Increased
Methylphenidate
Usage
for
Attention
Deficit
Disorder
in
the
1990s
Daniel
J.
Safer, MD*; Julie M. Zito, PhD; and Eric M. Fine, MD, MPHABSTRACT. Objective. To estimate the increased use
and the prevalence of methylphenidate (Ritalin)
treat-ment of youth with attention deficit disorder (ADD)
dur-ing the 1990s.
Design. Using time-trend findings from two large
population-based data sources, three pharmaceutical
da-tabases, and one physician audit, a best-fit estimate of
the usage and the usage trends for methylphenidate
treatment over the half decade from 1990 through 1995
was sought.
Setting. Five regions in the United States (US) and the nation as a whole.
Patients. Youths on record as receiving
methylpheni-date for ADD.
Results. The findings from regional and national
da-tabases indicate that on average, there has been a 2.5-fold
increase in the prevalence of methylphenidate treatment
of youths with ADD between 1990 and 1995. In all,
ap-proximately 2.8% (or 1.5 million) of US youths aged 5 to
18 were receiving this medication in mid-1995. The
in-crease in methylphenidate treatment for ADD appears
largely related to an increased duration of treatment;
more girls, adolescents, and inattentive youths on the
medication; and a recently improved public image of this
medication treatment.
Conclusion. The database findings presented serve to
correct exaggerated media claims of a 6-fold expansion of
methylphenidate treatment, although they do not clarify
the issue of the appropriateness of this treatment.
Pediatrics 1996;98:1084-1088; attention deficit disorder with hyperactivity, methylphenidate,
pharmacoepidemi-ology.
ABBREVIATIONS. DEA, Drug Enforcement Administration; US,
United States; ADD, attention deficit disorder; ARCOS,
Auto-mated Reports Consolidated Orders System.
Drug Enforcement Administration (DEA)
produc-tion quotas for methylphenidate (Ritalin) in the
United States (US) increased from I 768 kg in 1990 to
10 410 kg in mid-1995.1 Because approximately 90%
of methylphenidate is prescribed for children and
adolescents with attention deficit disorder (ADD),2’3 this 6-fold rise in production quotas led some
observ-ers recently to conclude that a very profound
in-From the *Departments of Psychiatry and Pediatrics Johns Hopkins
Uni-versity School of Medicine, the Baltimore County Health Department, and
the §Schools of Pharmacy and Medicine, University of Maryland at Balti-more, Baltimore, Maryland.
Received for publication Dec 27, 1995; accepted Mar 19, 1996.
Reprint requests to (D.J.S.) 7702 Dunmanway, Dundalk, MD 21222.
PEDIATRICS (ISSN 0031 4005). Copyright © 1996 by the American
Acad-emy of Pediatrics.
crease in stimulant treatment for youths with ADD
was in process.
No doubt methylphenidate treatment for ADD has
increased substantially since 1990, but this is not a
new development. Stimulant treatment for ADD
youths doubled every 4 to 7 years between 1971 and
l987. Although it declined somewhat between 1987
and 1990 in response to the anti-Ritalin law suit and media blitz campaign,5 it resumed its steady upward climb thereafter. However, the recent increase in the
methylphenidate treatment of ADD youths is not
6-fold, and the increase has not been evenly
distrib-uted to all subcategories of ADD youths receiving
the medication.
The findings presented herein-from a variety of
sources-will hopefully clarify the extent of recent
increases in methylphenidate treatment for ADD and
will also detail the apparent reasons for this.
DESCRIPTION OF THE METHYLPHENIDATE DATA
SOURCES REVIEWED
DEA Production Quotas
The DEA production quotas for prescribed
stimu-lant medications reflect a gross estimate of future use
based on numerous factors, such as Food and Drug
Administration estimates of need, drug inventories on hand, exports, and industry sales expectations.
Data Sources for Population-based Time Trends
The Baltimore County Health Department biennial
survey of public school students receiving
medica-tion for ADD has been systematically performed
since 1971 using a school nurse head count.46 All
school nurses complete a form listing the names of
the names of the students in their school who are
receiving medication for ADD, the name of the
med-ication(s), grade level, gender, special education
sta-tus, duration of treatment, and source of the
pre-scription. All schools have consistently completed
the forms, which in 1995 surveyed a student
popu-lation of 98 335 (of which 72% were white). The
recordings are done in April of odd numbered years, yielding a point prevalence result.
