Elsevier

Lung Cancer

Volume 57, Issue 3, September 2007, Pages 339-347
Lung Cancer

Validation of the Functional Assessment of Cancer Therapy—Lung Symptom Index-12 (FLSI-12)

https://doi.org/10.1016/j.lungcan.2007.03.021Get rights and content

Summary

We tested the reliability and validity of a brief symptom index for use with patients in the advanced stages of lung cancer. The Functional Assessment of Cancer Therapy—Lung Symptom Index-12 (FLSI-12) is a brief self-report measure that combines seven items addressing symptoms common in advanced-stage lung cancer with five symptoms or concerns that are relevant to most people with advanced-stage cancer. The index was administered prospectively to 92 advanced-stage lung cancer patients beginning at the initiation of chemotherapy and for 12 consecutive weeks. Reliability, convergent and concurrent validities, and responsiveness to change were determined and a minimally important difference (MID) was estimated. The index had good internal consistency (all Cronbach's α's > 0.70), moderate to high item-to-total correlations (93% ρ's  0.30), and correlated highly with a measure of overall quality of life (ρ's  0.50). Baseline scores differentiated patients with better versus worse clinical features (p's < .05). Prospective changes in index scores were sensitive to changes in performance status ratings (p's < .05). An MID of 3–4 points was estimated by combining guideline-, distribution-, and anchor-based methods. The results show that the FLSI-12 is a psychometrically sound measure and support its use as an endpoint in clinical trials of advanced-stage lung cancer.

Introduction

The assessment of patient-reported outcomes (PROs) such as symptoms of disease, complications of therapy, and other aspects of health-related quality of life (HRQL) are becoming an integral part of clinical trials of advanced lung cancer. Proper interpretation of clinical trials in advanced cancer often requires assessment of HRQL, including disease-related symptoms. Recently, regulatory authorities have expressed cautious interest in such patient-reported outcomes, with concern expressed regarding the interpretability and relevance of scores from multi-domain instruments. Focused symptom measures have been suggested [1].

Clinical researchers now have a number of standardized assessment options available to them. Two comprehensive measurement systems, that of the European Organization for Research and Treatment of Cancer (EORTC) and the Functional Assessment of Cancer Therapy (FACT), have developed modular approaches to assessing PROs in various cancer sites. The EORTCs core QLQ-C30 [2] is a general cancer measure that is often augmented by the disease-specific LC13 [3] in lung cancer clinical trials. Likewise, the FACT-General measure (FACT-G) [4] is often administered with the disease-specific Lung Cancer subscale (LCS) in lung cancer clinical trials. Together the FACT-G and LCS make up the total FACT-Lung (FACT-L) [5]. Both the EORTC and FACT-Lung modules provide comprehensive assessments of lung cancer symptoms, treatment-related complications (i.e., toxicities), and other functional aspects of HRQL. Both have also been frequently used in lung cancer clinical trials [6], [7] including those of novel targeted therapeutic agents [8]. However, regulatory agencies such as the US Food and Drug Administration (FDA) have grappled with the interpretability of multidimensional HRQL measures as they relate to approvable product label terminology. This concern has led the FDA to draft a guidance document recommending among other things the use of simpler, more targeted symptom assessments, especially in instances where a symptom-focused label claim is ultimately desired for a medical product [1]. Given this, there may be merit in developing and validating simplified assessments of advanced lung cancer symptoms for use in clinical research.

Targeted measures of lung cancer symptoms have been used in clinical lung cancer studies. The Lung Cancer Symptom Scale (LCSS) [9] is a brief measure that focuses entirely on common symptoms of lung cancer including: appetite loss, fatigue, cough, dyspnea, pain, hemoptysis, symptom distress, activity level limitations, and overall quality of life. While psychometrically sound and easy to administer, the LCSS contains no direct measurement of side effects or other complications associated with treatment. Minimally important score differences are also currently unavailable for this measure. Another brief and targeted measure is the previously mentioned LCS of the FACT-L [5]. It too focuses exclusively on symptoms of lung cancer namely: breathing difficulty, weight loss, clarity of thought, coughing, appetite loss, and tightness in chest. It is also psychometrically sound and easy to administer. A minimally important score difference has been estimated for the LCS [10] making it particularly attractive for the clinical trials setting.

While certainly useful as a stand-alone measure, the utility of the LCS could perhaps be enhanced with the addition of a finite set of lung cancer relevant items. In a recent survey of physicians and nurses at 17-member institutions of the National Comprehensive Cancer Network (NCCN), several “priority” symptoms were identified as being important targets of therapy across a range of advanced cancers including lung [11]. Among those issues not currently represented in the LCS were fatigue, pain, nausea, and contentment with quality of life. Furthermore, since documenting patient tolerability to treatment can be as important as documenting treatment efficacy, there may also be added value in incorporating an overall assessment of treatment side effects. Items assessing all of these issues are available in the “core” FACT-G measure. Hence, our interest in this analysis was to take the seven items of the base LCS measure and add five items from the FACT-G that address the expert-derived symptom targets and attention to overall side effect burden. We tested this FACT Lung Symptom Index-12 (FLSI-12) in a sample of advanced lung cancer patients undergoing treatment. Our primary objective was to test the psychometric utility of the revised measure (i.e., reliability, validity, and responsiveness to change over time). A secondary objective was to estimate the minimally important difference (MID) of the scale.

