With one of the highest rates of avoidable hospital-acquired conditions in the country, Yale-New Haven Hospital will be penalized by the federal government for the first time in its history. But physicians at Yale and the hospital are questioning the metric used to determine which hospitals to penalize.

Ranking in the top quartile for rates of hospital-acquired conditions — medical issues that arise during a patient’s stay at a hospital — across the United States, YNHH is one of 14 hospitals in Connecticut that will be penalized between October 2014 and September 2015 by the federal government. Medicare payments, which make up roughly a third of the hospital’s income, will be reduced by one percent at YNHH and 720 other hospitals across the country, costing them $373 million in total.

While the metric used to assess which hospitals to penalize was not developed at Yale, the YNHH Center for Outcomes Research & Evaluation, led by Harlan Krumholz ’80, has recently been asked to refine it.

In an attempt to curb avoidable adverse events in acute care hospitals, the Affordable Care Act, passed in 2010, includes strategies to improve inpatient care. To determine the rate of a hospital HACs, Medicare assesses three types of unanticipated events: blood infections caused by catheters, catheter-associated urinary tract infections and “serious complications,” which are caused by eight types of injuries, including blood clots, bed sores and falls.

“[Tracking HACs] is a blunt instrument,” said Joshua Copel, professor of obstetrics, gynecology, reproductive sciences and pediatrics at the Yale School of Medicine. He added that it is an inadequate measure of patient safety if used without other metrics.

Though YNHH scored poorly on Medicare’s metric, Vice President for Performance Management and YNHH Associate Chief of Staff Thomas Balcezak said it does not accurately reflect the actual safety of the hospital.

Of the more than 4,000 hospitals in the United States, roughly 400 are academic medical centers with major affiliations to a medical school, Balcezak explained. YNHH is one of the largest and most complex of these institutions, he added. Since YNHH handles patients who are sicker than those at most hospitals, the patients are more likely to acquire infections in the hospital, he said.

“We’re the hospital of last resort for many patients,” said Howard Forman, Yale professor of diagnostic radiology, economics and public health, adding that the high number of academic medical centers on the list indicates that the acuity of patients’ conditions is not being considered. Roughly half of all academic medical centers in the nation are on the list for Medicare penalties.

He also drew attention to New Haven’s urban location, noting that poorer populations also tend to be more predisposed to complications.

But unlike strict outcome measures, such as the number of patients that get readmitted after leaving the hospital, complications are more difficult to use as a measure of hospital efficacy, Forman said.

For instance, catheter-associated infections are often mild, and it is difficult for physicians to identify their sources, Forman said. He added that while the infection may go away after the catheter is removed, that in itself is not proof that the catheter caused the infection.

“There are many people in the medical community that think [these measures] are deeply flawed,” said Forman.

There are ways to improve these measures, Forman said, making them more data-driven and less privy to abuse. Balcezak agreed that cheating the system was a possibility, but he said he thinks most institutions are ethical and would not take advantage of the metric.

Krumholz noted that since the penalty is dependent on the results of self-reporting, hospitals that are better at keeping records may be more likely to penalized, as opposed to those with poorer performances. CORE’s work will center on correcting for these biases, Krumholz added.

Balcezak also cited obstacles that may predispose certain institutions to either over- or underreporting their HACs. In order to measure the rate of central-line infections, for example, hospitals need to have standardized methods of counting the number of infections that occur, as well as the total number of days patients spend with central lines inserted. Inadequate data collection methods lead to inaccurate rates, Balcezak said.

He added that an increased emphasis on tracking adverse events and calculating penalties could pull hospital administrators away from focusing on patient safety.

“Time spent looking at billing records instead of fixing underlying processes is a mistake,” Balcezak said.

According to Copel, the one percent reduction in Medicare payments may be very damaging to YNHH’s finances because the hospital normally operates with extremely narrow profits. Since hospitals typically operate with low single-digit profit margins, this penalty would make a huge difference to the amount of funding hospitals have to spend on patients, he added.

“[YNHH] is not like a retail store that has a big markup for profit,” Copel said.

YNHH’s total revenue in fiscal year 2013 was $2.4 billion.