Avoidable pitfalls in behavioral medicine outcome research

Abstract

To secure the role of behavioral medicine in health care, researchers continue to improve the quality of their outcome studies. Despite the availability of guidelines for designing high quality clinical trials, however, we have noted two, unfortunately common, flaws in behavioral medicine outcome research that undermine these efforts. The first issue is that researchers recruit medical patients whose scores on psychological target measures are not elevated at pretest. Data are presented from quantitative reviews of cardiovascular and cancer populations to illustrate the impact of this protocol decision. It is demonstrated how magnitude of change and corresponding statistical power are greatly reduced when patients with few problems are enrolled. The second issue pertains to the failure of researchers to measure psychological change when the actual model to be tested is a mediational model such that successful treatment of psychological distress is presumed to account for good long-term health outcomes. Such lack of attention to protocol design can result in misinterpretation of obtained effects and can lead to premature dismissal of psychological treatment opportunities for physical disease. We suggest how these flaws can be avoided in the protocol design stage.

Course of Distress in Breast Cancer Patients, Their Partners, and Matched Control Couples

“…The finding that differences in distress between patients and comparison-control women are small and decrease over time also indicates that interventions aimed at alleviating distress would require a very large number of unselected patients to detect a treatment effect relative to the decrease in distress that is likely to occur without a formal intervention. Therefore, instead of treating all patients, it may be much more effective to pre-select patients with elevated distress scores, consistent with recent recommendations [43,44]. The few studies to date that do find a treatment by group effect [45,46] tended to pre-select patients with elevated distress scores.…”

Impact of Mindfulness-Based Stress Reduction (MBSR) on attention, rumination and resting blood pressure in women with cancer: A waitlist-controlled study

“…Finally, reductions in SBP were observed only for participants with relatively higher SBP levels at the outset. Recruiting patients with elevated casual BP would have prevented the potential for a floor effect and increased power to detect intervention-related change in BP (Linden & Satin, 2007).…”

“…Given the potential of a floor effect leading to underestimation of BP reductions after the intervention (Fillingim et al, 1998;Linden & Satin, 2007), participants were assigned to ''Higher BP'' or ''Lower BP'' groups based on a median split conducted on week 1 casual SBP. Next, two separate ANCOVAs were run with group (MBSR vs. Control) and condition (Higher BP vs. Lower BP) as between-subjects factors, week 8 SBP and DBP as dependent variables, and week 1 SBP or DBP and age as covariates.…”

Intimate Relationships Affected by Breast Cancer: Interventions for Couples

“…Given that most interventions were offered to all patients regardless of their distress level, floor effects occur [38]. In addition, there is a strong association between baseline and outcome distress [41]. Badr and Krebs [19] suggest the identification of profiles of couples at risk who may benefit from couple-based interventions.…”