Week 4: Multidisciplinary Mental Health Care Research

Question 1

What is the focus in psychological and psychiatric care at the moment?

Answer 1

Unidimensional

Question 2

What are the consequences of having a unidimensional approach in psychological research?

Answer 2

Having a unidimensional approach in health care as well

Question 3

How do we turn to a more multidisciplinary approach in health care?

Answer 3

By also turning to a more multidimensional approach in research.

Question 4

What is ontology?

Answer 4

The theory of being; the view that we hold about how the world is organized; a pair of glasses that you put on and how you see the world through them
What is the truth? Is there 1 absolute truth or many truths - that also depends on how our world is organized

Question 5

What is epistemology?

Answer 5

The theory of knowledge - they way we as humans can understand this world; the way that we can obtain knowledge about this world

Question 6

Which model do we see in the medical model in psychiatry: epistemological or ontological?

Answer 6

We see both models

Question 7

What does ontology mean in the medical model? (ontological premise)

Answer 7

There is one universal truth in a predominantly physical world
Based on the ontological premise we assume that mental health issues are understood from a disease model
The symptomatology is caused by a certain disease - something wrong in our physiology

Question 8

The epistemic premise in the medical model?

Answer 8

We can explain mental health issues in a very systematic way;
Research: observe symptoms systematically; develop treatment protocols - give people an intervention; assuming people will respond to the same type of intervention in a similar way;
The assumption that the intervention can be understood causally: if we give a specific treatment, the person will respond in a certain way and the outcome of this will be a reduction in symptoms
The reduction in symptoms is a sign of a reduction or even a cure of the disease

Question 9

What influences the way we treat patients?

Answer 9

The way we do research

Question 10

Protocol treatment

Answer 10

A well-defined treatment, for which all the steps are well-defined and put in specific order; doesn’t depend on the person who distributes the treatment; it works regardless of the therapist (debated point)

Question 11

What is protocol treatment?

Answer 11

A well-defined treatment, for which all the steps are well-defined and put in specific order; doesn’t depend on the person who distributes the treatment; it works regardless of the therapist (debated point)

Question 12

What do we do in evidence-based treatment research?

Answer 12

We study protocol treatments in a way we can actually compare their efficacy; the idea that the protocol treatment as a whole is working regardless of the person distributing it

Question 13

Dose-response?

Answer 13

You give a certain dose to a certain response: if you give a higher dose, the response will be higher, if you give a lower dose; the response will be less
Directly related to the medical model: if we give you more medicine, the response will be higher etc.
In the same way, assuming that if we give a certain number of sessions, we will get a certain response: important to differentiate between mild and severe problems: need a different number of sessions

Question 14

Model of stepped care

Answer 14

Giving people a dose of treatment
Evaluating the outcome
If everything is okay, perfect
If it is not okay, we increase the dose
Evaluate again

Question 15

What are the 3 levels in the stepped care model?

Answer 15

First line: first portal of psychological care; for mild and straightforward problems; basic first line evidence-based treatment; if the treatment doesn’t work, they will go to the second line of care
Second line: more specialized treatments
Third line: hospitalization (institutionalized care)

Question 16

What is the symptom-reduction model based on?

Answer 16

Characteristics of epistemology and ontology;
Assumptions: we have a disease that causes symptoms; if we want to impact the disease, we have to reduce the symptoms; as an effect the treatments are focused on reducing the symptoms; The core of evidence-based treatment, because treatments are distributed if we have a scientific support that they work (are effective in reducing symptoms for most people)

Question 17

What does efficacy mean in scientific research?

Answer 17

Proof of the fact that the treatment works; reduces symptoms for the majority of people who go through the treatment

Question 18

What is a core principle in today’s organization of mental health care?

Answer 18

Evidence-based principle

Question 19

Evidence-based principle?

Answer 19

• core principle in the organization of health care
• guides the way we work in health care; we want to distribute treatments to people that are actually scientifically evaluated

Question 20

Evidence in psychotherapy research?

