Studii

interventionale

 Primele trialuri clinice randomizate

efectuate in anii 1940 sub egida Medical
Research Council.
trialuri cu evaluarea streptomcinei in
tratamentul tuberculozei (1948)
1950 - Cochran si Cox publica primul
tratat asupra trialurilor clinice randomizate
1965 Bradford Hill & Richard Doll Principles of medical statistics

Studii interventionale
Studii interventionale

Date obtinute de la
grupuri
Studiu de interventie
in comunitate

Date obtinute
individualizat
Trial clinic randomizat
Trial in teren

Doua tipuri de studiu unitatea evaluata si

expunerea este:
1. individualizata
2. obtinuta populational

Studii interventionale individualizate:

Trialuri clinice randomizate (terapeutice)
Trialuri in teren (trialuri de profilaxie primara)
subiectii sunt protejati printr-o masura profilactica
si nu vor dezvolta boala

Studii de interventie in comunitate - masura

educationala, profilactica la nivel populational

Trialul clinic randomizat (TCR)

Experiment stiintific utilizat pentru aprecierea
eficacitatii si eficientei serviciilor de sanatate
Masoara eficacitatea terapeutica a unor proceduri
terapeutice
Testarea produselor farmaceutice

Implica alocarea random (dependenta de

sansa) a interventiei asupra subiectilor
Randomizarea elimina eroarea sistematica de
selectie (R)

Investigatorul realizeaza controlul expunerii

(C)
Expunerea este specifica

Interventie
Interventie

+
Efecte
Efectepozitive
pozitivesisi
negative
negative

Alocare
Alocare
random
random

Fara
Fara
interventie
interventie

Esantion
Esantion

Obtinerea
consimtamantului
informat

TCR
Constituirea esantionului de studiu
(consimtamant informat)
Criterii de includere si excludere
Definitie de caz, preferabil cazuri noi de
boala
Masuratori initiale inainte de interventie

Randomizare
Aplicarea interventiei
Masurarea efectelor, repetitiv, end-point
Analiza rezultatelor

TCR
Avantaje
Cea mai buna dovada asupra relatiei cauza-efect
Singurul design posibil pentru introducerea oricarei
interventii medicale terapeutice
Raspuns precis la problema de sanatate evaluata

Dezavantaje

Cost: timp, resurse umane, bani

Bariere: etice, frecventa evenimentului
Poate aduce schimbari in practica medicala
Restrictia la un singur raspuns

Tipuri de trialuri clinice randomizate

Martori istorici - Comparatie cu rezultate
obtinute cu tratamentul conventional anterior
Trial clinic fara randomizare
- studiu simultan pentru grupul tratat si netratat
dar fara randomizare
TCR
- studiu cu randomizarea subiectilor pentru
interventie/non-interventie

TC fara randomizare
Martorii si subiectii tratati sunt urmariti simultan
Alocarea interventiei este non-random
Populatia --- Noua interventie ------------> Endpoint
Populatia--- Martori ------------------------> Endpoint

TC fara randomizare
Avantaje
Selectia interventiei bazata pe criterii clinice
Se castiga experienta cu interventia noua

Dezavantaje
Limitarea comparabilitatii
Studiile pe indivizi putini nu permit detectarea
diferentelor
Selectia tratamentului cu deviere in favoarea
unor subiecti

TCR

* "The Standard"
* Studiu in paralel
Noua interventie------> Endpoint
Populatia ---> Randomizarea
Martori------------------> Endpoint

TCR
Avantaje
Sansa egala la tratament
Elimina eroarea sistematica de selectie in
alocarea pe grup tratat/netratat

Comparabilitatea
Egalitate a prezentei
complianta si efect

factorilor

Validitatea testelor statistice

legati

de

Motive impotriva efectuarii

TCR
Filozofice
Etice
Logistice

Interpretarea efectelor terapeutice se face pe baza riscului relativ, la fel

ca n studiile de cohort dar expunerea este reprezentat de tratament
iar RR sub valoarea de 1 semnific beneficiu terapeutic (pentru c
riscul evenimentului negativ la expui este riscul la cei tratai i este
mai mic dect la cei netratai).

