What really happens when you mix medications? | Russ Altman

Prevodilac: Ivana Krivokuća Lektor: Tijana Mihajlović

So you go to the doctor and get some tests.

Odete kod doktora i obavite neke analize.

The doctor determines that you have high cholesterol

Doktor utvrdi da imate visok holesterol

and you would benefit from medication to treat it.

i da bi bilo dobro da uzimate lekove kako biste to lečili,

So you get a pillbox.

pa uzmete pilule.

You have some confidence,

Verujete, vaš lekar veruje da će to da pomogne.

your physician has some confidence that this is going to work.

Kompanija koja je izumela lek je obavila dosta ispitivanja,

The company that invented it did a lot of studies, submitted it to the FDA.

podnela ga na odobrenje Upravi za hranu i lekove.

They studied it very carefully, skeptically, they approved it.

Ispitali su ga veoma pažljivo, skeptično i odobrili ga.

They have a rough idea of how it works,

Imaju izvesnu predstavu o tome kako deluje,

they have a rough idea of what the side effects are.

o tome koje su nuspojave.

It should be OK.

Trebalo bi da bude u redu.

You have a little more of a conversation with your physician

Dodatno ste pričali sa svojim lekarom

and the physician is a little worried because you've been blue,

i on je malo zabrinut jer ste tužni,

haven't felt like yourself,

niste baš svoji,

ne uživate u stvarima u životu kao i obično.

you haven't been able to enjoy things in life quite as much as you usually do.

Your physician says, "You know, I think you have some depression.

Lekar vam kaže: „Znate, mislim da imate depresiju.

I'm going to have to give you another pill."

Moraću da vam dam druge pilule.“

So now we're talking about two medications.

Tako sad govorimo o dva leka.

This pill also -- millions of people have taken it,

Sa tim pilulama je isto - milioni su ih uzimali,

the company did studies, the FDA looked at it -- all good.

kompanija je ispitivala,

Uprava za hranu i lekove je pregledala, sve je u redu.

Think things should go OK.

Mislite da će sa ovim da bude sve u redu.

Think things should go OK.

Mislite da će i sa ovim da bude sve u redu.

Well, wait a minute.

Ipak, sačekajte malo.

How much have we studied these two together?

Koliko smo ova dva leka izučavali zajedno?

Well, it's very hard to do that.

Pa, to je vrlo teško uraditi.

In fact, it's not traditionally done.

Zapravo, to se po običaju ne radi.

We totally depend on what we call "post-marketing surveillance,"

Potpuno zavisimo od onoga što zovemo „postmarketinški nadzor“,

after the drugs hit the market.

nakon što lek bude pušten na tržište.

How can we figure out if bad things are happening

Kako možemo da otkrijemo da li se nešto loše dešava

between two medications?

između dva leka?

Three? Five? Seven?

Tri? Pet? Sedam?

Ask your favorite person who has several diagnoses

Pitajte svoju omiljenu osobu sa nekoliko dijagnoza

how many medications they're on.

koliko lekova uzima.

Why do I care about this problem?

Zašto je meni stalo do ovog problema? Jako mi je stalo do toga.

I care about it deeply.

I'm an informatics and data science guy and really, in my opinion,

Ja sam tip koji se bavi informatikom i naukom o podacima,

i prema mom mišljenju, jedina nada da razumemo ove interakcije

the only hope -- only hope -- to understand these interactions

is to leverage lots of different sources of data

je da usaglasimo mnogo različitih izvora podataka

in order to figure out when drugs can be used together safely

kako bismo otkrili kada se lekovi mogu bezbedno koristiti zajedno,

and when it's not so safe.

a kada baš i nije bezbedno.

So let me tell you a data science story.

Dopustite da vam ispričam priču o nauci o podacima.

And it begins with my student Nick.

Počinje sa mojim studentom Nikom.

Let's call him "Nick," because that's his name.

Zvaćemo ga „Nik“, jer mu je to ime.

(Laughter)

(Smeh)

Nick was a young student.

Nik je bio mladi student.

I said, "You know, Nick, we have to understand how drugs work

Rekao sam: „Znaš, Nik, moramo da razumemo kako lekovi deluju,

kako deluju zajedno i kako deluju zasebno,

and how they work together and how they work separately,

and we don't have a great understanding.

a ne razumemo mnogo o tome.

