Mirror, Mirror on the Wall
Are We Prescribing the Right Psychotropic
Medications to the Right Children Using the Right Treatment Plan?
Michael
S. Jellinek, MD
DURING THE period (1987-1996) reported by Zito et al,
1
researchers from the field of child and adolescent psychiatry and pediatrics
conducted studies to refine diagnostic criteria, applied these criteria to
determine the prevalence of child psychiatric disorders, and provided the
infrastructure for treatment studies. The research also led to greater
standardization of an "evaluation" that facilitated an expectation that a
formal diagnosis and treatment plan could be achieved in 1 or 2 hours. New
medications emerged that were studied in adults but used widely for children
as well. This clinical evolution was taking place in the context of dramatic
changes in the health care delivery system secondary to managed care cost
controls being applied for both general pediatric care and child mental
health services.2 The data
reported by Zito et al1 are a reflection, an
imperfect mirror, of the scientific, clinical, financial, and systems
changes that affected the care of children between 1987 and 1996 and
continue to be a substantial influence. What can we see in this mirror? How
could we understand this data set from a clinical perspective? From a
financial and systems perspective? And finally, what are our opportunities
and obligations to assure that we are providing comprehensive and high
quality treatment to the increasing number of children who are receiving
psychotropic medication?
CLINICAL PERSPECTIVE
The data indicate that 2 to 3 times more children were receiving
psychotropic medication of all types in 1996 than in 1987, with the most
rapid increase since 1991. In general terms, 6% of the children had been
prescribed at least 1 psychiatric medication in the final study year. Can we
rationalize such a high percentage and an accelerating trend? For example,
one could hope that the extensive efforts to define psychiatric disorders
such as attention-deficit/hyperactivity disorder (ADHD), the many studies
noting the response of ADHD to medication, the educational efforts in
pediatrics and child psychiatry, and the availability of longer-acting
stimulants all had an appropriate impact on practice. If true, the treatment
of ADHD could justify half of the overall psychotropic use. Given the age
range in the sample, the increasing recognition of adolescent depression,
and the efficacy/side effects profile of the selective serotonin reuptake
inhibitors, one can rationalize another share of the 6%. Finally, one could
add approximately 1% for severe disorders such as schizophrenia and bipolar
disorders (an increasingly popular diagnosis). The data reflect somewhat
higher utilization rates for younger ages and possibly more serious
disorders in Medicaid populations, which could support the hypothesis that
mental illness in parents and the stressors of poverty add to the prevalence
of child and adolescent psychiatric disorders in this subsample. An
optimistic clinical perspective suggests that these data reflect a broader
awareness of diagnostic criteria, less exclusive reliance on "talking
therapy," and thoughtful efforts to extrapolate "off-label" from adult-based
evidence to treat children and adolescents with serious mental health needs.
Yet there are some disturbing clinical trends that limit optimism. The
research supporting the treatment of prepubertal and early-adolescent
depression or anxiety in the period of the study was largely nonexistent.
Why did the use of tricyclic antidepressants stay constant despite the
availability and use of selective serotonin reuptake inhibitors and
longer-acting stimulants? Were these tricyclic antidepressants used for
second-line treatment of ADHD? The percentage of children and adolescents
treated with benzodiazepines was substantial (a 4-fold increase in the study
period), and there seemed to be a bias to diagnosing anxiety in children
served through the health maintenance organization. One worries about the
risks of using medications that are potentially sedating in an age group
whose highest morbidity and mortality are secondary to motor vehicle
accidents and whose trajectory on substance use is unknown. The use of mood
stabilizers also increased dramatically, and, though it is hard to tell from
the data, the trends suggest the use of multiple medications in combination
for the treatment of ADHD, depression, and bipolar disorder, all based on
clinical judgment with little research support.