The Maryland Medicaid data are based on
pre-scription reimbursement claims that have been
com-piled annually from 1988 through 1994 for
methyl-phenidate, converted to person-based files and
characterized by age, gender, county, and race. In
1994, there were 110 481 youths aged 5 to 14 enrolled
African-American.3 The totals represent an annual preva-lence rate.
Data Sources for Pharmaceutical Use Time Trends
The Automated Reports Consolidated Orders
Sys-tern (ARCOS) Database of the DEA maintains a
corn-puterized record of the bulk sales of controlled
sub-stances from pharmaceutical manufacturers to
wholesalers and retail chains. It is organized by zip code and state in terms of the weight in grams of the drug. Findings for the states are recorded in grams
per 100 000 persons and are rank-ordered. The
an-nual ARCOS data have been recorded since 1981, but
the latest available data (as of late 1995) were for
1993.
The National Prescription Audit obtained from
IMS America reports annual estimates of new and
refilled outpatient prescriptions dispensed in retail pharmacies. The findings of the systematic sampling are projected to a national figure for each medica-tion.8 The 1994 total represents the most recent avail-able data for methylphenidate.9
The Scott-Levin National Physician’s Drug and
Di-agnosis Audit is based on a 1-d/mo cross-sectional survey of 2400 US physicians in active office-based practice, including 25 varieties of medical specialists.
The data are grouped by age, gender, diagnosis, and
ethnicity. By projections, the data have been used to
estimate the annual prevalence of office visits for
ADD youth who were prescribed methylphenidate
in the US during the calendar years 1990 through
1994.2
The Rhode Island Division of Drug Control, using
duplicate prescriptions for controlled substances
from pharmacies, has reported the total number of
methylphenidate prescriptions in that state from
1988 through 1994. The findings of this database
reported thus far include the annual number of
pre-scriptions and the dosage units prescribed each
year. 1
Data Sources With a Single Assessment of Prevalence
State of Michigan triplicate prescription survey for
the months of February and March 1992 produced
methylphenidate data that were analyzed by
Rapp-ley and colleagues.11 The investigators assessed data for 32 608 youths who received 52 590 prescriptions
for methylphenidate in regard to age, gender, and
county of residence. The medication total resulted in a 2-month prevalence finding.
The New York State Health Department survey
using triplicate prescription data analyzed
methyl-phenidate usage in 62 New York counties by age,
gender, and county. The annual prevalence findings
on children aged 6 to 12 receiving methylphenidate were reported for the study year 1991.12
The Northwest Region Kaiser Permanente health
maintenance organization (HMO) dataset from
which the annual prevalence of methylphenidate
treatment was obtained in 1991 was part of a
par-tially computerized health record of the HMOs
380 000 enrollees. The HMO is located in Oregon and
the state of Washington, and 91 % of its enrollees are white.3
Synthesis of Results From Various Data Sources: The
Increased Use of Methylphenidate for ADD in the
1990s
The DEA Production Quotas
The DEA production quotas for methylphenidate
that show a 6-fold trend are misleading because
these quotas are not based on actual patient usage.
For example, annual production quotas for
methyl-phenidate between 1976 and 1986 did not increase
although strong evidence indicated a substantial
in-crease in its use over that period.4’
Population-based Time Trend Findings
A number of data sources have revealed trends in
methylphenidate treatment in the 1990s far more
precisely than the DEA estimates. Generally, the
composite data have revealed an approximately
2.5-fold increase between 1990 and 1995 in the
preva-lence of youths receiving methylphenidate
prescrip-tions. Zito and colleagues,3 using Maryland
Medicaid data, reported a 2.5-fold increase in the rate
at which enrollees, aged 5 to 14, were treated with
methylphenidate between 1990 and 1994. In
Balti-more County, MD, the rate at which all public school
students were treated with methylphenidate for
ADD increased 1.8-fold between 1991 and 1995
(Table I).
