Faulty gauge?

Read­mis­sions are down, but ob­ser­va­tional-sta­tus pa­tients are up— and that could skew Medi­care num­bers

Modern Healthcare - - COVER STORY - Joe Carl­son

The Obama ad­min­is­tra­tion and health pol­icy ex­perts have been tout­ing last year’s de­cline in Medi­care 30-day hos­pi­tal read­mis­sions as ev­i­dence that de­liv­ery and pay­ment re­forms de­signed to pre­vent un­nec­es­sary re­peat vis­its were start­ing to suc­ceed.

But one rea­son the read­mis­sion num­bers may be de­clin­ing is that hos­pi­tals in­creas­ingly are han­dling pa­tients on so-called out­pa­tient ob­ser­va­tion sta­tus, which in many cases is in­dis­tin­guish­able from in­pa­tient ad­mis­sion. Since ob­ser­va­tional-sta­tus pa­tients aren’t counted as ad­mis­sions, they aren’t counted as read­mis­sions if those pa­tients are hos­pi­tal­ized within 30 days. Sim­i­larly, if ob­ser­va­tional pa­tients had been hos­pi­tal­ized within 30 days prior to the ob­ser­va­tional-sta­tus treat­ment, that treat­ment wouldn’t be counted as a read­mis­sion.

The pos­si­ble in­ter­ac­tion of the higher rate of ob­ser­va­tion sta­tus cases and lower rates of read­mis­sions within 30 days is prompt­ing tough ques­tions about whether pol­i­cy­mak­ers and providers should rely so heav­ily on 30-day read­mis­sions as a key barom­e­ter of hos­pi­tal qual­ity and cost per­for­mance.

“Fun­da­men­tally, the ques­tion to me is, have we re­ally done a good job of pre­vent­ing read­mis­sions, or have we just re­as­signed peo­ple who would have been read­mit­ted to a dif­fer­ent sta­tus?” said Dr. Ashish Jha, pro­fes­sor of pub­lic health pol­icy at the Har­vard School of Pub­lic Health in Bos­ton.

The con­cern about the va­lid­ity of Medi­care’s 30-day read­mis­sions data comes just as the stakes for hos­pi­tals are ris­ing. Last Oc­to­ber, the CMS be­gan dock­ing hos­pi­tals’ pay by up to 1% of to­tal Medi­care fund­ing based on how many pa­tients re­turned for care within 30 days for heart fail­ure, heart at­tack and pneu­mo­nia, based on a pro­gram to cut pre­ventable read­mis­sions in the Pa­tient Pro­tec­tion and Af­ford­able Care Act.

Those penal­ties are set to dou­ble this fall, and then triple in Oc­to­ber 2014. The CMS is also ex­pected to add read­mis­sions for more con­di­tions in 2014, and is said to be con­sid­er­ing ex­tend­ing the read­mis­sions penal­ties to skilled-nurs­ing fa­cil­i­ties.

Out­pa­tient ob­ser­va­tion is a form of nona­cute care de­liv­ered to pa­tients in hos­pi­tal beds, and its use has ex­ploded in the past five years—grow­ing so fast that it may ex­plain the de­cline in read­mis­sions. In 2011, the num­ber of Medi­care out­pa­tient ob­ser­va­tion cases in hos­pi­tals rose by 230,000 claims, ac­cord­ing to an anal­y­sis of CMS data per­formed by the Amer­i­can Hos­pi­tal Directory at the re­quest of Mod­ern Health­care. Ob­ser­va­tion fig­ures for 2012 are not yet avail­able.

Ex­perts say the move to­ward greater use of out­pa­tient ob­ser­va­tion is due to the high rates of Medi­care pay­ment de­nials and ag­gres­sive au­dit­ing on short hos­pi­tal stays.

Mean­while, a study pub­lished last month in the Medi­care and Med­i­caid Re­search Re­view—an on­line jour­nal pub­lished by the CMS—found that the per­cent­age of Medi­care pa­tients read­mit­ted for any rea­son within 30 days de­clined to 18.4%, down from 19% in the pre­vi­ous five years. That trans­lated into a de­cline of about 70,000 fewer cases in 2012, as would have been seen if the five-year trend had con­tin­ued for a sixth year.

Pre­ventable read­mis­sions are a costly prob­lem for the CMS. Nearly 1 in ev­ery 5 Medi­care pa­tients re­turns to the hos­pi­tal within 30 days for ad­di­tional care, cost­ing the Medi­care hos­pi­tal trust fund about $18 bil­lion a year. Some por­tion of those re­hos­pi­tal­iza­tions is con­sid­ered pre­ventable though im­proved post-dis­charge care and co­or­di­na­tion, though ex­perts have strong dis­agree­ments about how much can be pre­vented, and by whom.

In a four-year study of hos­pi­tal read­mis­sion-re­duc­tion pro­grams in Mas­sachusetts, Michi­gan and Wash­ing­ton state, re­searchers at the In­sti­tute for Health­care Im­prove­ment in Cam­bridge, Mass., found that when to­tal ad­mis­sions drop, the num­ber of read­mis­sions de­clines as well. But the way the in­for­ma­tion was col­lected pre­vented re­searchers from quan­ti­fy­ing di­rectly how much im­pact ob­ser­va­tion rates had on read­mis­sion rates. “It’s a con­tribut­ing fac­tor,” said Pa­tri­cia Ruther­ford, a vice pres­i­dent with the in­sti­tute and coin­ves­ti­ga­tor on the study.