But the FDA has made available an amazing database.

Međutim, Uprava za hranu i lekove je objavila neverovatnu bazu podataka.

It's a database of adverse events.

To je baza podataka o neželjenim događajima.

They literally put on the web --

Bukvalno su je postavili na mrežu -

publicly available, you could all download it right now --

javno je dostupna, svi možete da je sada skinete -

hundreds of thousands of adverse event reports

stotine hiljada izveštaja o neželjenim događajijma

from patients, doctors, companies, pharmacists.

od pacijenata, doktora, kompanija, farmaceuta.

And these reports are pretty simple:

Ti izveštaji su prilično jednostavni.

it has all the diseases that the patient has,

Tu su sve bolesti koje pacijent ima,

all the drugs that they're on,

svi lekovi koje uzima

and all the adverse events, or side effects, that they experience.

i svi neželjeni događaji ili nuspojave koje doživljavaju.

It is not all of the adverse events that are occurring in America today,

To nisu svi neželjeni događaji koji se danas javljaju u Americi,

but it's hundreds and hundreds of thousands of drugs.

ali to su stotine i stotine hiljada lekova.

So I said to Nick,

Rekao sam Niku:

"Let's think about glucose.

„Razmotrimo glukozu.

Glucose is very important, and we know it's involved with diabetes.

Glukoza je veoma važna i znamo da je u vezi sa dijabetesom.

Let's see if we can understand glucose response.

Hajde da vidimo da li možemo da razumemo reakciju glukoze.“

I sent Nick off. Nick came back.

Ispratio sam Nika. Vratio se.

"Russ," he said,

„Ras“, rekao je,

"I've created a classifier that can look at the side effects of a drug

„Napravio sam klasifikator koji može da sagleda neželjene efekte leka

based on looking at this database,

na osnovu pregleda baze podataka

and can tell you whether that drug is likely to change glucose or not."

i može nam reći da li postoji šansa

da će taj lek promeniti nivo glukoze ili ne.“

He did it. It was very simple, in a way.

Uspeo je. Bilo je prosto, na neki način.

He took all the drugs that were known to change glucose

Uzeo je sve lekove za koje se zna da menjaju nivo glukoze

and a bunch of drugs that don't change glucose,

i gomilu lekova koji ne menjaju nivo glukoze

and said, "What's the difference in their side effects?

i zapitao se: „U čemu je razlika između njihovih nuspojava?

Differences in fatigue? In appetite? In urination habits?"

Razlike u osećaju premora? Apetitu? U pogledu vršenja mokrenja?“

All those things conspired to give him a really good predictor.

Sve to u sklopu mu je dalo veoma dobro sredstvo predviđanja.

He said, "Russ, I can predict with 93 percent accuracy

Rekao je: „Ras, mogu da predvidim sa 93 posto verovatnoće tačnosti

when a drug will change glucose."

kada će lek menjati nivo glukoze.“

I said, "Nick, that's great."

Rekao sam: „Nik, to je sjajno.“

He's a young student, you have to build his confidence.

On je mladi student, morate da mu podignete samopouzdanje.

"But Nick, there's a problem.

„Ipak, Nik, postoji problem.

It's that every physician in the world knows all the drugs that change glucose,

Činjenica je da svaki lekar na svetu zna sve lekove koji menjaju nivo glukoze

because it's core to our practice.

jer je to u suštini naše prakse.

So it's great, good job, but not really that interesting,

Tako da je to sjajno, odlično obavljeno, ali nije baš naročito zanimljivo,

definitely not publishable."

definitivno ne nešto što se može objaviti.“

(Laughter)

(Smeh)

He said, "I know, Russ. I thought you might say that."

Rekao je: „Znam, Ras. Pretpostavio sam da ćeš to reći.“

Nick is smart.

Nik je pametan.

„Pretpostavio sam da ćeš to reći, pa sam sproveo još jedan eksperiment.

"I thought you might say that, so I did one other experiment.

I looked at people in this database who were on two drugs,

Posmatrao sam ljude u ovoj bazi podataka koji uzimaju dva leka

and I looked for signals similar, glucose-changing signals,

i tražio sam slične signale, signale za promenu nivoa glukoze,

for people taking two drugs,

za osobe koje uzimaju dva leka,

where each drug alone did not change glucose,

pri čemu svaki lek sam po sebi ne menja glukozu,

but together I saw a strong signal."

ali vidim da zajednički daju jak signal.“

And I said, "Oh! You're clever. Good idea. Show me the list."