To our knowledge, no other population-based
time trend assessments of methylphenidate
treat-ment are available at this time. However, one physi-cian office visit audit and three time trend pharma-ceutical data base findings have been reported. In the
Scott-Levin National Physician Drug and Diagnosis
Audit, the rate of methylphenidate office visits for
5-TABLE 1. Prevalence of M ethylphenidate Treatment for ADD Based on Population Surveys
Percent of Youth Populations Prescribed Methylphenidate
Year Ages 5-14 Ages_5-17/18
Maryland
Medicaid3
Baltimore County Michigan NW Kaiser
Public Schools* Youths’4 Permanente3
Baltimore County
Public Schools*
Michigan
Youths#{176}
90 91 92 93 94 95
1.9 2.1 2.9 3.4 4.7
2.5 1.1
2.0 3.2
4.6
2.1
2.6
3.7
1.6
1991 1993 1995
N/T* i/ N/T (/( N/T
Elementary lliZ
43880
Middle _4,5.6_ 2.47
18508
Senior ___4__ 0.44
21347
SpeciaH’ _i4Z_ 8.09
1816
Totals iZZ6 2.08
85551
4.86 2.46
45817
3.44
51222
20951
2.73
22354
3.84
22648
0.59
24122
1.04
11.63
637
13.50
2.63 3.75
166
1427 2388
90843
* N/T, Number of students on methyiphenidate for ADD/total public school enrollment.
t Special education schools for the seriously handicapped.
:j: Of the total on medication for ADD, an average of 97% received stimulants and 91 % received methylphenidate.
3687
98335
to I 7-year-old youths increased 2.4-fold between
1990 and 1994.2 The National Prescription Audit of
IMS America reported a 3.2-fold increase in
methyl-phenidate prescriptions dispensed in the US from
1990 through l994, which is the same rate of
in-crease reported by the Rhode Island Duplicate
Pre-scription Program for the years, 1990 through 1994.10
And the ARCOS database that measures
methyl-phenidate distribution to retail registrants recorded a 2-fold increase in the bulk sales of that drug in the US
from 1990 through l993. The medication sales
in-creases suggest a higher utilization trend than that involving individuals treated with methylphenidate, a difference that will be discussed later.
Mt’tliii!plit’iiidatt’ Prt’valt’iict’ Fitidings
The percentage of youths aged 5 to 14 and 5 to 17
or 18 on methylphenidate for ADD is shown for the
population-based databases that we reported (Table
1). Using the Maryland Medicaid data, Zito and
col-leagues3 reported that the percentage of Medicaid
enrollees aged 5 to 14 receiving methylphenidate prescriptions rose from 1.9% in 1990 to 4.7% (5214/
1 10 481) in 1994 (Table 1). The Baltimore County
public school nurses’ headcount of students
medi-cally treated for ADD yielded very similar results. Of the total, 91 % of those receiving medication for ADD
were prescribed methylphenidate. The percentage of
the entire student body aged 5 to 14 receiving meth-ylphenidate rose from 2.5% in 1991 to 3.2% in 1993 to 4.6% in 1995 (Table 1).
The above cited rates are generally consistent with
the two single assessments of prevalence based on
triplicate prescription data. Kaufman12 reported that
the annual prevalence of methylphenidate usage for
youths aged 6 to 12 in 62 counties of New York State
was 2.6% in 1991, and Rappley and colleagues11
re-ported a 2.0% 2-month prevalence rate for Michigan
youths aged 5 to 14 prescribed methylphenidate
dur-ing February and March 1992. Only one moderately
dissimilar finding has been reported; it is the 1.1%
annual prevalence rate of methylphenidate
treat-ment for 5- to 14-year-old enrollees of the Northwest
Region Kaiser Permanente HMO in 1991 . The latter
data may reflect regional differences in treatment
criteria for ADD, patient population differences
(HMO vs Medicaid), and/or distinctive HMO and
general practice patterns.
Overall, there is a fairly consistent pattern of a
sizable rate of increase of methyiphenidate treatment
for ADD youths in the 1990s across databases.
Con-sequently, it seems appropriate to list probable rea-sons for the trend.
A/lore Youths Arc Staying on Ivh’dicatio,i for ADD Into Tiwir Tet’iis
ADD youths are nearly always initiated onto
stim-ulant medication treatment during their early
ele-mentary school years, and recently far more students have been staying on that medication into their
mid-and late secondary school years.6 Whereas the
pro-portion of students receiving stimulant medication
for ADD who were at the secondary school level
(grades 6 to 12) was only I 1 % in 1975, that propor-tion increased to 31 % in 1995 (Table 2).
