Comparison of Three Shortened Questionnaires for Assessment of Quality of Life in Advanced Cancer

Abstract

Objective:

Quality of life (QoL) assessment questionnaires can be burdensome to advanced cancer patients, thus necessitating the need for shorter assessment instruments than traditionally available. We compare three shortened QoL questionnaires in regards to their characteristics, validity, and reliability.

Methods:

A literature search was conducted to identify studies that employed or discussed three abridged QoL questionnaires: the European Organization for Research and Treatment of Cancer Quality of Life Core 15-Palliative Care (EORTC QLQ-C15-PAL), the Functional Assessment of Cancer Therapy-General-7 (FACT-G7), and the Functional Assessment of Chronic Illness Therapy-Palliative Care-14 (FACIT-PAL-14). Articles that discussed questionnaire length, intended use, scoring procedure, and validation were included.

Results:

The 7-item FACT-G7 is the shortest instrument, whereas the EORTC QLQ-C15-PAL and the FACIT-PAL-14 contain 14 and 15 items, respectively. All three questionnaires have similar recall period, item organization, and subscale components. Designed as core questionnaires, all three maintain content and concurrent validity of their unabridged original questionnaires. Both the EORTC QLQ-C15-PAL and the FACT-G7 demonstrate good internal consistency and reliability, with Cronbach's α ≥0.7 deemed acceptable. The developmental study for the FACIT-PAL-14 was published in 2013 and subsequent validation studies are not yet available.

Conclusion:

The EORTC QLQ-C15-PAL and the FACT-G7 were found to be reliable and appropriate for assessing health-related QoL issues—the former for palliative cancer patients and the latter for advanced cancer patients receiving chemotherapy. Conceptually, the FACIT-PAL-14 holds promise to cover social and emotional support issues that are not completely addressed by the other two questionnaires; however, further validation is needed.

Introduction

In advanced cancer trials, quality of life (QoL) is an increasingly important end point. QoL is a subjective, multidimensional construct that measures a patient's well-being by evaluating physical, emotional, social, and cognitive functions.¹ Because QoL examines a patient's own subjective well-being, it is most commonly and appropriately measured using a self-administered questionnaire. For cancer patients, most assessment instruments provide information on the patient's clinical status to assist in future clinical decision making and provide feedback on the effectiveness of previously administered treatments.²

Currently, two widely used and validated cancer QoL assessment instruments are the European Organization for Research and Treatment of Cancer Quality of Life Core 30 (EORTC QLQ-C30) with 30 items and the Functional Assessment of Cancer Therapy-General (FACT-G) with 27 items.³ The FACT-G questionnaire can be used in conjunction with the 19-item Functional Assessment of Chronic Illness Therapy-Palliative Care (FACIT-PAL) module for advanced cancer patients receiving palliative care, resulting in a total sum of 46 items. With such large numbers of items, these questionnaires may prove lengthy to patients.^4,5 In the advanced cancer setting, length of the administered questionnaire may be a more significant concern or burden.^4,5 Given the physical and emotional conditions of such patients, relieving patient burden remains a definite clinical aim.^4,5

To address issues related to questionnaire length, shortened questionnaires have been recently developed by both the EORTC and FACT organizations. These instruments are intended to comprehensively assess QOL in advanced cancer patients without substantially reducing measurement precision or core content validity.^6–8 Thus far, the 30-item EORTC QLQ-C30 was shortened to produce a 15-item EORTC QLQ-C15-PAL;⁸ the 27-item FACT-G was shortened to a 7-item FACT-G7;⁷ and the 46-item combination of the FACT-G and the FACIT-PAL was shortened to a 14-item FACIT-PAL-14.⁶ A direct comparison of these three tools may assist both health care providers (HCPs) in selecting the most appropriate instrument for meeting the needs of the patients and investigators in the selection of instruments for future clinical trials.

The purpose of this review is to compare the three shortened QoL tools available for advanced cancer patients: the EORTC QLQ-C15-PAL, FACT-G7, and FACIT-PAL-14. The study aims to compare and contrast the three tools in terms of characteristics, validity, and reliability.