Rekao sam: „O, pametan si. Dobra ideja. Pokaži mi spisak.“

And there's a bunch of drugs, not very exciting.

Tu je bila gomila lekova, ne naročito zanimljivo.

But what caught my eye was, on the list there were two drugs:

Ono što mi je privuklo pažnju je da su na spisku bila dva leka:

paroxetine, or Paxil, an antidepressant;

paroksetin ili Paksil, antidepresiv,

and pravastatin, or Pravachol, a cholesterol medication.

i pravastatin ili Pravakol, lek protiv holesterola.

And I said, "Huh. There are millions of Americans on those two drugs."

Rekoh: „Ha! Milioni Amerikanaca koriste ova dva leka.“

Zapravo, kako smo kasnije saznali,

In fact, we learned later,

15 million Americans on paroxetine at the time, 15 million on pravastatin,

15 miliona Amerikanaca uzimalo je paroksetin u to vreme

i 15 miliona pravastatin, a milion, prema našoj proceni, uzimalo je oba.

and a million, we estimated, on both.

So that's a million people

Dakle, to je milion ljudi

who might be having some problems with their glucose

koji možda imaju probleme sa glukozom

if this machine-learning mumbo jumbo that he did in the FDA database

ako ovo čudo od mašinskog učenja koje je on sproveo u bazi Uprave

actually holds up.

zaista drži vodu.

Ipak, rekao sam: „I dalje nije za objavljivanje,

But I said, "It's still not publishable,

because I love what you did with the mumbo jumbo,

mada mi se dopada to što si uradio sa tim čudesima, sa mašinskim učenjem,

with the machine learning,

but it's not really standard-of-proof evidence that we have."

ali to što imamo baš i nije odgovarajuća vrsta dokaza.“

So we have to do something else.

Moramo da uradimo nešto drugo.

Let's go into the Stanford electronic medical record.

Hajde da uđemo u elektronski medicinski zapis Stenforda.

Imamo njegovu kopiju koja je u redu za istraživanje,

We have a copy of it that's OK for research,

we removed identifying information.

uklonili smo informacije za identifikaciju,

And I said, "Let's see if people on these two drugs

i rekao sam: „Hajde da vidimo da li osobe koje uzimaju ova dva leka

have problems with their glucose."

imaju probleme sa glukozom.“

Now there are thousands and thousands of people

Postoje hiljade i hiljade ljudi

in the Stanford medical records that take paroxetine and pravastatin.

u stenfordskim medicinskim podacima koji uzimaju parokesetin i pravastatin.

But we needed special patients.

Međutim, bili su nam potrebni posebni pacijenti.

We needed patients who were on one of them and had a glucose measurement,

Bili su nam potrebni pacijenti

koji su uzimali jedan od tih lekova i imali izmerenu glukozu,

then got the second one and had another glucose measurement,

zatim dobili drugi lek i imali drugu meru glukoze,

all within a reasonable period of time -- something like two months.

a sve to u okviru prihvatljivog vremenskog perioda -

otprilike oko dva meseca.

And when we did that, we found 10 patients.

Kada smo to uradili, pronašli smo deset pacijenata.

However, eight out of the 10 had a bump in their glucose

Međutim, osmoro od tih deset je imalo porast glukoze

when they got the second P -- we call this P and P --

kada su dobili drugi P - nazivamo ih P i P -

when they got the second P.

kada su dobili drugi P.

Either one could be first, the second one comes up,

Bilo koji može biti prvi, zatim nastupa drugi,

glucose went up 20 milligrams per deciliter.

nivo glukoze raste za 1,1 mmol/l.

Just as a reminder,

Samo da podsetim,

you walk around normally, if you're not diabetic,

normalno se krećete, ako niste dijabetičar,

sa glukozom od oko 5.

with a glucose of around 90.

And if it gets up to 120, 125,

Ako dostigne 6,6 - 6,9,

your doctor begins to think about a potential diagnosis of diabetes.

vaš doktor će početi da pomišlja na potencijalnu dijagnozu dijabetesa.

So a 20 bump -- pretty significant.

