Date Published: January 31, 2019
Publisher: Public Library of Science
Author(s): Huibrie C. Pieters, Emily Green, Sally Khakshooy, Miriam Sleven, Annette L. Stanton, M Barton Laws.
It remains unclear how information about aromatase inhibitors (AI) impacts women’s decision-making about persistence with endocrine therapy.
To describe and compare how women treated for primary early stage breast cancer either persisting or not persisting with an AI received, interpreted, and acted upon AI-related information.
Thematic analysis was used to sort and compare the data into the most salient themes.
Women (N = 54; 27 persisting, 27 not persisting with an AI) aged 65–93 years took part in qualitative interviews.
Women in both subgroups described information similarly in terms of its value, volume, type, and source. Aspects of AI-related information that either differed between the subgroups or were misunderstood by one or both subgroups included: (1) knowledge of AI or tamoxifen prior to cancer diagnosis, (2) use of online resources, (3) misconceptions about estrogen, hormone replacement therapies and AI-related symptoms, and (4) risk perception and the meaning and use of recurrence statistics such as Oncotype DX.
Persisters and nonpersisters were similar in their desire for more information about potential side effects and symptom management at AI prescription and subsequent appointments. Differences included how information was obtained and interpreted. Interactive discussion questions are shared that can incorporate these findings into clinical settings.
In the US population, breast cancer is increasingly common in older women, with the median age of diagnosis at 61 years . The majority of breast cancers (60%–70%) express the estrogen or progesterone receptor or both. Therefore, following primary treatments, an endocrine therapy such as an aromatase inhibitor (AI) or tamoxifen is the standard of care for postmenopausal women with hormone receptor–positive breast cancer. In 2000, the National Institutes of Health consensus conference recommended 5 years of adjuvant tamoxifen for women with hormone receptor-positive tumors larger than 1 cm . Ensuing studies recommended the use of AIs for 5 years in postmenopausal women, with further studies suggesting endocrine therapy for up to 10 years in certain situations . However, despite the efficacy of AIs in reducing the risk of cancer recurrence, rates of discontinuation increase over time from 90% persisting at 1 year to only 50% at 5 years . A systematic review  has shown a mean of only 79% at 1 year and 56% at 5 years. Our work is focused on persistence, defined as the duration from initiation to discontinuation of therapy , in contrast to adherence which reflects taking the correct dose according to frequency . Nonpersistence rates appear to be especially high in older adults [6, 8–12], though findings are mixed .
Grounded theory informed by a constructivist worldview guided all aspects of this research. The philosophical tenets that underlie constructivist grounded theory, symbolic interactionism and pragmatism, are deeply steeped in how knowledge is perceived, understood and retained in ongoing interactions with self and others .
The 54 women were a mean age of 71.9 years (range = 65–93) at diagnosis. Half the sample was persisting with an AI at the time of interview, and half had prematurely discontinued the treatment. Non-persistence was intentional and no participant identified a comorbid condition, medication management, or financial barriers as a reason to discontinue endocrine therapy. Most participants self-identified as non-Hispanic white (n = 44) and were at least college graduates (n = 38). (See Tables 2 and 3 for sample characteristics).
As shown in recent randomized clinical trials targeting AI adherence [17–21], interventions that rely primarily on information strategies do not prevent discontinuation. With an aim to better understand why information strategies do not increase adherence and how they can be improved, here we present the complex and dynamic nuances of gathering and processing treatment-related information found in a sample of 27 women persisting with an AI and 27 intentionally not persisting. While this study was not intended to statistically compare the subgroups, the subjective perspectives bring novel insights for future quantitative research.