KT Intervention to Increase Use of Outcome Measures

Purpose/Specific Aims

1. Design two locally tailored multi-modal knowledge translation interventions using a mixed methods barrier assessment with input from the clinicians.
2. Compare the effectiveness of the interventions on physical therapists' use of standardized assessments/outcome measures.

• The fully supported group will have the intervention designed and implemented by the investigator (knowledge broker)
• The partially supported group will have the intervention designed and only partial implementation by the investigator (knowledge broker)

1. Determine the impact of the intervention on patients who were evaluated by PTs in both groups.

Objectives

1. To determine the immediate and longer term (6 months following the intervention) effectiveness of a theoretically informed multi-modal tailored intervention designed and implemented by a knowledge broker as compared to a tailored intervention designed, but only partially implemented by the knowledge broker on the use of a selected patient standardized assessment by physical therapists who work in rehabilitation.
2. To explore and compare patients' experiences of being evaluated by PTs in the fully supported and partially supported groups. The proposed study intends to compare two KT interventions designed by a KB using an fully supported and partially supported implementation groups. The outcomes will be measured through mixed methods to determine the immediate and 6 month retention of behavior change, impact of the intervention, and investigate the impact of the intervention on patients. Quantitative data analysis will be used through chart audit to determine documentation of outcome measures and a self-reported survey (Goal Attainment Scale). Qualitative analysis (focus groups) will be used to determine reasons for behavior change which can be used to guide future interventions. Long term analysis also helps investigators determine if the behavior was retained after the intervention was completed. Finally, few studies have investigated the impact of the intervention on the patients and an exploratory focus group will compare patients treated by physical therapists in the experimental and control groups.
3. Research Design and Methods This is a mixed methods cluster randomized controlled trial with 2 sites. Phase I: Barrier Assessment and Intervention Design The interventions will be designed by the knowledge broker with input from the clinicians through a barrier assessment. The barrier assessment will include a chart audit, a questionnaire and a focus group. The chart audit will be used to determine current use of standardized assessments. The questionnaire will be used assess current use, barriers, and facilitators to evidence based practice and using standardized assessments. The clinicians will be given a link to access and complete the barrier assessment survey anonymously. The barriers and facilitators to evidence based practice (EBP) and using outcome measures questionnaire was developed and validated in a previous study and designed using the Theoretical Domains Framework (TDF). The questionnaire contains a question on which outcome measure the participants would like to use more frequently, to help guide selection of the standardized assessment. The focus group will present chart audit and questionnaire data, confirm barriers and facilitators, and determine the strategies that will be used in the intervention based on the Theoretical Domains Framework consensus matrix.

Standardized assessments will be selected by the groups through the following process: Participants will complete a question on the questionnaire about which standardized assessments they would like to use more frequently. Those results will be tallied and two standardized assessment will be briefly presented to the experimental and active control groups during the first focus group. The two standardized assessments presented must be performance based, comprise of only a few steps, be easy to interpret and have at least adequate psychometric properties (reliability and validity testing) Each group will select a comparable outcome measure/standardized assessment based on the two standardized assessment the investigators selects to present. The group will be asked to select one assessment based on group consensus of 80%. Disagreement will be handled through discussion. Each group can choose different standardized assessments.

Phase II: Intervention The trial will be conducted in two inpatient rehabilitation hospitals within the same organization, Kessler (Chester and Saddle Brook locations). The fully supported group (enhanced arm) will receive a theoretically informed tailored multi-modal intervention designed and implemented by a knowledge broker (investigator) to increase the use of a selected standardized assessment. A knowledge broker (KB) is a researcher who brings evidence to the clinicians with the purpose to change practice. KBs are able to locally tailor interventions for the clinicians. The partially supported arm will also receive a theoretically informed tailored multimodal intervention that will be designed, and only partially implemented by the KB. The groups will be randomization, by site, after the barrier assessment is completed (Chester or Saddle Brook). Randomization by site was selected to reduce contamination among the therapists.

