Polic­ing the ethics po­lice

Re­search re­view boards face scru­tiny as feds pro­pose new rules

Modern Healthcare - - NEWS - By Sabriya Rice

In­sti­tu­tional re­view boards (IRBs), de­signed to serve as gate­keep­ers pro­tect­ing pa­tients from un­eth­i­cal hu­man re­search prac­tices, are un­der new scru­tiny from fed­eral reg­u­la­tors and ethi­cists. Crit­ics say stud­ies with se­ri­ous de­sign flaws that could lead to un­eth­i­cal prac­tices or par­tic­i­pant harm con­tinue to win ap­proval be­cause IRBs are too lax, have un­qual­i­fied mem­bers, are rid­dled with in­sti­tu­tional and per­sonal con­flicts of in­ter­est, and are over­whelmed by the vol­ume of stud­ies they must re­view.

Those in­volved with IRBs counter that while the sys­tem isn’t per­fect, the in­tegrity of hu­man re­search would be com­pro­mised with­out their in­volve­ment. They say IRBs have im­proved in terms of trans­parency and con­flict dis­clo­sure.

Last month, the con­sumer ad­vo­cacy group Pub­lic Cit­i­zen and the Amer­i­can Med­i­cal Stu­dent As­so­ci­a­tion com­plained to HHS’ Of­fice for Hu­man Re­search Pro­tec­tions (OHRP) about the eth­i­cal in­tegrity of two on­go­ing stud­ies an­a­lyz­ing the im­pact of longer-than-rec­om­mended work shifts for firstyear med­i­cal res­i­dents on 30-day pa­tient mor­tal­ity rates. Pub­lic Cit­i­zen and the stu­dent as­so­ci­a­tion called the stud­ies “highly un­eth­i­cal” for not seek­ing in­formed con­sent from the med­i­cal res­i­dents or the pa­tients stud­ied. The re­search de­sign, they said, forced hun­dreds of res­i­dents across the coun­try to work “dan­ger­ously long shifts,” plac­ing them and their pa­tients at risk of se­ri­ous harm. HHS said the OHRP is re­view­ing the com­plaints.

The lead in­ves­ti­ga­tors on those two stud­ies, from North­west­ern Univer­sity and the Univer­sity of Penn­syl­va­nia, say their stud­ies were eval­u­ated and ap­proved by IRBs at dozens of par­tic­i­pat­ing sites. They say there is no solid ev­i­dence about whether shorter res­i­dent hours are safer for pa­tients, which is why the stud­ies are needed.

“We just have never had prospec­tive, ran­dom­ized, high-level ev­i­dence to in­form our de­ci­sion­mak­ing,” said one of the study leads,

“First and fore­most, IRBs serve to en­sure that stud­ies are done eth­i­cally and that re­search sub­jects are pro­tected. But they are also serv­ing the re­searchers, as part­ners in ad­vanc­ing good science.” Elisa Hur­ley, ex­ec­u­tive di­rec­tor Pub­lic Re­spon­si­bil­ity in Medicine and Re­search

“Ul­ti­mately, re­searchers should be re­spon­si­ble for poor study de­signs, but they just say, ‘Well, the IRB ap­proved it.’ (IRBs) try their best, but the sys­tem really needs to be im­proved.” Dr. Robert Kl­itz­man, di­rec­tor graduate bioethics pro­gram Columbia Univer­sity

Dr. Karl Bil­imo­ria, vice chair­man for qual­ity in the surgery depart­ment at North­west­ern’s Fein­berg School of Medicine. “That’s why th­ese tri­als are so im­por­tant.”

In Septem­ber, the OHRP and 15 other fed­eral agen­cies pro­posed the first sub­stan­tial re­vi­sions in nearly 25 years to the Com­mon Rule, a set of reg­u­la­tions de­signed to pro­tect med­i­cal re­search par­tic­i­pants. The pro­posed re­vi­sions ad­dress ma­jor shifts in the med­i­cal re­search en­vi­ron­ment since the early 1990s, and would ap­ply to all U.S. clin­i­cal tri­als, re­gard­less of their fund­ing source. Two pro­vi­sions would di­rectly af­fect IRBs.

