IMPROVE Study Phase 2 Supporting Statement

FDA DOCUMENTATION FOR THE GENERIC CLEARANCE
ON DATA TO SUPPORT DRUG PRODUCT COMMUNICATIONS
(0910-0695)
TITLE OF INFORMATION COLLECTION: Educating Groups Influencing Generic
Drug Use (Survey)
DESCRIPTION OF THIS SPECIFIC COLLECTION
1. Statement of Need:
The Food and Drug Administration (FDA), Center for Drug Evaluation and
Research (CDER), Office of Generic Drugs (OGD) is seeking OMB approval
under the generic clearance 0910-0695 to conduct surveys for the project
“Educating Groups Influencing Generic Drug Use.”
Based on the supporting statement for generic clearance 0910-0695, 1 the purpose
of information collection under this generic clearance is “to provide tools to
assess the need for communications on specific topics and to assist in the
development and modification of communication messages to promote public
health and compliance with regulations.” The specific collection described in this
memo aims to assess the need for communications directed at healthcare
providers on the use of generic drugs for certain drug classes. Exhibit 1 illustrates
the full set of research phases for this project. Our previous submission addressed
only the focus groups in Phase 1, which informed the development of the survey
that is the subject of this information collection. Please note that this information
collection request concerns only Phase 2.
Exhibit 1. Overview of Research Phases
Phase 1:
Focus
groups to
inform
development of
surveys to
understand
barriers
and
facilitators
to generic
prescribing

1

Phase 3:
Development and
pilot testing
of
messages
to promote
generic
prescribing

Phase 2:
Surveys to
gain
insight into
barriers
and
facilitators
to generic
prescribing

http://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201509-0910-002
1

Phase 4:
Revision
of
messages
to promote
generic
prescribing

Congress passed the Generic Drug User Fee Amendments (GDUFA, Title III of
the Food and Drug Administration Safety and Innovation Act (Public Law 112144)) in July 2012. Under GDUFA, FDA obtains industry and public input to
create regulatory science initiatives regarding research on generic drugs that
advance public health. 2 Once marketed, certain generic drugs are often not
preferred over brand drugs 3 even though generic drugs generally cost less than
brand drugs. 4 Research to characterize the key influencers of generic drug use
(particularly for drug classes with low generic drug use), including their
knowledge base and perceptions toward generic drugs, is needed to effectively
design and deliver communications about generic drugs to the key groups
influencing consumer acceptance and use of generic drugs. To address this
regulatory science need regarding generic drugs, the FDA entered into a
cooperative agreement with the University of Chicago (UChicago) (Grant number
1U01FD005485-01).
As prescribers of brand or generic drugs, healthcare providers can influence
consumer use of generic drugs. Many healthcare providers have been slow to
adopt generic drugs 5 despite data suggesting clinical equivalence between generic
and brand drugs. 6
In primary care practice, both clinicians and patients express concerns regarding
use of certain generic drug classes such as antidepressants, oral contraceptives,
and cholesterol lowering drugs. 7 Through a cooperative agreement with
2

http://www.fda.gov/ForIndustry/UserFees/GenericDrugUserFees/ucm370952.htm
Scher, S. (2013) The Branded Advantage. Ophthamol Mgmt. July: p18
http://www.ophthalmologymanagement.com/printarticle.aspx?articleID=108618