The fully supported group will have 2 additional meetings with the Knowledge Broker (KB) to assist with implementation and barrier assessment. The KB will meet with the fully supported group six times to design and implement the intervention, and the KB will meet with the partially supported group four times to design and partially implement the intervention. The KB will encourage both groups to use the selected standardized assessment. The fully supported group will have four educational outreach visits over a four month period. Two of the four outreach visits will occur simultaneously with focus groups and provide feedback from chart audit data on current use of the selected standardized assessment and design the intervention by discussing an action plan to determine strategies to increase the use of the selected assessment. The investigator will implement the intervention in the two additional outreach visits. Strategies will be tailored to the group and may include: distribution of educational materials, an interactive educational meeting, and reminders. The partially supported arm will have two educational outreach visits where the intervention/action plan will be designed but not implemented by the investigator. The partially supported arm will be encouraged to self-implement the intervention during months 2 and 4.

Phase III: Outcome Assessment Outcomes will be assessed to determine documented and self-reported use of the selected standardized assessment through chart audit at baseline, immediately post intervention and 6 month follow-up. Focus groups will be conducted at post intervention and follow-up to explore and compare the impact of the intervention on behavior change, facilitators to change and barriers that remained among each group. In addition, four focus groups will be held with patients of the therapists' in each group to explore and compare their experience.

3.3 Sample Size Justification Physical Therapist- up to 40 PTs, 20 at each location will be invited to participate. The sample size is based on the number of PTs who work at Kessler and a power calculation based on pilot research.

Patients treated by the enrolled PTs- up to 24; 3-6 in each group.

3.3.1 Power Calculation: A power calculation was completed using the pilot study data. Based on the pilot study data, two groups of 6 will achieve an effect size of .7 and power of 81%.

Baseline data:

Group equivalency at baseline will be analyzed using the following variables: age, gender, years of experience, and educational level will be analyzed using either an unpaired t-test or Mann-Whitney U test. Continuous data (age, years of experience, use of standardized assessments) will be analyzed using an unpaired t-test if assumptions of normality are met and Levene's test of homogeneity is not significant. Level of significance will be set at .1. A Mann-Whitney U test will be used for ordinal data and data that are not normal. Variables on which the groups differ at baseline will be considered for inclusion in the final models as covariates.

Barrier assessment questionnaire data will be analyzed descriptively to guide intervention development. The baseline focus group will be transcribed, coded and themes will be created.

Documented use of the standardized assessment will be analyzed using a within-between, repeated measures ANOVA, with time and group as factors. If data are non-normal attempts will be made to transform the dependent variables to meet parametric standards. Another option will be to explore statistical procedures that model non-Gaussian distributions. Level of significance will be set at .05.

Self-reported use will be analyzed using the Goal Attainment Scale using the Friedman's test because data are ordinal. The Friedman's test will used for within group analysis only. Comparisons between groups will be reported descriptively or alternatively, a post-intervention Mann-Whitney U may be used. Level of significance will be set at .05.

Focus group: Each focus group will be recorded and transcribed and conventional content analysis will be conducted. After the focus group is transcribed, open coding will occur by coding notes, observations, and comments in the margins of potentially relevant information. Similar notes, observations and comments will be grouped together into categories or themes. Appropriate categories will be associated with Theoretical Domains Framework and the Consolidated Framework for Implementation Research (CFIR) into codes and rated. Experiences will be determined by satisfaction with the intervention, perceived facilitators for changing their behavior and barriers that remain. Several steps will be taken to establish trustworthiness of the qualitative data..

Realist Analysis: Chart audit data will be used as the Outcome and Qualitative codes from the CFIR will be entered as Mechanisms (Interventions) or Context. Codes and ratings will be compared with chart audit data to determine reasons for use or non-use.

Patient Focus group: The patient focus group, will be used to compare the experiences of patients treated by therapists in the intervention and control groups. Each focus group will be recorded and transcribed. Open coding will occur as I will write notes, observations, and comments in the margins of potentially relevant information of the transcribed focus group. Similar notes, observations and comments will be grouped together into categories or themes. Categories will be compared across groups.