Riscul relativ, RR = Riscul (decesului) la tratai /

Riscul (decesului) la cei netratai. Ex: RR = 0,15/
0,20 = 0,75

Reducerea absolut a riscului, RAR = R netratai - R

la tratai. Ex. RAR = 0,20-0,15 = 0,05 sau 5%
Reducerea riscului relativ, RRR = 1- RR Ex. RRR = 10,75 = 0,25 sau 25%
Numr de indivizi necesar de tratat pentru evitarea
unui eveniment negativ= NNT = 1/ RAR

Ex. NNT = 1/0,05 = 20 indivizi. De exemplu, hipolipemiantele

au efecte benefice la NNT de aproximativ 25, adic tratnd

Trialul in teren
Persoane sanatoase dar cu risc de face boala
Grupurile comparate identic cu TCR dar un grup este protejat celalalt nu
Masura este profilactica

Se evalueaza comparativ rezultatele pozitive (boala devine

rara in grupul in care se face profilaxie) si reactiile adverse

Eficacitate vaccinala
EV=Inevaccinati - I vaccinati / Inevaccinati

= 1- Ivaccinati/Inevaccinati= 1- RR

(I=incidenta)

Privire de ansamblu asupra studiilor clinice realizate cu vaccinul

Rota: date de eficacitate si siguranta
Programul global de studii clinice vaccinul Rota pe 5 continente
a inrolat peste >100,000 sugari pana la acest moment
Belgium
Germany
Finland
USA
Mexico
Panama
Costa Rica

Canada
Czech Republic
France

Venezuela

Spain

Brazil

Italy

Singapore
Bangladesh

Honduras

Hong Kong

Nicaragua

Taiwan

Dominican Rep.
Colombia
Peru
Chile
Argentina

Thailand
South Africa

Vietnam

Malawi

India
Korea

Vaccinul Rota eficacitate,imunogenicitate si siguranta

in Europa, studiul 036
124 centre in 6 tari din Uniunea Europeana

Vaccin rotaviral:
rezultate privind eficacitatea
Eficacitatea
vaccinului

Europa
faza III2
(dose 106.5 CCID50)

GERV grava

96%
(95% CI 90-99)

Spitalizari
datorate
GERV

Orice GERV

93%
(95% CI 82-100)

87%
(95% CI 80-92)

Spitalizari
datorate
oricarei GE

Vesikari T et al. PIDJ 200423:937-43

Vesikari T et al. ESPID, Basel, Switzerland May 35, 2006, Abstract 75

1
2

74%
(95% CI 46-89)

CCID 50 cell culture infecting dose 50%

ffu foci forming units

Vaccinuri rotavirus istoric

Dezvoltarea vaccinurilor rotavirus a inceput in anii 1970

Primul vaccin rotavirus a fost inregistrat in SUA in 1998:

Rotashield (Wyeth Lederle)
Rhesus-based tetravalent human reassortant vaccine (RRV-TV): vaccin
uman reasortant tetravalent
Retras in 1999 datorita legaturii cauzale cu invaginatia intestinala
(intussusception)

Asociatie temporala
clara

Murphy et al, N Engl J Med 2001 344 56472. Copyright 200x [2001] Massachusetts Medical Society.
All rights reserved

Vaccinarea cu vaccinul Rota

nu creste riscul de invaginatie intestinala
Grup cu Vaccin

Grup Placebo

cohorta evaluata pt siguranta

N=31,673

cohorta evaluata pt siguranta

N=31,552

Cazuri Inv.Int.
0 31 zile

Risc Relativ =

0.85 (0.30 ; 2.42)*

0 100 zile

Risc Relativ =

16

0.56 (0.25 ; 1.24)*

Fara risc crescut de invaginatie intestinala

Ruiz-Palacios G. et al. N. Engl. J. Med. 2006; 354: 11-22 Macias, abstract, ICAAC, 2005, Washington, USA (poster)

*95% CI

Studii de interventie in comunitate

Similar trialului clinic dar in conditii
mai apropiate de realitate
Studii interventionale in comunitati intregi

