November 9, 2022
1:30 pm / 3:00 pm
TITLE: Piloting a Standardized Psychosocial Assessment Tool (BATHE) in Genetic Counseling
DATE: Wednesday, 9 November 2022
TIME: 1:30–3:00 PM
LOCATION: Conference Room X303, Medical School Office Building, 1265 Welch Road, Stanford, CA
MaryAnn Campion (1)
Chloe Reuter (2)
Tia Moscarello (2)
Mimi Nguyen (1)
Beth Pollard (1)
(1) Department of Genetics
(2) Stanford Center for Inherited Cardiovascular Disease (SCICD)
The Data Studio Workshop brings together a biomedical investigator with a group of experts for an in-depth session to solicit advice about statistical and study design issues that arise while planning or conducting a research project. This week, the investigator(s) will discuss the following project with the group.
Significant rates of psychological distress have been found in patients receiving genetic counseling (GC) across a range of settings. These needs are often profound and unmet. Although genetic counselors (GCs) routinely assess patients’ emotional and psychological state, there is neither a universal definition of “psychosocial assessment” in genetic counseling nor are there any broadly-applicable assessment tools available. Those tools that do exist tend to be questionnaire-based, specific to a clinical indication, too time-consuming to be used clinically, require additional pre-appointment paperwork for patients, and/or were developed without deliberate attention to aspects of diversity. We can look to other healthcare settings to borrow tools which standardize the psychosocial assessment. One such tool developed in primary care is the BATHE method: a structured technique consisting of four questions that explore patients’ background (B), their emotional affect (A), the most troubling aspect (T), and how they are handling (H) the situation, all of which are paired with empathic responses (E). Evidence from the primary care literature shows that BATHE reduces patient anxiety and improves patient empowerment, and that providers find it concise and easy-to-learn, without increasing consultation time.
HYPOTHESIS & AIMS
For Aim 2 (Is the BATHE method feasible and acceptable among genetic counselors?), we have developed a 2hr workshop (for in-person or virtual delivery) to train genetic counselors on the BATHE method. We plan to lead 3–4 BATHE training workshops for ~200 practicing GCs (final number depending on power analysis) across diverse clinical specialties. We will use pre/post workshop questionnaires to gather the following data to assess the acceptability, appropriateness, and feasibility of BATHE among GCs.
We will collect data at three time points labeled T1 (prior to workshop), T2 (immediately following workshop), and T3 (two months post workshop). At time point T1, the data collected will include: clinical practice demographics and job satisfaction; self-reported practices of psychosocial assessment; and the modified GC Self-Efficacy Scale (Caldwell et al, 2018). At time point T2, the data collected will include: modified GC Self-Efficacy Scale (Caldwell et al, 2018); prospective acceptability, appropriateness, and feasibility (AAF); the Acceptability of Intervention Measure (AIM); Intervention Appropriateness Measure (IAM); and Feasibility of Intervention Measure (FIM; Weiner et al., 2017). The AAF, AIM, IAM, and FIM are measures of implementation outcomes that are often considered “leading indicators” of implementation success (Proctor et al., 2011). At time point T3, the data collected will include: modified GC Self-Efficacy Scale (Caldwell et al, 2018); changes in clinical practice; and retrospective acceptability, appropriateness, and feasibility (AAF). Here is a link to our planned data tables ( https://docs.google.com/spreadsheets/d/1Tnv_1Mj7Bip7Dsc1wT6SKSoVDJRX9n8ET8tuw-ZiOfs/edit#gid=0 ).
STATISTICAL ANALYSIS PLAN
These are the research questions that we hope to answer coupled with our best guess at the most appropriate statistical method(s) for each.
Q1: Is the BATHE method acceptable, appropriate, and feasible (AAF) among genetic counselors?
For this question, we will compute descriptive statistics for AAF at T2 and T3 and perform the paired samples t-test to determine whether there is a difference between participants’ prospective AAF (T2) and their retrospective AAF (T3).
Q2: What is the relationship between participants’ clinical practice demographics, job satisfaction, and psychosocial assessment practices, AND their AAF of BATHE?
Q2a: Is there a relationship between participants’ clinical practice demographic data (T1) and their prospective AAF (T2)?
Q2b: Is there a relationship between participants’ clinical practice demographic data (T1) and their retrospective AAF (T3)
Q2c: Is there a relationship between participants’ psychosocial assessment practices (T1) and their prospective AAF (T2)?
Q2d: Is there a relationship between participants’ psychosocial assessment practices (T1) and their retrospective AAF (T3)?
For Q2abcd, we plan to compute correlations for measurement data and to perform Pearson’s chi-squared test for categorical data.
Q3: What is the relationship between self-efficacy and AAF at the timepoints?
Q3a: What is the relationship between GCs’ self-efficacy and AAF of BATHE (T1, T2, T3) at any time point?
For Q3a, we will perform the paired samples t-test to determine whether participants’ self-efficacy scores change over time (compare T1, T2, and T3)?
Q3b: Is there a relationship between participants’ self-efficacy scores (T1 or T2) and their prospective AAF (T2)?
Q3c: Is there a relationship between participants’ self-efficacy scores (T1, T2, or T3) and their retrospective AAF (T3)?
For Q3bc, we will compute correlations for measurement data and perform Pearson’s chi-squared test for categorical data.
Q4: What is the relationship between changes of clinical practices (T3) and AAF of BATHE (T2 and T3)?
For Q4, we will compute correlations for measurement data and perform Pearson’s chi-squared test for categorical data.
1. Do the analyses above look correct? If not, what do you recommend?
2. Are we planning to run too many analyses?
3. When looking for associations between several variables, should we:
a. run separate individual correlations (for measurement data) and chi-square analyses (for categorical data)?
b. run an ANOVA if comparing multiple groups (e.g. genetic counselors representing different specialties)?
c. run something else?
4. For measurement data, do we run the same analyses if the data is on a sliding scale versus Likert versus numerical?
5. What tests should we run for “check all that apply” and/or when “other” is an option in the survey (in T1)?
6. For categorical data, does it matter how many categories participants can choose from?
7. What is the best way to conduct a power analysis to determine the minimum sample size?
8. How do we handle incomplete data (e.g. if we only get T1 and T2 on some participants)?
9. Can you recommend a statistician who might want to assist us along the way?