Transcript No Slide Title - Tulane University

Study Design
Design of Studies in STD Research
Objectives:
• Discuss the following study designs:
–
–
–
–
cross-sectional
case-control
Cohort
Clinical trial
• Discuss the components of study design:
– Study Design, population, time frame,
inclusion/exclusion, sample size, study flow diagram ,
outcome/predictors/confounders/effect modifiers, plan
of analysis, efforts to reduce threats to validity,
strengths/limitations
• Discuss some complicated issues in study
design
Study Designs
Descriptive
Analytic
correlational
case control
case report/
case series
cohort
cross-sectional
Experimental
clinical trial
community trial
Criteria for
Causality
•
•
•
•
•
Biological Credibility
Consistency of findings
Dose-response
Magnitude of the association
Time sequence
Cross sectional
D
__
D
E
__
E
Case Control
E
__
E
E
__
E
D
__
D
Cohort
D
E
__
E
__
D
D
__
D
Phases of a Clinical Trial
• Phase I - safety (pharmacokenetics - to
determine maximum tolerated dose)
• Phase II - Evidence of a response
• Phase III - Safety, efficacy
• Phase IV - Safety, Acceptability, Efficacy
Study Diagram - Classic
Randomized Controlled
O
Eligible Subject
Pool
Int
LTF/C
P/SC
O
R
LTF/C
Study Design - Cross-over
Study
Eligibles
E
E
C
C
R
Hypothesis Testing
• Hypothesis testing involves conducting
a test of statistical significance and
quantifying the degree to which
sampling variability may account for the
results observed in a particular study
• When designing data collection tools,
keep in mind your final analysis
Statistical Tests: 2 T-test
Measures of Association:
Odds Ratio, Relative Risk
Objectives should be stated in
terms of an hypothesis
• Null Hypothesis: There is no difference
Medication A will have not effect on disease
progression
• Two tailed Hypothesis: There is some difference
Medication A will have some effect on disease
progression
• One tailed Hypothesis: The difference is greater or
less
Medication A will reduce deaths due to disease X
Medication A will increase deaths due to disease X
Outcome of interest
• Write the research question in advance
• outcome variable:
– should be measurable in all subjects
– should be capable of unbiases
assessments
– should be ascertained as completely as
possible
Response or Outcome
variables
• You may have outcomes other than
hard endpoints
• surrogate markers
• quality of life
Follow-up Studies - Survival Analysis
• This analysis used when subjects are
entered over a period of time and have
various lengths of follow-up.
• Dichotomous endpoints
• Kaplan Meier or Product Limit
• Cox Proportional Hazard modeling
Intent-to-treat Analysis
• For persons who cross-over to the other
arm. You classify that person into the
arm they were originally assigned.
• Less biased results than “as treated”
because you maintain randomization.
• Only works if there is not a lot of
crossing over very early in the study
Reasons for withdrawal of
Subjects
• Ineligibility (misclassification,
imprisonment, moved)
• Noncompliance (adverse effects of
intervention, loss of interest, changes in
underlying conditions, substance usage)
Measurement
•
•
•
•
Outcome
Predictor
Confounder
Effect modifier
Validity and Reliability
Validity
x
Reliability
xx
Relative Risk
for a disease exposure
Drug use
No drug
use
STD
75
25
No STD
25
100
75
100
100
100
RR = 75/100 = 3.00
25/100
C.I. (2.10 - 4.29)
200
Odds Ratio Calculation
Drug use
No Drug use
STD
100
100
200
No STD
50
150
200
O.R. = (100*150)
(100*50)
Total
150
250
400
= 3.00
Confounding and/or interaction
(Kleinbaum, Kupper and Morgenstern)
HIV risk perception and self-protective
behaviors among high risk persons in
community settings
Patricia Kissinger, Ph.D.(1)
Nomi Fuchs, MPH (2)
Catherine Schieffelin, MPH (2)
Jane Herwehe, MPH (2)
DeAnn Gruber, MSW (2)
(1) Louisiana State University, HIV Outpatient Program
(2) Children's Hospital - Family Advocacy, Care and
Education Services (FACES)
Purpose
• The purpose of this study was to
examine HIV risk perception and selfprotective behaviors among high risk
people in community settings.