Unitatea observata si analizata este

grupul nu individul

Fazele studiului de interventie in

comunitate
Populatia A

Populatia B

Se observa frecventa factorului de risc sau a bolii

pentru o perioada de timp
Alocare intamplatoare
Interventia

Masurarea efectului

Martor

Masurarea efectului

Studiul de interventie in comunitate

Avantaje
Estimeaza realistic impactul masurii
Dezavantaje
Pierderea controlului specific experimentului
Foarte costisitor
Tendintele multianuale nu apar evidente

Selectia grupurilor populationale

Scopul este aprecierii masurii de interventie
in comunitate
Reprezentativitatea esantioanelor
comparate este importanta pentru
generalizarea rezultatelor
Comunitatile trebuie alocate random la
procedura profilactica/terapeutica evaluate
initial si urmarite identic.

Randomizarea
Randomizarea se aplica la nivel de
comunitate nu individual
Scopul randomizarii este de a minimaliza
efectul distorsionant al factorilor de
confuzie, adica de a asigura
comparabilitatea

Efecte
Indicatori monitorizati pe parcurs
Evaluari de participare, prezentari media etc

Schimbarea asteptata:
Prevalenta factorului de risc
Atitudini si obiceiuri
Utilizarea serviciilor
Incidenta bolii

Community Intervention Trial for

Smoking Cessation (COMMIT)

Proiect multicentric pentru a testa metode combinate de

control a tabagismului
Conceput pentru marii fumatori care sa fie ajutati in abandonul
si mentinerea abandonului fumatului
Conceput cu interventia media, organizatii mari, capabile sa
intervina in comunitate in acest scop
Inceput in 1989
11 perechi de comunitati
4 ani
Ratele de abandon ale fumatului apreciate in cohorte de mari
fumatori selectate random
Masuri intermediare de evaluare a activitatilor anti-tabac, a
comportamentului a inclus studii transversale si evaluari ale
serviciilor medicale si nemedicale de tip educational
Journal of the National Cancer Institute. 83(22):1620-8, 1991 Nov 20.

Surgical Mask vs N95 Respirator for Preventing

Influenza Among Health Care Workers: A Randomized
Trial
Loeb M, Dafoe N, Mahony J, et al
JAMA. 2009;302:1865-1871. Epub 2009 Oct 1
It became clear in April 2009 that a pandemic influenza with
influenza A (H1N1) would occur. Although the magnitude of the
pandemic was difficult to predict, it was clear that preparedness
by healthcare workers was urgently needed to preserve the
workforce and to prevent viral transmission to highly vulnerable
patients. The major mechanism of transmission is by aerosol
particles, so the major prevention method is wearing a mask,
either the N95 respirator that is favored by the CDC or the
surgical mask that has been deemed satisfactory by other
groups. In view of a lack of data to document benefit or relative
merit of different mask types, the Institute of Medicine (IOM)
convened a meeting on September 3, 2009 and issued a report
entitled Respiratory Protection for Healthcare Workers in the
Workplace Against Novel H1N1 Influenza A.[3] The conference
noted that the N95 respirators filter out 95%-99% of aerosol
particles, but worker tolerance of these is poor; the supply is
clearly inadequate; and most importantly, there are no quality
studies comparing the relative merits of these 2 masks for the
prevention of influenza.

Methods. The participants The purpose of the study was to compare

the relative merits of surgical masks and the fit-tested N95 masks
for protecting healthcare workers from influenza. were nurses
working in areas associated with extensive flu exposure at 8
hospitals in Ontario, Canada. These participants were randomly
assigned to wear either surgical masks or fit-tested N95 masks.
Relative merits were based on the frequency of influenza
demonstrated by polymerase chain reaction (PCR) or by
seroconversion.
Results. The study was stopped prematurely on April 23, 2009 due
to the Health Ministry's recommendation for sole use of the fittested N95 masks. At that point, analysis of data collected from
the 422 nurses showed nearly identical results for laboratoryconfirmed influenza rates, 24% in those wearing surgical masks,
and 23% for the N95 masks. The relative merits of surgical vs fittested N95 respirators for preventing influenza and other
respiratory viruses are summarized in Table 4.

