was associated with higher symptom scores. As well, women had consistently worse symptom scores than men. Worse outcomes in women have been observed in some studies,3, 42 but results are inconsistent.43 A recent review suggests that females have greater pain sensitivity and may have different treatment responses, although the mechanisms are unclear.44 This research area requires further investigation to determine whether women are more likely to report symptoms, less likely to receive adequate symptom treatment, or whether there is a biologic basis for worse outcomes.

Similarly, data regarding the relation of age to symptom is inconsistent. In our study, the adjusted odds of having moderate to severe pain were highest for those between the ages of 30 and 59 years. Some studies suggest that older patients are less likely to report cancer pain;45, 46 thus they are at highest risk for being inadequately assessed and treated,47 and pain intensity decreases with age.48, 49 However, some studies have found no relation between age and cancer pain.50-52 In addition, important factors that may influence the relation of age and pain, such as type of cognitive impairment and treatment, were not statistically controlled.53 Moreover, average symptom scores for anxiety, depression, and nausea were also highest for the 30-59-year age range. Further research is needed to determine the biological and/or systematic explanations for differences between age and severity of specific symptoms.

Our study is limited because patients who are included in the study are not systematically screened at regular intervals and assessments occur on an opportunistic basis. Moreover, not every patient or provider completes the ESAS or PPS tool. For patients with more than 1 assessment on the same day, we chose to keep the assessment with the worse score. Because this choice affected less than 0.5% of the assessments, it would not appreciably change the results. The high frequency of missing data for treatment intent limited some of our interpretations; and with existing data, it is difficult to describe exactly where each patient was in their continuum of care. However, because of the number and the broad range of cancer patients included, the results from this cross-sectional ambulatory cohort are relevant to physicians, as these patients are representative of whom oncologists will encounter in daily clinical practice.

To our knowledge, Ontario is the only place that has standardized symptom and performance status assessment into routine care at cancer centers at a population level. As such, this is the first and largest study to report ESAS and PPS scores in a large geographically based cohort. Among this ambulatory cancer population, several moderate to severe symptoms are prevalent, which represent targets for improved clinical care. Differences in outcomes for certain patient characteristics and cancer subgroups require further investigation. This research sets the groundwork for future research on patient outcomes using linked administrative healthcare data.


  1. Top of page
  2. Abstract
  6. Acknowledgements

The authors thank Susan King for her insight and explanations regarding the Symptom Management Reporting Database.


