Benz Scott L, Gravely S, Sexton TR, Brzostek S, Brown DL. Effect of Patient Navigation on Enrollment in Cardiac Rehabilitation. JAMA Intern Med. 2013;173(3):244-246. doi:10.1001/2013.jamainternmed.1042
Author Affiliations: School of Health Technology and Management and Department of Medicine, School of Medicine, Stony Brook Medicine, Stony Brook, New York (Drs Benz Scott, Gravely, Brzostek, and Brown); and College of Business, Stony Brook University, Stony Brook (Dr Sexton).
Globally, cardiovascular diseases are the leading causes of morbidity and mortality.1 Secondary prevention measures, such as outpatient cardiac rehabilitation (OCR), effectively reduce this burden.2,3 Randomized controlled trials and quantitative reviews demonstrate that OCR programs result in significant reductions in morbidity, mortality, and cost of care compared with usual care. Despite this evidence, referral to and enrollment in OCR is low, averaging 19% in the United States.4 It is important to identify new approaches to improve enrollment in OCR because there are limited intervention studies targeting participation among eligible cardiac patients.2,5,6
The objective of this study was to test the hypothesis that a patient navigation (PN) intervention would achieve significantly higher rates of OCR enrollment compared with usual care.
The study design was a randomized controlled trial approved by the human subjects committee at Stony Brook University. From May 2009 to June 2011, patients were screened on the general cardiology and thoracic surgery floors at Stony Brook University Hospital. Inclusion criteria were age 21 years or older and a diagnosis or procedure clinically indicated for OCR referral (myocardial infarction, heart failure, stable angina pectoris, percutaneous coronary intervention, coronary artery bypass graft surgery, or heart valve replacement or repair). Patients were excluded if they were not proficient in the English language, had a major noncardiac comorbidity with a poor prognosis, had a recent uncontrolled psychiatric or substance abuse disorder, were participating in another study that would interfere with the trial, were unable to provide a telephone number, or declined participation. All consenting patients were consecutively assigned to either PN or usual care groups using computer-generated block randomization.
Consistent with prior research on the use of lay health advisors in cancer care coordination,7 2 individuals with no prior clinical knowledge were trained to help patients navigate the inpatient to outpatient cardiac care system, with a particular focus on enrolling in a local OCR program. Each patient assigned to PN had a navigator meet with them prior to discharge. Patients in the PN intervention were educated about OCR (ie, the likely benefits of participation, the location of local programs, anddetails on how to access it). Navigators facilitated contact with an OCR program of the patient's choice. Individuals discharged prior to face-to-face navigation were mailed information to their home and a navigator reviewed it by telephone within 1 week. Approximately 10 days after hospital discharge, navigators called PN patients to encourage them to discuss OCR with a physician and to enroll. Those assigned to usual care received the standard discharge instructions.
Sociodemographic and clinical data were collected from medical charts and 2 in-depth telephone interviews at 4 and 12 weeks after hospitalization. The primary outcome was patient OCR enrollment, defined as having attended at least 1 OCR session. For patients who reported enrollment in an OCR program, the program was contacted to verify enrollment. For patients who did not complete the second telephone interview (9 usual care and 22 PN participants), the local OCR programs were contacted to determine enrollment status.
Categorical variables and outcomes were compared using a 2-tailed Fisher exact test, and continuous variables were compared using 2-tailed independent sample t tests. P < .05 (2-sided) was considered statistically significant. Statistical analysis was performed with SPSS version 19.0 (SPSS Inc).
Of the 599 inpatients screened for the trial, 181 consented to participate. Of the 181 trial participants, 3 died before the first telephone interview (within 1 month of hospital discharge) and were excluded from the analyses. Therefore, 178 patients (89 per group) had their cardiac rehabilitation enrollment verified and were included in the analyses.
The Table gives the sociodemographic and clinical variables at baseline. Overall, participants tended to be young (mean [SD] age, 60.4 [10.5] years), male (66.3%), white (86.5%), and married (60.1%); have insurance (84.3%); and have an education greater than high school (60.3%). Most patients had a prior myocardial infarction (83.1%), hypertension (79.8%), or hyperlipidemia (65.7%). Participants in the PN intervention were less likely to be married and more likely to have an education beyond high school. Of the 178 study participants, 21 of the 89 randomized to PN (23.6%) and 6 of the 89 randomized to usual care (6.7%) enrolled in an OCR program (P = .003).
Despite compelling evidence that participation in OCR is associated with reductions in morbidity and mortality as well as gains in quality of life and functional status, enrollment of eligible cardiac patients into OCR programs remains suboptimal. To our knowledge, this is the first randomized controlled trial to compare the effect of a PN intervention against usual care on increasing patient enrollment into OCR. The results demonstrated a 3-fold increase in patient enrollment in an OCR program compared with usual care.
In conclusion, lay individuals trained as cardiac patient navigators achieve significantly higher rates of OCR enrollment compared with usual care.
Correspondence: Dr Benz Scott, Graduate Program in Public Health, Stony Brook Medicine, Health Sciences Center, Level 3, Room 078, Stony Brook, NY 11794 (Lisa.Benzscott@stonybrookmedicine.edu).
Published Online: December 17, 2012. doi:10.1001/2013.jamainternmed.1042
Author Contributions: Dr Benz Scott had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Benz Scott and Brown. Acquisition of data: Benz Scott and Brzostek. Analysis and interpretation of data: Benz Scott, Gravely, Sexton, Brzostek, and Brown. Drafting of the manuscript: Benz Scott, Gravely, and Sexton. Critical revision of the manuscript for important intellectual content: Benz Scott, Gravely, Sexton, Brzostek, and Brown. Statistical analysis: Gravely and Sexton. Obtained funding: Benz Scott. Administrative, technical, and material support: Benz Scott, Brzostek, and Brown. Study supervision: Benz Scott and Brown.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was funded in part by General Clinical Research Center grant MO1RR10710 and a Targeted Research Opportunity Fusion Award with matching funds provided by the School of Medicine and the School of Health Technology & Management.