week 16: synthesizing research

Question 1

clinical outcomes research

Answer 1

the foundation for EBP

• to organize the research of clinical outcomes, a phased model has been proposed to organize into a coherent whole
• it provides a conceptual framework fro the progression of clinical-outcomes research from exploration to the point of refinement

Question 2

phase 1 of the outcomes research model

Answer 2

pilot the treatment or procedure of interest

• explorative, based on a tentative treatment protocol
• participants might not be representative, and external controls may be lacking
• designed to detect the presence or absence of a treatment effect

Question 3

phase 2 of the outcomes research model

Answer 3

happens if phase 1 showed premise of the treatment or procedure

• explorative, but with 4 goals
• –finalize operational definitions
• –define the exact population of interest
• –refine methodology
• –explore the treatment effect’s degree and permanency
• singe case and small group designs

Question 4

phase 3 of the outcomes research model

Answer 4

aims to test the critical hypothesis and answer the research question regarding the treatment efficiency

• large representative samples and includes eternal controls, comparison groups, etc
• typically pretest/posttest design with the comparison group receiving a different treatment

Question 5

phase 4 of the outcomes research model

Answer 5

aims to bridge the gap between research and practice

• –treatment efficacy on specific sub populations and other populations
• particularly important to collaborate with clinical personnel
• typically large group design

Question 6

phase 5 of the outcomes research model

Answer 6

the focus of this phase shifts to other treatment effectiveness issues:

• –cost benefit
• –consumer satisfaction
• –quality of life issues
• typically large group designs and usually does not include a comparison group

Question 7

systematic reviews

Answer 7

put the pieces together

• –combines phase 4 and 5 studies
• identified studies that have a common hypothesis, combines their results, and formulates conclusions based on the sum of the results

Question 8

goals of systematic reviews

Answer 8

• provide comprehensive summaries of research on a specific topic
• provide strong evidence for clinical decision making
• identify ideas for future research

Question 9

narrative approach for systematic reviews

Answer 9

• the traditional method of reviewing and synthesizing literature
• requires
• –thorough search of the relevant lit
• –qualitative analysis of the results of past studies
• –a conclusion based on synthesis of results
• limitations
• –researchers bias
• –narrative reviews are prone to different interpretations by different authors

Question 10

steps to a narrative approach of systematic review

Answer 10

• formulate a problem
• locate and select studies
• assess study quality
• collect data (qualitative and quantitative)
• analyze and present results
• interpret results

Question 11

the quantitative approach of systematic reviews (3 methods)

Answer 11

• one of the simplest methods for the quantitative synthesis of research is the cote counting method where you say positive, negative, or non significant findings
• –not sensitive to sample size effects and does not evaluate effect size
• the combined probability method which incorporates exact probability in the synthesis, accounting for different sample sizes
• –does not quantify effect size nor heterogenity among studies
• modern meta-analysis
• –interprets the results of several independent studies statistically
• –combines measures of effect size from the studies and achieves an overall measure of effect then tests the significance of this overall effect

Question 12

the best evidence approach of systematic review

Answer 12

aims to combine the best aspects of both narrative and quantitative approaches by having 4 parts

• intro
• method–describe and justify the selection criteria
• lit review–discuss effect size and explore concepts and method issues
• conclusion

Question 13

step 1 of the systematic review process

Answer 13

develop a hypothesis and eligibility criteria
—once parameters (population, intervention, outcomes of concern, study designs to review) are identified the hypothesis and questions to be answered should be apparent

Question 14

step 2 of the systematic review process

Answer 14

develop a search strategy and select studies for inclusion

• assemble a database to pick studies from
• assumes the entire body of lit has been identified including unpublished

Question 15

step 3 of the systematic review process

Answer 15

assess study quality

• might have to contact researcher about method
• use the threshold or quality weighting approach for the methodological quality

Question 16

the threshold approach to assess study quality

Answer 16

• sets an absolute threshold for each study to meet
• excludes more studies
• some have a continuum of obviously meets to obviously does not meet

Question 17

the quality weighting approach to assess study quality

Answer 17

• includes most studies and attempts to weigh them according to their quality–better quality means more weight in the conclusion
• three advantages
• –provides a larger poor of studies to represent available research
• –enables reviewers to examine the relationship between methodology and study outcomes
• bias from assigning weight to each study potentially so may want to use blinding

Question 18

step 4 of the systematic review process

Answer 18

collect data and convert study statistics to a common metric, typically effect size

• summary stats use effect size
• count data use repetition rate
• f, t, and chi squared stats
• combine effect sized to form a cumulative effect size
• –if qualitative then look for trend and use tables to compare results for similarities and differences

Question 19

step 5 of the systematic review process

Answer 19

analyze and present results

• larger studies should be given more weight because larger studies have smaller variance
• need to measure heterogeneity
• the interpretation of the cumulative effect depends on the validity of the database, calculation of cumulative effects, and the presence or absence of moderator variables

Question 20

measuring heterogeneity

Answer 20

• how the results of individual studies may be mathematically incompatible with the results of other studies
• – a degree of heterogeniety can be measured by several procedures but easiest to visually inspect mean, SD, CI
• or use Q statistic= significant means other variables affecting= moderator variable which is an independent variable other than the treatment variable which is causing the variance between studies

Question 21

statistical models based off of heterogeneity tests

Answer 21

• the fixed methods model which assumes that the variability of results is attributable to random variation
• the random effects model which assumes the variability between studies is attributable to random variation plus the effects of confounding variables
• –this model more often fits data from quasi-experimental studies

Question 22

step 6 of the systematic review process

Answer 22

interpret the results

• reviewers depend on the synthesis of results for their interpretation
• reviews need to be updated as often as possible