Abstract
Spirometry is necessary to diagnose chronic obstructive pulmonary disease (COPD), yet a large proportion of patients are diagnosed and treated without having received testing. This study explored whether the effects of interventions using the electronic health record (EHR) to target patients diagnosed with COPD without confirmatory spirometry impacted the incidence rates of spirometry referrals and completions. This retrospective before and after study assessed the impact of provider-facing clinical decision support that identified patients who had a diagnosis of COPD but had not received spirometry. Spirometry referrals, completions, and results were ascertained 1.5 years prior to and 1.5 years after the interventions were initiated. Inhaler prescriptions by class were also tallied. There were 10,949 unique patients with a diagnosis of COPD who were eligible for inclusion. 4,895 patients (44.7%) were excluded because they had completed spirometry prior to the cohort start dates. The pre-intervention cohort consisted of 2,622 patients, while the post-intervention cohort had 3,392. Spirometry referral rates pre-intervention were 20.2% compared to 31.6% post-intervention (p < 0.001). Spirometry completion rates rose from 13.2% pre-intervention to 19.3% afterwards (p < 0.001). 61.7% (585 of 948) had no evidence of airflow obstruction. After excluding patients with a diagnosis of asthma, 25.8% (126 of 488) patients who had no evidence of airflow obstruction had prescriptions for long-acting bronchodilators or inhaled steroids. A concerted EHR intervention modestly increased spirometry referral and completion rates in patients with a diagnosis of COPD without prior spirometry and decreased misclassification of disease.
Introduction
Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide [Citation1]. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) requires spirometry for the diagnosis and management of COPD [Citation2]. Despite this guidance, clinicians continue to diagnose COPD without obtaining spirometry. In the US, only approximately one third of patients with a new COPD diagnosis undergo spirometry [Citation3, Citation4]. Without spirometry, patients risk being misdiagnosed or overdiagnosed [Citation5–14]. In data from the multinational Burden of Obstructive Lung Disease (BOLD) study, a majority of patients (61.9%) with a physician-reported diagnosis of COPD had unobstructed postbronchodilator spirometry [Citation15].
Overdiagnosis of COPD is not without adverse effects. Even in patients with COPD, overtreatment with inhaled anticholinergics could portend an increased risk of myocardial infarction and cardiovascular death, and overprescribing inhaled corticosteroids has been associated with an increased risk of pneumonia and osteoporosis [Citation16–19]. In the absence of evidence-based justifications for therapy, overtreatment with long-acting inhalers can lead to unjustifiable medication side effects and significantly increases healthcare costs [Citation16, Citation20].
While numerous studies have highlighted the prevalence of diagnoses of COPD without spirometric confirmation, few have demonstrated successful systematic and low-maintenance approaches to mitigate this [Citation6, Citation21]. Many electronic health record (EHR) platforms have clinical decision support systems or reminder tools that support routine patient care and preventative health initiatives. These tools could conceptually be used to identify and address patients who are at risk for COPD overdiagnosis, but their efficacy in this context has not been studied. In March of 2017, we implemented a multi-pronged EHR-based intervention to address potential COPD overdiagnosis in our healthcare system. Through this study, we sought to determine the resultant impact of provider and patient directed EHR interventions on spirometry referrals and completions in our patient population.
Methods
Study design and setting
This retrospective before and after study was conducted at an urban, academic safety-net healthcare system that has over 18 ambulatory centers across Cuyahoga County in the state of Ohio with over 1.3 million outpatient visits per year.
Participants
We included patients who had at least 1 ambulatory primary care encounter in the preceding 2 years and either a problem list diagnosis or an encounter-based diagnosis of COPD within the last year. COPD was defined using International Disease Classification (ICD) 10 codes of J41-J44 [Citation22]. Patients who met inclusion criteria during an ambulatory primary care visit in the departments of Internal Medicine, Family Medicine, Medicine-Pediatrics, or Geriatrics between March 23, 2015 and September 23, 2016 were assigned to the pre-intervention group and patients meeting inclusion criteria during a visit between March 24, 2017 and September 24, 2018 were assigned to the post-intervention group. The date of the outpatient primary care visit during which patients met criteria for inclusion was regarded as the index encounter (time zero). Patients with any spirometry completed (within the institution or as ascertained from claims or health-information exchange data) prior to the start date of their respective cohort were excluded.