Answer 20

• we have to follow a specific methodological procedure, that gives the best evidence there is: we trust the methodology and that the results we get are the best proof of the efficacy of the treatment
• hierarchy of the methodological procedures: we trust some methods more than others
• randomized controlled trials: predominant in research

Question 21

The pyramid of the methodological procedures”

Answer 21

1. Randomized controlled trials (on top)
2. Meta-analysis: putting RCTs together
3. Cohort study
4. Case-control study
5. Case reports
6. Opinion/Research Agenda
   Everything below the RCT is not considered as serious as the RCTs

Question 22

What is the Gold standard of psychological research?

Answer 22

RCT

Question 23

What is RCT?

Answer 23

• systematic comparison of group level symptom development pre-post treatment
• an experimental group that gets the treatment
• a control group that doesn’t get the treatment (ideally gets nothing) - ethical concerns of waiting list or getting just placebo –> nowadays we use a treatment as usual group instead (parallel trials; 2 different treatments, but still considered as RCT)
• looking into change over time in the two groups: symptom level before and after the treatment: goal - for the symptoms to decrease

Question 24

Methodological requirements of RCT:

Answer 24

• comparable populations: homogeneous samples
• we randomize: try to avoid systematic differences between the conditions - make the samples as comparable as possible (homogeneous)
• well-defined outcome variable (dependent variable): no confounds; eligibility criteria for participation (for example the absence of comorbidity; only having the target symptomatology)
• systematic and well-defined intervention (independent variable) –> to make it systematic and comparable we use protocol treatment
• randomization: keep all possibly interfering factors as even over samples as possible (random allocation to interventions)
• keep expectancy effects limited: placebo control; in reality we use comparative trials with treatment as usual

Question 25

Westen et al. article (2006): core argument

Answer 25

The methodological requirements that we make in our design shape what we can find, the knowledge we can obtain
Methodology shapes findings

Question 26

What do Westn et al. (2004) argue for regarding the assumption of malleability and evidence-based research

Answer 26

assumption of malleability: people have the ability to change (we can influence their symptoms with our treatment) - very important to keep the therapy short in order to proof that the treatment is the one influencing the symptoms (if it goes for a long time, people move, change jobs etc, a lot of interfering factors); the longer the therapy, the more variability, the less one can draw causal conclusions –> preference for brief treatments (a natural consequence of efforts to standardize treatments to bring them under experimental control)

Question 27

Why do researchers prefer brief treatments?

Answer 27

To keep the confounding variables as little as possible; in longer treatments, life influences the cause; therefore researchers have less control to draw causal conclusions

Question 28

What is the problem with mainly using brief interventions instead of long ones?

Answer 28

We don’t have specific proof that short interventions work better than long ones.
But it is harder to research the efficacy of longer treatment because it does not fit the methodological requirements

Question 29

Why is the most evidence we obtain is focused on short-term treatments?

Answer 29

• assumption of malleability
• requirement for causality

Question 30

What is the consequence of using brief interventions?

Answer 30

They shape the results that we get

Question 31

Paradox of pure samples:

Answer 31

• pure samples: groups in which there is no comorbidity
• interventions to address such comorbid conditions will, as a simple result of methodological preconditions, never be identified if investigators routinely start with less complex cases and focus studies in the community on interventions previously validated in RCT

Question 32

Why don’t we want comorbidity in research?

Answer 32

Comorbidity is a confounding factor, therefore we cannot draw conclusions as clearly based on the findings

Question 33

The paradox of Manualization (Westen et al. 2004) (Truijens et al., 2019)

Answer 33

We use protocol treatments because this way we can clearly define the independent variable
If your treatment is not well-defined before you start, you cannot draw a causal conclusion
Manualization is a very important pre-condition
The evidence we obtain is for manualized
treatments
We get no research for treatments that are more eclectic
The consequence is that we think that a treatment is better if it is manualized; in reality, we cannot make this conclusion because we have not compared manualized to non-manualized treatments
Truijens et al. study: found no proof that manualized treatments are better

Question 34

Three paradoxes in Westen et al. article (2004)

Answer 34

• manualization
• pure samples
  -maleability

They are not necessarily requirements of practice, but are requirements of research

Question 35

Why it is important to think about the effects of research methods on clinical practice?

