RACE AND OUT­COMES

Higher mor­tal­ity rates seen for white pa­tients, but read­mis­sions higher for mi­nori­ties, re­port finds

Modern Healthcare - 100 Top Hospitals - - FRONT PAGE - By Linda Wil­son

Higher mor­tal­ity rates seen for white pa­tients, but read­mis­sions higher for mi­nori­ties.

Race plays a role in how pa­tients fare on some of the out­come mea­sures in­cluded in Medi­care’s pay-for-per­for­mance pro­grams. That was a key find­ing of an in-depth anal­y­sis from Tru­ven Health An­a­lyt­ics’ Cen­ter for Health­care An­a­lyt­ics, re­leased ex­clu­sively to Mod­ern Health­care. Specif­i­cally, Tru­ven Health found that as a hos­pi­tal’s pro­por­tion of white pa­tients in­creases, so does its 30-day mor­tal­ity rate for heart fail­ure. But they did not find a sig­nif­i­cant re­la­tion­ship be­tween the pro­por­tion of white pa­tients at a hos­pi­tal and 30-day mor­tal­ity rates for heart at­tack and pneu­mo­nia.

Con­versely, as a hos­pi­tal’s pro­por­tion of blacks—or, to a lesser ex­tent, other mi­nori­ties— in­creases, so does its rate of 30-day read­mis­sions for heart fail­ure, heart at­tack and pneu­mo­nia.

To ar­rive at those find­ings, Tru­ven re­searchers an­a­lyzed data in two steps. They looked at hospi-

tal-level data from the CMS web­site, Hos­pi­tal Com­pare. They di­vided pa­tients into three cat­e­gories: white, black and other—a small per­cent­age of the to­tal pop­u­la­tion, in­clud­ing His­pan­ics, North Amer­i­can Na­tives, Asians and oth­ers.

This is where re­searchers found the cor­re­la­tion be­tween the racial com­po­si­tion of a hos­pi­tal’s pa­tient pop­u­la­tion and its per­for­mance on 30-day mor­tal­ity and read­mis­sion rates.

Re­searchers then an­a­lyzed dis­charge-level data, us­ing Medi­care Provider Anal­y­sis and Re­view data from the third quar­ter of 2007 through the sec­ond quar­ter of 2010 to an­a­lyze 30-day mor­tal­ity rates, and 2010 data from the CMS’ Stan­dard An­a­lyt­i­cal Files to track 30day read­mis­sions. What they found: Whites had a higher rate of 30-day mor­tal­ity for heart fail­ure than blacks or other mi­nori­ties, while blacks and other mi­nori­ties had higher rates of 30-day read­mis­sion for heart at­tack, heart fail­ure and pneu­mo­nia than whites.

But the find­ing on 30-day mor­tal­ity for con­ges­tive heart fail­ure puz­zled re­searchers. Given the im­pact of racial dis­par­i­ties in health­care, they ex­pected blacks to die at a faster rate. U.S. Cen­sus data re­vealed a pos­si­ble ex­pla­na­tion: Blacks had higher over­all death rates than whites be­tween ages 35 to 84. By the time blacks are old enough to qual­ify for Medi­care, their num­bers have been de­pleted. Mean­while, the death rate for whites catches up be­cause “in the end, ev­ery­body dies,” says David Fos­ter, lead sci­en­tist at Tru­ven’s Cen­ter for Health­care An­a­lyt­ics.

Tru­ven’s find­ings come on the heels of a sim­i­lar study pub­lished last year in the Jour­nal of the Amer­i­can Med­i­cal As­so­ci­a­tion. In that study, Har­vard Univer­sity re­searchers found that elderly blacks had higher rates of 30-day read­mis­sion for heart at­tack, heart fail­ure and pneu­mo­nia than elderly whites.

Their work also re­vealed that pa­tients dis­charged from mi­nor­ity-serv­ing hos­pi­tals—in­sti­tu­tions in the top 10% based on the pro­por­tion of blacks in the to­tal pa­tient pop­u­la­tion—had higher rates of read­mis­sion. Like Tru­ven, they also an­a­lyzed Medi­care ad­min­is­tra­tive data.