Tako da porast od 1,1 prilično ima značaja.

I said, "Nick, this is very cool.

Rekao sam: „Nik, ovo je veoma zanimljivo,

But, I'm sorry, we still don't have a paper,

ali, žao mi je, i dalje nemamo rad,

because this is 10 patients and -- give me a break --

jer ovo je deset pacijenata i, ma daj,

to nije dovoljno pacijenata.“

it's not enough patients."

So we said, what can we do?

Stoga smo se zapitali šta možemo da uradimo.

And we said, let's call our friends at Harvard and Vanderbilt,

Rešili smo da pozovemo naše prijatelje sa Harvarda i Vanderbilta -

who also -- Harvard in Boston, Vanderbilt in Nashville,

Harvarda u Bostonu i Vanderbilta u Nešvilu -

who also have electronic medical records similar to ours.

koji imaju elektronske medicinske podatke slične našim.

Let's see if they can find similar patients

Hajde da vidimo da li mogu da nađu slične pacijente

with the one P, the other P, the glucose measurements

sa jednim P, drugim P, merama glukoze

in that range that we need.

u opsegu koji nam je potreban.

God bless them, Vanderbilt in one week found 40 such patients,

Bog ih blagoslovio, Vanderbilt je pronašao 40 takvih pacijenata za nedelju dana,

same trend.

utvrđena je ista tendencija.

Harvard found 100 patients, same trend.

Harvard je pronašao 100 pacijenata, ista tendencija.

So at the end, we had 150 patients from three diverse medical centers

Tako smo na kraju imali 150 pacijenata iz tri različita medicinska centra

that were telling us that patients getting these two drugs

koji su nam ukazivali da pacijenti koji uzimaju ova dva leka

were having their glucose bump somewhat significantly.

imaju donekle značajan porast glukoze.

More interestingly, we had left out diabetics,

Što je još zanimljivije, izostavili smo dijabetičare,

because diabetics already have messed up glucose.

jer dijabetičari već imaju poremećenu glukozu.

When we looked at the glucose of diabetics,

Kada smo pogledali glukozu dijabetičara,

it was going up 60 milligrams per deciliter, not just 20.

bila je povišena za 3,3 mmol/l, ne samo 1,1.

This was a big deal, and we said, "We've got to publish this."

Ovo je bila velika stvar i rekli smo: „Moramo da objavimo ovo.“

We submitted the paper.

Predali smo rad.

It was all data evidence,

Sasvim je obuhvatao dokaze zasnovane na podacima,

data from the FDA, data from Stanford,

podacima iz Uprave za hranu i lekove, podacima iz Stenforda,

data from Vanderbilt, data from Harvard.

iz Vanderbilta i Harvarda.

We had not done a single real experiment.

Nismo sproveli nijedan pravi eksperiment.

But we were nervous.

Ipak, bili smo nervozni.

So Nick, while the paper was in review, went to the lab.

Zato je Nik otišao u laboratoriju dok je rad bio pod razmatranjem.

We found somebody who knew about lab stuff.

Našli smo nekog ko se razumeo u laboratorijske stvari.

I don't do that.

Ja to ne radim.

I take care of patients, but I don't do pipettes.

Brinem se o pacijentima, ali ne koristim pipete.

They taught us how to feed mice drugs.

Naučili su nas kako da miševima dajemo lekove.

We took mice and we gave them one P, paroxetine.

Uzeli smo miševe i dali im jedan P, paroksetin.

We gave some other mice pravastatin.

Nekim drugim miševima smo dali pravastatin,

And we gave a third group of mice both of them.

a trećoj grupi miševa smo dali oba.

And lo and behold, glucose went up 20 to 60 milligrams per deciliter

I gle čuda, glukoza se popela za 1,1 do 3,3 mmol/l kod miševa.

in the mice.

Rad je prihvaćen samo na osnovu informatičkih dokaza,

So the paper was accepted based on the informatics evidence alone,

but we added a little note at the end,

ali smo na kraju dodali malu belešku

saying, oh by the way, if you give these to mice, it goes up.

u kojoj smo naveli da, uzgred, ako ovo date miševima, poveća se.

That was great, and the story could have ended there.

To je bilo sjajno, priča se mogla tu završiti.

But I still have six and a half minutes.