But Dr. Robert Kl­itz­man, di­rec­tor of the graduate bioethics pro­gram at Columbia Univer­sity, said the pro­posed rules don’t do enough to hold IRBs more ac­count­able and don’t ad­dress what he calls short­com­ings in the qual­i­fi­ca­tions and ex­per­tise of IRB mem­bers.

Kl­itz­man, who in­ter­viewed 46 chair­men, mem­bers and ad­min­is­tra­tors from 34 IRBs for his new book, The Ethics Po­lice? The Strug­gle to Make Hu­man Re­search Safe, said con­flicts of in­ter­est on IRBs may arise from per­sonal or col­le­gial re­la­tion­ships. “We know very lit­tle about how IRBs make de­ci­sions,” he said. “There is no feed­back loop, they are not ac­count­able to any­one, and there is no ex­ter­nal ap­peals process or re­quired train­ing to par­tic­i­pate.”

In Septem­ber, re­searchers from the Univer­sity of South Florida and the Univer­sity of Vir­ginia pub­lished an ar­ti­cle in the Euro­pean jour­nal Acta In­for­mat­ica Med­ica ques­tion­ing whether mem­bers of IRBs fully understand study de­sign. They sur­veyed 148 in­di­vid­u­als with med­i­cal or doc­toral de­grees who par­tic­i­pate on IRBs at three dozen U.S. re­search uni­ver­si­ties. Ex­am­ples of sur­vey ques­tions in­cluded ask­ing whether a ran­dom­ized con­trol trial needed a placebo con­trol group; defin­ing the pur­pose of a Phase I trial; and ex­plain­ing the pur­pose of ran­dom­iza­tion.

Over­all, 26.5% of those sur­veyed an­swered all of the ques­tions cor­rectly. On the six ques­tions spe­cific to ran­dom­ized tri­als, 34% got a per­fect score.

The re­searchers called their find­ings “alarm­ing,” say­ing they showed that “those who are charged with over­sight in the pro­tec­tion of re­search in­tegrity may not be

ad­e­quately equipped to ful­fill their du­ties.”

Such find­ings could un­der­mine the con­fi­dence of re­searchers and pa­tients who rely on IRBs to pro­vide rig­or­ous safety and ethics re­views of pro­posed stud­ies. “Too of­ten (IRBs) ac­cept what­ever the in­ves­ti­ga­tor is telling them and don’t ask the right ques­tions,” said Dr. Michael Carome, di­rec­tor of Pub­lic Cit­i­zen’s health re­search group. “There­fore, they fail to de­tect the flaws in re­search de­sign, the eth­i­cal prob­lems and the risks.”

There also are wor­ries about “ap­proval fatigue” among IRB mem­bers based on the sheer vol­ume of stud­ies they re­view each year. Ac­cord­ing to the OHRP, there are 3,547 reg­is­tered IRBs that re­view re­search sup­ported by HHS or that are reg­u­lated by the Food and Drug Ad­min­is­tra­tion. Those IRBs re­view more than 675,000 re­search pro­to­cols an­nu­ally.

There are many more IRBs at aca­demic cen­ters, hos­pi­tals, health sys­tems and for-profit re­search groups that are not re­quired to reg­is­ter with the fed­eral gov­ern­ment, but there are no data on ex­actly how many ex­ist.

