To the Editor: The Special Article by Landon and colleagues(March 1 issue)1 on improving the management of chronic diseaseat community health centers illustrates the importance of identifyingappropriate outcomes when measuring the effectiveness of interventionsto improve processes of care. Establishing a more realisticschedule than that used in this study for assessing the effectof the program might have yielded a different picture. Sufficienttime needs to be allowed for the measurement of clinical outcomes,particularly regarding outcomes of patients with chronic disease.Process interventions to improve outcomes in chronic diseasehave been shown to be associated with an increase in healthcare utilization during the first year. This increase oftenreflects preexisting needs of the patient that had not beenmet; it is often not until the second year that a measurabledecrease in health care utilization is noted.2 Physicians andothers working to establish evidence-based interventions inthe community can identify appropriate outcomes by partneringwith families, community stakeholders, and local institutions.Implementation designs that incorporate the collection of locallymeaningful outcomes data into realistic, community-sensitivetimetables have been reported to result in effective and sustainableprograms.3
Matthew D. Sadof, M.D. Baystate Children's Hospital Springfield, MA 01199 matthew.sadof{at}bhs.org
References
Landon BE, Hicks LS, O'Malley AJ, et al. Improving the management of chronic disease at community health centers. N Engl J Med 2007;356:921-934. [Free Full Text]
Mastal M, Palsbo S. Measuring the effectiveness of managed care for adults with disabilities. Hamilton, NJ: Center for Health Care Strategies, December 2005. (Accessed May 17, 2007, at http://www.chcs.org/usr_doc/CCOMeasures_final.pdf.)
Sadof MD, Boschert KA, Brandt SJ, Motyl A. After the funding is gone: an analysis of predictors of sustainability efforts at the inner city asthma intervention sites. Ann Allergy Asthma Immunol 2006;97:Suppl 1:S31-S35. [Web of Science][Medline]
To the Editor: The examination by Landon et al. of the qualityof care at community health centers adds to an impressive literature;although this study covered too short a period to capture healthoutcomes, earlier research has documented such effects. Theauthors do not report on the broader policy context of thiswork, however. Health centers face a staggering increase inthe number of uninsured patients. Yet not only has the Bushadministration eliminated all funding for quality-improvementcollaboratives, but its proposals for the fiscal year 2008 budgetcall for deep reductions in Medicaid (the most important sourceof funding for health centers) and seek no appropriations foreither quality improvement or health-information technology.Moreover, the administration has recently begun to withholdaccess to data on health center performance that were previouslypublic under the Uniform Data System and that provide importantinformation on the deep challenges confronting health centersand their communities.
Sara Rosenbaum, J.D. George Washington University School of Public Health and Health Services Washington, DC 20006 sarar{at}gwu.edu
To the Editor: If the outcome is not improved, it is illogicalto conclude that the process is improved. All that has beendemonstrated is that doctors and nurses are able to jump throughthe hoops mandated by expert committees. When "hierarchicalregression models" and other elements of scientific sophisticationare removed, one is left stating that the operation was a successbut the patient still died. Landon et al. cite numerous articlesin the medical literature of this same genre.
Mitchell T. Smolkin, M.D. Waynesboro Primary Care Waynesboro, VA 22980 mts8s{at}virginia.edu
To the Editor: Landon et al. find that a quality-improvementprogram was better for improving care processes for patientswith asthma, diabetes, and hypertension than for improving intermediateoutcomes. The Translating Research into Action for Diabetesstudy has reported similar findings.1 According to that study,across 67 physician groups, variation in the intensity of diseasemanagement of diabetes was strongly associated with variationin care processes but not in intermediate outcomes. Scores forprocess-based quality were unrelated to scores for intermediateoutcomes.2 Disease management reduced disparities related torace or ethnic group in processes but not in intermediate outcomes.3
If process measures are easier to affect than are outcomes,we should perhaps focus on processes for measuring, providingfeedback, and providing incentives — but only those processesthat have already been rigorously linked to improved outcomes.4Rates of testing (e.g., for low-density lipoprotein cholesterol)and other unproven process measures may be easy to improve butoffer little clinical benefit. Conversely, evidence-based processes,such as aspirin use or smoking cessation programs, have obviousvalue for patients. Finally, intermediate outcomes are improvingover time, despite our inability to relate the changes to specificinterventions. A better understanding of what is driving theseimprovements is needed.