3

Alloway, RR, Isaacs R, Lake K, Hoyer P, First R, Helderman H, Bunnapradist S, Leichtman A, Bennett
MW, Tejani A, Takemoto SK. (2003) Report of the American Society of Transplantation Conference on
Immunosuppressive Drugs and the Use of Generic Immunosuppressants. A J Transpl 3: 1211.
Liow K, Barkley GL, Pollard JR, Harden CL, Bazil CW. (2007) Position statement on the coverage of
anticonvulsant drugs for the treatment of epilepsy. Neurology 68: 1249.
American Thyroid Association, The Endocrine Society, and American Association of Clinical
Endocrinologists. (2004) Joint Statement on the U.S. Food and Drug Administration’s Decision Regarding
Bioequivalence of Levothyroxine Sodium. Thyroid 14: 486.
4
Avoidable Costs in U.S. Healthcare: The $200 Billion Opportunity from Using Medicines More
Responsibly. Report by the IMS Institute for Healthcare Informatics.
http://www.imshealth.com/deployedfiles/imshealth/Global/Content/Corporate/IMS Institute/RUOM2013/IHII_Responsible_Use_Medicines_2013.pdf
5
Shrank WH, Liberman JN, Fischer MA, Girdish C, Brennan TA, Choudhry NK. Physician Perceptions
about Generic Drugs. Ann Pharmacother. 2011;45(1):31-8.
6
Kesselheim AS, Misono AS, Lee JL, et al. Clinical Equivalence of Generic and Brand-Name Drugs Used
in Cardiovascular Disease: A Systematic Review and Meta-Analysis. JAMA: the journal of the American
Medical Association. 2008;300(21):2514-2526.
7
Bolton JM, Dahl M, Sareen J, Enns MW, Leslie WD, Collins DM, Alessi-Severini S. A population-based
study of the use of selective serotonin reuptake inhibitors before and after introduction of generic
equivalents. Can J Psychiatry. 2012;57(4):223-9.

2

UChicago, the barriers and facilitators to consumer use of these drug classes will
be explored by conducting surveys of primary care physicians and nurse
practitioner prescribers.
2. Intended use of Information:
Information collected in these surveys will be used to improve understanding of
the barriers and facilitators to prescribing generic drugs, with a particular interest
in knowledge gaps regarding generic drugs.
Information collected in the Phase 2 surveys will be used to inform development
of educational messages regarding generic drugs (Phase 3). By understanding
why healthcare providers may sometimes not prescribe generic drugs, effective
educational messages may be developed.
3. Description of Respondents:
The survey will be disseminated through the American College of Physicians
(ACP) and the American Association of Nurse Practitioners (AANP).
ACP is the largest medical-specialty organization and the second largest physician
group in the United States. General internist primary care physicians typically
provide comprehensive longitudinal care and coordinate complex treatment for
adults. The Internal Medicine Insider Research Panel is a community of
approximately 1100 U.S. ACP members and 300 non-members who participate in
research surveys distributed by the ACP Research Center.
AANP is a union of the American Academy of Nurse Practitioners and the
American College of Nurse Practitioners. It is currently the largest full-service
national professional membership organization for nurse practitioners (NPs) of all
specialties. AANP maintains the only national NP database of all practicing NPs.
NPs are authorized to prescribe medications in all fifty states and Washington,
D.C.
The AANP Network for Research (AANPNR) is a practice-based research
network open to all members and allows them to participate in research surveys
Committee on Gynecologic Practice, American College of Obstetricians and Gynecologists. ACOG
Committee Opinion No. 375: Brand versus generic oral contraceptives. Obstet Gynecol. 2007;110 (2 Pt
1):447-8.
Green JB, Ross JS, Jackevicius CA, Shah ND, Krumholz HM. When choosing statin therapy: the case for
generics. JAMA Intern Med. 2013;173(3):229-32.
Drug Benefit News. As Competitors Encroach, Pfizer Seizes A Few More Glory Days With Lipitor Promo.
http://www.silverlink.com/assets/pdfs/silverlinknews/dbn020411.pdf
3

distributed by the AANP. The AANPNR consists of approximately 1400 NPs
from a variety of practices.
The ACP and AANP serve as collaborators in this study under the
1U01FD005485-01 collective agreement. The well-established networks and
infrastructures of the ACP and AANP guarantee the success of this study by
ensuring feasibility and a representative recruitment pool that can be surveyed
with minimal logistical burden.
4. Date(s) to be Conducted:
The survey will be launched and conducted electronically in January, 2017.
5. How the Information is Being Collected:
Recruitment
This survey will utilize ACP and AANP’s preexisting research panel of
participants. Both panels select using stratified random sampling to ensure they
are representative of membership within the US across multiple demographics.
Participants fill out profiles to confirm their eligibility and ensure the panels is
representative of the organization’s membership.
These profiles include
information including member ID numbers, names, gender, member class,
primary specialty, primary practice site, and time spent in patient care. These
profiles are updated by the organizations regularly. Non-members are recruited to
the panel using a similar profile database to be representative of the larger
provider population.
Using their respective database of information on panel participants, ACP and
AANP will work with UChicago to identify eligibility requirements for survey
participation (see attached screening criteria) to ensure participants have adequate
experience with patients and the survey subject matter. The eligibility screening
process is conducted using a preexisting database and bears no burden on
participants.
Eligible participants will receive an invitation email with the survey URL. All
survey communication will be from the respective organizations’ CEO or Director
of the Research Center. Emails will contain the ACP/AANP logo and will be
formatted to match standard survey practices of the respective organization.
One week after the initial invitation email, up to 10 email reminders will be sent
at intervals of 4-21 days to participants who have not yet completed the survey.
The survey will be closed and all contact regarding the survey will end when an
approximate response rate of 50% is reached or when all planned contact has been
completed.

