Transcript 48x36 Poster Template

A Novel Method of Identifying and Studying Children Receiving
Psychotropic Medications
Christine Wolf, Robert Hilt, Kent Koprowicz, Mary Chandler, Mark Gunther, Matthew Russell. Axio Research, Seattle, Washington.
Abstract
Psychotropic medications are increasingly prescribed to children with mental health disorders
despite the absence of prospective data confirming safety and efficacy in the pediatric population
(1,2,3). Factors impeding prospective studies in children include high costs, difficulty recruiting
children into randomized controlled trials, and the challenge of enrolling study participants who
are representative of children seen in community practice settings.
In cooperation with Costco Wholesale Corporation’s national pharmacy chain, parents of children
who were recently prescribed psychotropic medications were identified and recruited for a Web
based survey that included demographic information, drug history and assessments of diagnoses,
side effects and efficacy. Survey completers were solicited for identical surveys 6 and 18 months
later. Our primary goal was to assess the feasibility of pharmacy generated invitations targeting
parents as a method of collecting observational information about children with psychiatric
disorders. Secondary goals were to analyze the prevalence of specific drug classes and of multiple
simultaneous psychiatric medications.
Of the 1500 targeted households, 16% completed an initial survey and 9% completed a 6-month
follow-up survey. Among parents logging in, 88% completed the survey. Completers were
demographically similar to the Costco customer population.
A pharmacy-generated invitation can effectively recruit parents of children using psychotropic
medications, produces a reasonable response rate despite inclusion of a lengthy instrument, and
can identify a representative sample drawn from a community setting.
Methods (cont.)
Informed consent and patient privacy protection
The 1,500 invitational letter recipients were identified by Costco’s IT department. Letters were sent
out by Costco’s mailing house. Identities of letter recipients were not disclosed to study
investigators.
The invitational letter stated only that the research studied medications in children. Upon logging
onto the study Web site by using a unique code included in the letter, the reader viewed a list of
psychotropic medications and was invited to complete the survey if their child was taking one or
more of them. Excluding the drugs of interest from the invitational letter implied that parents of
children taking any type of drug were the target audience, preventing any unintended letter
recipient from learning about a child’s medication history.
The first survey page was an informed consent. Clicking a box indicating agreement brought up
the survey questions. Upon completing the survey, participants had the option to provide e-mail
and postal addresses. Participant contact information was stored unlinked from survey responses
and could be viewed only by the database administrator and a study administrator who answered
calls and e-mails and sent Costco gift cards to study completers. Study investigators could not
view contact information of study participants.
The invitational letter, study methods, and online consent document were reviewed and approved
by the Seattle Children’s Hospital IRB.
Objectives
•
•
•
To determine whether parents of children prescribed psychotropic medications will respond
to a pharmacy-generated invitation to provide information about their children via the Web
To determine whether participants will complete a lengthy survey
To determine whether participants will provide contact information allowing longitudinal
follow up
Results (cont.)
• 82% of Costco gift cards awarded to participants were redeemed
• 97% of completers provided voluntary contact information
• Consistent with the Costco customer profile, participants were 88% white, 5% Hispanic, 3%
Asian, 1% Black, 1% American Indian/Alaskan Native, and 2% declined to answer.
• 68% of respondent’s answered questions about a child on one of our medications of interest,
19% answered questions for a child on 2 medications and 13% answered questions for a child
on 3 or more medications.
The study administrator received numerous
communications
from
parents
expressing
appreciation for our work and gratitude for
having an opportunity to help. In response to the
initial letter with the incorrect login code, several
parents called or e-mailed the administrator to ask
for help in solving technical challenges. Survey
responses continued for several months after both
the initial letter and after the 6-month e-mail
invitation to repeat the survey, from individuals
who had clearly saved the letter and login code for
that period of time. An 18-month follow-up
survey is ongoing.
Limitations
Participants in this study were volunteers, and may not represent the population of all parents of
children using psychotropic medications. Volunteered information, especially that which is private
and may evoke uncomfortable emotions, may be less accurate than information collected in a
prospective study. The socioeconomic status of Costco members is slightly higher than the US
population, which may influence study findings.
Disclosure
CW- None, RH- None, KK- None, MC- None, MG- None, MR- None
Methods
Conclusions
Target population
Parents of the 1,500 children aged 6-18 who most recently filled a prescription for a psychotropic
medication at a Costco pharmacy in WA, OR or CA.
Broad description of study methods
An invitational letter was sent to 1,500 parents of children who had filled a prescription for
psychotropic medication within the previous 6 months. The letter provided instructions for
accessing the study Web site, where parents could complete a survey about their child’s problems
and experiences with the medications.
Letter recipients logged onto the Web site using a unique code included in the letter and selected a
password. The password and code allowed re-access so that participants could complete the
survey in more than one sitting. Participants who completed the survey and voluntarily provided
contact information received a $25 Costco gift card and were invited to repeat identical surveys at 6
and 18 months.
Content of the survey
• Demographic information: child’s age, grade in school, family income, participant type (mother,
father, or legal guardian)
• Psychotropic medications used by the child: drug name, duration of use, type of prescribing
provider, parent perceived diagnoses, side effects and effectiveness
• Child Behavior Checklist for Ages 6-18 (CBCL, ASEBA©): a validated instrument for assessment
of children with psychiatric disorders, the CBCL includes 140 Likert scale items evaluating
aggression, hyperactivity, bullying, conduct problems, defiance and violence
Mailing Error
An invitational letter was initially sent to 1,000 households. The study database administrator
provided Costco with a list of 1,000 unique login codes to be included in the letters. Due to a
miscommunication, the mailing house printed the same login code on the 1,000 letters. The study’s
Web site security design linked login codes with participant selected passwords. Therefore, the
first parent to login completed the survey, but subsequent parents who attempted to login could
not do so because the login code in their letters was already associated with the password selected
by the first participant. Study investigators became aware of the problem when the study
administrator received several communications from letter recipients who were unable to log in.
To remedy the problem, Costco sent the 1,000 letter recipients an IRB-approved letter of apology
and a new login code. The corrected letters were mailed within three weeks of the original letter.
Since the primary objective of the study was to assess willingness to participate, the inconvenience
introduced by the mailing error could have affected the participation rate. Therefore, the database
administrator generated 500 additional login codes and Costco sent letters to 500 more customers
who qualified for the study.
Results
The overall survey completion rate was 16%.
Among the original 1,000 homes receiving the
apology and corrected letter, 13% completed the
survey. Among the 500 homes with no mailing
error, the response rate was 23%. Other key
findings related to response rates were:
• 88% of parents initiating the survey also
completed it, despite the length of > 30
minutes
• 55% of those completers also completed an
identical follow-up survey 6 months later
after a single e-mailed reminder
The 16% response rate (23% among participants not exposed to the mailing error) was a good
response rate for a Web survey, especially one that required 30 minutes to complete; 24% has been
reported as an expected response rate for Web surveys (4).
We have stated only brief results here, since our focus was to demonstrate the feasibility of this
novel method of collecting information in the pediatric psychiatric population. Our success
demonstrates that a Web survey, even one that is lengthy and seeks sensitive information, is a
feasible method of collecting parent-reported data.
We were surprised at the level of interest and desire to participate suggested by the
communications received by the study administrator. Several participants thanked us for studying
the problem. Letters recipients who confronted initial difficulties logging in called saying they
wished to participate because the subject of the study was important to them. The survey content
itself may have stimulated participation. We believe that for participants, the desire to help their
child or other children was a stronger motivator than the $25 gift card.
The survey results provide rare insight into how psychotropic medications are actually used
outside a controlled environment and into the prevalence of multiple medication use. Future plans
include expanding the survey to the customer base of another national pharmacy chain, and using
observational data generated from the survey to formulate questions for prospective research.
References
1.
2.
3.
4.
Raghavan R, Zima BT, Andersen RM, Leibowitz AA, Schuster MA, Landsverk J. Psychotropic medication use in a
national probability sample of children in the child welfare system. J Child Adolesc Psychopharmacol. Feb
2005;15(1):97-106.
Roberts R, Rodriguez W, Murphy D, Crescenzi T. Pediatric drug labeling: improving the safety and efficacy of
pediatric therapies. JAMA. Aug 20 2003;290(7):905-911.
McClellan JM, Werry JS. Evidence-based treatments in child and adolescent psychiatry: an inventory. J Am Acad
Child Adolesc Psychiatry. Dec 2003;42(12):1388-1400.
Leong FTL, J.T. A. The Psychology Research Handbook: A Guide for Graduate Students and Research Assistants.
2005.
Acknowledgements
We wish to acknowledge Seattle Children’s Hospital, Costco Wholesale
Corp. and ASEBA.