Multilevel intervention research holds the promise of more accurately representing real-life situations than single-level intervention research and, thus, with the proper research design and measures, being more likely to facilitate the effective and efficient resolution of complex health-care systems problems (1-3). Taking a multilevel approach to cancer care interventions, however, creates measurement challenges and opportunities. Some measurement issues are similar to those arising in single-level interventions, whereas some are unique to multilevel interventions (4,5). The purpose of this article is to examine the state of measurement in multilevel cancer intervention research and to suggest how opportunities in the field may be addressed by drawing on the measurement literature in other fields.One of the unique aspects of measurement in multilevel intervention research is that consideration of within and between level effects is needed. For multilevel interventions, it is not sufficient to measure effects at the different levels of intervention (eg, patient and healthcare team), cross-level effects must also be taken into consideration (eg, the effect of patients on the team and the team on the patients). Furthermore, some measurement issues may be more complex when conducting multilevel intervention research, and some measurement rules of thumb may be uniquely associated with specific levels of analysis (eg, group or organization levels of analysis) (6,7). Because multilevel interventions are based on systems theory [whether general systems (8-10), social ecological systems (11,12), or complex adaptive systems (13) models], they require identification of key constructs and measures by level and consideration of interactions within and across levels (6). Thus, multilevel intervention research benefits from thoughtful planning and design, an interdisciplinary approach, mixed methods (eg, a combination of objective factors, surveys, interviews, observations) measurement and analysis, and careful attention to the congruence of theory, constructs, and measures.Through a systematic review of the literature, this article examines the state of measurement in multilevel cancer care intervention research, identifies measurement opportunities in the field, and draws upon other fields to provide direction about measures and measurement to multilevel cancer care intervention researchers.
Literature ReviewWe conducted a systematic literature review focused on cancer care intervention research and measures published in the past 5 years. This contrasts with Stange et al. in this monograph (14), which focuses on the broad spectrum of health care and prevention research over a 10-year period (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010). Thus, our review provides a comprehensive overview of a more recent set of cancerspecific intervention studies.