乌鸦传媒

Skipping adjustment calculations for do-not-resuscitate (DNR) rates at hospitals, or omitting DNR rates altogether, may have skewed hospital quality scores, and could gloss over potentially high-quality decision-making or negligent medical practices, researchers reported.

Without accounting for DNR status, it looked like hospitals with higher rates of patients with DNR orders also had increased mortality rates (adjusted odds ratio 1.17, 95% CI 1.04-1.32); however, if DNR rate adjustments were made, hospitals with higher DNR rates had lower mortality (adjOR 0.79, 95% CI 0.70-0.89), , of Boston University School of Medicine, and colleagues, reported in .

"Although the results of this study are relatively straightforward, the interpretation and implications are anything but," , of NYU School of Medicine, wrote in an accompanying editorial.

"Other studies have shown that DNR patients have higher mortality at all levels of severity of illness, which in theory should not be the case since DNR status should be relevant only in catastrophic situations," Horwitz wrote. "Therefore, either excluding patients with early DNR status from quality measures or risk adjustment for DNR status has the potential to obscure a tendency by hospitals to provide a lower standard of care for such patients even if that was not the patient's explicit desire."

Horwitz noted that there is no easy resolution to this conundrum. With the data that currently exists, it's not possible to disentangle whether the increased mortality rates among DNR patients and the increased mortality rates at hospitals with more DNR patients reflects the careful attention paid to patient wishes for less aggressive care, or deaths that resulted from a "failure to rescue."

Horwitz argued that the current state of data reporting and collection has made it impossible to determine whether there actually is an increase in mortality for DNR patients, and increased mortality rates in hospitals with more of these patients. And this has substantial policy implications, for both the current system and any potential changes to the system.

"Qualitative studies might help, as might a more data-driven and patient-centered approach to DNR discussions," Horwitz wrote. "Accounting for DNR status is likely to encourage the health care system to pursue less-aggressive care for frail elderly patients, which may be more patient-centered and improve quality of life but is also likely to result in some degree of excess mortality for DNR patients. However, continuing the status quo incentivizes a more-aggressive overall approach to care, potentially saving some lives but also potentially causing some patients to undergo more interventions than they might otherwise have chosen," Horwitz wrote.

Walkey's team looked at the Project California Sate Inpatient Database, which has a unique, and validated, feature to capture DNR orders that have been written within the first 24 hours of hospitalization. Out of the 90,644 adult pneumonia patients treated across 303 California hospitals in 2011, roughly 5% were admitted. Patients were 52% women, 59% white, and an average age of 73.

Previous research has shown that early DNR orders strongly predict in-hospital and post-hospitalization mortality, according to the authors.

Walkey's group used a random DNR slope model to account for between-hospital variation in associations between hospital DNR and hospital mortality rates, and to diminish potential variation from hospitals that elect to introduce DNR orders at lower levels of illness severity compared with other hospitals.

The researchers split the cohort into quartiles. The median early DNR rate was 16% (interquartile range 9%-22%). The proportion of between-hospital variation for DNR orders unexplained by measured clinical covariates was 21% (95% CI 19%-25%).

Overall, patients who had DNR orders had a higher risk of mortality compared with patients who did not have DNR orders (adjOR 3.74, 95% CI 3.55-3.93).

Without accounting for DNR status, higher hospital DNR rates were associated with increased patient mortality (adjOR 1.17, 95% CI 1.04-1.32). However, after adjustments for DNR rates, and between-hospital variation in the association between DNR status and mortality rates, hospitals with higher DNR rates had lower mortality (adjOR 0.79, 95% CI 0.70-0.89). And out of the 27 hospitals in the high-mortality quartile, only 14 of these remained outlier after DNR status adjustments were calculated.

Without accounting for patient DNR status, hospitals with higher DNR rates appeared to be associated with lower survival, resulting in hospitals with higher DNR rates receiving worse mortality quality rankings, but after risk adjustments for patient DNR status, hospitals with higher DNR rates demonstrated better survival scores, the authors wrote.

Walkey and colleagues proposed that current methods of comparing hospitals, which do not account for patient DNR status, penalize potentially high-quality hospitals that admit a larger proportion of patients who had chosen to forego resuscitation. "Therefore, accounting for DNR status in programs that compare hospital mortality outcomes may substantially affect publicly reportable hospital rankings and hospital reimbursements."

The authors reported limitations to the analysis, including potential misclassification bias in the data set, and its restriction to hospitals in California.

Comment