Strong fed­eral role needed to or­ga­nize pro­duc­tive use of pa­tient data

Modern Healthcare - - NEWS -

The era of big data is ap­proach­ing. It’s hard to un­der­es­ti­mate its po­ten­tial for rev­o­lu­tion­iz­ing the prac­tice of medicine, de­vel­op­ing and im­prov­ing tech­nolo­gies and up­grad­ing the de­liv­ery sys­tem’s over­all ef­fi­cacy and ef­fi­ciency. Yet none of that will hap­pen un­less the govern­ment plays a prom­i­nent role in or­ga­niz­ing the col­lec­tion and use of the mas­sive amounts of pa­tient in­for­ma­tion that will soon be within reach.

To gauge the im­por­tance of the fed­eral role, look no farther than the $22.5 bil­lion it will spend on physi­cians and hos­pi­tals to adopt elec­tronic health records. It ac­cel­er­ated a process that, sadly, had lagged be­hind other in­dus­tries. But now, dig­i­ti­za­tion of med­i­cal records is ir­re­versible—adop­tion of­fi­cially passed the halfway point last month. That cre­ates a golden op­por­tu­nity for re­searchers and an­a­lysts to gather pri­vacy-pro­tected data and put it to pro­duc­tive use.

Within hos­pi­tals, sys­tem of­fi­cials will have the ca­pac­ity to quickly an­a­lyze vari­a­tions in prac­tice pat­terns, de­ter­mine what works best, and en­cour­age prac­ti­tion­ers to make ad­just­ments that im­prove pa­tient out­comes. Of­fi­cials can pin­point the causes of high read­mis­sion rates, spot med­i­ca­tion er­rors and iden­tify and bet­ter man­age high-cost pa­tients.

More broadly, the era of big data al­lows providers and pay­ers to merge their de-iden­ti­fied pa­tient records into huge cloud-based data­bases. Given ac­cess to this trove of in­for­ma­tion, re­searchers could run ret­ro­spec­tive clin­i­cal tri­als, con­duct com­par­a­tive ef­fec­tive­ness re­search and an­a­lyze the real world out­comes of highly touted tech­nolo­gies to see if they ac­tu­ally lived up to their prom­ise.

No area will ben­e­fit more from the era of big data than on­col­ogy. Nearly 200 forms of can­cer strike more than 1.2 mil­lion Amer­i­cans ev­ery year and cause more than 500,000 deaths. Yet no one col­lects or an­a­lyzes the out­comes from the vary­ing treat­ment reg­i­mens that peo­ple re­ceive.

Ear­lier this month, Pub­lic Health Eng­land an­nounced it would be­gin col­lect­ing and ge­net­i­cally an­a­lyz­ing tu­mor sam­ples while track­ing treat­ment reg­i­mens and out­comes for ev­ery sin­gle can­cer pa­tient in its National Health Ser­vice. The data­base will al­low re­searchers to an­a­lyze how ev­ery can­cer sub-type re­sponds to dif­fer­ent reg­i­mens, and ad­just fu­ture treat­ments ac­cord­ingly. So what is hap­pen­ing here? A few years ago, the National Can­cer In­sti­tute formed a Can­cer Bio­med­i­cal In­for­mat­ics Grid (Ca BIG) to col­lect sim­i­lar in­for­ma­tion. But the ef­fort fell apart be­cause few in­sti­tu­tions chose to par­tic­i­pate, and the agency lost in­ter­est.

The Amer­i­can So­ci­ety of Clin­i­cal On­col­ogy, the on­col­o­gists’ pro­fes­sional so­ci­ety, re­cently an­nounced its own datamin­ing ef­fort for breast can­cer with 150,000 med­i­cal records. That’s a pit­tance com­pared to the quar­ter mil­lion Amer­i­can women newly di­ag­nosed with breast can­cer ev­ery year.

Not ev­ery fed­eral agency has fum­bled the ball. The Food and Drug Ad­min­is­tra­tion launched its Sen­tinel pro­ject, which is col­lect­ing more than 125 mil­lion pa­tient drug records to look for safety prob­lems in ap­proved drugs. Pro­ject man­agers are al­ready churn­ing out stud­ies. But cre­at­ing a cloud-based data­base for drug re­search is easy com­pared to the com­plex pay­ment and pa­tient records from the broader health­care sys­tem. That will re­quire col­lab­o­ra­tion be­tween prac­tic­ing physi­cians, hos­pi­tals and post-acute-care set­tings on the care side, and the claims data from pub­lic and pri­vate pay­ers.

Given that hun­dreds of con­sul­tants and com­pa­nies are now of­fer­ing big data an­a­lyt­ics to their clients, it raises the specter of a ca­coph­ony of com­pet­i­tive and non-com­mu­nica­tive “so­lu­tions.” It will take the strong hand of govern­ment to en­sure the era of big data in health­care doesn’t take as long to ar­rive as the onset of EHRs.



Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.