Međutim, imam još šest i po minuta.

(Laughter)

(Smeh)

So we were sitting around thinking about all of this,

Sedeli smo tako i razmišljali o ovome

and I don't remember who thought of it, but somebody said,

i ne sećam se ko se setio toga, ali neko je rekao:

„Pitam se da li pacijenti koji uzimaju ova dva leka

"I wonder if patients who are taking these two drugs

are noticing side effects of hyperglycemia.

primećuju nuspojave hiperglikemije.

They could and they should.

Mogli bi da osete i trebalo bi.

How would we ever determine that?"

Kako bismo uopšte to ustanovili?“

We said, well, what do you do?

Rekli smo, pa, šta ćeš uraditi?

You're taking a medication, one new medication or two,

Uzimaš jedan lek, jedan ili dva nova leka

and you get a funny feeling.

i dobiješ čudan osećaj.

What do you do?

Šta onda radiš?

You go to Google

Odeš na Gugl

and type in the two drugs you're taking or the one drug you're taking,

i uneseš dva leka koja uzimaš ili jedan lek koji uzimaš

and you type in "side effects."

i uneseš „nuspojave“.

What are you experiencing?

Šta je to što doživljavate?

So we said OK,

Rekli smo, u redu,

let's ask Google if they will share their search logs with us,

hajde da pitamo Gugl da li hoće da podele sa nama unose pretraga,

so that we can look at the search logs

tako da možemo da ih pogledamo

and see if patients are doing these kinds of searches.

i vidimo da li pacijenti sprovode takve pretrage.

Google, I am sorry to say, denied our request.

Gugl je, nažalost, odbio naš zahtev.

So I was bummed.

Tako da sam se baš osećao loše.

I was at a dinner with a colleague who works at Microsoft Research

Bio sam na večeri sa kolegom koji radi na istraživanjima u Majkrosoftu

and I said, "We wanted to do this study,

i rekao sam: „Hteli smo da sprovedemo istraživanje

Google said no, it's kind of a bummer."

i Gugl je odbio, baš bezveze.“

He said, "Well, we have the Bing searches."

Odgovorio je: „Pa, mi imamo pretrage sa Binga.“

(Laughter)

(Smeh)

Yeah.

Da.

That's great.

To je sjajno.

Now I felt like I was --

Sada sam se osećao kao da -

(Laughter)

(Smeh)

I felt like I was talking to Nick again.

Kao da ponovo pričam sa Nikom.

He works for one of the largest companies in the world,

Radi u jednoj od najvećih kompanija na svetu

i već pokušavam da učinim da se oseća bolje.

and I'm already trying to make him feel better.

But he said, "No, Russ -- you might not understand.

Međutim, rekao je: „Ne, Ras, možda me nisi razumeo.

We not only have Bing searches,

Ne samo da imamo pretrage sa Binga,

but if you use Internet Explorer to do searches at Google,

već ako koristiš Internet Eksplorer da bi pretraživao na Guglu,

Yahoo, Bing, any ...

Jahuu, Bingu, gde god,

Then, for 18 months, we keep that data for research purposes only."

tada čuvamo te podatke 18 meseci samo u svrhe istraživanja.“

I said, "Now you're talking!"

Uzviknuo sam: „To je već druga priča!“

This was Eric Horvitz, my friend at Microsoft.

To je bio Erik Horvic, moj prijatelj sa Majkrosofta.

So we did a study

Tako smo sproveli istraživanje

where we defined 50 words that a regular person might type in

gde smo odredili 50 reči koje bi bilo koja osoba mogla uneti

if they're having hyperglycemia,

ako imaju hiperglikemiju,

like "fatigue," "loss of appetite," "urinating a lot," "peeing a lot" --

kao što su „premor“, „gubitak apetita“, „učestalo mokrenje“, „često piškanje“ -

forgive me, but that's one of the things you might type in.

oprostite, ali to je jedna od stvari koje biste mogli uneti.

Dakle, imali smo 50 fraza koje smo nazvali „dijabetskim rečima“.

So we had 50 phrases that we called the "diabetes words."

And we did first a baseline.

Prvo smo ustanovili polaznu liniju.

And it turns out that about .5 to one percent

Ispostavilo se da oko 0,5 do 1 posto

of all searches on the Internet involve one of those words.

svih pretraga na internetu obuhvata jednu od ovih reči.

