Linderman DJ, Koff PB, Freitag TJ, Min S, Vandivier RW. Effect of Integrated Care on Advanced Chronic Obstructive Pulmonary Disease in High-Mortality Rural Areas. Arch Intern Med. 2011;171(22):2059-2061. doi:10.1001/archinternmed.2011.576
Author Affiliations: Divisions of Pulmonary Sciences and Critical Care Medicine (Drs Linderman and Vandivier) and Health Care Policy and Research (Dr Min), University of Colorado, Denver, Anschutz Medical Campus, Aurora; and University of Colorado Hospital, Aurora (Mss Koff and Freitag).
Guideline-based care only reaches approximately 50% of patients with chronic illnesses, including chronic obstructive pulmonary disease (COPD),1 suggesting that implementation strategies are flawed. Inadequate care for COPD is likely to be greatest in rural areas where access to care is limited and COPD-related mortality is high.2 Disease education and management programs have been developed to enhance patient knowledge, improve delivery of guideline-based care, and maximize early identification and treatment of exacerbations. In a small, prospective randomized clinical trial, we showed that a proactive model of integrated care (PIC) results in dramatic improvements of quality of life and effectively identifies COPD exacerbations.3 PIC differs from other self-management or integrated care models by combining disease-specific education and self-management principles with a simple remote monitoring platform (Health Buddy; Robert Bosch Healthcare, Palo Alto, California). We hypothesized that our novel disease management program would improve quality of life and increase guideline-based care4 in a predominantly rural group of patients with exceptionally high COPD-related mortality.
One hundred patients with advanced COPD were enrolled in a 12-week longitudinal, cohort study. Subjects resided in 1 of 16 Colorado counties with the highest COPD-related mortality (90-129 deaths per 100 000),2 well above the 2005 national average (64.3 deaths per 100 000). Fourteen counties were rural or frontier (≤6 persons per square mile). Details of the management, monitoring and educational program have been published.3 Assessments were made at enrollment and study completion, including quality of life as measured by the St George's Respiratory Questionnaire (SGRQ), guideline-based therapies, BODE index (body mass index [B], the degree of airflow obstruction [O] and dyspnea [D], and exercise capacity [E]), and health care utilization. Univariable analysis was performed to see which component(s) of PIC had the most impact on improvement in quality of life. Variables with P < .15 were then entered into multiple linear regression to determine independently significant factors. Finally, worsening symptoms were flagged by the Health Buddy, and the symptoms most commonly associated with an acute exacerbation were noted.
Baseline characteristics for the 100 enrollees, 82 completers, and 18 noncompleters are given in eTable 1. At baseline, subjects' mean FEV1 was 43% predicted, and they had 53 pack-years of tobacco use; long-acting inhaler use ranged from 34% to 57%, and postexercise oxygen saturations were low at 85.5%. Of the 82 completers, 60 (73%) resided in rural counties.
The Table shows that PIC improved SGRQ by 11.6 units, increased inhaled medication use by 18 to 23 percentage points, and improved oxygen use and saturations. PIC increased the 6-minute walk distance (6MWD) by 45 m, improved the BODE index by 1.14 units, and decreased COPD-related emergency department visits and hospitalizations. In addition, smoking rates tended to decrease from 24.4% to 18.3% (P = .06).
Multiple linear regression analysis (eTable 2) showed that only the initiation of a long-acting β-agonist (LABA) medication and an improvement in 6MWD showed significant effects on improvement in SGRQ (β = −10.3 [P = .03] for LABA initiation; β = −0.06 [P < .001] for 6MWD improvement). However, LABA initiation and 6MWD improvement only explained 27% of the variation in SGRQ change.
Twenty-four subjects had 31 documented COPD exacerbations. As given in eTable 3, of the 68 red and yellow flags that occurred in the 3 days prior to an exacerbation, shortness of breath was the most commonly flagged category (see eTable 4 for flag definitions). Oxygen desaturation, difficulty with activity and worsening cough or sputum production were also commonly flagged. A decline in forced expiratory volume in 1 second (FEV1) and wheezing were flagged the least.
There is a dearth of data on outcomes for rural patients with COPD, with only a handful of publications specifically evaluating rural patients with chronic respiratory illnesses and no studies targeting patients with COPD in high-mortality areas. Our results suggest that programs focusing on optimization of care in areas with limited access to health care may have an impact that is equivalent to the introduction of a new medication.
While other disease management studies have demonstrated improvements in quality of life,3,5- 8 none have demonstrated an improvement to the degree shown herein. We found that the addition of LABA and improvements in 6MWD were mediators of this impressive improvement in quality of life. However, they only partially explain this improvement. We suspect that daily interaction with the Health Buddy, with its educational and monitoring components, was also an important factor.
Guideline-based care is a cornerstone of COPD management with many positive effects.4 For example, long-acting bronchodilators and inhaled corticosteroids improve quality of life, reduce exacerbations, and decrease hospitalizations.4 PIC led to substantial increases in the use of long-acting bronchodilators and inhaled corticosteroid. PIC also improved the use and titration of long-term oxygen therapy and tended to decrease smoking rates. Although participation in pulmonary rehabilitation was very low, patients performed 6-minute walks daily. This may have led to the observed increases in 6MWD and SGRQ, which are similar to that seen in both traditional and home-based pulmonary rehabilitation programs.
In conclusion, PIC significantly improved quality of life and guideline-based care, identified COPD exacerbations, and reduced health care utilization in a predominantly rural population with high COPD-related mortality.
Correspondence: Dr Linderman, Division of Pulmonary Sciences and Critical Care Medicine, COPD Center, University of Colorado, Denver, Anschutz Medical Campus, Research Bldg 2, 12700 E 19th Ave, PO Box C272, Aurora, CO 80045 (Derek.Linderman@ucdenver.edu).
Author Contributions: All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Koff and Vandivier. Acquisition of data: Koff, Freitag, and Vandivier. Analysis and interpretation of data: Linderman, Koff, Min, and Vandivier. Drafting of the manuscript: Linderman, Koff, and Vandivier. Critical revision of the manuscript for important intellectual content: Linderman, Koff, Freitag, Min, and Vandivier. Statistical analysis: Min and Vandivier. Obtained funding: Koff and Vandivier. Administrative, technical, and material support: Koff and Freitag. Study supervision: Vandivier.
Financial Disclosure: Ms Freitag is currently employed by Connections365, a telehealth solution company, which uses the Bosch Health Buddy 3 (the device used in the present research into the care of patients with COPD in rural Colorado) and has designed and deployed a telehealth kiosk. Ms Koff has consulted for Nonin Medical, Connection365, American Association for Respiratory Care, and Robert Bosch Healthcare. As of July 1, 2010, Ms Koff has been employed by Robert Bosch Healthcare as an account manager.
Funding/Support: This research was supported by grants to Dr Vandivier from the Cancer, Cardiovascular Disease, and Pulmonary Disease (CCPD) Competitive Grants Program of the Colorado Department of Public Health and Environment and from the University of Colorado Hospital.
Previous Presentation: This work was presented in abstract form at the American Thoracic Society International Conference; May 19, 2010; New Orleans, Louisiana.
Additional Contributions: We thank health care providers throughout rural Colorado for their help identifying study subjects. Linda Gunnison, RRT, and Shannon James, BSN, assisted in recruiting and managing the patients in this study, and Debra Diaz helped create and manage the patient database.