  1. Top of page
  2. Abstract
  6. Acknowledgements

This study was conducted with the support of the Ontario Institute for Cancer Research through funding provided by the Government of Ontario. This study was supported by the Institute for Clinical Evaluative Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results, and conclusions reported in this article are those of the authors and are independent from the funding sources. No endorsement by ICES or the Ontario MOHLTC is intended or should be concluded.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Carr D, Goudas L, Lawrence D, et al. Management of cancer symptoms: pain, depression, and fatigue. Evid Rep Technol Assess (Summ). 2002: 1-5.
  • 2
    Ripamonti CI, Farina G, Garassino MC. Predictive models in palliative care. Cancer. 2009; 115: 3128-3134.
  • 3
    Cleeland CS, Gonin R, Hatfield AK, et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med. 1994; 330: 592-596.
  • 4
    Mitera G, Zeiadin N, Kirou-Mauro A, et al. Retrospective assessment of cancer pain management in an outpatient palliative radiotherapy clinic using the Pain Management Index. J Pain Symptom Manage. 2010; 39: 259-267.
  • 5
    Patrick DL, Ferketich SL, Frame PS, et al. National Institutes of Health State-of-the-Science Conference Statement: Symptom Management in Cancer: Pain, Depression, and Fatigue, July 15-17, 2002. J Natl Cancer Inst. 2003; 95: 1110-1117.
  • 6
    Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care. 1991; 7: 6-9.
  • 7
    Anderson F, Downing GM, Hill J, Casorso L, Lerch N. Palliative performance scale (PPS): a new tool. J Palliat Care. 1996; 12: 5-11.
  • 8
    Downing M, Lau F, Lesperance M, et al. Meta-analysis of survival prediction with Palliative Performance Scale. J Palliat Care. 2007; 23: 245-252.
  • 9
    Dudgeon DJ, Knott C, Chapman C, et al. Development, implementation, and process evaluation of a regional palliative care quality improvement project. J Pain Symptom Manage. 2009; 38: 483-495.
  • 10
    Dudgeon DJ, Knott C, Eichholz M, et al. Palliative Care Integration Project (PCIP) quality improvement strategy evaluation. J Pain Symptom Manage. 2008; 35: 573-582.
  • 11
    Cancer Care Ontario. Cancer System Quality Index. 2009.
  • 12
    Robles SC, Marrett LD, Clarke EA, Risch HA. An application of capture-recapture methods to the estimation of completeness of cancer registration. J Clin Epidemiol. 1988; 41: 495-501.
  • 13
    Clarke EA, Marrett LD, Kreiger N. Cancer registration in Ontario: A computer approach. In: JensonOM, ParkinDM, MacLennanR. Cancer Registration Principles and Methods. Lyon, France: IARC; 1991: 246-257.
  • 14
    Iron K, Zagorski BM, Sykora K, Manuel DG. Living and Dying in Ontario: An Opportunity for Improved Health Information. ICES Investigative Report 2008. Toronto, Ontario, Canada: Institute for Clinical Evaluative Sciences; 2008.
  • 15
    C anadian Institute for Health Information. Data Quality Of The Discharge Abstract Database Following The First Year Implementation of ICD-10-CA/CCI-Final Report. Ottawa, CIHI; 2004.
  • 16
    Nekolaichuk C, Watanabe S, Beaumont C. The Edmonton Symptom Assessment System: a 15-year retrospective review of validation studies (1991-2006). Palliat Med. 2008; 22: 111-122.
  • 17
    Richardson LA, Jones GW. A review of the reliability and validity of the Edmonton Symptom Assessment System. Curr Oncol. 2009; 16: 55.
  • 18
    Chang VT, Hwang SS, Feuerman M. Validation of the Edmonton Symptom Assessment Scale. Cancer. 2000; 88: 2164-2171.
  • 19
    Selby D, Cascella A, Gardiner K, et al. A single set of numerical cutpoints to define moderate and severe symptoms for the Edmonton Symptom Assessment System. J Pain Symptom Manage. 2010; 39: 241-249.
  • 20
    Ho F, Lau F, Downing MG, Lesperance M. A reliability and validity study of the Palliative Performance Scale. BMC Palliat Care. 2008; 7: 10.
  • 21
    Lau F, Maida V, Downing M, Lesperance M, Karlson N, Kuziemsky C. Use of the Palliative Performance Scale (PPS) for end-of-life prognostication in a palliative medicine consultation service. J Pain Symptom Manage. 2009; 37: 965-972.
  • 22
    Wilkins R. PCCF + Version 3G Users Guide: Automated Geographic Coding Based on the Statistics Canada Postal Code Conversions Files. Ottawa, Ontario, Canada: Statistics Canada; 2001.
  • 23
    Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992; 45: 613-619.
  • 24
    Chow E, Hird A, Wong R, et al. Validation of meaningful change in pain scores in the treatment of bone metastases. J Pain Manage. 2009; 2: 407-412.
  • 25
    Serlin RC, Mendoza TR, Nakamura Y, Edwards KR, Cleeland CS. When is cancer pain mild, moderate or severe? Grading pain severity by its interference with function. Pain. 1995; 61: 277-284.
  • 26