4

Emails to participants will follow the standard format and structure used by the
ACP and AANP (see attached example). Emails will include basic information
about the survey topic, expected length of time to complete the survey, privacy
information, and consent process. As with other AANP and ACP panel surveys,
participants’ submission of the survey is considered implied consent.
Participation is voluntary and participants may choose to not participate at any
time.
Survey
Given the eligibility criteria for the survey, the target sample size for the ACP
survey and AANP survey is 850 participants and 900 participants, respectively
with an approximate goal response rate of 50% based on past surveys
administered by these organizations. Participants will access the survey via a
URL provided in the recruitment emails. The survey will include an introductory
section (see attached example) including the statement, “Submission of this
survey is considered implied consent.”
The survey is divided into two parts- a section of Likert questions focused on
generic skepticism and general perceptions and a vignette section where
participants answer Likert questions in response to a series of proposed
prescribing scenarios.
Individual participants will be randomized to one of four blocks. Each block has
one of four different combinations of [A – WHOM] and [C – PATIENT DRUG
PREFERENCE] that are held constant over two vignettes. Two sets of identical
questions per vignette will vary [B – MESSAGE]. [D – DRUG] will vary
between vignettes. The order of the vignettes and question sets will be
randomized across blocks.
Variable Options
[A – WHOM]
• a 0 FDA – “from/by the FDA”
• a 1 professional societies – “from/by your professional society”
[B – MESSAGE]
• b 0 “equally as effective as”
• b 1 “bioequivalent to”
[C – PATIENT DRUG PREFERENCE]
• c 0 neutral – “has never expressed a preference for brand or generic drugs”
• c 1 brand name preference – “expressed concern that the generic drug will
not work for her”
[D – DRUG]
• d 0 “antidepressants”
• d 1 “oral contraceptives”

5

Vignette Structure
One of your patients comes to your clinic for a medication refill. She is currently
taking a brand name [D – DRUGb]. She has no complaints and is doing well. In
previous visits, the patient [C – PATIENT DRUG PREFERENCEa].
Recently you received a notification from [A – WHOMa]. The message
highlighted the importance of prescribing generic [D – DRUGb] since they are [B
– MESSAGEa] brand name [D – DRUGb].
Exhibit 2. Survey Design
VIGNETTE 1
BLOCK 1 (A 0 C 0 ) a 0 b 0 c 0 d 0 a 0 b 1 c 0 d 0
BLOCK 2 (A 0 C 1 )
BLOCK 3 (A 1 C 0 )
BLOCK 4 (A 1 C 1 )

a0b0 c1d0
a1b0 c0d0
a1b0 c1d0

a0b1 c1d0
a1b1 c0d0
a1b1 c1d0

VIGNETTE 2
a0b0 c0d1 a0b1 c0d1
a0b0 c1d1
a1b0 c0d1
a1b0 c1d1

a0b1 c1d1
a1b1 c0d1
a1b1 c1d1

Survey
Draft: https://qtrial2016q3az1.az1.qualtrics.com/jfe/form/SV_8iVlufkwzVUp
aF7
The ACP and AANP will use its own established survey platforms to survey its
members, thus we anticipate some slight differences in survey formatting. The
ACP utilizes Questback for survey administration, while the AANP utilizes
FluidSurveys or Qualtrics. UChicago has worked extensively with the ACP and
AANP to ensure consistency and equipoise not only between the ACP and AANP
surveys, but also with other surveys administered by these organizations.
6. Confidentiality of Respondents:
UChicago and FDA will comply with safeguards for ensuring participant
information is kept private to the extent permitted by law. All data will be
collected with an assurance that the respondents' answers will remain confidential.
The study instrument will contain a statement that their responses will be kept
confidential.
Personally identifiable information, including membership ID numbers, names,
and emails, are used by ACP and AANP in order to contact participants and recontact non-respondents for this survey. All identifiable data is passwordprotected and stored on a secured platform according to each organization’s
respective data security and confidentiality protocols. Only research staff
working on this project will have access to identifiable data. Once survey
responses are collected by ACP/AANP, a deidentified data set is sent to
UChicago for combined analysis.