The OHRP re­quires reg­is­tered IRBs to in­clude at least five mem­bers with var­i­ous spe­cific back­grounds, in­clud­ing at least one sci­en­tist, such as a physi­cian or doc­toral-level re­searcher. Also, the board gen­er­ally in­cludes other health­care pro­fes­sion­als such as nurses. In ad­di­tion, IRBs must have “un­am­bigu­ously non-sci­en­tific” mem­bers, such as ethi­cists, at­tor­neys or com­mu­nity lay­men mem­bers. The fed­eral agency does not sys­tem­at­i­cally mon­i­tor or eval­u­ate re­view boards. But it says mem­bers must be “suf­fi­ciently qual­i­fied” in ex­per­tise, di­verse in gen­der and eth­nic­ity, and re­flect “com­mu­nity at­ti­tudes.” It also says mem­bers should not be al­lowed to re­view pro­to­cols where they have a con­flict of in­ter­est. The OHRP can act when there are “sub­stan­tive writ­ten al­le­ga­tions or in­di­ca­tions of non­com­pli­ance with HHS reg­u­la­tions.”

Crit­ics noted a num­ber of con­tro­ver­sies over stud­ies that they say raise ques­tions about the stan­dards, trans­parency and ex­per­tise of IRBs. In 2005, the Univer­sity of Penn­syl­va­nia agreed to pay $517,496, and Chil­dren’s Na­tional Med­i­cal Cen­ter in Wash­ing­ton, D.C., paid $514,622 to the fed­eral gov­ern­ment to re­solve al­le­ga­tions of false

state­ments and claims as­so­ci­ated with a gene ther­apy study con­ducted by the two in­sti­tu­tions. The fam­ily of an Ari­zona teen who died dur­ing the trial said the re­searchers did not dis­close all of the study’s an­tic­i­pated tox­i­c­i­ties. That clin­i­cal trial con­tin­ued for a year be­fore it was shut down, even though sev­eral pa­tients ex­pe­ri­enced sig­nif­i­cant liver tox­i­c­i­ties.

Pub­lic Cit­i­zen pre­vi­ously crit­i­cized a Na­tional In­sti­tutes of Health-funded trial known as the Sur­fac­tant, Pos­i­tive Pres­sure, and Oxy­gena­tion Ran­dom­ized Trial, which ran from 2005 to 2009. Pre­ma­ture ba­bies were ran­dom­ized to re­ceive one of two oxy­gen reg­i­mens to see which was less likely to cause blind­ness. Crit­ics said the par­tic­i­pat­ing aca­demic med­i­cal cen­ters, in­clud­ing the Univer­sity of Alabama, Brown, Duke, Stan­ford and Yale, did not fully in­form par­ents of the trial’s risks, such as reti­nal dam­age, neu­ro­logic in­jury and death.

The OHRP ex­am­ined that study’s pro­to­cols and con­cluded that its re­searchers had failed to ad­e­quately warn prospec­tive par­ents about the risks to their new­borns posed by par­tic­i­pa­tion in the study. NIH Di­rec­tor Dr. Fran­cis Collins later co-au­thored a com­men­tary in the New Eng­land Jour­nal of Medicine dis­agree­ing with the OHRP’s find­ings.

In an­other re­search con­tro­versy cited by IRB crit­ics, law­suits were filed against the Kennedy Krieger In­sti­tute over a study from the 1990s on the haz­ards of pe­di­atric ex­po­sure to lead paint. The in­sti­tute is af­fil­i­ated with Johns Hop­kins Univer­sity, whose IRB ex­am­ined and ap­proved the study pro­to­col.

The plain­tiff in a case from 2001 claimed that as part of that study, in­sti­tute re­searchers tested his home for lead. But re­searchers did not re­veal for nine months that they found high lead lev­els in that home. The plain­tiff also said the study’s con­sent doc­u­ments did not clearly dis­close that par­tic­i­pat­ing chil­dren might ac­cu­mu­late mea­sur­able lev­els of lead in their blood. The court found that the re­search pro­to­col was not within an ac­cept­able level of risk be­cause par­tic­i­pants who were ex­posed to high lead lev­els did not re­ceive treat­ment. But Kennedy Krieger won dis­missal of a sim­i­lar law­suit last year in Bal­ti­more.