    Bradley N, Davis L, Chow E. Symptom distress in patients attending an outpatient palliative radiotherapy clinic. J Pain Symptom Manage. 2005; 30: 123-131.
  • 27
    Cheung WY, Le LW, Zimmermann C. Symptom clusters in patients with advanced cancers. Support Care Cancer. 2009; 17: 1223-1230.
  • 28
    Hofman M, Ryan JL, Figueroa-Moseley CD, Jean-Pierre P, Morrow GR. Cancer-related fatigue: the scale of the problem. Oncologist. 2007; 12( suppl 1): 4-10.
  • 29
    Massie MJ. Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr. 2004; 57-71.
  • 30
    Teunissen SC, Wesker W, Kruitwagen C, de Haes HC, Voest EE, de Graeff A. Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manage. 2007; 34: 94-104.
  • 31
    Astradsson E, Granath L, Heedman PA, Starkhammar H. Cancer patients hospitalised for palliative reasons. Symptoms and needs presented at a university hospital. Support Care Cancer. 2001; 9: 97-102.
  • 32
    Brandt HE, Ooms ME, Deliens L, van der WG, Ribbe MW. The last two days of life of nursing home patients-a nationwide study on causes of death and burdensome symptoms in The Netherlands. Palliat Med. 2006; 20: 533-540.
  • 33
    Chow E, Fan G, Hadi S, Filipczak L. Symptom clusters in cancer patients with bone metastases. Support Care Cancer. 2007; 15: 1035-1043.
  • 34
    Fadul NA, El Osta B, Dalal S, Poulter VA, Bruera E. 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.
  • 35
    Fan G, Hadi S, Chow E. Symptom clusters in patients with advanced-stage cancer referred for palliative radiation therapy in an outpatient setting. Support Cancer Ther. 2007; 4: 157-162.
  • 36
    Riechelmann RP, Krzyzanowska MK, O'Carroll A, Zimmermann C. Symptom and medication profiles among cancer patients attending a palliative care clinic. Support Care Cancer. 2007; 15: 1407-1412.
  • 37
    Salminen E, Clemens KE, Syrjanen K, Salmenoja H. Needs of developing the skills of palliative care at the oncology ward: an audit of symptoms among 203 consecutive cancer patients in Finland. Support Care Cancer. 2008; 16: 3-8.
  • 38
    Sigurdardottir KR, Haugen DF. Prevalence of distressing symptoms in hospitalised patients on medical wards: A cross-sectional study. BMC Palliat Care. 2008; 7: 16.
  • 39
    Barbera L, Paszat L, Qiu F. End-of-life care in lung cancer patients in Ontario: aggressiveness of care in the population and a description of hospital admissions. J Pain Symptom Manage. 2008; 35: 267-274.
  • 40
    Barbera L, Taylor C, Dudgeon D. Why do cancer patients visit the emergency department near the end of life? CMAJ. 2010; 182: 563-568.
  • 41
    Brazil K, Howell D, Bedard M, Krueger P, Heidebrecht C. Preferences for place of care and place of death among informal caregivers of the terminally ill. Palliat Med. 2005; 19: 492-499.
  • 42
    Husain AF, Stewart K, Arseneault R, et al. Women experience higher levels of fatigue than men at the end of life: a longitudinal home palliative care study. J Pain Symptom Manage. 2007; 33: 389-397.
  • 43
    Miaskowski C. Gender differences in pain, fatigue, and depression in patients with cancer. J Natl Cancer Inst Monogr. 2004; 139-143.
  • 44
    Fillingim RB, King CD, Ribeiro-Dasilva MC, Rahim-Williams B, Riley JL 3rd. Sex, gender, and pain: a review of recent clinical and experimental findings. J Pain. 2009; 10: 447-485.
  • 45
    Bernabei R, Gambassi G, Lapane K, et al. Management of pain in elderly patients with cancer. SAGE Study Group Systematic Assessment of Geriatric Drug Use via Epidemiology. JAMA. 1998; 279: 1877-1882.
  • 46
    Walsh D, Donnelly S, Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1,000 patients. Support Care Cancer. 2000; 8: 175-179.
  • 47
    Melzack R. The tragedy of needless pain: A call for social action. In: DubnerR, GebhartGF, BondMR, eds. Proceedings of the Fifth World Congress On Pain. Amsterdam, Netherlands: Elsevier Science; 1988: 1-11.
  • 48
    Caraceni A, Portenoy RK. An international survey of cancer pain characteristics and syndromes. IASP Task Force on Cancer Pain International Association for the Study of Pain. Pain. 1999; 82: 263-274.
  • 49
    Morris JN, Mor V, Goldberg RJ, Sherwood S, Greer DS, Hiris J. The effect of treatment setting and patient characteristics on pain in terminal cancer patients: a report from the National Hospice Study. J Chronic Dis. 1986; 39: 27-35.
  • 50
    Gagliese L, Jovellanos M, Zimmermann C, Shobbrook C, Warr D, Rodin G. Age-related patterns in adaptation to cancer pain: a mixed-method study. Pain Med. 2009; 10: 1050-1061.
  • 51
    Stein WM, Miech RP. Cancer pain in the elderly hospice patient. J Pain Symptom Manage. 1993; 8: 474-482.
  • 52
    Vigano A, Bruera E, Suarez-Almazor ME. Age, pain intensity, and opioid dose in patients with advanced cancer. Cancer. 1998; 83: 1244-1250.
  • 53
    Gagliese L. Pain and aging: the emergence of a new subfield of pain research. J Pain. 2009; 10: 343-353.