The find­ings from both stud­ies add fuel to an on­go­ing de­bate about whether to in­cor­po­rate race into the risk-ad­just­ment method­ol­ogy that the CMS uses to cal­cu­late out­come mea­sures, such as 30-day mor­tal­ity and read­mis­sion rates. While the CMS mod­els in­clude risk ad­just­ments to ac­count for the im­pact of age, sever­ity of ill­ness and case mix, the agency does not ad­just for race.

Tru­ven’s Fos­ter says the lack of risk ad­just­ment for race is un­fair be­cause some hos­pi­tals will fare worse on 30-day read­mis­sion or mor­tal­ity rates based solely on the racial com­po­si­tion of their pa­tients “That is the whole point of do­ing the ad­just­ments—it lev­els the playing field and takes those dif­fer­ences out,” he says.

The is­sue has got­ten more at­ten­tion lately be­cause the CMS plans to im­ple­ment per­for­mance-based re­im­burse­ment pro­grams that in­clude th­ese out­come mea­sures. This Oc­to­ber, the CMS will launch the hos­pi­tal read­mis­sions re­duc­tion pro­gram, which as­sesses penal­ties for ex­ces­sive read­mis­sions. In Oc­to­ber 2013, the agency plans to add 30-day mor­tal­ity to the method­ol­ogy it uses to as­sess hos­pi­tals’ per­for­mance as part of the value-based pur­chas­ing pro­gram, which de­buts in Oc­to­ber 2012.

That is why the Amer­i­can Hos­pi­tal As­so­ci­a­tion has pub­licly com­mented on the CMS’ riskad­just­ment method­olo­gies. For ex­am­ple, in a let­ter last year, the AHA urged Dr. Don­ald Ber­wick, who was the CMS ad­min­is­tra­tor at the

time, to in­clude race and lim­ited English pro­fi­ciency in the risk-ad­just­ment method­ol­ogy for the hos­pi­tal read­mis­sions re­duc­tion pro­gram.

The AHA also has ex­pressed con­cern about the im­pact of dual-el­i­gi­bles, or those en­rolled in both Medi­care and Med­i­caid, and a group that is of­ten a proxy for so­cio-eco­nomic sta­tus in re­search stud­ies. So­cio-eco­nomic sta­tus and race “tend to go together. It is hard to sep­a­rate the two,” says Nancy Fos­ter, the AHA’s vice pres­i­dent of qual­ity and pa­tient-safety pol­icy.

None­the­less, of­fi­cials at the CMS re­jected the idea of ad­just­ing read­mis­sion mea­sures to ac­count for dis­par­i­ties in out­comes based on race in its fi­nal rule for the hos­pi­tal read­mis­sions re­duc­tion pro­gram. To ex­plain their de­ci­sion, CMS of­fi­cials wrote, “Dif­fer­ences in the qual­ity of health­care re­ceived by cer­tain racial and eth­nic groups may be ob­scured if the mea­sures riskad­just for race and eth­nic­ity. Also, risk-ad­just­ing for pa­tient race, for in­stance, may sug­gest that hos­pi­tals with a high pro­por­tion of mi­nor­ity pa­tients are held to dif­fer­ent stan­dards of qual­ity than hos­pi­tals treat­ing fewer mi­nor­ity pa­tients.”

What does the CMS’ cur­rent stance on the is­sue mean for hos­pi­tal ex­ec­u­tives? “Ad­just­ing for race and un­der­stand­ing racial dis­par­i­ties is a com­plex topic,” says Dr. An­thony Slonim, ex­ec­u­tive vice pres­i­dent and chief med­i­cal of­fi­cer at West Orange, N.J.-based Barn­abas Health, which in­cludes two hos­pi­tals on this year’s 100 Top list: Com­mu­nity Med­i­cal Cen­ter in Toms River, N.J., and Clara Maass Med­i­cal Cen­ter in Belleville, N.J. “There is so much con­found­ing that goes on when you are look­ing at some­thing generic like white vs. black that you need more con­text to make it ac­tion­able. What it re­quires for me is to dig deeply into my con­text of care and bet­ter un­der­stand the cir­cum­stances.”

At Barn­abas, he adds, “Race has not been a ma­jor cat­e­gory for us. I could set up a pro­gram for African-Amer­i­cans, but that does me less good than iden­ti­fy­ing the health lit­er­acy prob­lem and mak­ing sure that all pa­tients, re­gard­less of race, know how to take their medicines.”