Here’s how the ob­ser­va­tional-sta­tus and 30-day read­mis­sions is­sues may in­ter­act.

Amy Deutschen­dorf, se­nior di­rec­tor of uti-

liza­tion and clin­i­cal re­source man­age­ment at Johns Hop­kins Health Sys­tem, Bal­ti­more, of­fered the ex­am­ple of an 85-year-old pa­tient with coro­nary stents and long car­diac his­tory who comes to the hos­pi­tal with faint­ing spells known as “syn­cope.”

De­spite this pa­tient’s risk for heart ar­rhyth­mias and the signs of a symp­tom that he may be on the verge of se­ri­ous car­diac event, hos­pi­tals are more likely to ad­mit him to out­pa­tient ob­ser­va­tion or treat him in the emer­gency depart­ment than ad­mit him as an in­pa­tient for syn­cope—even though the bat­tery of test­ing he would un­dergo would likely be the same in any event.

“We can’t ad­mit that pa­tient,” Deutschen­dorf said. “We would have three years ago.”

If the man were ad­mit­ted to the hos­pi­tal in the 30 days be­fore or af­ter the visit, putting him in ob­ser­va­tion could make the hos­pi­tal look bet­ter on pa­per—even though the ob­ser­va­tional-sta­tus clas­si­fi­ca­tion could have fi­nan­cial and qual­ity-of-care reper­cus­sions for the pa­tient.

No one nec­es­sar­ily ac­cuses hos­pi­tals of con­sciously send­ing pa­tients into ob­ser­va­tion as a way to avoid 30-day read­mis­sions penal­ties. Whether the move was trig­gered by high rates of Medi­care pay­ment de­nials and ag­gres­sive au­dit­ing on short hos­pi­tal stays or some other fac­tor, hos­pi­tal of­fi­cials doubt it is a pre­med­i­tated strat­egy.

Dr. Mark Wil­liams, chief of hos­pi­tal medicine for North­west­ern Univer­sity Fein­berg School of Medicine, said he would be sur­prised if the trend was the re­sult of any de­lib­er­ate plan. “I’m sure there is some­one in the U.S. who is do­ing it, but I’ve not seen cases of peo­ple who are talk­ing about it,” he said.

Nancy Foster, vice pres­i­dent for qual­ity and pa­tient safety for the Amer­i­can Hos­pi­tal As­so­ci­a­tion, said there does ap­pear to be some sta­tis­ti­cal re­la­tion­ship be­tween fall­ing read­mis­sion rates and ris­ing use of ob­ser­va­tion. It would be un­for­tu­nate, she said, if Medi­care read­mis­sions penal­ties on hos­pi­tals were caus­ing some clin­i­cians to use ob­ser­va­tion sta­tus more. But she said that un­in­tended con­se­quences can hap­pen even with well-in­ten­tioned met­rics that are de­signed around broad goals that are sup­ported by the ev­i­dence.

“When we get down to the nit­tygritty of a par­tic­u­lar set of mea­sures be­ing ap­plied in a par­tic­u­lar way, that’s when you dis­cover that no mea­sure is per­fect,” Foster said. “And you may be in­vok­ing penal­ties and pres­sur­ing hos­pi­tals and clin­i­cians to make changes that the science shows are not in the pa­tients’ best in­ter­est. And that is the tricky thing.”

Even if read­mis­sion rates may have fallen last year in part due to ris­ing use of ob­ser­va­tional sta­tus, she added, that doesn’t mean hos­pi­tals haven’t made im­por­tant strides in re-en­gi­neer­ing their care pro­cesses to avoid pre­ventable read­mis­sions.

The CMS de­clined to pro­vide comment for this ar­ti­cle.

Mean­while, 14 Medi­care ben­e­fi­cia­ries or their es­tates are su­ing the CMS in U.S. Dis­trict Court in Con­necti­cut, say­ing Medi­care poli­cies are en­cour­ag­ing hos­pi­tals to put pa­tients in ob­ser­va­tional care in­ap­pro­pri­ately, and some­times retroac­tively. The de­ci­sions force pa­tients to pay 20% Medi­care Part B co­pays and the cost of pre­scrip­tions, plus the full cost of re­ha­bil­i­ta­tion care, which ran up­ward of $30,000 for some of the plain­tiffs. Pa­tients who are ad­mit­ted, and thus cov­ered un­der Medi­care Part A hos­pi­tal­iza­tion ser­vices, have their re­hab care paid for by the govern­ment.

Other el­derly pa­tients can’t af­ford the high costs for re­hab care that come with ob­ser­va­tion, and there­fore they may forgo the nec­es­sary skilled nurs­ing, said Ali Bers, an at­tor­ney for the lit­i­gants at the Cen­ter for Medi­care Ad­vo­cacy. “The lack of abil­ity to get needed fol­low-up care can af­fect peo­ple’s health,” she said.

A judge is ex­pected to rule soon on whether to grant the CMS’ mo­tions to dis­miss the law­suit, and also whether to grant the case class-ac­tion sta­tus for se­niors across the coun­try af­fected by ob­ser­va­tion de­ci­sions.

Ob­ser­va­tional-sta­tus pa­tients may un­dergo the same treat­ment as in­pa­tients, but they would not count as read­mis­sions if they re­turned within 30 days.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.