So that's our baseline rate.

To je naša osnova.

If people type in "paroxetine" or "Paxil" -- those are synonyms --

Ako ljudi unesu „paroksetin“ ili „Paksil“ - to su sinonimi -

and one of those words,

i jednu od ovih reči,

the rate goes up to about two percent of diabetes-type words,

dolazi do porasta od oko dva posto za reči koje odgovaraju dijabetesu

if you already know that there's that "paroxetine" word.

ako već znate da je prisutna ta reč „paroksetin“.

If it's "pravastatin," the rate goes up to about three percent from the baseline.

Ako je u pitanju „pravastatin“, porast je oko tri posto u odnosu na polaznu liniju.

If both "paroxetine" and "pravastatin" are present in the query,

Ako su u upitu prisutni i „paroksetin“ i „pravastatin“,

it goes up to 10 percent,

porast je oko 10 posto,

a huge three- to four-fold increase

ogromno trostruko do četvorostruko povećanje

in those searches with the two drugs that we were interested in,

u tim pretragama sa dva leka koja su nas zanimala

and diabetes-type words or hyperglycemia-type words.

i reči vezanih za dijabetes ili hiperglikemiju.

We published this,

Objavili smo ovo

and it got some attention.

i pridobilo je pažnju.

The reason it deserves attention

Razlog zbog kojeg zaslužuje pažnju

is that patients are telling us their side effects indirectly

je što nam pacijenti indirektno govore o svojim nuspojavama

through their searches.

kroz svoje pretrage.

We brought this to the attention of the FDA.

Izneli smo ovo pred Upravu za hranu i lekove.

They were interested.

Bili su zainteresovani.

They have set up social media surveillance programs

Postavili su programe za nadgledanje društvenih medija

to collaborate with Microsoft,

kako bi sarađivali sa Majkrosoftom,

which had a nice infrastructure for doing this, and others,

koji je imao finu infrastrukturu za sprovođenje ovog, i drugima,

to look at Twitter feeds,

da bi pregledali unose na Tviteru,

to look at Facebook feeds,

unose na Fejsbuku,

da bi pregledali unose pretraga,

to look at search logs,

to try to see early signs that drugs, either individually or together,

da bi pokušali da uoče rane znake da lekovi, bilo zasebno ili zajedno,

are causing problems.

stvaraju probleme.

What do I take from this? Why tell this story?

Šta ovde smatram značajnim? Zašto sam ispričao ovu priču?

Well, first of all,

Pa, pre svega, sada imamo nadu u podatke velikih i malih razmera

we have now the promise of big data and medium-sized data

to help us understand drug interactions

koji će nam pomoći da razumemo interakcije lekova

and really, fundamentally, drug actions.

i ono što je zaista u osnovi, dejstva lekova.

How do drugs work?

Kako lekovi deluju?

This will create and has created a new ecosystem

Ovo će stvoriti i stvorilo je novi ekosistem

for understanding how drugs work and to optimize their use.

za razumevanje dejstva lekova i njihovo najoptimalno korišćenje.

Nick went on; he's a professor at Columbia now.

Nik je nastavio sa ovim; danas je profesor na Kolumbiji.

He did this in his PhD for hundreds of pairs of drugs.

Ovo je sproveo u svom doktoratu na stotinama parova lekova.

He found several very important interactions,

Otkrio je nekoliko veoma važnih interakcija

and so we replicated this

i tako smo ovo ponovili

and we showed that this is a way that really works

i pokazali da je ovaj način zaista delotvoran

for finding drug-drug interactions.

u pronalaženju interakcija između lekova.

However, there's a couple of things.

Međutim, u igri je još par stvari.

We don't just use pairs of drugs at a time.

Ne koristimo samo parove lekova u isto vreme.

As I said before, there are patients on three, five, seven, nine drugs.

Kao što sam već rekao,

ima pacijenata koji uzimaju tri, pet, sedam, devet lekova.

Have they been studied with respect to their nine-way interaction?

Jesu li oni izučavani imajući u vidu njihovu devetostruku interakciju?