Methods

A literature search was conducted using Ovid MEDLINE (1980 to May 2013), Ovid EMBASE (1980 to 2013 week 20), the Cochrane Central Register of Controlled Trials (May 2013) and CINAHL (1980 to May 2013) to identify studies that discussed the EORTC QLQ-C15-PAL, the FACT-G7, or the FACIT-PAL-14. Search terms included “quality of life,” “assessment,” “module,” “questionnaire,” in conjunction with “EORTC QLQ-C15,” “PAL,” “FACT-G7,” or “FACIT-PAL 14” to obtain relevant literature. Articles that discussed the development and/or use of at least one of these tools were included. English articles that included the following information were examined: questionnaire length, subscales covered, intended use, scoring procedure, reliability, and validation.

Results and Discussion

Characteristics

A direct comparison of the three questionnaires' characteristics is summarized in Table 1. The three abbreviated questionnaires (the EORTC QLQ-C15-PAL, FACT-G7, and FACIT-PAL-14) all aim to address patient burden and at the same time preserve the core content of the lengthy original instruments. Whereas the EORTC QLQ-C15-PAL and the FACIT-PAL-14 are used specifically for palliative care, the FACT-G7 is intended to be used across a broad spectrum of advanced cancers. Both the EORTC QLQ-C15-PAL (15 items) and the FACIT-PAL-14 (14 items) are at least double the length of the FACT-G7 (7 items). In their most recent versions, all three shortened questionnaires are intended to be administered on their own as core questionnaires. However, one has the option of administering them with additional cancer-specific subscales or questions.^6–8

Table 1.

Comparison of the Characteristics of the EORTC QLQ-C15-PAL, the FACT-G7, and the FACIT-PAL-14

	EORTC QLQ-C15-PAL	FACT-G7	FACIT-PAL-14
Use	Used as a core questionnaire instead of the EORTC QLQ-C30; used in palliative care settings	Used as a core questionnaire instead of the FACT-G; used for advanced cancer patients receiving chemotherapy	Used as a questionnaire instead of the FACIT-PAL; used in palliative care settings
Number of items	15	7	14
Response options	Likert scale (1–4 or 1–7 inclusive)	Likert scale (0–4 inclusive)	Likert scale (0–4 inclusive)
Recall period	Past 7 days	Past 7 days	Past 7 days
Item format	All questions	All statements	All statements
Item organization	Subscale items dispersed throughout the questionnaire; no specific order	Subscale items dispersed throughout the questionnaire; no specific order	Subscale items dispersed throughout the questionnaire; no specific order
Assessed subscales	• Overall QoL • Physical functioning • Emotional functioning • Fatigue • Nausea/vomiting • Pain • Dyspnea • Insomnia • Appetite loss • Constipation	• Physical well-being • Emotional well-being • Functional well-being	• Physical well-being • Emotional well-being • Functional well-being • Social/family well-being • Additional concerns
Subscale components	Both single-item and multi-item subscales	Both single-item and multi-item subscales	Both single-item and multi-item subscales
Scoring	1–4 numerical scores are converted to 0–100 scale. Higher symptom scores reflect greater distress, but higher functional scores represent better functioning. Higher overall QoL scores also represent better overall QoL	0–4 numerical scores are summed on a scale of 0–28. Higher scores correspond to better overall QoL	0–4 numerical scores are summed on a scale of 0–56. Higher scores correspond to better overall QoL

QoL, quality of life.

All questionnaires have a recall period of the previous week and use a Likert scale for response. The EORTC QLQ-C15-PAL provides patients with four choices (1=not at all; 2=a little; 3=quite a bit; 4=very much), whereas both the FACT-G7 and the FACIT-PAL-14 give patients five choices (0=not at all; 1=a little; 2=somewhat; 3=quite a bit; 4=very much). Items from each questionnaire are presented in no specific order. One difference between the three tools is the way in which the items are presented. Whereas the FACT-G7 and the FACIT-PAL-14 present statements, the EORTC QLQ-C15-PAL asks questions. For example, the EORTC QLQ-C15-PAL asks, “Have you felt nauseated?” whereas the FACT-G7 and FACIT-PAL-14 both direct the item as a statement: “I have nausea.”