6

All results from the study will be reported only in the aggregate. Data will be
kept for six years after the close of the study. At the end of this period, written
documentation will be destroyed according to proper University of Chicago
protocol and channels. Any digital data will be destroyed using commercial
software applications designed to remove all data from the storage device.
7. Amount and Justification for any Proposed Incentive:
Compensation to subjects participating in focus groups is proposed in accordance
with the ACP and the AANP standard practices regarding surveys. To maintain
equipoise with other surveys, physician and nurse practitioner survey respondents
are proposed to receive a $10 Amazon gift card.
The incentive is not a reward or salary. Rather, it is a stimulus to ensure adequate
participation, high-quality data, and that participants are reasonably representative
of the ACP and AANP membership.
8. Questions of a Sensitive Nature:
There will be no questions of a sensitive nature asked of participants.
9. Description of Statistical Methods:
Survey data will be analyzed using descriptive statistics to summarize means,
medians and measures of spread (confidence intervals, standard deviations). In
addition, whether differences in perceptions about generic substitutions exist by
drug classes, or between PCPs and NPs, will be ascertained.
Data will be analyzed in aggregate. We will use basic cross tabs and frequencies
to identify trends and patterns in the data. We will test research hypotheses with
ANOVAs, linear regression, chi-square tests, and correlations, as appropriate for
the type of data (discrete, continuous, binary) and the research question.
We conducted a power analysis to assess whether the expected sample size of 425
will be sufficient to detect the hypothesized differences. Of primary interest is
whether there is a main effect of MESSAGE (effective, bioequivalent) or
PATIENT DRUG PREFERENCE (neutral, brand). The response choices are on a
5-point Likert scale, but the distribution of responses is expected to be skewed.
For this reason, a binary outcome was considered for the power calculations.
Specifically, the outcome was whether a participant answered extremely likely or
likely. For MESSAGE, the hypothesized difference is 60% (effective) versus
40% (bioequivalent), and the analysis would involve a within-participant
comparison. To generate a conservative estimate of power, the smallest possible
sample size of 106 (i.e., the number in each of the four blocks) was used with a
two-sided alpha = 0.05 and gave 91% power assuming a correlation between

7

proportions of 0.25, based on McNemar’s test. Even if the correlation is weaker
(0.1), the power would be 86%. For PATIENT DRUG PREFERENCE, the
hypothesized difference is 50% (neutral) versus 30% (brand), and the analysis
would involve a between-participant comparison. To generate a conservative
estimate of power, the smallest possible sample size of 106 per group (i.e., the
number in each of the four blocks) was used with a two-sided alpha = 0.05 and
gave 85% power based on a Pearson’s chi-squared test. Of note, more powerful
models that would leverage the full sample size of 425 would actually be utilized
for analysis (e.g., standard logistic regression model with robust standard error
estimates or generalized estimating equations (GEE) regression model).
BURDEN HOUR COMPUTATION (Number of responses (X), estimated response or
participation time in minutes (/60) = annual burden hours):
The survey takes approximately 12 minutes to complete.
Type/Category
of Respondent

No. of Respondents

General internist 425
primary
care
physicians (ACP)
Nurse
425
practitioners
(AANP)
Total
850

Participation
Time
(minutes)
12

Burden
(hours)
85

12

85

12

170

REQUESTED APPROVAL DATE: By November 1, 2016 (or earlier, if possible).
Approval later than early November would likely hinder the planned survey deployment
and delay Phases 3 and 4.
NAME OF PRA ANALYST & PROGRAM CONTACT:
Domini Bean
Paperwork Reduction Act Staff
[email protected]
301.796.5733
Lucie Yang, MD, PhD, MBA
Director (Acting), Division of Quality Management Systems
Office of Generic Drugs
301.796.5112
[email protected]
FDA CENTER: Center for Drug Research and Evaluation

8