Lever­ag­ing Data for Bet­ter Pa­tient Care

So­lu­tions that re­duce vari­ance of di­ag­noses, im­prove qual­ity of care, and de­liver bet­ter pa­tient out­comes

Modern Healthcare - - Regional News - Dar­rell Johnson Vice Pres­i­dent Data Science Medtronic

Look­ing at how data can im­prove health­care — both at a macro and mi­cro level — has been a strat­egy at Medtronic for more than 60 years. Dar­rell Johnson, vice pres­i­dent of data science at Medtronic, dis­cusses the near-and long-term ben­e­fits of data-driven so­lu­tions for clin­i­cians and pa­tients alike.

What is the big­gest chal­lenge fac­ing health sys­tems to­day?

DJ: In to­day’s health­care sys­tem, there is a grow­ing prob­lem re­lated to sup­ply and de­mand.

The uptick in de­mand stems from many fac­tors, but per­haps most sig­nif­i­cantly, an ag­ing pop­u­la­tion with high rates of mul­ti­ple chronic dis­eases. Amer­i­cans with five or more chronic con­di­tions make up 12% of the pop­u­la­tion but ac­count for 41% of to­tal health care spend­ing.

The sup­ply short­age likely comes by way of hospi­tal sys­tems strug­gling to make ends meet. Sixty per­cent of hospi­tal ex­penses are at­trib­uted to la­bor costs, yet, a re­cent sur­vey re­ported nearly 90% of physi­cians sur­veyed only spend about 40% of their time on di­rect pa­tient care.

To close the gap be­tween what pa­tients need and what health­care sys­tems can pro­vide, we’re de­vel­op­ing so­lu­tions de­signed to help clin­i­cians re­duce vari­ance of di­ag­noses, im­prove qual­ity of care, and de­liver bet­ter pa­tient out­comes.

What role does data play?

DJ: Lever­ag­ing data to im­prove health­care has been a long-stand­ing pri­or­ity at Medtronic. To­day, there is a need for ad­di­tional real-world ev­i­dence to sup­port con­trolled pre- or post-mar­ket clin­i­cal stud­ies. In an un­con­trolled en­vi­ron­ment, there are many fac­tors that could im­pact out­comes. Front-line clin­i­cians need an un­der­stand­ing of the spe­cific pa­tient co­hort and these ad­di­tional fac­tors to make in­formed care de­ci­sions, cre­ate bet­ter care path­ways, and im­prove pa­tient out­comes.

With a con­tin­ued fo­cus on data stan­dard­iza­tion through EHRs, there have been tech­nol­ogy ad­vances over the past 10 years to fa­cil­i­tate real-world data sourc­ing and ev­i­dence gen­er­a­tion. Col­lab­o­rat­ing with oth­ers, Medtronic seeks to combine EHR data with non-stan­dard­ized data, pro­vid­ing anal­y­sis and in­sights re­lated to clin­i­cal out­comes.

What are the near-term op­por­tu­ni­ties for data to help clin­i­cians?

DJ: One of our big­gest pri­or­i­ties is to help au­to­mate and con­tex­tu­al­ize data col­lec­tion. Clin­i­cians don’t want a high volume of noisy data. They want highly trusted data, pack­aged with ac­tion­able in­sights. Our goal is to first ap­ply ad­vanced an­a­lyt­ics, ar­ti­fi­cial in­tel­li­gence (AI), and our de­vice tech­nol­ogy to fa­cil­i­tate static and dy­namic data col­lec­tion. From there, we col­lab­o­rate with our cus­tomers to cre­ate com­pre­hen­sive re­port­ing tools that can be lever­aged by clin­i­cians, pay­ers and hospi­tal sys­tems.

What is the long-term po­ten­tial?

DJ: Re­search sug­gests it takes an av­er­age of 17 years for the ma­jor­ity of physi­cians to adopt — in clin­i­cal prac­tice — new ther­apy guide­lines. At Medtronic, we be­lieve real-world clin­i­cal prac­tice out­come ev­i­dence can help shorten time to adop­tion, which will lead to re­duced vari­a­tion and higher qual­ity of care. This out­come ev­i­dence pro­vides deeper in­sight on a pa­tient pop­u­la­tion and can help clin­i­cians de­velop more com­pre­hen­sive, per­son­al­ized, and in­te­grated care path­ways.

We are still in the early phases of cre­at­ing the ideal lon­gi­tu­di­nal pa­tient record and pro­vid­ing clin­i­cians with the right data at the right time. But we are trending in the right di­rec­tion. With the shared goal of im­prov­ing pa­tient out­comes, we con­tinue our work with oth­ers to link mul­ti­ple data sources with valu­able clin­i­cal in­sights. The cur­rent chal­lenge is twofold:

How do we em­power clin­i­cians with ac­tion­able, lon­gi­tu­di­nal data from pa­tients liv­ing with mul­ti­ple chronic con­di­tions? And how can we ef­fi­ciently in­te­grate that data into clin­i­cians’ ev­ery­day work­flow?

How can com­pa­nies like Medtronic help clin­i­cians and hospi­tal sys­tems build a closed-loop feed­back sys­tem so in­di­vid­ual in­sights are shared at a broader, more im­pact­ful level and all health­care stake­hold­ers ben­e­fit?

The op­por­tu­ni­ties to make an im­pact on pa­tient out­comes through data are ex­cit­ing, and we know col­lab­o­ra­tions will be key to get­ting us there.


1. https://www.cdc.gov/nchs/health_pol­icy/adult_chron­ic_­con­di­tions.htm

2. https://www.rand.org/pubs/tools/TL221.html

3. https://revcy­clein­tel­li­gence.com/news/hos­pi­tals-tar­get-la­bor-cost­slay offs-to-re­duce-health­care-costs

4. https://www.ncbi.nlm.nih.gov/pmc/ar­ti­cles/PMC3886458/

5. http://ajcc.aac­n­jour­nals.org/con­tent/25/3/194.full.pdf

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.