Pulmonary Rehabilitation* Joint ACCP/AACVPR Evidence-Based Clinical Practice Guidelines

Andrew L. Ries, MD, MPH, FCCP (Chair); Gerene S. Bauldoff, RN, PhD, FCCP; Brian W. Carlin, MD, FCCP; Richard Casaburi, PhD, MD, FCCP; Charles F. Emery, PhD; Donald A. Mahler, MD, FCCP; Barry Make, MD, FCCP; Carolyn L. Rochester, MD; Richard ZuWallack, MD, FCCP; Carla Herrerias, MPH


CHEST. 2007;131(5):4S-42S. 

In This Article

Methodology and Grading of the Evidence for Pulmonary Rehabilitation

In 1997, the ACCP and the AACVPR released an evidence-based clinical practice guideline entitled "Pulmonary Rehabilitation: Joint ACCP/AACVPR Evidence-Based Guidelines."[2,3] Following the approved process for the review and revision of clinical practice guidelines, in 2002 the ACCP Health and Science Committee determined that there was a need for reassessment of the current literature and an update of the original practice guideline. This new guideline is intended to update the recommendations from the 1997 document and to provide new recommendations based on a comprehensive literature review. The literature review and development of evidence tables were conducted by Carla Herrerias, MPH, the ACCP Clinical Research Analyst. The joint ACCP/AACVPR expert panel used the evidence to develop graded recommendations.

The guideline panel was organized under the joint sponsorship of the ACCP and the AACVPR. Andrew Ries, MD, MPH, FCCP, Chair of the 1997 panel, served as Chair of the new panel. Panel members were evenly distributed between and selected by the two organizations with a goal of making the panel multidisciplinary and geographically diverse. Drs. Casaburi, Mahler, Make, and Rochester represented the ACCP, and Drs. Bauldoff, Carlin, Emery, and ZuWallack represented the AACVPR. Five panel members (Drs. Carlin, Casaburi, Emery, Mahler, and Make) had served on the previous guideline panel. In addition to several conference calls, the panel met for one 2-day meeting to review the evidence tables and become familiar with the process of grading recommendations. Writing assignments were determined by members’ known expertise in specific areas of pulmonary rehabilitation. Each section of the guideline was assigned to one primary author and at least one secondary author. Sections were reviewed by relevant panel members when topics overlapped.

At several stages during the guideline development period, panel members were asked to disclose any conflict of interest. These occurred at the time the panel was nominated, at the first face-to-face meeting, the final conference call, and prior to publication. Written forms were completed and are on file at the ACCP.

The 1997 practice guideline on pulmonary rehabilitation focused on program component areas of lower and upper extremity training, ventilatory muscle training, and various outcomes of comprehensive pulmonary rehabilitation programs, including dyspnea, quality of life, health-care utilization, and survival. Psychosocial and educational aspects of rehabilitation were examined both as program components and as outcomes.

For this review, the panel decided to focus on studies that had been published since the previous review, again concentrating on stable patients with COPD. Since there have been many advances and new areas of investigation since the previous document was written, the panel decided to expand the scope of this review rather than just update the previous one. Topics covered in this document include the following:

  • Outcomes of comprehensive pulmonary rehabilitation programs: lower extremity exercise training; dyspnea; health-related quality of life (HRQOL); health-care utilization and economic analysis; survival; psychosocial outcomes; and long-term benefits from pulmonary rehabilitation;

  • Duration of pulmonary rehabilitation;

  • Postrehabilitation maintenance strategies;

  • Intensity of aerobic exercise training;

  • Strength training in pulmonary rehabilitation;

  • Anabolic drugs;

  • Upper extremity training;

  • Inspiratory muscle training (IMT);

  • Education;

  • Psychosocial and behavioral components of pulmonary rehabilitation;

  • Oxygen supplementation as an adjunct to pulmonary rehabilitation;

  • Noninvasive ventilation;

  • Nutritional supplementation in pulmonary rehabilitation;

  • Pulmonary rehabilitation for patients with disorders other than COPD; and

  • Summary and recommendations for future research.