Interventions
Three EHR-based interventions were implemented in March of 2017 to address the issue of COPD misdiagnosis. First, a health maintenance reminder was created to prompt primary care providers to consider spirometry testing for patients assigned a diagnosis of COPD but who had not previously had spirometry testing. This was embedded into an existing reminder list that is routinely accessed by providers and patients (via the patient portal) which also included timed and potentially recurring vaccination and cancer screening reminders. Second, the existing spirometry referral order was revised to make it easier for any provider to electronically and directly schedule spirometry without necessitating a phone call. Third, patients without prior spirometry testing received messages directly through the patient portal which provided them the ability to directly self-schedule spirometry.
Data collection
All data was extracted from the Epic EHR (Epic Corporation, Verona, WI). Demographic data included age, sex, weight, race, and ethnicity. Comorbid conditions were determined using appropriate ICD 10 codes assigned up to 10 years prior to the cohort start date (ICD 10 J45 for asthma, J30 - J32 for sinus disease and I20-I28, I30-I39, I40- I52 for cardiovascular disease). Pulmonary function testing was ascertained using Current Procedural Terminology (CPT) codes assigned at our institution or identified via claims data or health information exchange. A referral to and completion of pulmonary function testing was considered only if it occurred within 6 months following a primary care visit. Airway obstruction was classified as a forced expiratory volume in one second (FEV1) to forced vital capacity (FVC) ratio (FEV1/FVC) less than the lower limit of normal (the 5th percentile) following the administration of bronchodilators [Citation23]. When post-bronchodilator testing was not completed, pre-bronchodilator values were used. Prescriptions for long-acting inhaled bronchodilators, including long acting muscarinic antagonists, long acting beta agonists, or inhaled corticosteroids, were also obtained from the EHR. Prescriptions were considered active if they were ordered within 6 months of a patient’s index visit.
Outcomes
The primary outcomes were the rate of spirometry referrals and completions in each cohort within a six-month period from the index encounter. Secondary outcomes included the rates of obstruction in patients who completed testing.
Analyses
Proportions were compared using the chi-squared test. A multiple variable logistic regression model was constructed from both cohorts to determine the odds for non-obstructed spirometry testing results after adjustment for patient age, sex, weight, race, ethnicity, smoking history and comorbid conditions. We also conducted a post-hoc interrupted time series analysis to determine if detected changes in rates of referral pre- and post-intervention were related to secular trends. Data analysis was accomplished using R version 3.5.1 [Citation24]. The study was approved by the Human Subjects Division of The MetroHealth System (IRB19-00261).
Results
Between March 23, 2015 and September 24, 2018, there were 10,949 unique patients eligible for inclusion. 4,895 patients (44.7%) were excluded because they had completed spirometry prior to the cohort start dates. The pre-intervention cohort consisted of 2,662 patients (35.9% of all patients system-wide with a diagnosis of COPD and a primary care visit during that time period) and there were 3,392 patients in the post-intervention cohort (41.2% of all patients with a diagnosis of COPD and a primary care visit during that time period). Baseline demographic features for study participants are noted in . A greater proportion of patients in the post-intervention cohort were referred for spirometry (31.6% vs 20.2%, p < 0.001) and completed spirometry (19.3% vs 13.2%, p < 0.001) compared to patients in the pre-intervention cohort (, and ). Among the patients who completed spirometry, discrete results were missing in 25 patients in the pre-intervention cohort and 33 patients in the post-intervention cohort. Chart review revealed these to be due to erroneous medical record number linkage or inability to complete testing in accordance with lab standards. Among those with spirometry results, 76 patients in the pre-intervention cohort and 65 in the post-intervention cohort did not have post-bronchodilator testing. When tabulating only completed results, there was a non-significant difference between the cohorts in the proportion of studies without evidence of airflow obstruction (58.3% vs. 63.5%, p = 0.07).
Figure 1. Cumulative spirometry referrals over time before the intervention vs. After the intervention.
Figure 2. Cumulative spirometry completion rates over time before the intervention vs. After the intervention.
Table 1. Study cohort demographics.
Table 2. Pulmonary function referral, completion and results.
Among the 948 total patients in the study who underwent spirometry, 585 (61.7%) had no evidence of airway obstruction. A post-hoc interrupted time series analysis showed that the increase in spirometry referral was not related to prior secular trends in referral rates over time (supplementary material).
In a multiple variable analysis consisting of patients from both pre and post-intervention cohorts, patients were more likely to have no evidence of airway obstruction if they were female (Adjusted odds ratio [aOR] = 0.54, 95% Confidence interval [CI] = 0.41 − 0.72), were Black (aOR = 0.55, 95% CI = 0.40 − 0.75), had a BMI over 30 (aOR = 0.40, 95% CI = 0.30 − 0.54), or had comorbid conditions including cardiovascular disease (aOR = 0.79, 95% CI = 0.59 − 1.06) or sinus disease (aOR = 0.59, 95% CI = 0.39 − 0.88) (). A diagnosis of asthma, however, was associated with a greater likelihood of having airway obstruction (aOR = 2.10, 95% CI = 1.42 − 3.12).
