We read with great interest the recent article by Piatt et al. (1). The implementation of intervention strategies that effectively close the gap between evidence-based treatment goals and clinical outcomes remains a challenge in daily practice, particularly in the management of complex diseases like diabetes. The Task Force on Community Preventive Services of the Center for Disease Control and Prevention found strong evidence supporting disease and case management interventions for diabetes (2), many of which constitute examples of the essential elements of the chronic care model (CCM) (3). However, the optimal components of disease management interventions and the effectiveness of multiple- versus single-component CCM interventions in diabetes are not yet established.

In their study, Piatt et al. illustrate how the six components of the CCM may be incorporated in the design of a diabetes intervention program, but the outcomes of the intervention strategies under investigation (CCM versus provider only versus usual care) remain unclear and cannot be evaluated without adequately correcting for differences between groups. Due to the study’s small sample size, randomization of the 11 practices and its 119 patients into three groups fails to correct for group nonequivalency. Particuarly, differences in ethnicity, insulin use, baseline HbA1c, cholesterol levels, and Diabetes Knowledge Test scores need to be accounted for. For example, preliminary calculations for differences in insulin use at baseline (χ2 without controlling for nesting, which would require a larger sample) indicate that at least 360 patients would be needed to have a power of 0.8 to detect a 15% difference between groups.

Propensity scores (4) allow investigators to correct for differences in baseline characteristics among study groups when randomization is not an option. This methodological technique derives a coarse balancing factor from the collection of baseline characteristics that enables meaningful direct comparisons among study groups. Propensity scores could be used to achieve a less biased estimate of the treatment effects of the CCM intervention when compared with provider-only intervention and with usual care in the study by Piatt et al. We look forward to the results of this study using this statistical approach and to its contribution to the question on the effect of multiple- versus single-component CCM interventions on diabetes outcomes in primary care.

1.
Piatt GA, Orchard TJ, Emerson S, Simmons D, Songer TJ, Brooks MM, Korytkowski M, Siminerio LM, Ahmad U, Zgibor JC: Translating the chronic care model into the community: results from a randomized controlled trial of a multifaceted diabetes care intervention.
Diabetes Care
29
:
811
–817,
2006
2.
Centers for Disease Control and Prevention: Strategies for reducing morbidity and mortality from diabetes through health-care system interventions and diabetes self-management education in community settings: a report on recommendations of the Task Force on Community Preventive Services.
MMWR Recomm Rep
50
:
1
–24,
2001
3.
Bodenheimer T, Wagner EH, Grumbach K: Improving primary care for patients with chronic illness.
JAMA
288
:
1775
–1779,
2002
4.
Rosenbaum PR, Rubin DB: The central role of the propensity score in observational studies for causal effects.
Biometrika
70
:
41
–55,
1983