Am J Health Behav.™ 2007;31(Suppl 1):S19-S26 S19 The Causal Pathways Linking Health Literacy to Health Outcomes Michael K. Paasche-Orlow, MD, MA, MPH; Michael S. Wolf, PhD, MPH Michael K. Paasche-Orlow, Assistant Profes- sor of Medicine, Section of General Internal Medi- cine, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts and Boston University Medical Center, Boston, MA. Michael S. Wolf, Assistant Professor and Direc- tor, Health Literacy and Learning Program, Insti- tute for Healthcare Studies, Feinberg School of Medicine, Northwestern University, Chicago, IL. Address correspondence to Dr Paasche-Orlow, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts and Boston University Medical Center, 91 East Concord Street, Suite 200, Boston, MA 02118. E-mail: mpo@bu.edu Objective: To provide an evidence- based review of plausible causal pathways that could best explain well-established associations be- tween limited health literacy and health outcomes. Methods : Through analysis of current find- ings in medical and public health literature on health literacy we derived a conceptual causal model. Results: Health literacy should be viewed as both a patient and a system phenomenon. Three dis- tinct points along a continuum of health care are suggested to be influenced by health literacy: (1) access and utilization of health care, (2) patient-provider relation- ship, and (3) self-care. Conclusions: The conceptual model organizes what has been learned to date and underscores promising areas of fu- ture inquiry and intervention. Key words: causal pathways, health literacy, literacy, concep- tual model Am J Health Behav. 2007;31(Suppl 1):S19-S26 T hough limited health literacy has been shown to be associated with worse health outcomes, 1 the causal pathways are not entirely known, 2,3 and several projects have focused on explain- ing potential mechanisms. 3-8 This paper provides a review of the current evidence and proposes a conceptual model describ- ing the systemic, interactional, and self- care mechanisms by which limited health literacy is most likely to lead to worse health outcomes. By doing so, we begin to highlight some of the most promising areas for intervention research as well as important gaps in our current under- standing of the pathways linking literacy and health. This is one step in what will need to be an iterative process of model specification and clarification. It is our hope that investigators will increasingly pursue research designs and analytic approaches to refine the model so that the most valid and useful explanations of the relationship between health literacy and outcomes can inform professional re- sponses to the problem in the many di- verse contexts of health care. 9 It should first be recognized that lim- ited health literacy is strongly associated with other socioeconomic indicators, in- cluding educational attainment, race/ ethnicity, and age. 10 Such associations make it difficult to discern the indepen- dent effect of health literacy from the complex relationships known between these latent and evolving traits that are also interrelated themselves. Yet many of the determinants of disease, in general,