Another way to illustrate the proportionately
greater increase of methylphenidate treatment for
older ADD children is as follows. At the elementary
school level, the percentage of Baltimore County
public elementary school students who were given
medication for ADD rose 5-fold between 1971 and
1995 (from 1.07% to 5.23%). In middle schools, the
increase was 7-fold from 1975 through 1995 (0.59% to
4.25%), and in senior high schools, the increase was
6-fold from 1983 through 1995 (0.22% to 1.21%).46
When these population trends are compared, the
average annual ADD medication rate increases in
public high schools have been more than twice those
in the public elementary schools.
More Studt’,its Wit/i ADD but Wit/tout Notable Hypcractn’iti/ Are Being Placed on Stimulant Medication
In the mid- and late 1970s, an assessment of
teacher ratings of the classroom behavior of students
referred for ADD treatment revealed that 7% of the
total who were given stimulant medication had been
identified as having attention problems but no
nota-ble degree of hyperactivity. In the mid-1980s, that
inattentive subpopulation increased to 18% of the
total given medication)3 Most learning-disabled
dents have an uncommonly high level of classroom inattentiveness on teacher ratings,14 and in recent years, a sizable number of these inattentive youths have been placed on stimulant medication.13
Increased Use of Medication for Girls With ADD
The proportion of girls on medication for ADD has
increased at all school levels in Baltimore County
over the past 24 years. Recent increases in this regard at the elementary school level have been modest, but
such has not been the case for public secondary
school female ADD students. In 1981, 1983, and 1985,
the female to male ADD medication ratio for middle
school students stood at 1:12; 1:10, and 1:10.6 But in
1991, 1993, and 1995, the female/male medication
gender ratio narrowed to 1:7, 1:6, and 1:5. Even the
Baltimore County high school female/male ADD
medication gender ratio narrowed to 1:5 by 1995.
A Growing Positive Public Image of Medicating Youths for ADD
During the anti-Ritalin media blitz of 1987 to 1990,
many parents of ADD youths decided not to
medi-cate their ADD children fueled by concerns about
possible side effects as reported by the media.
Fur-thermore, a number of physicians became hesitant
about prescribing stimulants for ADD youths
be-cause of the risk of anti-Ritalin law-suits that had
been initiated.5 However, by the mid-1990s all the
anti-Ritalin lawsuits generated in the late 1980s had
failed.15 Furthermore, major magazines and
newspa-pers which had carried anti-Ritalin stories during the media blitz exhibited a far more positive perspective On the subject 5 years later.
Are Most ADD Youths on Stimulant Medication in 1995?
Epidemiologic studies in the late 1980s identified
6% to 7% of youths aged 5 to 14 as having ADD with
hyperactivity.1618 The rate of ADD with hyperactiv-ity peaks at 8% to 9% at ages 6 to 9 and decreases to a 5% average at ages 12 to 1619,20 The rate at which
stimulant medication for ADD was prescribed for
Baltimore County students in 1995 peaked at 6% to
7% from the 2nd to the 5th grades and decreased to
an average of 3% from grades 7 to 10. If one subtracts
the nonhyperactive stimulant-medicated youths
from the total and assumes that the remaining
youths meet the research criteria used to identify
ADD with hyperactivity and that the published
re-search diagnostic rates established elsewhere match
the Baltimore County population, then 50% to 60% of
Baltimore County students with ADD and
hyperactiv-ity received stimulant medication treatment at the
elementary and middle school level in mid-1995.
Obviously, further health services assessments are
needed to provide more precise treatment/disorder prevalence data.
Do Baltimore County Public School and Maryland Medicaid Medication Usage Figures Reflect US Methyl phenidate Usage?
Every data source cited here has limitations
be-cause of its regional and/or selective nature.
