Psychotherapy Research Methods Flashcards
there is no _________________ in psychotherapy
one size fits all
4 components of psychotherapy research
what works
for whom
in what context
and why
stages of intervention development
basic science
creation and preliminary testing
pure efficacy
real world efficacy
effectiveness
implementation and dissemination
stage 0 of intervention development
basic science
- things to go into intervention/why the intervention will work
- identify problems that need treatment (brain, family history, etc.)
- understand treatment gap/things that are/aren’t working
stage 1 of intervention development
creation and preliminary testing
- stage 1a:
- developing intervention components with clear explanation for replication
- done with patient input (focus groups to make changes/feedback)
- by the end should have materials for trials
- stage 1b
- feasibility and testing
- can we do this, get people to sign up?
- often done in lab, goal is to see if investing is worth it
stage 2 of intervention development
pure efficacy
- controlled/internal conditions
- efficacy = how well it works under perfect controlled conditions
stage 3 of intervention development
real world efficacy
- trial in more external environment/real-world setting
- ex. train then closely monitor therapists
stage 4 of intervention development
effectiveness
- understand what happens once you take some controls off
- less strict
- ex. train then don’t monitor therapists
stage 5 of intervention development
implementation and dissemination
- how to get interventions to be used clinically?
- delay between research and implication b/c…
- people do not like change
- environment isn’t necessary for implication (where feasibility plays in)
prospective treatment assignment
we have more than 1 group and assigned before condition takes place
if there is prospective treatment assignment, we are conducting an _________ study. If there is NOT prospective treatment assignment, we are conducting an _________ study
experimental; observational
key difference in experimental and observational studies
ex: observe cause and effect
obs: no manipulation
observational study
precursor for intervention/seeing if study needs to take place
more external validity
used when experimental cannot be (studies to do with smoking, children, etc.)
cheaper and shorter time frame
If there is a comparison group, we are conducting a(n) _______ study. If there is NOT a comparison group, we are conducting a(n) ________ study.
analytical; descriptive
types of observational studies
descriptive
analytical
descriptive study
establishing phenomenon
first study type in new research area
tells frequency, progression over time, correlations, etc.
cannot conclude cause and effect
how results are biased in observational studies?
confirmation bias
- looking for one thing and only measure/record this
selection bias
how do we determine the type of analytical study?
when were treatment and outcome identified/assessed?
cohort study
treatment then outcome
studying effectiveness
- follow the groups forward in time to determine if they experience outcomes
no assignment takes place/no manipulation
steps: identify groups then measure outcome
cohort study strengths
treatment comes before outcome (establishes temporal precedence)
less prone to recall bias (longer ago an event took place, the least likely you are to remember details)
provides estimates of incidence of outcomes overtime
cohort study limitations
cost
rare outcomes are hard to observe
studies may need to be very long to observe outcomes
case control study
outcome then treatment assessed
assess whether there were differences in treatment exposure retrospectively
- chart reviews
- self-report
- interviews
case control study strengths
useful for rare outcomes
can save time and money
case control study limitations
difficult to select an appropriate control group
recall bias
cannot tell you how prevalent the outcome or treatment is - only odds of experiencing both
cross-sectional study
outcome and treatment assessed at the same time
use on timepoint to assess both outcome and treatment exposure
can provide estimates of frequency or prevalence of an outcome or treatment
no temporal precedence and subject to recall bias
rate
frequency of an event in the population over a defined period of time
proportion
frequency of an event without a defined time period
ratio
number of people in one condition relative to the number in another
absolute risk
probability of outcome
odds ratio
likelihood of membership in one group, given membership in another (also relative)
ex. given membership in the treatment group (vs. nontreatment group, how likely are you to also be in the outcome group
relative risk
ratio representing how often the outcome happens in the treatment group, relative to the untreated group
types of experimental studies
randomized-control trial
non-randomized-control trial
key difference in experimental studies and why does it matter
if the treatment conditions are randomly assigned
matters because of:
- selection bias and open-label trial
randomized-control trial
control for what works for who
can claim cause and effect
non-randomized control trial
used in early stages of a trial
more exposed to bias
cannot claim cause and effect
inclusion criteria for randomized-control trials (RCTs)
diagnosis
clinical staging
prior treatment and iatrogenic comorbidity
clinical staging for RCTs and specific stages
ways to grade out severity of disorder symptoms/symptoms experienced
symptoms on continuum
stages:
0: no symptoms
1: some symptoms that aren’t specific
2: moderate-severe symptoms, meet criteria for diagnosis
3: some level of remission
4: persistent/severe, not going away or getting better
iatrogenic comorbidity
had other treatments that didn’t work or potentially make things worse
past treatment experiences impact future reactions and so you don’t repeat any treatments
recruitment for RCTs
how are participants identified?
is the sample representative?
can become biased (socio-economically and culturally)
control groups for RCTs
no treatment or waitlist
minimal attention
treatment as usual
- standard treatment that is currently being used in field
attention placebo control
- rigorous, doing as much as we can to mimic the intervention
other active treatment
additional controls needed when pharmacotherapy is added
designs for RCTs
parallel treatment
- most common! groups assigned then followed simultaneously
adaptive
- changing course of movement throughout treatment based on how they respond, aka “stepped care”
dismantling
- tries to figure out active ingredients, pull apart to figure out further which produces the larger effect/most essential part of treatment
assessments for RCTs
treatment allocation should be concealed from those administering assessment
pre and post treatment, consider length follow up
must be sensitive to change
incremental validity
patient-reported vs. observer-rated
assess for adverse effects, not just desired effects
demand effects
the more you include, the easier a patient may guess what you are studying and sway/bias their answers
incremental validity
what we are measuring while adding something new
patient-reported vs observer-rated
subjective or objective: who’s perspective is represented in outcome?
value both things: feelings and observations
assessing for adverse effects, not just desired effects
focusing on people who didn’t respond to treatment that may be affecting averages
outcomes for RCTs
define a priori
what level of improvement will mean the treatment worked?
- remission (no longer meet criteria for diagnosis)
- reduction of symptoms (still meet criteria)
- longevity of change
priori
established before start to gauge what a success is considered to be in an experiment
drift
expect therapists to naturally drift from adhering strictly to treatment overtime because:
- find things that work with patients
- become less rigorous or just forget steps/get distracted
the more ________ = the more confident one can be in the results of the study
rigorous
efficacy vs. effectiveness
efficacy:
- emphasis on internal validity
- very controlled
effectiveness:
- less control
- emphasis on external validity
- comorbidities that influence treatment
empirically-supported treatments
stringent criteria for study designs
- active control, large sample, etc.
subject to limitations of RCTs
- efficacy vs. effectiveness
- manualized interventions
may be difficult to adapt or not applicable to a wide range of patients
evidence-based practice has 3 components
best research evidence
clinical expertise
patient values and preferences