Methods
• Street intercept and in-depth interviews
were conducted from August 1997 to
June 1998
• Inclusion:
– Sexually active people
– aged 15-35
– living in six communities of New Orleans
with the highest gonorrhea rates.
Results
• Of 1133 respondents, 97% were African
American, 37.4% were 15-18 years of age.
• 46.2% reported an HIV risk behavior, 66.5%
reported condom use, and 69.9% reported ever
having been tested for HIV.
• Many respondents (39%) perceived themselves to
be at no risk, but reported engaging in an HIV risk
behavior
• Adolescents and persons who had been HIV
tested were most likely to have this discrepancy.
Results con’t
• Among the 524 persons who reported an HIV
risk behavior, 19-35 year olds were less likely
to use condoms and adolescent men were
less likely to have been HIV tested.
• In-depth interviews revealed diverse reasons
for failure to perceive oneself at risk and
failure to be HIV tested including optimistic
bias, risk group identity, hierarchy of risk and
fear.
Table 2.
Factors associated with a discrepant
responsea (N=1072)
% discrepant
Adjusted
O.R. (95% C.I.)
Age
15-18
19-35
44.9
36.2
1.58 (1.20-2.09)**
1.00
Gender
Women
Men
39.2
38.7
1.02 ( .79-1.33)
1.00
42.5
36.1
1.16 ( .88-1.52)
1.00
39.3
38.3
1.37 (1.01-1.85)*
1.00
Used a condom
last sexual act
Yes
No
Been HIV tested
Yes
No
Table 3. Factors associated with self protective
behaviors among persons reporting an HIV
risk behavior (N=524)
Condom use
Adjusted O.R.
(95% C.I.)
Age
15-18
19-35
Gender
Women
Men
Self-assessed HIV risk
Yes
No
Assessed partner's risk
Yes
No
**p < .01
Ever been HIV
tested
Adjusted O.R.
(95% C.I.)
1.00
.16 ( .10- .25)**
.48 ( .32 - .72)** 1.00
.68 ( .46- 1.01)
1.00
1.00
.34 ( .22 - .52)**
1.26 ( .65-2.30)
1.00
2.01 ( .94-4.29)
1.00
.58 ( .31 - 1.10)
1.00
.58 (.27-1.24)
1.00
Table 4 Association between
reported risk behavior and
self-assessed risk
Self-perceived
at risk
Perceived
partner(s)' at
risk
Among those Agreement
reporting high between self
risk behavior reported risk and
assessed risk
K (95% C.I.)
15.6%
.095 (.057-.135)
19.1%
.130 (.099-.171)
Kappa .10 (95% C.I. 06-.14) indicating poor reliability
Non-experimental (analytic)
study designs
• Conducted because of ethics, cost or
convenience
• Two primary types:
– Cohort
– Case-control
Experimental Designs
• Experiment – a set of observations, conducted under
controlled circumstances, in which the scientist
manipulates the conditions to ascertain what effect
such manipulation has on the observations.
• Ideally only one factor is examined (however,
biological variation exists)
– Clinical Trials – (individual in a special environment are
randomized)
– Field Trials – (individuals in the community are
randomized)
– Community Interventions – (whole communities are
randomized)
Field Trials
• Differ from clinical trials in that subjects
have not yet gotten disease
– (1955) Salk vaccine for Polio
– (1975) Vitamin C in preventing the
common cold)
– (1982) MRFIT – a field trial of several
primary preventives of MI (N=12,866 and
cost $115 million)
Community Intervention and
Cluster Randomized Trials
• Community intervention is an extension of a
field trial that involves intervention on a
community-wide basis
– (eg. Mass media campaigns)
– (eg. Fluoridated water)
• Cluster randomization - groups of
participants are randomized. The larger the
cluster, the less that is accomplished by
randomizing.