Medi-cation surveys indicate that urban youths receive
more stimulant medication than rural youths,#{176} that there are major differences (5-fold at each extreme) in
methyiphenidate cumulative consumption from
state to state,7 that public school students receive
medication for ADD at a far higher rate than do
parochial and private school students,4 and that
chil-dren in less affluent areas receive medication for
ADD at a higher rate than do children residing in
more affluent areas.4 Thus, although Maryland
re-cently ranked 28th and 34th among the 50 states in
the cumulative consumption of methylphenidate
ac-cording to the DEA’s ARCOS database, its per capita
use has consistently been higher than the national
average.7 Also limiting the projection of Maryland’s
Medicaid and its school-based findings to the US as
a whole is the state’s distinct socioeconomic profile, eg, high urban density, numerous pediatric special-ists.
A further concern about generalizations from the
population-based findings presented is that the
ma-jority were based on an annual prevalence rate that
included many youths who had only briefly tried
stimulant medication. This explains the lower
per-centages of youth on methylphenidate based on the
point prevalence findings from Baltimore County
and the 2-month prevalence findings from Michigan
(Table 1).
Nonetheless, one can approximate this country’s
methylphenidate prevalence rate for youths if the
findings on Table I are adjusted for family income,
prescription payment sources, and rural versus
ur-ban and private versus public school differences. The
resultant estimate is that between 3% and 4% of US
youths, aged 5 to 14 and between 2.5% and 3% of
youths aged 5 to 18 were receiving methylphenidate
treatment in mid-1995. If 3.5% of the 38 million US
youths aged 5 to 14 and 2.8% of the 52 million aged
5 to 1821 were receiving methylphenidate, the total
US youth population on this medication in
mid-1995-for each respective age range-would be 1.3
and I.5 million. These projections are twice those
made in 1987. Better data are needed to confirm this estimate.
Do Prescription Totals Relate to the Number of Youths on Met hyl phenidate?
Recently, the National Prescription Audit of IMS
America and the Rhode Island Duplicate
Prescrip-tion Program reported that methylphenidate
pre-scriptions increased 3.2-fold in the US from 1990
through 19949,10 However, these reports should not
cast doubt on the lower (approximately 2-fold)
in-crease of ADD youths receiving methylphenidate
over that time period because across datasets the
increased prescription rate was consistently more
than the increased rate of medicated youths.3’22 The relatively greater increase in prescriptions could re-flect longer utilization, more frequent
administra-tion, and increased managed care restrictions on the
quantity of medication allowed per prescription. If the 5:1 average (from 1990 through 1994)
meth-ylphenidate prescription to medicated individual
ra-tio from the MD Medicaid data3 is applied to the
at Viet Nam:AAP Sponsored on August 30, 2020 www.aappublications.org/news
National Prescription Audit’s estimated prescription
total for methylphenidate in the US in 1994
(7 817 000), then 1.6 million individuals received
methylphenidate that year (including about I .4 mil-lion 5- to 18-year-old youths with ADD).
Is Medication Treatment for ADD Overused at This Time?
This frequently posed question can be clarified
somewhat by a review of recent findings on
methyl-phenidate usage, but the data presented are
insuffi-cient to settle the matter. The appropriate use of
stimulant medication for ADD can be viewed from a
symptomatic or a broader perspective. From the
symptomatic perspective, it is clear that at least 75%
of ADHD youths exhibit a measurable and useful
lessening of behavioral and attention difficulties af-ter stimulant treatment-in clinical trials,23 multi-year research studies24 and in two small naturalistic stud-ies performed in the early 1970s and mid-1980s.13’25 A
broader perspective goes beyond positive symptom
changes to include the following dimensions: school pressures to prescribe, parental attitudes, the exten-sion of stimulant treatment to increasingly include adults and preschoolers as well as learning-disabled
and conduct-disordered youths with ADD with
hy-peractivity, lingering concerns about drug safety and
potential stimulant abuse, some imprecise
commu-nity diagnostic and treatment practices, and the
present lack of research evidence that stimulant
treatment changes the long-term outcome of ADD.
Obtaining a comprehensive picture of community
practice patterns and a multidimensional outcome
perspective on stimulant treatment requires a mix of
experimental and naturalistic studies. Spurring such
research may not appear particularly necessary at
this time, but it will undoubtedly be more important
if methylphenidate usage again doubles before the
year 2000.
