Hospital Frailty Risk Score Not Up to Snuff for Patients With COPD Exacerbation

Use of the population-based Hospital Frailty Risk Score (HFRS) may misclassify frailty in hospitalized patients with chronic obstructive pulmonary disease (COPD) exacerbation compared with use of the beside Clinical Frailty Scale (CFS), according to a cross-sectional study.

Using the CFS as a criterion standard measurement, the HFRS had a sensitivity of 27% and a specificity of 93% to detect frailty versus non-frailty among 99 patients, reported Sunita Mulpuru, MD, MSc, of the University of Ottawa in Canada, and co-authors.

In addition, the optimal probability threshold of the HFRS was found to be a score of 1.4 points or higher. “HFRS values greater than or equal to 1.4 had a sensitivity of 69% and specificity of 57% to detect frailty as classified by the CFS,” the authors wrote in JAMA Network Open.

“Clinical interventions such as comprehensive pulmonary rehabilitation, if initiated early in the trajectory of frailty, have the potential to treat, improve, and even reverse the degree of frailty among individuals with COPD,” Mulpuru and team noted. “Although administrative data measures come with advantages of large numbers and routine applicability without additional clinical resource use, they come with potential limitations in terms of the relevance and availability of contributing data points and the lack of contextual clinical judgment.”

According to the CFS, 34 patients in the study were moderately to severely frail, 18 were mildly frail, 33 were vulnerable, and 14 were not frail.

Discrepancies between the two assessment methods were observed for each degree of frailty seen in the study.

For the 33 patients considered vulnerable by the CFS, only five were considered to be in the intermediate-risk frailty group by the HFRS, while of the 18 patients determined to be mildly frail by the CFS, four were classified in the intermediate-risk frailty group by HFRS. Meanwhile, 59% of the 34 patients categorized as moderately or severely frail by the CFS were put in the low-risk frailty group by HFRS.

“On the basis of our results, the HFRS administrative data-based frailty assessment tool cannot fully replace the bedside assessment of frailty by CFS, specifically in the cohort of patients deemed vulnerable, for whom interventions such as pulmonary rehabilitation may be important,” Mulpuru and colleagues wrote. “The HFRS is more specific at detecting frailty (using a score >5 points) and can be easily calculated on a population level with administrative data. However, the poor sensitivity calls into question whether the HFRS would provide value as a widespread screening tool for frailty in the population of hospitalized patients with COPD exacerbation.”

They noted that using the HFRS among hospitalized patients with COPD exacerbation “may result in missed opportunities to provide interventions, such as pulmonary rehabilitation and care planning earlier in the chronic disease trajectory, which can potentially improve the quality of life for patients with COPD.”

“To improve bedside frailty recognition and detection among individuals with COPD, research should focus on adapting existing validated bedside frailty assessment tools to optimize detection of frailty in younger populations, explore the performance of these tools in male vs female patients, and study whether early detection of frailty combined with optimal interventions can subsequently improve clinical outcomes in the chronic disease trajectory,” they concluded.

For this study, Mulpuru and team included 99 patients with COPD exacerbation in the respiratory ward of a single tertiary care academic hospital in Ottawa from December 2016 to June 2019. All used a clinical care pathway for COPD.

Mean patient age was 70.6, 57% were women, and 40.4% had heart disease.

Mulpuru’s group found that mean age increased as the degree of frailty increased (66.9 years among those who were well or managing well vs 72.2 years among those who were moderately to severely frail). There was also a higher proportion of women in each group as the degree of frailty increased (28.6% among those who were well or managing well vs 73.5% among those who were moderately to severely frail).

The authors acknowledged several limitations to their study. For one, only COPD patients using the COPD clinical pathway during hospitalization were analyzed, which may have introduced selection biases. Furthermore, the HFRS is a validated tool for patients who are older than 75, and the study participants were largely younger.