Case control studies (including single case studies and case series) Flashcards
(26 cards)
What are single case studies?
Experimental methods that allow inferences to be made about effectiveness of intervention for that one person.
What are the advantages of single case studies?
Allow discrimination of specific treatment effects from non-specific treatment effects, such as spontaneous recovery, the ‘charm’ effect and support, social participation, motivation.
Single case studies allow initial exploration of innovative treatments. Read about a new treatment can do a single case study on your client.
They are widely used in the intervention literature for some client groups e.g. PWA and and well-developed experimental methods for approaching therapy issues.
What is N=1 RCT
In the N=1 RCT, patients undertake pairs of treatment periods in which they receive a target treatment in one period of each pair and a placebo/alternative in the other. The N = 1 RCT continues until both the patient and clinician conclude that the patient is, or is not, obtaining benefit from the target intervention.
What are N=1 RCTs unsuitable for?
short-term problems/for therapies that cure (such as surgical procedures)- need to be able to look at it over a long period.
For therapies that act over long periods of time or prevent rare or unique events (such as stroke, myocardial infarction, or death).
Only possible when patients and clinicians have the interest and time required to ‘try out’ different interventions.
Advantages of N=1 RCT
Are highly feasible, just your patient 1-1.
Can provide definitive evidence of treatment effectiveness in individual patients
May lead to long-term differences in treatment administration.
Contrast with population studies which provide only evidence about what’s most effective on average.
Allow for exploration of different treatments for clients with chronic conditions. (may have found 2 RCTs arguing for different therapies, patient saying idk what I prefer, can try which one works best for them, doing own RCT with your client).
Have better compliance and higher patient satisfaction
What are the Single Case Study Experimental Designs
- Repeated Baseline (AB design)
- Multiple Baseline
- Item-specific design
- Crossover design
- Combined design
Single case study experimental designs- REPEATED BASELINE (AB design). What is this?
Repeated baseline designs contrast performance on a single task across time. Treatment is preceded by repeated testing of the process targeted in treatment e.g. naming
Typical ABA format (A-period of no therapy, B-period of therapy). Test-treat-test. If treatment is effective, significant gains will be found following period B, compared to period A.
Pros of a repeated baseline design
Easy to use
Can consider multiple phases of treatment
Allows measurement of any spontaneous recovery
Looks at durability of treatment over time- how long the treatment will be effective for, how long you’ll see progress without going stagent.
Cons of repeated baseline design
Cannot distinguish between specific and non-specific treatment effects.. use multiple baseline instead/ repeated baseline with control.
May see practice effects on tests
Predictions of spontaneous recovery may be a problem
Single case study experimental designs- MULTIPLE BASELINE (including control). What is it?
Contrasts performance on a number of tasks to show effectiveness of intervention. Control built in as assessing multiple skills/tasks at baseline.
Need strong predictions about what will change and what will not.
Control tasks must be unrelated and expected to stay unaffected by treatment
Pros of baselines including control
Easy to use
Controls for ‘Charm effects’ because have treatment AND control.
Cons for baseline including control
Assumes that spontaneous recovery will be equal for all tasks.
Assumes tasks have equal impairment and levels of difficulty
Requires selection of independent tasks
Single case study experimental designs- ITEM SPECIFIC DESIGN. What is it?
Select 3 groups of items- treated items, practice and untreated (control).
A patient’s performance is tested on a particular task and then items randomly assigned to treatment and control sets.
Prior to therapy, must make predictions as to whether generalisation to untreated items is expected. After treatment, performance on treated/untreated items is contrasted. Either item-specific effect OR
if performance on treated and untreated is equal, there may be generalised improvement, spontaneous recovery or non-specific treatment effects.
How do you assign items?
- Random assignment
Assign the items into n groups by any random method – shuffle cards, draw from hat, random numbers in excel =rand ( ).
or
- Random assignment balanced by pre-test accuracy
Group the items by pre-test performance e.g., items correct on both pre-tests, items correct in test1 but not in test 2 or vice versa and items not correct in either pre-test. Then assign each of the items within each group randomly.
Pros of item-specific design
Easy to use
Well controlled
Cons of random assignment
need stronger predictions of therapy effects as using multiple levels of control- from literature
Requires many items
Method of allocation to groups may affect results
CROSS-OVER DESIGN- what is it?
This is similar to N=1 RCT.
There are two phases of treatment preceded and followed by testing of performance on two unrelated processes (or two different sets of items). Process A (or set A) treated initially followed by treatment for process B (or set B). If treatment is effective, significant gains in process A will follow the first phase of treatment with no change in process B. Significant gains in process B will be seen following the second phase of treatment.
Pros of Cross-over design
Very well controlled
Controls for ‘Charm’ effects
Cons of cross-over design
*Can be difficult to set up
*Extends over a long period of time/lots of sessions
*Need independent process or items
Evaluating single case studies: Three key questions:
- Are the results of the study valid?
- What are the results?
- Will the results help me caring for my own client/patients?
Scale for single case studies is the SCED Scale.
What is a case series
Case series are single case studies across multiple participants. Participants need to be as similar as possible and the same experimental design (some assessment measures/treatment) applied across all participants. It can be consecutive or non-consecutive cases. Case series allow replication of therapy across individuals – an exploration of whether effects are generalisable; and an exploration of factors that may influence outcome of treatment - do some people benefit more from this treatment than others? Why?
* Severity
* Pre-treatment profile
* Personal factors
Critical appraisal questions of aims and methods
Are aims and research questions clearly stated?
Is there a theoretical rationale for intervention?
Is the review of the literature clearly focuses?
Are methods used appropriate to adress the research question?
Critical appraisal questions for participant
Who are the participants? age. SES, severity, motivation. Should be relevant to your client
Who provided the intervention- normal practitioners or some sort of “special provision”