The literature review was based on the scope of the work as outlined in the previous section. The literature search was conducted through a comprehensive MEDLINE search from 1996 through 2004, and was supplemented by articles supplied by the guideline panel as well as by a review of bibliographies and reference lists from review articles and other existing systematic reviews. The literature search was limited to articles published in peer-reviewed journals only in the English language, and on human subjects. Inclusion criteria primarily included a population of persons with a diagnosis of COPD determined either by physical examination or by existing diagnostic criteria; however, those with other pulmonary conditions (eg, asthma or interstitial lung disease) were also included. The search included randomized controlled trials (RCTs), metaanalyses, systematic reviews, and observational studies. The search strategy linked pulmonary rehabilitation or a pulmonary rehabilitation program with each key subcomponent, as listed in section on "Scope of Work." To locate studies other than RCTs, such as systematic reviews and metaanalyses, those key words were used in searching MEDLINE and the Cochrane databases. Informal review articles were included only for hand searching additional references. For the purpose of this review, pulmonary rehabilitation was defined operationally as studies involving exercise training plus at least one additional component. Associated outcomes across all components were dyspnea, exercise tolerance, quality of life and activities of daily life, and health-care utilization. An initial review of 928 abstracts was conducted by the ACCP Clinical Research Analyst and the Research Specialist. Full articles (a total of 202) were formally reviewed and abstracted by the Clinical Research Analyst, and a total of 81 clinical trials were included in all evidence tables. RCTs were scored using a simplified system that was based on methods of randomization, blinding, and documentation of withdrawals/loss to follow-up. This system follows a method that is based on a 3-point scale, which rates randomization (and appropriateness), blinding (and appropriateness), and tracking of withdrawals and loss to follow-up. Studies were graded on a scale of 0 to 5.[28] No formal quantitative analysis was performed due to the wide variation in methodologies reported in studies. Given the length of time required to prepare the final manuscript after the conclusion of the systematic literature review in December 2004, from which the tables were constructed, the committee was allowed to include reference to selected articles published in 2005 and 2006 in the text if the additional information provided by the newer publications was felt to be important.

The ACCP system for grading guideline recommendations is based on the relationship between the strength of the evidence and the balance of benefits to risk and burden ( Table 1 ).[29] Simply stated, recommendations can be grouped on the following two levels: strong (grade 1); and weak (grade 2). If there is certainty that the benefits do (or do not) outweigh risk, the recommendation is strong. If there is less certainty or the benefits and risks are more equally balanced, the recommendation is weaker. Several important issues must be considered when classifying recommendations. These include the quality of the evidence that supports estimates of benefit, risks, and costs; the importance of the outcomes of the intervention; the magnitude and the precision of estimate of the treatment effect; the risks and burdens of an intended therapy; the risk of the target event; and varying patient values.

The strength of evidence is classified, based on the quality of the data, into the following three categories: high (grade A); moderate (grade B); and low (grade C). The strongest evidence comes from well-designed RCTs yielding consistent and directly applicable results. In some circumstances, high-quality evidence can be the result of overwhelming evidence from observational studies. Moderate-quality evidence is based on RCTs with limitations that may include methodological flaws or inconsistent results. Studies other than RCTs that may yield strong results are also included in the moderate-quality category. The weakest type of evidence is that from other types of observational studies. It should be noted that the ACCP Health and Science Policy Committee has endorsed the principle that most relevant clinical studies provide evidence, even though the quality of that evidence is varied. Therefore, the reasons for excluding studies should be documented.

Table 2 describes the balance of benefits to risk and burden, and the level of certainty based on this balance. As stated above, the more certain the balance, or lack thereof, the stronger the recommendation. Patient and community values are important considerations in clinical decision making and are factored into the grading process. In situations in which the benefits clearly do or do not outweigh the risks, it is assumed that nearly all patients would have the same preferences. For weaker recommendations, however, there may not be consistency in patient preferences.

In addition to recommendations, the committee included several statements when it thought that there was insufficient evidence to make a specific recommendation. These statements are included along with the recommendations but are not graded.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.