Development of the Anemia Impact Measure (AIM): A Disease-Specific Patient Reported Outcome (PRO) Instrument to Measure Anemia Symptoms and Their Impact on Functioning in Cancer Patients Receiving Chemotherapy

Introduction: Several patient based measures are available to assess symptom experience in cancer patients with anemia due to myelosuppressive chemotherapy. However, existing measures have not been based on extensive patient qualitative data, as recommended in the FDA draft guidance on PRO measures. We developed a measure of anemia symptoms and the impact of these symptoms on daily functioning based on a conceptual framework, which outlined relevant concepts and domains (groupings of items and concepts) using extensive input from adult cancer patients with chemotherapy-induced anemia (CIA).

Methods: The initial conceptual framework was based on a literature review, expert input from a PRO working group, and interviews with community oncologists and oncology nurse practitioners. Six focus groups and 24 interviews were conducted in 7 sites across the U.S. using 44 patients with breast, colorectal, small cell or non-small cell lung cancer, non-Hodgkin’s Lymphoma, or multiple myeloma, and CIA (hemoglobin [Hb] < 11 g/dL within 2 weeks of enrollment, with documented anemia symptoms). Ten additional interviews were conducted in patients receiving chemotherapy with no history of CIA and those successfully treated for CIA to explore differences between symptoms unique to CIA and those due to chemotherapy. Semi-structured concept elicitation guides with open-ended questions were based on the literature and informed by the PRO working group and clinicians. A smaller focus group size of 4 patients on average enabled individual questions and group discussion. Patients ranked symptoms and their impact in order of importance at the end of the session. Qualitative coding and analysis were used to identify key concepts employing a standard software program, Atlas.ti 5.0. Preliminary item generation of the Anemia Impact Measure (AIM) was based on content analysis of transcripts and informed by expert input. Cognitive interviews with 16 additional CIA patients were then conducted to evaluate patients’ understanding of the questionnaire. The AIM was reviewed by the PRO working group and clinicians for content validity. Modifications and reductions were based primarily on cognitive interviews and also on content validity assessment.

Results: Most CIA patients were white (76.1%) and female (82.6%), and the average age was 60 years (range 39–69 years). Most patients had breast cancer (54.4%) or lung cancer (17.4%). Average Hb was 10.7 g/dL (SD 0.5, range 8.6–10.9 g/dL). Tiredness was rated as the most important symptom of anemia by 83% of all CIA patients. Weakness, shortness of breath, and symptoms associated with lightheadedness and dizziness were among the symptoms ranked next in order of importance by CIA patients, but were not reported by non-CIA patients. Consistent with the conceptual framework and in-depth patient qualitative data, the final AIM contains:

1. a daily diary measuring relevant CIA symptoms (9 items measuring tiredness, getting tired easily, weakness, exhaustion, lack of energy, dizziness, lightheadedness, and breathlessness with exertion, including one item measuring intensity of worst tiredness, and

2. a weekly questionnaire measuring relevant impacts of CIA on functioning (29 items covering the following four domains: activities of daily living, social/leisure activities, cognitive function, and emotional function).

All items in the AIM have a 5-point Likert-type response options ranging from 1 (not at all) to 5 (extremely) except one item measuring intensity of tiredness which uses a numeric rating scale from 0 to 10. Content validity ratings from 17 clinicians including those on the PRO working group indicated that all items included in the AIM were relevant for CIA patients.

Conclusion: The AIM assesses important patient-perceived symptoms of CIA and their impact on functioning and was developed using extensive patient input. Cognitive interviews and content validity assessment found that the AIM is relevant, easy to understand, and complete. Future studies are planned to evaluate the reliability, validity, and responsiveness of the AIM in patients with CIA.