Using SMART design to improve symptom management among cancer patients: A study protocol

    loading  Checking for direct PDF access through Ovid


In clinical practice, a clinician selects treatments based on the available evidence. If the initial treatment does not work, clinical logic includes allowing more time, adding supplemental modalities, or switching to different treatments. However, these decisions are often not evidence‐based, and the decision rules may be implicit and difficult to replicate (Song, DeVito Dabbs, & Ward, 2016). The Precision Medicine initiative is designed to overcome the “one size fits all” approach to health care and identify treatments best suited for individuals (National Institutes of Health, 2015). The National Institutes of Health (NIH) has encouraged use of novel research designs, such as the sequential multiple assignment randomized trial (SMART), which includes sequences of treatments as opposed to single fixed treatments (Collins, Murphy, & Bierman, 2004). The dynamic model of sequencing alternative treatments depending on observed success is ideally suited for the temporal and concurrent nature, and varying etiologies of multiple symptoms that present complex challenges to symptom management science (Kroenke, 2001).
The burden of symptoms resulting from cancer and its treatment contribute to diminished health related quality of life (HRQOL), as has been well‐documented (Badger, Segrin, & Meek, 2011; Brant, 2016; Cleeland et al., 2013). Existing static symptom management interventions deliver a predetermined dose at specific intervals and are tested against controls in standard randomized controlled trials (RCTs). Although overall efficacy of an intervention may be established, heterogeneity may still exist in patient responses, and moderators of treatment outcomes may be identified that define groups of patients who benefit from interventions differentially (Kraemer, Wilson, Fairburn, & Agras, 2002; Sikorskii et al., 2015). While the identification of moderators is one step toward accounting for heterogeneity, the next step of intervention sequencing and tailoring is needed to advance intervention science (Knobf et al., 2015) and the science of cancer symptom management in particular. In this SMART design approach, we rigorously test the adjustment of intervention type and/or duration through sequencing that is based on patient response.
    loading  Loading Related Articles