Surgical Mask (n
= 212)

N95 Mask (n =
210)

Laboratory-confirmed influenza

50 (24%)

48 (23%)

-- RT-PCR proven

6 (3%)

4 (2%)

-- HA titer increased 4-fold to

2009 (H1N1)

17 (8%)

25 (12%)

Other respiratory viruses

20 (9%)

22 (11%)

Influenza-like illness

9 (4%)

2 (1%)

Absenteeism

42 days

39 days

Laboratory and Clinical

Outcomes

Conclusion. The investigators concluded that "the

use of the surgical masks compared with the N95
respirators resulted in noninferior rates of
laboratory-confirmed influenza."

Perspective. First, the study points out an obvious defect in the

medical care scientific agenda. It is acknowledged that the mask
issue is very important because nosocomial influenza and
protection of the workforce are a very high priority, but precious
few studies provide any data on the relative merits of these 2 very
different masks. Nevertheless, the IOM, the CDC, and the
Occupational Safety and Health Administration all endorsed
routine use of the N95 masks despite the fact that these masks
were substantially more expensive; a major supply shortage was
predicted; and tolerance by healthcare workers who had to wear
them was poor. The study summarized above seems to
demonstrate a case for surgical masks, but has not persuaded
these authoritative sources.
Reason for selection. This article was included because it deals
with a rather fundamental issue in infection control -- one that is in
great need of a tiny slice of the billions spent on biomedical
research.

Probiotics in the critically ill: A systematic review of the

randomized trial evidence; Petrof EO, Dhaliwal R,
Manzanares W, Johnstone J, Cook D, Heyland DK Critical
Care Medicine (Sep 2012)
OBJECTIVE: Critical illness results in changes to the microbiology of the
gastrointestinal tract, leading to a loss of commensal flora and an overgrowth
of potentially pathogenic bacteria. Administering certain strains of live
76

bacteria (probiotics) to critically ill patients may restore balance to the

microbiota and have positive effects on immune function and gastrointestinal
structure and function. The purpose of this systematic review was to evaluate
the effect of probiotics in critically ill patients on clinical outcomes.

Probiotics in the critically ill: A systematic review of the randomized trial

evidence; Petrof EO, Dhaliwal R, Manzanares W, Johnstone J, Cook D,
Heyland DK Critical Care Medicine (Sep 2012)

DESIGN: Systematic review

MEASUREMENTS AND MAIN RESULTS: We searched
computerized databases, reference lists of pertinent
articles, and personal files from 1980 to 2011. We
included randomized controlled trials enrolling
critically ill adults, which evaluated probiotics
compared to a placebo and reported clinically
important outcomes (infections, mortality, and length
of stay). A total of 23 randomized controlled trials met
inclusion criteria.

 Probiotics were associated with reduced infectious

complications as documented in 11 trials (risk ratio
0.82; 95% confidence interval 0.69-0.99; p = .03). When
data from the seven trials reporting ventilatorassociated pneumonia were pooled, ventilatorassociated pneumonia rates were also significantly
reduced with probiotics (risk ratio 0.75; 95% confidence
interval 0.59-0.97; p = .03). Probiotics were associated
with a trend toward reduced intensive care unit
mortality (risk ratio 0.80; 95% confidence interval 0.591.09; p = .16) but did not influence hospital mortality.
Probiotics had no effect on intensive care unit or
hospital length of stay. Compared to trials of higher
methodological quality, greater treatment effects were
observed in trials of a lower methodological quality.

Probiotics in the critically ill: A systematic review of the

randomized trial evidence; Petrof EO, Dhaliwal R, Manzanares
W, Johnstone J, Cook D, Heyland DK Critical Care Medicine
(Sep 2012)
CONCLUSIONS: Probiotics appear to reduce infectious complications
including ventilator-associated pneumonia and may influence intensive
care unit mortality. However, clinical and statistical heterogeneity and
imprecise estimates preclude strong clinical recommendations. Further
research on probiotics in the critically ill is warranted.

Medicina bazata pe dovezi

O abordare in formularea
deciziei clinice prin care se cauta
date din literatura medicala, se
evalueaza critic rezultatele
cercetarilor si apoi se alege cea
mai adecvata interventie de
adoptat.