Health lit­er­acy fig­ures promi­nently into a pro­gram Barn­abas launched this year to re­duce read­mis­sions among pa­tients with con­ges­tive heart fail­ure. All heart fail­ure pa­tients re­ceive at least one phone call af­ter dis­charge; those at higher risk for read­mis­sion get ad­di­tional phone calls and, in some cases, home-care vis­its.

The fo­cus of all of the per­son­al­ized at­ten­tion is to en­sure that pa­tients un­der­stand their dis­ease and all of their dis­charge in­struc­tions, in­clud­ing their med­i­ca­tion reg­i­men.

Barn­abas launched the pro­gram at Com­mu­nity Med­i­cal Cen­ter and Kim­ball Med­i­cal Cen­ter, Lake­wood, N.J. Clara Maass Med­i­cal Cen-

ter is sched­uled to roll out the pro­gram next.

Ex­ec­u­tives at Ad­vo­cate Health Care, Oak Brook, Ill., de­vel­oped an elec­tronic tool to as­sess the im­pact of race and other fac­tors on pa­tients’ risk of be­ing read­mit­ted within 30 days.

Ad­vo­cate has three hos­pi­tals on the cur­rent 100 Top list: Ad­vo­cate Illi­nois Ma­sonic Med­i­cal Cen­ter, Chicago; Ad­vo­cate Christ Med­i­cal Cen­ter, Oak Lawn; and Ad­vo­cate Good Sa­mar­i­tan Hos­pi­tal, Down­ers Grove.

Ad­vo­cate an­a­lyzed 200,000 pa­tients’ med­i­cal records to de­velop the tool, which in­cludes 25 vari­ables across broad cat­e­gories, such as de­mo­graph­ics, med­i­ca­tions, med­i­cal re­source uti­liza­tion and med­i­cal mor­bidi­ties.

Within de­mo­graph­ics, one of the vari­ables they found was, in­deed, race. “African-Amer­i­cans were at an in­creased risk of be­ing read­mit­ted,” says Dr. Rishi Sikka, vice pres­i­dent of clin­i­cal trans­for­ma­tion at Ad­vo­cate. Other de­mo­graphic vari­ables as­so­ci­ated with an in­creased risk of read­mis­sion in­clude pa­tients who are ad­mit­ted from a skilled­nurs­ing fa­cil­ity or have Medi­care as their pri­mary payer.

In the other cat­e­gories, risk fac­tors in­clude med­i­ca­tions, such as war­farin and in­sulin, and cer­tain dis­eases, such as heart fail­ure and cancer. Pa­tients who were hos­pi­tal­ized within the past 12 months or used a lot of med­i­cal ser­vices also were con­sid­ered at higher risk.

Sikka says pa­tients’ risk of read­mis­sion is eval­u­ated us­ing the tool at the be­gin­ning of their hos­pi­tal stay and pe­ri­od­i­cally there­after. Pa­tients earn a to­tal score based on the num­ber of points they ac­cu­mu­late for var­i­ous risk fac­tors. Pa­tients deemed to be at risk for read­mis­sion qual­ify for a se­ries of in­ter­ven­tions, in­clud­ing the new­est: a tran­si­tion coach. The coaches, who are not yet avail­able at all of the sys­tem’s hos­pi­tals, fol­low pa­tients as­signed to them for 30 days af­ter dis­charge with phone calls and home vis­its.

“It is prob­a­bly not pos­si­ble to give a pro­gram to ev­ery­body be­cause some of the in­ter­ven­tions to im­prove a pa­tient’s tran­si­tion in care can be ex­pen­sive. You re­ally need a way to be able to say: Who is at the great­est risk? Let’s tar­get those pa­tients who are at the great­est risk,” Sikka says.

Ad­vo­cate Health Care has de­vel­oped an elec­tronic tool to as­sess the im­pact of race and other fac­tors on pa­tients’ risk of be­ing read­mit­ted within 30 days.

Health lit­er­acy will fig­ure promi­nently into a pro­gram Clara Maass Med­i­cal Cen­ter is plan­ning to roll out.

A pro­gram at Com­mu­nity Med­i­cal Cen­ter aims to re­duce read­mis­sions among pa­tients with con­ges­tive heart fail­ure.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.