Is COPD management getting better or worse? How would we know without benchmarking?

Nearly 1 in 10 Canadians 35 years of age or older have chronic obstructive pulmonary disease (COPD).¹ In Canada, COPD is the leading cause of hospitalization and the third leading cause of mortality.¹ Cigarette smoking is the most important modifiable risk factor for COPD, accounting for 70 to 80% of all cases.² Curiously, despite the massive reduction in smoking rates in Canada (from 50% in 1965 to 13% in 2015³) and significant advancement in COPD therapeutics and management over the past 30 years, the population prevalence of COPD and its associated hospitalizations and mortality have increased during this time.⁴ Over the next 20 years, the burden of COPD is expected to double. Why is this happening?

The answer is not entirely clear. One major reason for this ambiguity is that there are no good quality indicators to evaluate COPD management and, thus, no easy way to monitor the quality of care provided to our patients with COPD. Given the large improvements in therapeutics and the vigorous science behind currently recommended treatment strategies (e.g. pulmonary rehabilitation), the assumption is that overall COPD care has improved significantly in the community, which in turn should lead to better health outcomes for our patients. However, this logic only holds true if data from experimental studies and trials are efficiently translated and integrated into practice. However, it is now well known that in many cases scientific knowledge gets lost in translation, leading to major delays in implementation (or in certain cases, no implementation at all).⁵ Quality assessment is thus crucial in monitoring clinical practice, identifying potential barriers to clinical implementation and enabling durable practice changes. According to the Institute of Medicine, there are 6 crucial domains of health quality: 1) safety (ensuring that no harm comes to patients), 2) effectiveness (ensuring that effective therapies are applied to patients who will most likely benefit from these therapies and avoiding them in those who will experience harm); 3) patient-centerdness (ensuring that management is provided in a respectful and responsive manner to patients and ensuring that patients are central to the decision-making process at all levels); 4) timeliness (ensuring that wait-times for therapies or diagnostics are minimized); 5) efficiency (ensuring that there is no waste in resources); and 6) equity (ensuring that access to quality care is blind to the personal characteristics of the patients or where they reside).⁶ However, before these domains can be implemented, benchmarking is first needed to understand the standard of current state of care relative to “best practice” based on high quality evidence. This in turn will lead to the identification of challenges and pitfalls in achieving best practice and ascertainment of potential solutions to surmount these shortcomings.

In this issue of the Journal, Gershon and colleagues⁷ report on the development of a new quality indicator for COPD that can be used to “benchmark” clinical practice. The authors used a modified RAND Appropriateness Method to derive this instrument. The RAND Appropriateness Method was originally developed in the mid-1980s to determine whether or not there was over or under-use of medical and surgical procedures.⁸ There are 2 essential components to this method: 1) a thorough literature search and evidence synthesis (e.g. systematic review and meta-analysis) of the topic of interest; and 2) rating of the recommendations from the literature search by each member of an expert panel using a scale of 1 to 9. A score of 1 indicates that the expected harm greatly outweighs the potential benefits of the intervention and a score of 9 reflects the polar opposite (the potential benefits greatly outweigh the expected harm). A score of 5 denotes ambivalence in which the potential benefits are perfectly balanced by the expected harm. A score of 5 may also confer “ignorance” where the expert is unable to render a judgement owing to inadequate information or lack of knowledge. There are usually 2 rounds to the rating system: the first is performed independently by each member and the second occurs in a face-to-face meeting where the experts discuss their scores, and through iteration arrive at a consensus score in which each recommendation is categorized as “appropriate” (a consensus score of 7–9), “uncertain” (a consensus score of 4-6) or “inappropriate” (a consensus score of 1–3).⁸ If a consensus score cannot be reached because of disagreements, “uncertain” category is then assigned to that recommendation. The RAND Appropriateness Method has been used in multiple clinical settings and in almost all disease areas. A quick search of PubMed reveals over 500 peer-reviewed publications that have used this instrument for quality assessment.

What were the notable findings by Gershon et al.⁷? The expert panel ranked spirometry for diagnosis confirmation as the top quality indicator. This is both surprising and disappointing. Despite the widespread promulgation of guidelines by local, national and international expert panels, which have stated unequivocally that a diagnosis of COPD cannot be made in the absence of spirometry, there is ample evidence that this “edict” is widely ignored in clinical practice. In fact, 40 to 50% of patients with “COPD” in the community have never had spirometry-confirmation.⁹ There are many barriers to spirometry in the community including poor access, long wait-times (in some cases months to years), and inadequate reimbursement for the procedure. It may be that we have reached a ceiling for spirometry use in the community as many of these challenges appear to be insurmountable in our current health care setting. Other (more novel) approaches may be needed including the use of artificial intelligence and computerized algorithm to identify high-risk patients from thoracic computed tomographies (CT) or simple blood or breath tests as biomarkers of COPD. The expert panel also ranked highly smoking cessation interventions as an indicator of quality care. Although the overall rates of smoking have fallen dramatically to about 10% in most communities in Canada, rates of current smoking among those with COPD are much higher, estimated to be 30 to 50%.¹⁰ Thus, there cannot be any complacency in tackling smoking cessation in this patient population. As these patients are often recalcitrant or recidivist smokers, they may require more than general smoking cessation counseling or nicotine replacement therapy. Smoking in these patients should be considered as a “chronic disease”¹¹ and treated in the same manner as other chronic conditions. Finally, it is noteworthy to highlight the role of therapeutics that experts have prioritized in this report. Contrary to the prevailing trends in the community where inhaled corticosteroids in combination with a long acting beta-2 agonists are used as first line agents for COPD patients,¹² experts endorse the use of long-acting bronchodilators (and not inhaled corticosteroids) as a clear mark of high quality. During exacerbations, the experts endorse the use of oral corticosteroids over the use of antibiotics.

COPD is a growing health and economic burden in Canada. Gershon and colleagues⁷ have given the community an excellent quality indicator tool to benchmark our current practice and most importantly highlight potential deficiencies in care. Spirometry under-use, smoking cessation and steroid over-use for maintenance therapy are all major areas that require special attention for enabling better and more personalized therapy for millions of Canadians with COPD.

Acknowledgments

D.D. Sin is a Tier 1 Canada Research Chair in COPD and holds the De Lazzari Family Chair at HLI.

Disclosure statement

D.D. Sin has received research funding from AstraZeneca (AZ), Boehringer Ingelheim (BI), Merck and have received honoraria for sitting on advisory boards of AZ, BI, Regeneron, Sanofi-Aventis and Novartis and for speaking engagements from AZ, BI and Novartis.