In terms of age-related clinical concerns, stimulants were used with
equal frequency for the 5- to 9-year-old and the 10- to 14-year-old age
ranges, which means few had a change in the need for stimulants, somewhat of
a surprise, or any changes were matched by those newly diagnosed after age
10 years. In terms of gender-related concerns, boys aged 10 to 14 years were
treated for depression proportionately more than girls. Last, in the
"Comment" section, the authors give somewhat disturbing data showing
substantial differences in use by area of the country and on the basis of
reimbursement (which parallels many other studies reporting regional
differences in medical utilization that are hard to explain).
Zito and coauthors1 anticipate well the
likely frustration of the clinical readership. We do not know if the
diagnoses are valid. Who made the diagnosis? We do not know the dosage of
medication, length of treatment, adherence, or outcome of treatment. We do
not know about the use of multiple medications. We do not know the context
of treatment. Did the plan include counseling, therapy, work with the
family, or school-based interventions?
FINANCIAL PERSPECTIVE
Although the impact of managed care controls and "carve-outs" (mental
health benefits subcontracted out to for-profit behavioral health care
companies) is hard to characterize in this data set, the use of medications
increased steeply at a time when for-profit behavioral managed care
companies expanded their influence.3,
4 Under these plans, psychotherapy is tightly
managed by requiring frequent reauthorizations or limiting the number of
therapists available. Medication visits are reimbursed at twice or more the
rate per minute than therapy time, and less rigorous authorization is
required. In many for-profit behavioral health carve-outs, the cost of the
medication itself is charged to the medical, not behavioral, budget. Thus,
the incentives are aligned to use medication. The health plan has lower
psychiatric costs as care is carved out and managed, the behavioral health
plan makes a profit by limiting inpatient stays and therapy, and the mental
health providers maintain or increase their income by seeing several
patients for medication visits an hour. Since 1986, for-profit managed
behavioral health care has spread to Medicaid programs, and there have been
several contract cycles with states being tempted to take lower and lower
bids. A possible consequence is that poverty populations with intrinsic
higher prevalence rates and fewer resources for "out-of-plan" options may
have an even higher use of psychotropic medications.
In addition, the financial factor is the role of pharmaceutical
companies, which see the opportunity of a large market for psychotropic
medications, primarily in adults, but also in children. Sponsoring more
medication trials, vigorous marketing at professional meetings, detailing
pediatricians, neurologists, and child psychiatrists, and, most recently,
direct-to-consumer advertising, all contribute to increasing use of
psychotropic medication.
SYSTEMS PERSPECTIVES
The clinical and financial factors that influence the delivery of mental
health services for children work in synergy.5
Pediatricians, neurologists, and child psychiatrists are facing increasing
patient and family demand; there are fewer options other than medications;
reimbursement favors the use of medications; and pharmaceutical marketing
efforts support (and track) their medication decisions. The shortage,
geographical distribution, and difficulty accessing child psychiatrists add
further pressure on the pediatrician to prescribe psychotropic medication.
No single decision or participant in this system was or is evil. The
researchers and clinicians improving diagnostic criteria have genuinely
advanced the field. The pharmaceutical industry did develop new medications,
market them, and return a profit to investors. Those organizations that
manage medical care were under pressure to limit the rising costs of health
care and saw carving out behavioral health care as a cost-saving measure
consistent with their mission of utilization management. The behavioral
health companies and their investors saw a business model emerging that
could reduce utilization (that had been bloated by an increase in for-profit
marketing of child and adolescent psychiatric beds),3
save some dollars for their customer (the health maintenance organization or
Medicaid), and return a profit to their investors. Unfortunately, the
current set of influences have few, if any, checks and balances.