IRBs of­ten are tasked with mon­i­tor­ing on­go­ing stud­ies af­ter their ini­tial ap­proval, re­view­ing changes to re­search pro­to­cols and in­formed con­sent doc­u­ments, and eval­u­at­ing prob­lems that arise. But they are not al­ways re­quired to do so, and the FDA says its guidance for IRB fol­low-up re­view is non­bind­ing.

Last month’s con­tro­versy over the North­west­ern and Univer­sity of Penn­syl­va­nia res­i­dent work-hours stud­ies grew out of con­cern that pa­tients would be placed at greater risk by res­i­dents work­ing longer hours than rec­om­mended by the Ac­cred­i­ta­tion Coun­cil for Graduate Med­i­cal Ed­u­ca­tion.

“Any­thing that in­volves putting pa­tients at se­ri­ous risk, es­pe­cially by go­ing against rec­om­men­da­tions … de­mands more con­sent from pa­tients,” said Arthur Caplan, di­rec­tor of the di­vi­sion of med­i­cal ethics at New York Univer­sity’s Lan­gone Med­i­cal Cen­ter.

But the flap con­founded other ex­perts. Dr. Mark Fried­berg, a se­nior nat­u­ral sci­en­tist at the RAND Corp. who fo­cuses on med­i­cal qual­ity and per­for­mance im­prove­ment, said he was sur­prised by the crit­ics’ as­sump­tion that longer work hours au­to­mat­i­cally meant in­creased risks to pa­tients. Shorter work hours mean more pa­tient hand­offs be­tween res­i­dents, which height­ens the risk of mis­com­mu­ni­ca­tion and med­i­cal er­rors, he said. “It’s not clear-cut. That’s why you have to do the science.”

Re­searchers dis­agree over whether the pro­posed re­vi­sions to the Com­mon Rule af­fect­ing IRBs will be pos­i­tive or neg­a­tive. One con­tro­ver­sial rec­om­men­da­tion ad­vo­cates us­ing a sin­gle in­sti­tu­tional re­view board for mul­ti­site re­search stud­ies rather than re­ly­ing on an IRB at each site.

Supporters of the rec­om­men­da­tion say it will re­duce bur­den­some re­quire­ments that slow down re­search. De­lays are of­ten caused by “mul­ti­ple IRBs re­quest­ing mi­nor changes to pro­posed re­search with lim­ited, if any, gains in the pro­tec­tion of hu­man sub­jects,” Ni­cholas Bau­roth, vice chair­man of the IRB at North Dakota State Univer­sity, and his col­leagues said in writ­ten com­ments on the rule.

But Drs. Ann Arvin and Harry Green­berg, the dean and as­so­ciate dean for re­search at Stan­ford’s School of Medicine, wrote that the pro­posed change ad­vo­cat­ing a sin­gle IRB for mul­ti­site stud­ies would mean that an in­di­vid­ual institution could not ef­fec­tively safe­guard hu­man sub­jects en­rolled at its site. Fur­ther­more, they said the change would lead re­searchers to en­gage in “IRB shop­ping … for a more-rapid or less-rig­or­ous re­view.”

The pro­posed rules are open for pub­lic com­ment through Jan. 6. More than 450 com­ments had been posted as of Dec. 9.

Elisa Hur­ley, ex­ec­u­tive di­rec­tor of the not-for-profit Pub­lic Re­spon­si­bil­ity in Medicine and Re­search, said her group’s main con­cern is that the Com­mon Rule changes could re­duce pa­tient-safety pro­tec­tions by al­low­ing re­searchers more free­dom to in­de­pen­dently de­cide whether their stud­ies need ex­ter­nal IRB re­view. Her group of­fers a cer­ti­fi­ca­tion pro­gram for IRB mem­bers.

IRBs, she said, play an im­por­tant role that shouldn’t be down­sized. The reg­u­la­tions “over­look the range of eth­i­cal is­sues that an IRB can un­cover … even for re­search that, on its face, looks like it’s min­i­mal risk. In­ves­ti­ga­tors them­selves may not be in the best po­si­tion to de­ter­mine those risks,” Hur­ley said.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.