Study Protocol
•
•
•
•
•
•
•
•
•
Rationale and background
Objectives
Study Design
Inclusion/Exclusion
Definitions (intervention, measurements,
adherence)
Study Flow chart
Sample Size calculation
Plan of analysis (interim analysis)
Appendices
– Questionnaires
– Consent forms
– Instructions to interviewers
Example of a flow chart for
randomization
Example of a comparison table to demonstrate
that randomization was successful
Incidence vs. Prevalence
• In infectious diseases of short duration,
incidence may be close to prevalence
• In chronic diseases, prevalence will be
far greater than incidence
• Monitor disease burden by prevalence
• Monitor efficacy of programs by
incidence
Calculate an Incident Rate
Jan July Jan July Jan July Jan July Jan July Jan time at
1976 1976 1977 1977 1978 1978 1979 1979 1980 1980 1981 risk
Sub A *---------------------2.0
Sub B
*---------------------------------x
3.0
Sub C *--------------------------------------------------------- 5.0
Sub D
*--------------------------------------4.0
Sub E
*---------------------------x
2.5
Total Years at risk
16.5
* = initiation of study
ID=___cases/___person-years
-- =Time followed
x = development of disease
Measures of Associaton
• Since clinical trials are prospective and
the intervention precedes the outcome,
a relative risk is calculated.
• Covariates and confounders can be
either controlled for in the design or
adjusted for in the analysis
Is PID more common among
HIV-infected women
•
•
•
•
•
•
Research Question
Population
Inclusion/exclusion
Study Design
Type of analysis and Unit of analysis
What are the predictors, confounders, and
outcomes of interest
• Findings
• Limitations/Strengths
Difficulties with this study
• Definition of a case
• Choice a proper control
• Detection bias
A microbicide to prevent HIV
among women
•
•
•
•
•
•
Research Question
Population
Inclusion/exclusion
Study Design
Type of analysis and Unit of analysis
What are the predictors, confounders, and
outcomes of interest
• Findings
• Limitations/Strengths
Difficulties with this study
•
•
•
•
Ethical dilemma
Exposure is altered by study itself
Choice of cases and controls
Sample size considerations
An HPV vaccine to prevent
HPV among women
•
•
•
•
•
•
Research Question
Population
Inclusion/exclusion
Study Design
Type of analysis and Unit of analysis
What are the predictors, confounders, and
outcomes of interest
• Findings
• Limitations/Strengths
Difficulties with this study
•
•
•
•
Misclassification bias possible
Population to study difficult to find
Sample size
Generalizability
Study Design
• Statement of hypothesis
• Population
– Sampling
– Inclusion/Exclusion
• Time frame
• Design
• Measurement
– Predictors
– Confounders
– outcome
• Analysis plan
– Sample size
– Dummy Tables
– Analyses to be done
• Efforts to minimize threats to validity
• Strengths and limitations
Study Designs
Descriptive
Analytic
correlational
case control
case report/
case series
cohort
cross-sectional
Experimental
clinical trial
community trial
Confounding
E
D
C
E=exposure
C=confounder
D=disease
Strategies for Partner
Treatment for STD control
By
Patty Kissinger, Ph.D.
Objectives
•
•
•
•
Background
Prior Studies
Present Studies
Policy implications
Why treat partners?
• Primary prevention - to break the chain of
transmission
– Healthy men don’t access health care
– Many STDs are asymptomatic
• Secondary prevention - to prevent complications
of the disease
– STD infections increase the risk of HIV
– Recurrence can cause serious health consequences
Recurrent chlamydia
• Causes PID, ectopic pregnancy,
infertility and chronic pelvic pain
• Many women are re-infected by an
untreated partner
• Strategies for partner treatment are
necessary
Basic Reproductive Rate of
Infection
(Anderson and May)
Ro=  D c
Ro is the basic reproductive rate of
infection

is the transmission coefficient
D
is the infectious period
c
is the mean rate of sexual partner
change
Sexual Networks
X
X
X
X
X
X
X
Methods of Partner Treatment
• Partner referral
• Partner notification
• Patient delivered partner treatment
Problems with Partner
Referral
• Studies of chlamydia demonstrate that
only 25-40% of named male partners
were treated.