NEXT STEPS
Given the scale of current psychotropic medication utilization, we have a
responsibility to know what we are doing and the quality of our efforts. We
need to ask the mirror on the wall this question: Are we prescribing the
right psychotropic medications to the right children using the right
treatment plan? Specifically, the work of Zito and colleagues1
reinforces our obligation to apply a quality assurance and research
framework to our current pattern of utilization:
1. Major third party providers and their contracted behavioral health
care companies should report utilization data on an annual basis. How many
children are being prescribed psychotropic medications, for which diagnosis,
and at what age? Who is making these mental health diagnoses and prescribing
the medication? How many mental health visits does the plan provide for
patients receiving psychotropic medication? How many visits are for
medication? How many for other treatments? This is a partial list of
high-level data that should be available from any health care or corporate
computing system and is likely already known on a quarterly basis for
budgeting purposes. Agreeing on a reporting format should be relatively
easy. Having this data available would immediately update the study by Zito
and coauthors, highlight regional differences, and give direction for more
specific quality assurance activities.
2. Based on this regional and national data set, health maintenance
organizations should be asked to initiate, using their internal resources,
one major quality assurance study each year that focuses on children's
mental health. For example, such a study could include a review of
adolescent boys being treated for depression and, for those using
medication, the dosage ranges, duration, and adherence to treatment. This
quality assurance study could include a comparison of the plan's current
practice to pathways established by the Academy of Pediatrics or the Academy
of Child and Adolescent Psychiatry, including an assessment of communication
with schools, work with families, and evaluation of the child's functioning
as an outcome measure of treatment.
3. In terms of a national research agenda, the study by Zito and
coauthors documents an enormous allocation of resources to the use of
psychotropic medications for children and the exposure of millions of
developing children to the effects of these medications. There is a clear,
ongoing need for health services studies that accurately capture the full
cost of child mental health disorders across the health, mental health,
school, social service, and juvenile justice systems. Shifting costs to the
pharmacy budget or the local school budget or to a court clinic makes little
overall sense, and the lack of coordination is a hidden administrative and
human cost. We also must support basic research on the biological impact of
psychotropic medications on the developing child. As Leckman and Young6
recently stated, "Despite this promise [of benefit], the widespread use of
selective serotonin reuptake inhibitors to treat prepubertal children with
depression or anxiety may be associated with untoward outcomes for some
children. We may ignore the long term effects of these agents on [central
nervous system] development at our peril." I am concerned about what we do
not know regarding individual medications and combinations of medications.
The article by Zito et al1 has given us a
mirror, albeit imperfect, of the increasing use of psychotropic medication
in children and adolescents. We need to have a more complete view in an
ongoing, up-to-date time frame. We need a health plan–by–health plan quality
assurance effort and a health services and basic research program. We
undertake such activities in many areas of medical care, more for adults
than for children. Will we value our children sufficiently to ask if are we
prescribing the right psychotropic medications to the right children using
the right treatment plan?
Author/Article Information
Michael S. Jellinek, MD
Department of Psychiatry
Massachusetts General Hospital
55 Fruit St–Bulfinch 351
Boston, MA 02114
(e-mail: mjellinek@partners.org)
REFERENCES
1.
Zito JM, Safer DJ, dos Reis S, et al.
Psychotropic practice patterns for youth: a 10-year perspective.
Arch Pediatr Adolesc Med.
2003;157:17-25.
ABSTRACT | FULL
TEXT | PDF
| MEDLINE
2.
McArthur JH, Moore FD.
The two cultures and the health care revolution.
JAMA.
1997;277:985-989.
MEDLINE
3.
Jellinek MS, Nurcombe B.
Two wrongs don't make a right.
JAMA.
1993;270:1737-1739.
MEDLINE
4.
Jellinek MS, Little M.
Supporting child psychiatric services using current managed care approaches:
you can't get there from here.
Arch Pediatr Adolesc Med.
1998;152:321-326.
ABSTRACT | FULL
TEXT | PDF
| MEDLINE
5.
Jellinek MS.
Changes in the practice of child and adolescent psychiatry: are our patients
better served?
J Am Acad Child Adolesc Psychiatry.
1999;38:115-117.
MEDLINE
6.
Leckman JF, Young C.
Nonadrenergic and serotonergic neuroendocrine responses in prepubertal,
peripubertal, and postpubertal rats pretreated with desipramine and
sertraline.
J Am Acad Child Adolesc Psychiatry.
2002;41:1007-1009.