Section snippets

Patients and procedure

From December 2002 through 2004, we enrolled 92 advanced-stage lung cancer patients into a prospective observational study of patient-reported health status. All patients were recruited from the Kellogg Cancer Care Center (KCCC) of Evanston Northwestern Healthcare (Evanston, Highland Park, and Glenbrook Hospitals). Eligibility criteria included: (1) newly diagnosed or recurrent stages III or IV non-small cell lung cancer (NSCLC) or extensive-stage small cell lung cancer (SCLC); (2) initiation

Sample description

Baseline socio-demographic and clinical characteristics of the sample are shown in Table 3. One of the 92 lung cancer patients enrolled was subsequently deemed ineligible; hence data from 91 patients were analyzed. Median age of patients was 70.5 years, most were white (90%) and female (54%). The sample was largely diagnosed with stage IV NSCLC. Median time since diagnosis was a little over 1 month. Most patients rated themselves as having good performance status (PS) at the study outset (80%

Discussion

Our expanded version of the original LCS-7, the FLSI-12, performed quite well in a sample of advanced-stage lung cancer patients followed prospectively after treatment. The index had excellent internal reliability, showed convergence with physical and functional domains of HRQL, and demonstrated concurrent validity. Baseline index scores were able to differentiate patient groups with better versus worse health status or prognostic features. The FLSI-12 was also sensitive to changes in clinical

Conflict of interest

None declared.

Acknowledgements

Funding for this manuscript was provided by Bristol Myers Squibb Company. The sponsor had no role in the study design, collection, analysis or interpretation of data, drafting of the manuscript, or decision to submit the manuscript for publication.

References (27)

  • H. Langendijk et al.

    The prognostic impact of quality of life assessed with the EORTC QLQ-C30 in inoperable non-small cell lung carcinoma treated with radiotherapy

    Radiother Oncol

    (2000)
  • C.-H. Chang et al.

    Real-time clinical application of quality-of-life assessment in advanced lung cancer

    Clin Lung Cancer

    (2002)
  • Center for Drug Evaluation and Research. Draft guidance for industry patient-reported outcome measures: use in medical...
  • Cited by (15)

    • A randomized trial of weekly symptom telemonitoring in advanced lung cancer

      2014, Journal of Pain and Symptom Management
      Citation Excerpt :

      These AUC analyses indicated no difference in symptom burden between the groups for the total score (P = 0.246) or PSI (P = 0.563), again with effect sizes of one-tenth of SD or less. To examine potential responders and nonresponders to symptom management, change scores from baseline were calculated for the 12 FLSI time points and classified as improved, worsened or unchanged based on the four-point minimally important difference.80 The proportion of participants in each change group was compared between groups at each week using Mantel–Haenszel Chi-square tests for ordinal data.

    • An investigation of symptom burden and quality of life in Chinese chemo-naïve advanced lung cancer patients by using the Instrument-Cloud QOL System

      2014, Lung Cancer
      Citation Excerpt :

      All QOL assessment was conducted before the anti-cancer treatment. The disease-related symptoms in lung cancer were quantified using seven items in LCS and two items in FACT-G, which were all included in FACT-Lung Symptom Index-12 (FLSI-12) [14,15]. The nine symptoms were shortness of breath, weight loss, unclear thinking, cough, appetite loss, tightness in chest, breathing difficulty, pain and lack of energy (supplementary Table 1).

    • Interdisciplinary palliative care intervention in metastatic non-small-cell lung cancer

      2013, Clinical Lung Cancer
      Citation Excerpt :

      This tool is composed of 5 subscales: physical, social/family, emotional, functional well-being, and the lung cancer symptom index. Each item is rated on a 5-point Likert scale.35-37 The Functional Assessment of Chronic Illness Therapy-Spirituality Subscale is a 12-item tool that assesses the spiritual well-being of patients with cancer using a 5-point Likert scale.

    • Mathematical coupling may account for the association between baseline severity and minimally important difference values

      2010, Journal of Clinical Epidemiology
      Citation Excerpt :

      In an example of the clinically driven approach, Eton et al. used the Eastern Cooperative Oncology Group Performance Status Rating scale to identify three types of patients: those who improved, those who were stable, and those who deteriorated over time. MID values for the Functional Assessment of Cancer Therapy–Lung Symptom Index-12 (FACT-LSI12) were then produced by calculating mean change scores on the FACT-LSI12 for each of the three groups [8]. The use of MID reference groups defined according to clinical measures has been criticized for prioritizing criteria that have no known relationship to the patient experience [9].

    View all citing articles on Scopus
    View full text