Answer 35

Because in the end these methods/assumptions become a dominant way of approaching people in clinical practice

Question 36

What is the basic philosophy of psychology science?

Answer 36

Doing experimental research, drawing causal conclusions, systematically observing symptoms and treatment effects
Based on the model of the natural sciences, but in reality social sciences are different from the natural ones

Question 37

The discourse of Erklaren (Explanation)

Answer 37

• basic philosophy of science entails:
• ontology: there is a real, universal, regular outside (lawlike) world
• epistemology: that we can explain by deducing universal patterns of phenomena
• The goal is to explain phenomena causally
  We want to find an explanation (how do phenomena work) by understanding a causal relationship
• we use an experimental method: compare, trace processes and from that we logically deduce logical causal conclusion
  Method: experimental, deductive, top-down (we hypothesize, we do the research and at the end we can go back to deduce what we hypothesized), measurement (systematic and rational way of comparing) and calculation

Question 38

Measurement in psychological research

Answer 38

Gold standard: we measure symptoms before and after treatment and we conclude if the treatment was effective
Vital for conducting a scientific research
Systematic assessment: use of standardized symptom-focused measures
Symptom development: We use symptoms measurements that are scored by patients themselves: self-reported, self-scored (before and after treatment)
Summing the scores of our individuals (at the beginning and at the end)
“The data” will tell the story of treatment efficacy

Question 39

Self-report questionnaires

Answer 39

• we ask people to rate a certain statement (how much they agree or disagree)
• we quantify content: certain experiences are put on a certain scale (giving them numbers from 0 to 5)
• data is scores, we sum them up and analyze them

Question 40

Efficacy number as an outcome of aggregation and analyses

Answer 40

Efficacy =

1. sum all the scores on questionnaires per individual
2. aggregation of pre- and post- scores over sample
3. aggregate difference scores per condition
4. between-group difference (we compare: significance test)

Note: based on symptom reduction model - critiqued, because we are looking at means: 1 and 10 or 8 and 9 make the same mean; we could have zero effect + a very big effect making an average mean: does not represent reality

Question 41

Critiques of the symptom reduction model?

Answer 41

we are looking at means: 1 and 10 or 8 and 9 make the same mean; we could have zero effect + a very big effect making an average mean: does not represent reality

Symptom reduction: good for getting an overview of the problem, but clinically is not very helpful; we limit the measurement to pre- and post- treatment; snapshot research: one moment of time compared to another moment in time; can make quite a difference given the exact moment you measure this

Question 42

Alternative for snapshot research?

Answer 42

1. Beyond singular/linear outcome variables: looking at more dimensions:
   - domain based: e.g. Research Domain Criteria
   - Complex Dynamic Systems: e.g. network theory in diagnostics
2. Using more types of design: beyond the gold standard research design:
3. Beyond measurement:
   - mixed methodology approach
   - qualitative stance

Question 43

What type of research is snapshot research?

Answer 43

Unidimensional research

Question 44

Why did they develop the Research Domain Criteria

Answer 44

• probelm: the one single outcome idea doesn’t work anymore, it’s linear, problem with comorbidity, doesn’t tell us enough to help us understand mental health
• alternative: dimensional, multi-level, developmental

Question 45

Research Domain Criteria

Answer 45

dimensional, multi-level, developmental model
Several domains:
- negative valence
- positive valence
- cognitive systems
- systems for social processes
- arousal/regulatory systems
- sensorimotor systems
The basic assumption is that all these domains work together whenever we talk about mental health

Note: this model is also critiqued, because in practice the emphasis is on the bottom domains (biological and neurological systems); and the domain on the top (the more psychological ones) are considered to be catalysts, triggers or mediators

Question 46

Multidisciplinary research

Answer 46

• enforced by RDoC
• It is interdisciplinary research: not just psychiatric framework, but a collaboration between social work, sociologists, anthropologists, neurologists, policy makers, medical staff, psychologists etc.
• different kinds of methods are invited, which is not set a priori (different from the other methods)

Question 47

What is the advantage of RDoC?