In regard to content, the EORTC QLQ-C15-PAL mainly addresses a patient's physical issues, such as pain, weakness, or constipation. Conversely, the only item on the EORTC QLQ-C15-PAL that addresses psychosocial aspects of care is the item, “Did you feel depressed?” Although development of the EORTC QLQ-C15-PAL recognized the importance of social functions in palliative care, social function items in the EORTC QLQ-C30 did not work well in a palliative setting, as they asked patients to rate “normal” contact with family and friends, which was disrupted when a patient was in the hospital or hospice care.⁸ Accordingly, they were not included in the EORTC QLQ-C15-PAL.

Social and emotional supports are also less emphasized in the FACT G-7. As the FACT G-7 is intended as a rapid assessment instrument for advanced cancer patients who are receiving chemotherapy,⁸ the rapid assessment of symptom management may take priority over the assessments of social and emotional supports. This hypothesis is reflected in the fact that none of the seven total items in the FACT G-7 address social or emotional supports. Ultimately, the brevity of having only seven items in the FACT G-7 helps address patient burden for these advanced cancer patients.

In contrast, the FACIT-PAL-14 places greater emphasis on emotional and social support than the EORTC QLQ-C15-PAL and the FACT G-7. Most notably, the FACIT-PAL-14 examines items such as, “I get emotional support from my family” and “I am able to openly discuss my concerns with the people closest to me.” Whereas the EORTC QLQ-C15-PAL and the FACT G-7 focus more on health-related symptoms, the FACIT-PAL-14 also assesses social and emotional support needs, which can aid HCPs in involving the patient's family and communicating the patient's needs to family members.

In addition, it should be noted that each question in the EORTC QLQ-C15-PAL asks patients to rate solely negative concerns. On the other hand, the FACT G-7 and FACIT-PAL-14 may be advantageous in assisting patients to consider a more holistic view of their QoL by focusing on both positive aspects (e.g., “I am able to enjoy life”) and negative aspects (e.g., “I feel sad”) of their QoL by providing a balance of item types.

During the development of the FACIT-PAL-14, energy and emotional support items were both highly rated by patients and HCPs (32% of patients and 63% of HCPs listed energy as a top-10 item; 41% of patients and 27% of HCPs listed emotional support as a top-10 item).⁶ Both these scales are included in the FACIT-PAL-14 but not in the EORTC QLQ-C15-PAL. Although energy was included as an item in the FACT G-7, emotional support was not. Because QoL is a subjective measure by definition, patient responses are extremely important. Considering the high percentage of patients that rated energy and emotional support questions as top-10 questions of importance, developers of the EORTC QLQ-C15-PAL indicated that they planned to develop new items to address “role and social functioning” that were not found in the original unabridged instrument. However, to date, these changes have not been documented in the literature.

The raw 1 to 4 scores of the EORTC QLQ-C15-PAL are converted to a 0 to 100 scale. For symptoms, higher scores represent greater distress, but for functional scales and overall QoL, higher scores represent better functioning and overall QoL.⁹ The FACT-G7 raw 0 to 4 scores are summed on a scale of 0 to 28; higher scores correspond to better overall QoL. Finally, the FACIT-PAL-14 raw 0 to 4 scores are summed on a scale of 0 to 56. Again, higher scores indicate a better overall QoL.

Validity and reliability

Each of these three abridged questionnaires has improved content validity through a rigorous development process to include items directly nominated by cancer patients and their HCPs. They also elevate concurrent validity through selection of only core items well validated in the original unabridged versions. Additionally, multitrait analyses showed that the translated versions of the EORTC QLQ-C15 demonstrated good convergent and discriminant validity.^10–12

Other than the FACIT-PAL-14, for which reliability data have not been published, the other two shortened questionnaires have both demonstrated good internal consistency and scalability.^7,8 In terms of internal consistency, Cronbach's α ≥0.7 was deemed acceptable. Specific Cronbach's α and other relevant statistics are detailed below in sections “EORTC-QLQ-C15” and “FACT-G7.” Regarding scalability, Spearman correlations between single items were published for the EORTC-QLQ-C15, with Spearman correlation coefficients (CC) >0.4 being considered supportive of scalability. Specific subscale values are detailed in the section “EORTC-QLQ-C15.” Because QoL issues are subjective, patients' feelings and concerns are considered to be most relevant.⁷ With the original unabridged assessment instruments having been widely validated, the developers of the three abridged versions all aimed to obtain concurrent validity based on retaining the validity of the original instruments. To that end, each of the three abridged questionnaires has selected items straight from their respective original versions. For example, it is for this reason that the EORTC QLQ-C15-PAL has not added items relating to social, emotional, and spiritual issues, which were not present in the original EORTC-QLQ-C30.⁸ Essentially, each abridgement was a rigorous prioritization process involving appropriate patients and their HCPs to assess the most important QoL concerns to them based on the existing items from the unabridged instrument.