Joe V. Selby, M.D. Kaiser Permanente Northern California Oakland, CA 94612 joe.selby{at}kp.org
Carol M. Mangione, M.D. University of California Los AngelesSchool of Medicine Los Angeles, CA 90095
Robert B. Gerzoff, M.S. Centers for Disease Control and Prevention Atlanta, GA 30341
References
Mangione CM, Gerzoff RB, Williamson DF, et al. The association between quality of care and the intensity of diabetes disease management programs. Ann Intern Med 2006;145:107-116. [Free Full Text]
Ackermann RT, Thompson TJ, Selby JV, et al. Is the number of documented diabetes process-of-care indicators associated with cardiometabolic risk factor levels, patient satisfaction, or self-rated quality of diabetes care? The Translating Research into Action for Diabetes (TRIAD) study. Diabetes Care 2006;29:2108-2113. [Free Full Text]
Duru OK, Mangione CM, Steers NW, et al. The association between clinical care strategies and the attenuation of racial/ethnic disparities in diabetes care: the Translating Research Into Action for Diabetes (TRIAD) Study. Med Care 2006;44:1121-1128. [CrossRef][Web of Science][Medline]
Kerr EA, Krein SL, Vijan S, Hofer TP, Hayward RA. Avoiding pitfalls in chronic disease quality measurement: a case for the next generation of technical quality measures. Am J Manag Care 2001;7:1033-1043. [Web of Science][Medline]
The authors reply: Both Sadof and Rosenbaum suggest that wemight have observed improvements in intermediate outcomes, givenmore time. Conceptually, we agree that, to the extent that outcomesof care are directly related to process interventions, moretime than the period of our study might be needed to observemeaningful improvements in clinical outcomes such as mortalityor the incidence of acute myocardial infarction. There is noreason to expect, however, that the intermediate outcomes weassessed (e.g., control of glycated hemoglobin and control ofhypertension) would require such a lag. In addition, as we state,the 1-year postintervention assessment period began 1 year afterthe completion of the intervention, a timing consistent withthat suggested by Sadof.
Smolkin argues that improvements in the processes of care aremeaningless if they are not accompanied by improvements in outcomes.With the exception of asthma, the intermediate outcomes we assessedexamine the control of important risk factors. Given the requiredtime frame and sample size, we could not assess clinical outcomessuch as the incidence of cardiovascular disease or mortality,but we would expect that these outcomes would ultimately beaffected by improvements in the processes of care. Moreover,many of the process measures we examined are strongly linkedto these meaningful clinical outcomes (e.g., daily aspirin use)but are not directly related to the intermediate outcomes weassessed. Selby and colleagues studied the association betweenvarious care-management techniques and the quality of care ofpatients with diabetes and reported results similar to ours.1We agree with their suggestion that quality-improvement effortsshould focus on evidence-based processes of care that have beenrigorously linked to important clinical outcomes.
Finally, Rosenbaum provides important information on the broadpolicy context and the challenges facing community health centers.We agree that such centers are an important cornerstone of effortsto provide a safety net for millions of Americans and that everyeffort should be made to provide adequate funding to meet theneeds of the underserved populations they care for.
Bruce E. Landon, M.D., M.B.A. LeRoi S. Hicks, M.D., M.P.H. Edward Guadagnoli, Ph.D. Harvard Medical School Boston, MA 02115 landon{at}hcp.med.harvard.edu
References
Mangione CM, Gerzoff RB, Williamson DF, et al. The association between quality of care and the intensity of diabetes disease management programs. Ann Intern Med 2006;145:107-116. [Free Full Text]
Previous