Answer 47

Multidimensional content (ontology: the world is not built by only 1 unidimensional factors that we want to understand, but is build of different phenomena working together to form a bigger phenomenon) asks for a pluralist (allowing different perspectives to come together) research approach & mixed methodology (epistemology)

Question 48

Networks (Borsboom & Cramer, 2013)

Answer 48

• not assuming that depression and anxiety are two separate disorders
• their domains are highly related, they interact
• one symptom of one disorder may interact with one from the other: or may influence other disorders as well

Question 49

Why are efficacy designs critiqued?

Answer 49

• inclusion criteria
• pre- and post- snapshots
• limitation to 1 outcome variable

Alternative: rise of efficiency research

Question 50

What is efficency treatment?

Answer 50

We have an intervention; we look at the symptoms pre and post, but we use less strict inclusion criteria, we focus more on the process and we use multiple variables

Downside: less causal, cannot deduce such a strong causal relationship
Advantage: closer to what is happening in clinical practice
–> builds to process research

Question 51

Process research

Answer 51

• if we measure pre- and post- we assume the process is linear; the result is a mean between the two
• the process can be non-linear for a person (not that important for statistics, but is very clinically relevant; it is important to know what is happening to the person and be able to address it)
• important to also measure during the treatment: a couple of points

Question 52

Reliable clinical change (Jacobson & Truax)

Answer 52

• we don’t look only at the statistically relevant results
• formulated a new way of calculation
  Reliable Change Index: differentiate between the different results of people
• recovery
• improvement
• no change
• deterioration

Gives rise to individual differences research

Question 53

Individual differences research

Answer 53

• we should also look at what the patterns mean on an individual level, not just at the majority
• individual differences allow us to learn which people are prone to deterioration and how to address this

Question 54

What does personalized measurement do?

Answer 54

• allows people to formulate their own outcome variables: things that they want to change during treatment - when you measure them, you have an outcome variable that is very individually based, it’s not that generalizable to the rest of the population, but is very clinically relevant

Question 55

What is the goal of measurements?

Answer 55

To understand people, experience of their symptoms, their meaning and the difference that they make for them

Question 56

What does it mean to go beyond measurement?

Answer 56

• approaching the numbers that people score in questionnaires as words (Truijens)
• numbers don’t allow you to tell your whole story

Question 57

What is called “noise” in research?

Answer 57

When the participant doesn’t adhere to the given instructions; the data is not reliable anymore; cannot be compared to the other participants anymore

Question 58

Qualitative research: voices in data (Truijens)

Answer 58

We ask the participants to answer in pre-structures, quantified and aggregated structures, questionnaires
downside: we limit the way people can express their experience; we assume that the participants understand the questionnaire the way we intended

Who is the “average patient”?
- the statistical mean is not a person
- we have an a priori idea of how the research will work out: can we learn something not-anticipated?
- noise can be part of the clinical process
- numbers are words too - part of the clinical story

Question 59

The discourse of Verstehen (Understanding)

Answer 59

We adhere to an ontology where people make meaning of phenomena in their lived experience (making meaning the whole day just by communicating and living) - you always make meaning by your own personal experience (using pre-understanding on the subject that you have)
Epistemology: we want to understand how people make this meaning: by listening, analyzing and synthesizing what people tell us: in clinical practice we understand how the patient makes meaning by a listening method of obtaining narratives
Working from the narrative itself and finding patterns of how people make meaning; working from the bottom up; we don’t have a pre-defined hypothesis
Goal: understand how people understand phenomena
Mixed method: can use numbers as well, but look at them as a part of a story

Question 60

Multidimensional ontology

Answer 60

• as many people we have, as many truths we have: everyone views the world from their own experience
• but there is a common (collective) understanding: based on common culture etc.
• Bachelard: is there is no question, there would be no knowledge; you are asking the question, you are choosing the method, you are doing the research from your own perspective

Question 61

Paradigm of Social Constructivism

Answer 61

• meaning is socially constructed and is mediated by language
• observation is never neutral: researchers are always part of generated knowledge (e.g. by choice of methods, which shape the knowledge obtained)

Question 62

What is phenomenology?