Table 3. Results of multivariable logistic regression.
Among the 1,016 patients who received messages via their patient portals to schedule their spirometry, 85 (8.4%) completed spirometry during the post-intervention period, and 61 (6.0%) scheduled the test on their own through the patient portal interface.
Inhaler prescriptions were examined between tested patients with and without evidence of airway obstruction. Among patients with airway obstruction without comorbid asthma, 44.3% were prescribed a long-acting inhaler prior to testing whereas only 25.8% of patients without evidence of airway obstruction were prescribed an inhaler prior to testing (p < 0.001). The majority of patients were prescribed either a long-acting muscarinic antagonist or an inhaled corticosteroid/long-acting beta agonist ().
Table 4. Inhaler class comparisons in patients without a diagnosis of asthma.
Discussion
In this before and after study of patients with a diagnosis of COPD who received primary care within an urban, safety-net hospital system, the implementation of three simultaneous EHR-based interventions across the system was associated with significant increases in the rates of both spirometry referrals and completions. These novel interventions were implemented systemically across a healthcare system over an extended period of 1.5 years. Testing uncovered a sizeable burden of overdiagnosis and potentially overtreated patients in our diverse patient population.
In keeping with data published from other studies, most patients in our cohort who underwent testing (61.7%) had no evidence of obstruction on post-bronchodilator spirometry and therefore were overdiagnosed [Citation5, Citation15, Citation25]. At the individual level, the significance of a diagnosis of COPD in the absence of post-bronchodilator obstruction has been openly debated [Citation26]. Smokers have been shown to have clinical symptoms, resource utilization and/or subclinical radiographic features associated with COPD, even when obstruction has been ruled out [Citation27]. The management of such patients is invariably at the discretion of the treating physician, but there is a potentially greater risk of inappropriate prescription of long-acting inhalers in a way that is not otherwise guided by objective data and evidence-based indications. Importantly, overdiagnosis can lead to missed opportunities for the diagnosis of an alternative cause of a patient’s symptoms. The purpose of our intervention was to obtain spirometry when it was otherwise missing after the diagnosis of COPD was made, presumably based on symptoms and clinical presentation alone. Further work will be required to determine what the individual impact of either obstructed or non-obstructed spirometry results might be on these patients.
Overdiagnosis was more common among women, never smokers, the morbidly obese and those with a diagnosis of asthma or sinus disease, consistent with prior data [Citation25]. In addition, we found that the likelihood of non-obstructed spirometry was higher in self-identified Black patients. The underpinnings and implications of this racial disparity warrant further evaluation, particularly through the lens of the likely misestimation of “normal” values based on self-identified constructs of race [Citation28].
Our finding that patients who had non-obstructive spirometry were significantly less likely to have been on long-acting bronchodilators or inhaled steroids suggests that providers were not as confident in the diagnosis that was documented in the EHR. Many of these diagnoses may be erroneous and otherwise disregarded by primary care providers, but EHRs have an unfortunate potential to propagate such misdiagnoses [Citation29]. Even then, our finding that 25.8% of non-obstructed patients in our cohort were prescribed therapy for COPD indicates that the number of potentially overtreated patients in our cohort was not trivial.
While many studies have highlighted the prevalence of COPD diagnoses without spirometry, our study is the first to evaluate a standardized clinical-decision support-based approach to addressing this gap. By leveraging the EHR and patient portal for this purpose, the process is integrated into standard clinical care and is automated, requiring little to no upkeep. At least from the provider’s perspective, such prompting has been shown to be helpful in closing numerous important care gaps [Citation30–32]. Extending such “health maintenance” prompts to the patient portal can also boost gap closure [Citation33]. While our study did leverage direct patient portal outreach, only 1,016 of 3,392 patients had an active patient portal and a fraction of these patients completed testing (85 of 1,016, or 8.4%). Most of these completions were self-scheduled (61 of 85, or 71.8%). The limited success of patient portal outreach in our unique patient population is unfortunately consistent with prior published reports [Citation34].
More than half of our patients who underwent testing were overdiagnosed with COPD, and approximately one quarter of those without a diagnosis of asthma were potentially overtreated. Extrapolating our data to a national estimate of approximately 16 million individuals diagnosed with COPD suggests that there are potentially millions of overdiagnosed patients, hundreds of thousands of whom may be overtreated and could potentially be identified using approaches like ours.
The strengths of our study include the use of reproducible and pragmatic prompts in a diverse real-world patient population over an extended period of time. The generalizability of our study is limited by its reliance on a single site, non-randomized pre-post-design, and inability to specifically attribute the increase of spirometry referrals and completions to any one of the 3 EHR-based interventions that were implemented simultaneously.
Interpretation
EHR-based interventions can increase spirometry referral and completion rates in patients with a clinical diagnosis of COPD who did not have confirmatory spirometry. Most potentially overdiagnosed patients who underwent spirometry were not obstructed, and up to 1/4th of those without a diagnosis of asthma may be overtreated. Further research is warranted to understand the clinical implications of uncovering COPD overdiagnosis and its effect on overtreatment.
Declaration of interest statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
| Abbreviations | ||
| aOR | = | Adjusted Odds Ratio |
| BOLD | = | Burden of Obstructive Lung Disease Study |
| CI | = | Confidence Interval |
| COPD | = | Chronic Obstructive Pulmonary Disease |
| CPT | = | Current Procedural Terminology |
| EHR | = | Electronic Health Record |
| FEV1 | = | Forced Expiratory Volume in one second |
| FVC | = | Forced Vital Capacity |
| GOLD | = | Global Initiative for Chronic Obstructive Lung Disease |
| ICD | = | International Disease Classification |
| ICS | = | Inhaled Corticosteroid |
| LABA | = | Long-acting Beta agonist |
| LAMA | = | Long-acting Muscarinic antagonist |
| OR | = | Odds Ratio |
Supplemental Material
Download PDF (208.4 KB)Acknowledgements
A previous version of this manuscript was presented at the American Thoracic Society 2020 International Conference. All authors have seen and approved this manuscript. The manuscript does not report the results of a clinical trial.
Additional information
Funding
References
- Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med. 2006;3(11):e442. DOI:10.1371/journal.pmed.0030442
- Vestbo J, Hurd SS, Agusti AG, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2013;187(4):347–365. DOI:10.1164/rccm.201204-0596PP
- Han MK, Kim MG, Mardon R, et al. Spirometry utilization for COPD: how do we measure up? Chest. 2007;132(2):403–409. DOI:10.1378/chest.06-2846
- Lee TA, Bartle B, Weiss KB. Spirometry use in clinical practice following diagnosis of COPD. Chest. 2006;129(6):1509–1515. DOI:10.1378/chest.129.6.1509
- Josephs L, Culliford D, Johnson M, et al. COPD overdiagnosis in primary care: a UK observational study of consistency of airflow obstruction. NPJ Prim Care Respir Med. 2019;29(1):33.
- Diab N, Gershon AS, Sin DD, et al. Underdiagnosis and overdiagnosis of chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2018;198(9):1130–1139. DOI:10.1164/rccm.201804-0621CI
- Spero K, Bayasi G, Beaudry L, et al. Overdiagnosis of COPD in hospitalized patients. Int J Chron Obstruct Pulmon Dis. 2017;12:2417–2423. DOI:10.2147/COPD.S139919
- Berry CE, Yawn BP. COPD overdiagnosis, underdiagnosis, and treatment. Chronic Obstr Pulm Dis. 2016;3(1):491–497.
- Miller MR, Levy ML. Chronic obstructive pulmonary disease: missed diagnosis versus misdiagnosis. BMJ. 2015;351:h3021. DOI:10.1136/bmj.h3021
- Buffels J, Degryse J, Liistro G, et al. Differential diagnosis in a primary care population with presumed airway obstruction: a real-life study. Respiration. 2012;84(1):44–54. DOI:10.1159/000332836
- Walters JA, Walters EH, Nelson M, et al. Factors associated with misdiagnosis of COPD in primary care. Prim Care Respir J. 2011;20(4):396–402. DOI:10.4104/pcrj.2011.00039
- Jones RC, Dickson-Spillmann M, Mather MJ, et al. Accuracy of diagnostic registers and management of chronic obstructive pulmonary disease: the devon primary care audit. Respir Res. 2008;9:62. DOI:10.1186/1465-9921-9-62
- Sichletidis L, Chloros D, Spyratos D, et al. The validity of the diagnosis of chronic obstructive pulmonary disease in general practice. Prim Care Respir J. 2007;16(2):82–88. DOI:10.3132/pcrj.2007.00016
- Tinkelman DG, Price DB, Nordyke RJ, et al. Misdiagnosis of COPD and asthma in primary care patients 40 years of age and over. J Asthma. 2006;43(1):75–80. DOI:10.1080/02770900500448738
- Sator L, Horner A, Studnicka M, et al. Overdiagnosis of COPD in subjects with unobstructed spirometry: a BOLD analysis. Chest. 2019;156(2):277–288. DOI:10.1016/j.chest.2019.01.015
- White P, Thornton H, Pinnock H, Georgopoulou S, et al. Overtreatment of COPD with inhaled corticosteroids-implications for safety and costs: cross-sectional observational study . PLoS One. 2013;8(10):e75221. DOI:10.1371/journal.pone.0075221
- Suissa S, Patenaude V, Lapi F, et al. Inhaled corticosteroids in COPD and the risk of serious pneumonia. Thorax. 2013;68(11):1029–1036. DOI:10.1136/thoraxjnl-2012-202872
- Singh S, Loke YK, Furberg CD. Inhaled anticholinergics and risk of major adverse cardiovascular events in patients with chronic obstructive pulmonary disease: a systematic review and Meta-analysis. JAMA. 2008;300(12):1439–1450. DOI:10.1001/jama.300.12.1439
- Price DB, Trudo F, Voorham J, et al. Adverse outcomes from initiation of systemic corticosteroids for asthma: long-term observational study. JAA. 2018;Volume 11:193–204. DOI:10.2147/JAA.S176026
- Spyratos D, Chloros D, Michalopoulou D, et al. Estimating the extent and economic impact of under and overdiagnosis of chronic obstructive pulmonary disease in primary care. Chron Respir Dis. 2016;13(3):240–246. DOI:10.1177/1479972316636989
- Lucas AE, Smeenk FJ, Smeele IJ, et al. Diagnostic accuracy of primary care asthma/COPD working hypotheses, a real life study. Respir Med. 2012;106(8):1158–1163. DOI:10.1016/j.rmed.2012.03.002
- National Center for Health Statistics. Classification of Diseases, Functioning, and Disability. 2020. https://www.cdc.gov/nchs/icd/index.htm. Accessed 6/16/2020.
- Quanjer PH, Stanojevic S, Cole TJ, ERS Global Lung Function Initiative, et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations. Eur Respir J. 2012;40(6):1324–1343. DOI:10.1183/09031936.00080312
- R: a language and environment for statistical computing [computer program]. Vienna, Austria: R Foundation for Statistical Computing; 2013.
- Ghattas C, Dai A, Gemmel DJ, et al. Over diagnosis of chronic obstructive pulmonary disease in an underserved patient population. Int J Chron Obstruct Pulmon Dis. 2013;8:545–549. DOI:10.2147/COPD.S45693
- Vanfleteren L, Andersson AE, Fabbri LM. COPD: What’s in a name?: mismatch of diagnostic labels and required physiologic features. Chest. 2019;156(2):195–196. DOI:10.1016/j.chest.2019.05.008
- Woodruff PG, Barr RG, Bleecker E, et al. Clinical significance of symptoms in smokers with preserved pulmonary function. N Engl J Med. 2016;374(19):1811–1821. DOI:10.1056/NEJMoa1505971
- Kumar R, Seibold MA, Aldrich MC, et al. Genetic ancestry in lung-function predictions. N Engl J Med. 2010;363(4):321–330. DOI:10.1056/NEJMoa0907897
- Graber ML, Siegal D, Riah H, et al. Electronic health Record-Related events in medical malpractice claims. J Patient Saf. 2019;15(2):77–85. DOI:10.1097/PTS.0000000000000240
- Bright TJ, Wong A, Dhurjati R, et al. Effect of clinical decision-support systems: a systematic review. Ann Intern Med. 2012;157(1):29–43. DOI:10.7326/0003-4819-157-1-201207030-00450
- Linder JA, Ma J, Bates DW, et al. Electronic health record use and the quality of ambulatory care in the United States. Arch Intern Med. 2007;167(13):1400–1405. DOI:10.1001/archinte.167.13.1400
- Sequist TD, Gandhi TK, Karson AS, et al. A randomized trial of electronic clinical reminders to improve quality of care for diabetes and coronary artery disease. J Am Med Inform Assoc. 2005;12(4):431–437. DOI:10.1197/jamia.M1788
- Wright A, Poon EG, Wald J, et al. Randomized controlled trial of health maintenance reminders provided directly to patients through an electronic PHR. J Gen Intern Med. 2012;27(1):85–92. DOI:10.1007/s11606-011-1859-6
- Perzynski AT, Roach MJ, Shick S, et al. Patient portals and broadband internet inequality. J Am Med Inform Assoc. 2017;24(5):927–932. DOI:10.1093/jamia/ocx020