The instrument and sampling framework was developed by the George Washington University (GW) Mullan Institute for Health Workforce Equity as part of the Mental and Substance Use Disorder Practitioner Data grant funded by the Substance Abuse and Mental Health Services Administration (SAMHSA), grant number H79FG000028.
Employer Survey
The Survey of Employers of Behavioral Health Providers will be administered to a single representative of each of the 2,800 member organizations of the National Council of Behavioral Health (NCBH) that provide direct services. Survey participants will complete the questionnaire for their respective organization.
Provider Survey
The Survey of Clinical Behavioral Health Providers will be fielded via email to a random sample of 5,000 actively licensed clinical behavioral health providers in four professions (1,250 each): licensed psychologists, licensed clinical social workers, licensed marriage and family therapists, and licensed professional counselors.
A computer-generated automated process will be used to randomly select 1,250 invitees per profession drawing from the full list of licensed clinical providers included in the state licensing board lists where email addresses are available. To date, emails are available for the following states but more will be added as data becomes available:
16 states for licensed psychologists (AR, FL, KS, MI, MN, MS, ND, NE, NY, OH, OR, RI, TX, UT, WI, WV)
12 states for licensed clinical social workers (FL, KS, LA, MI, MN, NE, NY, OH, OR, RI, UT, WI)
10 states for licensed professional counselors (FL, KS, MI, NE, NY, OH, OR, RI, UT, WI)
11 states for licensed marriage and family therapists (FL, KS, IA, MI, NE, NY, OH, OR, RI, UT, WI)
Using all available state licensure rosters that include email contact information ensures the broadest geographic representation for the sampling frame for this email survey and will include potential respondents from all of the U.S. Census regions. Additionally, the survey will capture a range of scope of practice (SOP) guidelines and Medicaid reimbursement policies. For example, while Florida has liberal SOP guidelines for behavioral health counselors with the SOP explicitly allowing all four professions of interest – licensed psychologists, licensed clinical social workers (LCSW), marriage and family therapists (MFT), and licensed professional counselors (LPC) - to diagnose, Michigan has more restrictive SOP guidelines, only explicitly allowing LCSWs and psychologists to diagnose. Similarly, while some states reimburse several of the four professions for diagnosis, other states do not reimburse any, and therefore surveying providers with emails across all available states will capture the diversity of Medicaid reimbursement policies. For example, Florida’s Medicaid guidelines allow reimbursement for diagnosis services for both LCSWs and psychologists whereas Michigan’s Medicaid program does not reimburse any of the four professions for diagnosis.
B2. Information Collection Procedures
Eligible participants will be contacted via email using the email addresses obtained from: 1) NCBH for the employer survey and 2) state licensing boards for the provider survey. The surveys will be administered using Qualtrics, a secure survey software to conduct health workforce surveys (https://www.qualtrics.com/) – and covers the full range of survey administration from initial outreach via email, tracking response rates, targeting follow up emails only to those who have yet to complete the survey, and collecting the survey responses and storing them in a secure research ready file format that can be used for analysis. The Qualtrics software will enable the research team to generate customized emails for each respondent that includes a unique, anonymous link that participants can either click on or paste into the browser and can be completed on a phone or over the internet. Using the Qualtrics email system, up to four reminders will be sent to non-respondents after the initial invitation. (Participants who have already responded will not receive these reminders.) The email invitations will include a description of participant rights, including the right not to participate in the survey. The first page of the online survey will include a more detailed description of participant rights and ask them to affirm that they agree to participate in the survey before proceeding to the questions. Skip patterns will facilitate navigation through the survey by only displaying items that apply to the participant, based on information already entered in the system.
To increase response rates for the provider survey, the email invitation will also note that the first 1,500 respondents to complete the survey will receive a $20 incentive as a thank you for participating. (There is no incentive for the employer survey.) The research team will use a system called Forte Participant Payments (https://forteresearch.com/payments/), a secure online tool used by academic institutions and health systems for managing payments to research participants. Once a participant has successfully completed and submitted the survey, they will be redirected to the Forte system to claim their gift card. If participants choose to withdraw from the study, they will no longer be eligible to receive the gift card incentive.
Data Management
This study poses minimal risk to subjects. However, a possible risk to subjects participating in the 2021 Behavioral Health Workforce Surveys is the loss of privacy or confidentiality and the risk of identification. The information being collected is not sensitive and identifiers will be removed prior to analysis. The identifiable version will be destroyed after we obtain the deidentified data set.
Steps taken to minimize risks and to protect subjects’ welfare:
Data will be saved on GW Box, the university's enterprise file sharing service for online cloud storage. Only members of the research team on record with the IRB with a GW Net ID and password will have access to the Box folder.
Study ID and subject information will be kept in separate, encrypted files. Any personally identifiable information from the surveys will be sent over from the survey administrator to one research team member who will de-identify the data in GW Box and assign research IDs to each subject and then share the deidentified surveys with other research personnel on GW Box. Following the deidentification process, the survey research lead will destroy the raw, identifiable version of the survey.
Survey results will only be reported at the aggregate level.
The surveys were developed in partnership with GW, NCBH, and the University of Michigan’s Behavioral Health Workforce Research Center who provided significant input to ensure both surveys only included highly relevant questions and that the questions were clear and straightforward to ensure respondents would be able to readily answer.
NCBH, who is providing the employer survey mailing list, has historically had strong response rates for their member surveys. They have agreed to help promote the employer survey with their members to ensure their members are familiar with the survey and its value.
For the provider survey, the first 1,500 participants to complete the survey will receive a $20 incentive as an additional strategy for increasing response rates. This amount has been shown be a good level given the declining marginal improvement in response rates for higher incentives. 1 Past health workforce studies conducted by Mullan Institute research teams in collaboration with professional associations using $20 incentives have generated response rates of 25-35%. It is expected the response rate to be similar for this survey. Qualtrics software will track which respondents have completed the survey so we can generate response rates.
The surveys were reviewed by representatives of SAMHSA, NCBH Providers, and the University of Michigan’s Behavioral Health Workforce Center. Feedback helped to clarify terminology and language and eliminate unnecessary questions.
The estimates of time to complete the newly revised instruments are based on pilot tests. The mean time-to-complete figures reported by each tester was used as the final burden estimate for both surveys, which was consistent with Qualtrics software estimates of the time to complete.
Leah
E. Masselink, PhD
Assistant Professor
Department of Health
Policy and Management
Milken Institute School of Public
Health
The George Washington University
950 New Hampshire
Avenue NW, Suite 600
Washington, DC 20052
(p) 202.994.7183
(c)
240.903.1672
[email protected]
1 https://academic.oup.com/poq/article-abstract/56/4/442/1845084
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Erikson, Clese Elaine |
| File Modified | 0000-00-00 |
| File Created | 2021-04-14 |