REFERENCES
1. Drug Enforcement Administration. Yearly Aggregate Production Quotas.
Washington, DC: Drug Enforcement Administration Office of Public Affairs; 1995
2. Williams L, Lerner M, Swanson J. Prevalence of office visits for ADD:
Gender differences over the past five years (1990 to 1994). Presented at
the National Institute of Mental Health (NIMH) Conference on Gender Differences in ADHD; November 16, 1994; Bethesda, MD
3, Zito JM, Riddle MA, Safer DJ, et al. Pharniacoepidesniology of Youth nit/s
Treatments for Mental Disorders. Psychop/sarniacol Bull. 1995;31:540.
(Abstract)
4. Safer DJ, KragerJM. A survey of medication treatment for hyperactive/
inattentive children. JAMA. 1988;260:2256-2258
5, Safer DJ, Krager JM. Effect of a media blitz and a threatened law suit on stimulant treatment. JAMA. 1 992;268:1004-1007
6. Safer DJ, Krager JM. The increased rate of stimulant treatment for hyperactive/inattentive students in secondary schools. Pediatrics. 1994;
94:462-464
7, Drug Enforcement Administration. ARCOS Database 1986-1993.
Wash-ington, DC: Department of Justice; 1995
8. Strom BL, Morse ML. Use of computerized databases to survey drug
utilization in relation to diagnosis. Acta Med Scand Suppl. 1988;721:13-20 9, Batoosingh KA. Ritalin prescriptions triple over last 4 years. Cliii
Psy-chiatric Next’s. 1995;23:1-2
10. Rhode Island Duplicate Prescription Program, 1990-1994. Presented at the
annual conference of the National Association of State Controlled Sub-stances Authorities; November 2, 1995; Albuquerque, NM
11. Rappley MD, Gardiner JC, Jetton JR. Houang RT. The use of methyl-phenidate in Michigan. Arch Pediatr Adolesc Med. 1995;149:675-679 12. Kaufman G. Methyiphenidate Findings from Next’ York’s Triplicate
Prescrip-fiats Data. Presented at the annual conference of the National
Associa-tion of State Controlled Substances Authorities; November 2, 1995; Albuquerque, NM
13. Safer DJ, Krager JM. Hyperactivity and inattentiveness: school assess-ment of stimulant treatment. Cliii Pediatr. 1989;28:216-221
14. Rutter M, Tizard J, Whitmore K. Education, Health and Be/savior. New
York, NY: Wiley; 1970
15. Safer DJ. The impact of recent lawsuits on methylphenidate sales. Clin
Pediatr. 1994;33:166-169
16. Anderson JC, Williams S. McGee R, Silva P. DSM III disorders in
pre-adolescent children. Arc/s Gen Psyc/siatrs,i. 1987;44:69-76
17. Bird HR. Canico G, Stipec M. Estimates of the prevalence of childhood
maladjustment in a community survey in Puerto Rico. Arc/i Gets
Psychi-atri/. 1988;45:1120-1126
18. Offord DR. Boyle MH, Szatmari P. et al. Ontario child health study. Arc/i Gets Psychiatry. 1987;44:832-836
19. August GJ, Realmuto GM, Crosby RD. MacDonald AW. Community-based multiple-gate screening for children at risk for conduct disorder.
IAbnorin C/iild Psycho!. 1995;23:521-544
20. Szatmari P. Offord DR. Boyle MH. Ontario child health study: preva-lence of attention deficit disorder with hyperactivity. I C/iild Psyc/sol Psyc/iiatri,’. 1989;30:219-230
21. US Bureau of the Census. United States Population Estimates by Age. Sex
and Race 1990-1994. Washington, DC: US Bureau of the Census:
Popu-lation Division (PPL-21); 1995
22. Swanson JM, Lerner M, Williams L. More frequent diagnosis of
atten-tion-deficit hyperactivity disorder. N Engl JMed. 1995;333:944
23. Barkley R. A review of stimulant drug research with hyperactive
chil-dren. JC/dId Psyc/sol Psychiatry. 1977;18:137-165
24. Hechtman L, Abikoff H. Multi-modal Treatment Plus Stimulants x’s.
Stissi-ulant Treatment in ADHD Children. Presented at the 42nd Annual
Meet-ing of the American Academy of Child and Adolescent Psychiatry; October 20, 1995; New Orleans, LA
25. Sleator E, Von Neumann A, Sprague R. Hyperactive children: a
continuous long term, placebo controlled follow-up. JAMA. 1974;229:
