CERSI Compounding PRA SSB 2019

U.S. Food & Drug Administration
Obtaining Information for Evaluating Nominated Bulk Drug Substances for Use in
Compounding Drug Products Under Section 503B of the Federal Food, Drug, and Cosmetic Act.
OMB Control No. 0910-NEW
SUPPORTING STATEMENT—Part B: Statistical Methods
1. Respondent Universe and Sampling Methods
Research will generally rely on qualitative methods and is not intended to yield results that are
projectable to the national population. As described below, research will target individuals with
specific types of clinical expertise to provide their expert opinions, as well as a pre-identified
group of parents of children with Autism Spectrum Disorder.
UMD-CERSI Expert Interviews and Focus Groups
Researchers will conduct in-depth interviews and focus groups with an estimated 150
experts/specialists annually to obtain their opinions on the use of compounded drugs containing
specific bulk drug substances in patient care. Experts will be identified through various means,
including contacts at the University of Maryland School of Medicine as well as referrals obtained
through contacts at applicable national medical professional organizations.
UMD-CERSI Expert Questionnaire
After completion of the interviews and focus groups, a tailored questionnaire will be utilized to
obtain a more detailed understanding of how compounded drugs containing specific bulk-drug
substances are used in patient care. Researchers will engage applicable national medical
professional organizations for each medical specialty to facilitate questionnaire distribution and
participation. Researchers will seek to obtain participation of at least 50 experts from each
specialty group (a total of 750 clinicians) annually.
JHU-CERSI Parent Questionnaire
Researchers will use a questionnaire to obtain information from parents of children with autism
spectrum disorder (ASD) on the use of compounded drugs containing any of six bulk drug
substances in these patients. Recruitment will be conducted through SPARK (Simons Foundation
Powering Autism Research for Knowledge) and their online autism research registry, which has

20,000 individuals with ASD and their families. Researchers will seek to recruit approximately
1,000 families from SPARK (consisting of parents and caregivers of children with ASD).
SPARK registry participants have previously consented to be in SPARK and have agreed to
sharing of de-identified data with outside researchers. Since JHU faculty or staff will not have
access to the registry or private information, SPARK staff will query their registry to find parents
and caregivers with at least one child with ASD. Once identified, eligible families will be
emailed an invitation to complete the online ASD Treatment Questionnaire.
2. Procedures for the Collection of Information
UMD-CERSI Expert Interviews and Focus Groups
Potential participants will be contacted via email to schedule the interview or focus group. Each
interview and focus group will last one or two hours, respectively, and will include questions
regarding the participants’ knowledge and experience using a bulk substance. Interviews will be
conducted in-person if the participant is local, but researchers will also conduct interviews by
telephone to support a more geographically diverse group of participants. The interviews and
focus groups will be conducted using a semi-structured approach which will allow the participant
to expand on his or her previous experience, as needed.
UMD-CERSI Expert Questionnaire
The questionnaire will be administered using a web-based platform and will be distributed via
email. The questionnaire will include a combination of free-text and prompted responses and will
include questions regarding his or her knowledge, experience, frequency, and indications of use
for each substance. We estimate the questionnaire will take respondents approximately 30
minutes to complete.
JHU-CERSI Questionnaire
Information on the six bulk drug substances that are the subject of JHU-CERSI’s research will be
collected as part of a broader questionnaire on medical and behavioral treatments used to treat
ASD, including both conventional therapies and complementary and alternative medicine
(CAM). Participants will be asked to identify and answer questions about specific interventions
they have used to treat their children with ASD. These outcome measures will be used to identify
the treatments used most often and the characteristics of individuals who use them.
The questionnaire includes a short series of basic demographic questions, followed by questions
about the child’s co-occurring medical conditions. The following sections present extensive lists

2 

of medical and behavioral therapies used to treat ASD and prompt the respondent to select all
that they have tried on his or her child. The lists include the six bulk drug substances of interest.
For each treatment selected, the respondent will be asked to answer an additional series of
questions detailing the use and experience with the treatment. Estimated average completion
time for the questionnaire is 30 minutes, though will vary by child and will depend on the
number of treatments a child has used.
Researchers will examine the response rates of participation by comparing the number of
SPARK participants invited to complete the questionnaire to the number who agreed to
participate. We will also examine if any one item on the questionnaire has limited response rates,
and work to identify factors associated with response/ non-response.
Conventional statistical techniques for response data, such as descriptive statistics, t-tests, chisquare tests, and regression models will be used to analyze the data. Researchers will examine
use of the six compounds of interest and will examine phenotypic and demographic indicators of
endorsement of use.
3. Methods to Maximize Response Rates and Deal with Nonresponse
As noted, research methods are primarily qualitative, and respondent groups for questionnaires
consist of samples of convenience and individuals within specialty organizations with specific
expertise. Efforts will be made to maximize response rates, but results will not be nationally
representative.
UMD-CERSI Expert Interviews and Focus Groups
Researchers will conduct outreach through local and national networks until approximately 10
specialists from each of 15 medical specialties are identified and participate. Thematic analysis
of each interview and focus group will be completed utilizing NVivo to identify common
indications, dosing, and experience using each substance.
UMD-CERSI Expert Questionnaire
Bi-weekly reminder emails will be sent to obtain a target participation of approximately 50
experts from each specialty group (a total of 750 experts). Quantitative and qualitative analyses
will be conducted to evaluate both response frequency as well as common response themes.
JHU-CERSI Questionnaire

3 

Up to three reminders will be sent over a 4-week period until recipients respond. Recipients who
have not responded after 6 weeks will be considered to have refused participation. The 6-week
time frame for the project has been validated by SPARK in prior online research invitations.
4. Test of Procedures or Methods to be Undertaken
Small adjustments may be made to questionnaires based on information obtained through
interviews and focus groups with experts.
5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing
Data
UMD-CERSI
Statistical analysis beyond simple response rate frequency is not anticipated. The following
individuals participated in research design and will participate in results analysis:
Ashlee Mattingly, PharmD, BCPS, Assistant Professor, University of Maryland,
Baltimore.
Email: [email protected]; Phone: (410) 706-1361
James Polli, PhD, Professor, University of Maryland, Baltimore
Email: [email protected]; Phone: 410-706-8292
Stephen Hoag, PhD, Professor, University of Maryland, Baltimore
Email: [email protected]; Phone: (410) 706-6865
Lisa Lebovitz, JD, Assistant Dean of Academic Affairs and Assessment, University of
Maryland Baltimore
Email: [email protected]; Phone: (410) 706-3457
JHU-CERSI
Simple statistical analysis will be conducted of JHU-CERSI’s parent questionnaire results. The
following individuals participated in research design and will participate in results analysis:
Heather E. Volk, PhD, Assistant Professor, Departments of Mental Health and
Environmental Health and Engineering; Associate Director, Wendy Klag Center for

4 

Autism and Developmental Disabilities, Johns Hopkins Bloomberg School of Public
Health
Email: [email protected]; Phone: 410-614-9404
Luke Kalb, PhD, Assistant Professor, Department of Mental Health, Johns Hopkins
Bloomberg School of Public Health. Assistant Professor, Center for Autism and Related
Disorders, Department of Neuropsychology, Kennedy Krieger Institute
Email: [email protected], [email protected]; Phone: 443-923-7630

5