Yes, we can do pair-wise, A and B, A and C, A and D,

Da, možemo da uzemo parove, A i B, A i C, A i D,

but what about A, B, C, D, E, F, G all together,

ali šta ako A, B, C, D, E, F i G zajedno,

being taken by the same patient,

ako ih uzima isti pacijent,

perhaps interacting with each other

možda međusobno ulaze u interakciju

in ways that either makes them more effective or less effective

na načine koji ih čine bilo više ili manje efikasnim

or causes side effects that are unexpected?

ili stvaraju neočekivane nuspojave?

We really have no idea.

Zaista nemamo predstavu.

It's a blue sky, open field for us to use data

To je ogromno otvoreno polje u kome možemo koristiti podatke

to try to understand the interaction of drugs.

da bismo pokušali da razumemo interakciju lekova.

Two more lessons:

Još dve lekcije.

I want you to think about the power that we were able to generate

Želim da razmislite o moći koji smo uspeli da proizvedemo podacima

with the data from people who had volunteered their adverse reactions

od ljudi koji su dobrovoljno prijavili svoje neželjene reakcije

through their pharmacists, through themselves, through their doctors,

preko njihovih farmaceuta, njih samih, njihovih doktora,

the people who allowed the databases at Stanford, Harvard, Vanderbilt,

ljudi koji su dozvolili pristup bazama podataka

na Stenfordu, Harvardu, Vanderbiltu kako bi bile korišćene u istraživanju.

to be used for research.

People are worried about data.

Ljudi su zabrinuti zbog podataka.

They're worried about their privacy and security -- they should be.

Zabrinuti su zbog svoje privatnosti i bezbednosti i treba da budu.

We need secure systems.

Potrebni su nam bezbedni sistemi.

But we can't have a system that closes that data off,

Ipak, ne smemo imati sistem koji blokira pristup tim podacima

because it is too rich of a source

jer su previše bogat izvor

of inspiration, innovation and discovery

inspiracije, inovacije i otkrića

for new things in medicine.

za nove stvari u medicini.

And the final thing I want to say is,

Poslednje što želim da kažem

in this case we found two drugs and it was a little bit of a sad story.

je da smo ovde otkrili dva leka i to je bila pomalo tužna priča.

The two drugs actually caused problems.

Dva leka su stvarala probleme.

They increased glucose.

Povećavala su nivo glukoze.

They could throw somebody into diabetes

Mogli su da prouzrokuju dijabetes kod nekoga

who would otherwise not be in diabetes,

ko inače ne bi imao dijabetes,

and so you would want to use the two drugs very carefully together,

tako da biste hteli da koristite ta dva leka vrlo pažljivo zajedno,

perhaps not together,

možda ne zajedno,

make different choices when you're prescribing.

doneti drugačije odluke prilikom propisivanja lekova.

But there was another possibility.

Međutim, postoji još jedna mogućnost.

We could have found two drugs or three drugs

Mogli smo da otkrijemo dva ili tri leka

that were interacting in a beneficial way.

koji ulaze u interakciju na povoljan način.

We could have found new effects of drugs

Mogli smo naći nova dejstva lekova

that neither of them has alone,

koje nijedan od njih nema zasebno,

but together, instead of causing a side effect,

već zajedno, umesto da uzrokuju nuspojavu,

they could be a new and novel treatment

mogu biti novi način lečenja

for diseases that don't have treatments

za bolesti koje se ne leče

or where the treatments are not effective.

ili gde način lečenja nije delotvoran.

If we think about drug treatment today,

Ako razmislite o lečenju lekovima danas,

all the major breakthroughs --

svim većim otkrićima -

for HIV, for tuberculosis, for depression, for diabetes --

kod HIV-a, tuberkuloze, depresije, dijabetesa -

it's always a cocktail of drugs.

uvek je tu mešavina lekova.

And so the upside here,

Dobra strana u ovome,

and the subject for a different TED Talk on a different day,

kao i tema nekog drugog TED govora nekog drugog dana

is how can we use the same data sources

je kako možemo da koristimo iste izvore podataka

to find good effects of drugs in combination

da bismo otkrili dobra dejstva lekova u kombinaciji

that will provide us new treatments,

koji će nam pružiti nova lečenja,

new insights into how drugs work

nove uvide u to kako lekovi deluju

and enable us to take care of our patients even better?

i omogućiti nam da se još bolje staramo o našim pacijentima.

Thank you very much.

Hvala vam mnogo.

(Applause)

(Aplauz)