By having patients self-nominate their most important QoL issues and concerns, all three abridged questionnaires focused their content validity to an even shorter, representative list of concerns. For the sake of brevity, the FACT-G7 only has seven items that were all highly rated by patients. Although there may be fewer items, content validity may be affected only minimally, as the included items are all core items that are the most important in priority by the patients involved.

The FACIT-PAL-14 adopted all seven items of the FACT-G7 to maintain comparability and enhance content validity. Furthermore, seven additional items that were rated by >25% of patients as being top issues from the FACIT-PAL list of 46 items were also included. Specifically, the FACIT-PAL-14 includes two additional items, “I get emotional support from my family” and “I am openly able to discuss my concerns with the people closest to me,” which may provide a more expansive social and emotional scope to its assessment of QoL when compared with the FACT-G7. Future psychometric validation, however, is needed to confirm whether the FACIT-PAL-14 can act as a core instrument for palliative care assessments and outcome research.

There are other types of validation that are important to consider for each instrument. The following sections explore these relevant constructs for each abridged instrument. Due to differential data availability, the reporting of validity and reliability information has not been standardized.

EORTC QLQ-C15-PAL

Consistent validation of the EORTC QLQ-C15-PAL is documented in the literature across many countries, including Japan,¹⁰ Korea,¹¹ Poland,¹³ and Mexico.¹²

Single item and multiscaling analyses performed in Mexico by Suarez-del-Real and colleagues demonstrated good convergent and discriminant validity of the EORTC QLQ-C15-PAL.¹² Reliability and clinical validity of the questionnaire were also demonstrated.

Evidence for scalability was obtained with the Spearman CC, with correlation ≥0.4 being considered highly significant.¹² Better global health/QoL scale scores were associated with better physical function, less fatigue, less pain, and better appetite (CC 0.46, 0.43, 0.36, 0.37, respectively; p<0.001 for all correlations). Multi-item correlation was evaluated through Cronbach's α, with a correlation ≥0.7 considered acceptable.¹⁴ Three of the four multi-item scales (physical=0.88, emotional=0.74, and pain=0.77) showed Cronbach's α values of >0.7, indicating strong overall correlation.¹² Although the fourth multi-item scale, the fatigue scale, only had an accompanying α=0.67, investigators in a Korean study noted that even this coefficient should be acceptable, because it may reflect a reduction of items in the scale.¹¹

Reliability was also evaluated by comparing the difference of scale scores over time. Some scores showed consistency, whereas others showed improvement or deterioration, which could be due to changing physical conditions of patients participating in the re-test.¹²

FACT-G7

A study by Yanez and coworkers determined the reliability of the FACT-G7.⁷ Oncology patients who helped develop the FACT-G7 via patient surveys were also required to complete the FACT-G, the single-item Eastern Cooperative Oncology Group Performance Status Rating (ECOG PSR), and the EuroQol group's five-item health status measure (EQ-5D). In a separate study by Butt el al., an Internet panel of 2000 individuals from the United States was recruited to obtain general population reference values for the rapid index.¹⁵ Online panelists were provided with a link to a secure website and were asked selected FACT items.¹⁵

Cronbach's α was calculated in both the oncology patient sample and the general population sample for the items of the FACT-G7. Cronbach's α of 0.74 and 0.80 for the oncology patient sample and the general U.S. population sample demonstrated good overall internal consistency and reliability.⁷ Again, as with the QLQ-C15-PAL responses, Cronbach's α ≥0.7 was considered an acceptable correlation. For the oncology patient sample, the FACT-G7 items were highly correlated with the physical well-being (PWB) and functional well-being (FWB) subscales of the FACT-G (CCs of 0.80 and 0.82, respectively), moderately correlated with the emotional well-being (EWB) subscale and the EQ-5D index (CCs of 0.62 and 0.60, respectively), and weakly correlated with the social well-being (SWB) subscale (CC of 0.29).⁷ The moderate correlation with EWB and weak correlation with SWB are expected and convey discriminant validity. Specifically, of the seven items of the FACT-G7, there is only one EWB item (“I worry that my condition will get worse”) and no SWB items.⁷

