We thank Dr Clark for his comments and would like the opportunity to respond to his interpretation of the study results. As he notes, the data from the study were the results of a nonrandomized interventional trial. Although a randomized clinical trial is the most powerful study design, it was not possible for us to undertake given budgetary constraints at our institution. Alternatively, we included 2 comparison groups (historical and concurrent controls) and used 2 different severity adjustment techniques, the All Patient Refined Diagnosis Related Groups (APR-DRG)1 and the Cleveland Health Quality Choice (CHQC) Coalition severity adjustment model.2 It should be recognized that the latter model is considered state of the art and includes chart-based measures such as vital signs, laboratory values, and electrocardiogram and chest radiograph findings.
Cooper GS. Impact of a Guideline-Based Management on Outcomes of Very Old Persons With Heart Failure Living in Nursing Homes—Reply. Arch Intern Med. 2001;161(18):2264-2265. doi: