Accurately Identifying Heart Failure in Patients With COPD

A cross-sectional retrospective chart review of patients with chronic obstructive pulmonary disease (COPD) and heart failure from primary care practices in British Columbia, Canada, showed that by combining information gleaned from reviewing electronic medical records (EMRs) and examination charts, clinicians could more comprehensively and more accurately define cases of heart failure in their patients with COPD. Up to one-third of patients with COPD have comorbid heart failure, and these conditions together often result in undertreatment of patients due to inherent diagnostic and therapeutic challenges.

“This combination of conditions is also associated with greater morbidity and mortality than either condition alone,” the authors wrote. “To assess the quality of the care this high-risk population receives, an accurate case definition is necessary.”

They evaluated the success of the novel definition via its sensitivity, specificity, positive predictive value, and negative predictive value. Their findings appeared in a recent issue of CMAJ Open.

Eighteen primary care practices were invited to participate in the study. They were recruited from the Canadian Primary Care Sentinel Surveillance Network (CPCSSN), which covers more than 1500 primary care providers in most provinces (except Saskatchewan) and 17 EMRs. Ultimately, 3 practices (encompassing 9 physicians) said yes, consenting to provide their patient charts for review, which took place from September 1 through December 31, 2018.

Overall, heart failure was found in 23.2% of all study patients (N = 311) who had COPD, and it was defined according to national and international guidelines.

The authors note that Maritime Family Practice Research Network developed an initial case definition of heart failure and the CPCSSN suggested adjustments consisting of several definitions, based upon the data provided in the different EMRs, International Classification of Diseases diagnosis codes, and heart failure–specific medications. They then compared these definitions against each of the cases of heart failure, comprising mostly elderly (mean [SD] age, 83.6 [10.9] years) women (58.3%).

Their results on better identifying heart failure among patients with COPD were positive for the 4 measures previously mentioned, although some varied more than others:

• Specificity ranged from 87% to 96.6%
• Sensitivity ranged from 75% to 94.4%
• Negative predictive value ranged from 80.1% to 98.1%
• Positive predictive value ranged from 68.4% to 92.9%

Their preferred definition involved 1 heart failure billing code found in the EMR and a combination of medication codes, for which the 4 measures show:

• Sensitivity was 93.3% (95% CI, 89.4%-96.1%)
• Sensitivity was 90.3% (95% CI, 81.0%-96.0%)
• Positive predictive value was 80.2% (95% CI, 69.9%-88.3%)
• Negative predictive value was 97.0% (95% CI, 93.8%-98.8%)

Including levels of brain natriuretic peptides did not alter their accuracy findings or preferred definition.

The authors noted that high sensitivity improves case findings because “it more completely captures a population, increases the incidence and prevalence, and enhances generalizability,” which enables more accurate estimates of the burden of heart failure and reduces bias, while high positive predictive value “improves identification of true cases.”

“This comprehensive case definition improves upon previous primary care heart failure definitions to include medication codes and laboratory data, along with previously used billing codes,” the authors concluded. “A case definition for heart failure was derived and validated and can be used with data from EMRs to identify heart failure in patients with COPD in primary care accurately.”

The major strengths of their study are that it can inform the management of patients with COPD and heart failure in primary care practices and will permit epidemiological estimates to be made of the prevalence of heart failure in COPD across Canada.

Still, they recommend validation of their findings in future studies and additional practice settings, due to their data only coming from 1 province. “The case definition may perform differently with other clinicians and in other settings,” they stated.

Reference

Vijh R, Wong ST, Grandy M, et al. Identifying heart failure in patients with chronic obstructive lung disease through the Canadian Primary Care Sentinel Surveillance Network in British Columbia: a case derivation study. CMAJ Open. 2021;9(2):E376-E383. doi:10.9778/cmajo.20200183