Validity of the FACT-G7 was further supported through examination of its relationship between performance status classification and the ECOG PSR, which is used to assess how a patient's disease is progressing and its effect on the daily living abilities of the patient. The FACT-G and FACT-G7 followed similar trends in successfully distinguishing across ECOG PSR groups: [F (2, 256)=91.16, p<0.001] for the FACT-G, [F (2530)=87.43, p<0.001] for the FACT-G7 among the National Comprehensive Cancer Network (NCCN) oncology patient samples, and [F (21,997)=331.47, p<0.001] for the FACT-G7 among the general population sample.⁷

FACIT-PAL-14

At the time of data collection, there were no published validation findings for the FACIT-PAL-14, presumably due to the recent (2013) development of the scale.

Responsiveness and clinical utility

Responsiveness and clinical utility are important features of an instrument. These properties are used for both assessing the strength of the questionnaire but also for the selection of a questionnaire that will be useful in clinical practice.

All of the abridged questionnaires retain their face validity because all items from the abridged questionnaires are taken from their respective longer questionnaires. Concurrent validity is likely to be comparable in the abridged questionnaires, but this remains to be tested. Most importantly, because the abbreviated item sets were selected by patients in terms of their perception of relevance and importance, the abridged questionnaires are likely to reflect the most important needs of the patients. Future research is needed to confirm this hypothesis. Indeed, determination of responsiveness is important as responsiveness can be seen as a type of longitudinal validity; an instrument with good responsiveness should be able to accurately detect changes in clinical status of patients.¹⁶

Because the EORTC-QLQ-C15 and the FACT-G7 both exhibit Cronbach's α >0.7, they demonstrate similar internal consistency compared with their longer instrument counterparts. Cronbach's α values are not yet available for the FACIT-PAL-14.

Conclusion

All three abridged questionnaires are potentially useful to the field of palliative cancer care, due to their brevity and their link to longer, well-established parent instruments. By selecting only items from well-validated original questionnaires, these abridged questionnaires may retain much of their content and concurrent validity. By asking eligible cancer patients to prioritize their most important concerns from previous standardized instruments, these abridged questionnaires establish themselves as appropriate, relevant, core assessment tools in their respective intended patient populations. Both the FACT-G7 and the EORTC QLQ-C15-PAL have undergone appropriate validation that demonstrated good internal consistency and reliability; validation results for the FACIT-PAL-14 are pending.

The three shortened questionnaires were designed for slightly different purposes, with each having its own particular characteristics. With only 7 items, the FACT-G7 has the fewest number of items of all three instruments. Such brevity enables it to serve as a rapid assessment of QoL issues for advanced cancer patients undergoing chemotherapy. In contrast, the 15-item EORTC QLQ-C15-PAL is intended for cancer patients solely under palliative care. Because the original unabridged version, the EORTC QLQ-C30, does not have certain items deemed important for palliative care, the developers of the EORTC QLQ-C15-PAL indicated they planned to develop new items, which had not been documented in the literature at the time of data collection for this study. The FACIT-PAL-14 is also intended for cancer patients under palliative care. It has included all 7 items from the FACT-G7 and another 7 items that cover further QoL issues, such as emotional and social supports, which were less emphasized in the other two abridged questionnaires. Conceptually, the FACIT-PAL-14 shows promise to be an effective QoL assessment tool for palliative care patients without significant patient burden; however, published validation findings in the near future are warranted.

Footnotes

Acknowledgments

We thank the generous support of the Bratty Family Fund, Michael and Karyn Goldstein Cancer Research Fund, Joseph and Silvana Melara Cancer Research Fund, and the Ofelia Cancer Research Fund.