• Partners
– not told
– refuse to come for testing/Rx
Problems with Partner
Notification
• Confidentiality
• Expensive and time consuming
– Almost 800,000 cases of chlamydia and
400,000 cases of gonorrhea were reported
in the US in 2001
• Not all partners are named
• Hard to find partners
Problems with Partner
Treatment
• Safety
– Allergies
– Pregnant women
• Liability
– Physician
– Nurses
– Institutions
• Fear of uncontrolled antibiotic use
– Fear of selling medication
– Fear of stocking up on medicine
Empirical data in favor of PDPM
• Retrospective cohort in New Orleans (Kissinger et al.,
Sex Trans Inf 1998; 74:331-333)
• Correlational in Sweden (Ramsted et al 1991; 2:116118)
• Cross-sectional in San Francisco (Hammer et al.
National STD Conf 2000; Wisconsin)
• Cross-sectional in Washington (Golden et al STD
1999; 26:543-547).
• Randomized trial in Uganda (Nuwaha et al. STD
2001; 105-110
Multi-centered Trial – Infertility
Prevention Program
•
•
•
•
1787 women aged 14-34
Eight cities
Randomized to PDPM versus PR
Tested at 1 and 4 months using LCR or
PCR
• Given 1 gm azithromycin
• Outcome was recurrence
Second Study - Prospective Study
Strategy
Partner
referral
Patient
delivered
(n/N)
%
(108/726) 15
(87/728)
12
RR
95%CI
p value
1
--
--
0.8
(0.62-1.05)
0.102
Issues with the study
•
•
•
•
Loss-to-follow-up
Low power
Persistence versus recurrence
Powder form of medication
PDPM seems reasonable
• At the time of treatment for their own
chlamydial infection, a majority of
women have a partner who remains
untreated (Golden, 2001)
• Most patients with STDS prefer to notify
the partner themselves (Golden, 2001)
• Men generally perceive practical
obstacles to obtaining treatment
(Fortenberry, 1997)
Present Studies PA0008 –
Female trichomonas trial and male
urethritis study
• Testing three methods: partner referral,
booklet referral, PDM
• Male urethritis – quasi-experimental
– Delgado
• Female trichomonas – randomized trial
– 01 Family Planning
• Baseline and follow-up visit
– ACASI interviews
– STD testing
Booklet referral
Patient Delivered Partner
Medicine
• For Trichomonas (1 gram of
metronidazole)
• For Male urethritis (1 sachet of
azithromycin 1gram sachet and 1 dose
of cefixime 400 mg orally)
• Directly observed medication for index
Outcomes measures
• How many partners are treated (index
patient-report)
• How many partners show up to clinic
saying that they have been referred by
an index partner
• Recurrence rates
– InPouch
– BD urines
Trichomonas
Referral
Booklet
PDM
Patients
64
61
61
Partners
68
68
73
Ratio
1.06
1.11
1.20
Follow-up
rate
Desired
66.2
85.3
75.3
113
113
113
% of
desired
enrolled
56.6
54.0
54.0
Male urethritis
Referral
Booklet
PDM
Partners
282
237
207
Index
141
121
111
Ratio
2.0
1.96
1.86
Follow-up
rate*
Desired
66.9
81.0
64.0
182
182
182
% of
desired
enrolled
77.5
66.5
61.0
Interim Analysis
Partner took the ARM 1
medicine
Male Urethritis 36.8
ARM 2
ARM 3
45.1*
77.0**
Trichomonas
60.3
90.1*
73.9
*P<0.05, **P<0.01
Policy implementation issues
•
•
•
•
More evidence?
Practice protection
Need to educate
Financial support