Answer 62

• how people make meaning; how phenomena appear to people and how they makes sense of them
• in order to understand we have to listen

Question 63

What is hermeneutics?

Answer 63

The theory of language, the theory of meaning, how meaning is composed

Question 64

How do we understand something about phenomena?

Answer 64

By understanding how we and others make meaning of said phenomena

Question 65

What is meaning?

Answer 65

a constellation of interpretations, shaped by general, cultural, and societal stories, but ultimately regard the way in which the person understands these stories and is affected by them
- the starting point for understanding something is personal narrative

Question 66

Compare the theory of listening to the principles of Recovery paradigm (Slade)

Answer 66

• giving primacy to values and preferences of individuals
• complexity over simplification: not just looking for meaning to understand in a very simple and broad way, but we want to see how the specific person understands
• we assume that meaning is social, relational, personal and dynamic

The understanding paradigm, which gives rise to the theory of listening boils down to the recovery paradigm principles

Conclusion: the recovery paradigm vs the medical approach boil down to different philosophies of science; based on different principles, which applies that if you want to adhere to the recovery paradigm you have to think about how you think about the world and how you want to do science

Question 67

What is the moral of the most common “story” in psychological science according to Westen et al. article (2004)

Answer 67

Only science can distinguish good interventions from bad ones

Question 68

What is different in the “story” Westen et al. (2004) are putting forward?

Answer 68

It adds a second complementary moral: Unqualified statements and dichotomous judgments about validity or invalidity in complex arenas are unlikely to be scientifically or clinically useful and as a field we should attend more closely to the conditions under which certain empirical methods are useful in testing certain interventions for certain disorders

Question 69

What do Westen et al. argue for?

Answer 69

A more nuanced story about efficacy and treatment of choice than sometimes seen in the scientific literature

Note: not claiming their own narrative is bias-free

Question 70

What do Westen et al. do not argue against?

Answer 70

Not advocating against evidence-based practice (but asking for a more nuanced story)

Question 71

What are Westen et al. suggesting is time for in their article?

Answer 71

Time has come for a thoroughgoing assessment of the empirical status of not only the data but also the methods used to
assign the appellations empirically supported or unsupported.

Question 72

How do Westen et al. argue for their point of view?

Answer 72

• they examine the empirical basis of the assumptions that underlie the methods used to establish empirical support for psychotherapies
• then they reexamine the data supporting the efficacy of a number of the treatments currently believed to be empirically supported
• they offer suggestions for reporting hypotheses, methods, and findings from controlled clinical trials and for broadening the methods used to test the clinical utility of psychosocial interventions for particular disorders
• arguing from data (how to collect and interpret the data, so that we maximize our chances of drawing accurate inferences)

Question 73

What type of text is the Westen et al. article?

Answer 73

Critical towards the established “story” of efficacy and treatment choice
Arguing for a more nuanced story

Question 74

What is the target audience of the Westen et al. article?

Answer 74

The scientific community

Question 75

What happened in 1995 with the publication of the
first of several task force reports by the American Psychological
Association?

Answer 75

Suggested training professionals exclusively in the use of empirically validated therapies
The report distinguished ESTs from the less structured, longer-term treatments conducted by most clinicians

Question 76

Which characteristics do ESTs and the research models used to validate them, share in common?

Answer 76

• treatments are typically designed for a single Axis I disorder
• patients are screened to maximize homogeneity of diagnosis, and minimize co-occurring conditions that could increase variability of treatment response
• treatments are manualized
• treatments’ duration is brief and fixed to minimize within-group variability
• outcome assessment focuses primarily on the symptom that is the focus of the study

These characteristics are aimed at maximizing the internal validity of the study (“cleanness” of the design)

Question 77

What is considered a valid experiment according to the general assumption?

Answer 77

• randomization
• manipulating a small set of variables
• control of potentially confounding variables
• standardizing procedures as much as possible

–> drawing relatively unambiguous conclusions about cause and effect