Author Disclosure Statement

Dr. Andrew Bottomley is an author of the EORTC Quality of Life Group (QLG) measurement system. These tools are provided free to academics, but the EORTC charges a user fee for the use of the EORTC QLQ-C30 in industry-sponsored research to cover the cost of future psychometric validation, testing, and translation of health-related (HR) QoL tools. David Cella is president of FACIT.org, which licenses all FACT and FACIT tools for use in international pharmaceutical clinical trials.

References

Aaronson

, Ahmedzai

, Bergman

et al.: The European organization for research and treatment of cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Instit, 1993; 85:365–376.

Chow

, Hoskin

, van der Linden

, Bottomley

, Velikova

: Quality of life and symptom end points in palliative bone metastases trials. Clin Oncol (R Coll Radiol), 2006; 18:67–69.

Luckett

, King

, Butow

, Oguchi

, Rankin

et al.: Choosing between the EORTC QLQ-C30 and FACT-G for measuring health-related quality of life in cancer clinical research: issues, evidence and recommendations. Ann Oncol, 2011; 22:2179–2190.

Greisinger

, Lorimor

, Aday

, Winn

, Baile

: Terminally ill cancer patients. Their most important concerns. Cancer Pract, 1997; 5:147–154.

Lyons

, Bakitas

, Hegel

, Hanscom

, Hull

, Ahles

: Reliability and validity of the Functional Assessment of Chronic Illness Therapy-Palliative Care (FACIT-Pal) Scale. J Pain Symptom Manage, 2009; 37:23–32.

Zeng

, Bedard

, Cella

, Thavarajah

, Chen

, Zhang

, et al.: Preliminary results of the generation of a shortened quality-of-life assessment for patients with advanced cancer: The FACIT-PAL-14. J Palliat Med, 2013; 16:509–515.

Yanez

, Pearman

, Lis

, Beaumont

, Cella

: The FACT-G7: A rapid version of the functional assessment of cancer therapy-general (FACT-G) for monitoring symptoms and concerns in oncology practice and research. Ann Oncol, 2012; 24:1073–1078.

Groenvold

, Petersen

, Aaronson

, Arraras

, Blazeby

, Bottomley

, et al.: The development of the EORTC QLQ-C15-PAL: A shortened questionnaire for cancer patients in palliative care. Eur J Cancer, 2006; 42:55–64.

Lien

, Zeng

, Nguyen

, Cramarossa

, Culleton

, Caissie

, Lutz

, Chow

: Comparison of the EORTC QLQ-C15-PAL and the FACIT-PAL for assessment of quality of life in patients with advanced cancer. Expert Rev Pharmacoecon Outcomes Res, 2011; 5:541–547.

10.

Miyazaki

, Suzukamo

, Shimozuma

, Nakayama

: Verification of the psychometric properties of the Japanese version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 15 palliative (EORTC QLQ-C15-PAL). Qual Life Res, 2012; 21:335–340.

11.

Shin

, Choi

, Miyashita

, Choi

, Kang

, Baik

, et al.: Cross-cultural application of the Korean version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 15-Palliative Care. J Pain Symptom Manage, 2011; 41:478–484.

12.

Suarez-del-Real

, Allende-Perez

, Alferez-Mancera

, Rodriguez

, Jimenez-Toxtle

, Mohar

, Onate-Ocana

: Validation of the Mexican-Spanish version of the EORTC QLQ-C15-PAL questionnaire for the evaluation of health-related quality of life in patients on palliative care. Psychooncology, 2011; 20:889–896.

13.

Leppert

, Majkowicz

: Validation of the Polish version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire - Core 15 - Palliative Care in patients with advanced cancer. Palliat Med, 2013; 27:470.

14.

Fayers

, Machin

: Quality of Life. The Assessment, Analysis and Interpretation of Patient Reported Outcomes, 2nd ed. Chichester, England: John Wiley and Sons Ltd, 2007.

15.

Butt

, Peipert

, Webster

, et al.: General population norms for the Functional Assessment of Cancer Therapy-Kidney Symptom Index (FKSI). Cancer, 2013; 119:429–437.

16.

De Vet

HCW

, Beurskens

AJHM

, Bezemer

, Bouter

: Reproducibility and responsiveness of evaluative outcome measures. Theoretical considerations illustrated by an empirical example. Int J Health Technol Assess Health Care, 2001; 17:479–487.