Risk Factors in Preventable Adverse Drug Events in Pediatric Outpatients
STEPHANIE O. ZANDIEH, MD, MS, DONALD A. GOLDMANN, MD, CAROL A. KEOHANE, RN, BSN, CATHERINE YOON, MS,
DAVID W. BATES, MD, MSC, AND RAINU KAUSHAL, MD, MPH
ities in children who experienced an adverse drug event (ADE) in the ambulatory setting.
We conducted a prospective cohort study of pediatric patients <21 years seen during 2-month study periods
from July 2002 to April 2003 at 6 office practices in Boston. The primary outcome measure was ADEs. Descriptive analysis of
patient characteristics and types of ADEs experienced was followed by multivariate analysis to determine risk factors associated
with presence of a preventable ADE.
A total of 1689 patients receiving 2155 prescriptions were analyzed via a survey and chart review. Overall, 242
children (14%) experienced an ADE, of which 55 (23%) had a preventable ADE and 186 (77%) had a non-preventable ADE.
In multivariate analysis, children with multiple prescriptions (odds ratio, 1.46; 95% CI, 1.01-2.11) were at increased risk of
having a preventable ADE, controlling for parental education, racial/ethnic, English proficiency, practice type, and duration of
Children with multiple prescriptions are at increased risk of having a
preventable ADE. Further attention should be directed toward improved communication
among healthcare providers and patients. (J Pediatr 2008;152:225-31)
cause of adverse events.3Adverse events caused by medication errors occur in about 1% of
pediatric hospitalizations4; less is known about the ambulatory care setting.
Medication errors are defined as errors in any step of the medication process, and
each step is particularly complex and error-prone for children. For example, adult
medications typically have 1 standard dose with 1 type of preparation (ie, tablet or
capsule). In contrast, pediatric medications involve weight-based dosing requiring calcu-
lations and often multiple preparations (ie, infant drops, elixir, chewable tablets, and
capsules). At the administration and monitoring stages, young children cannot reliably
self-administer or monitor medication use. Moreover, younger children lack the commu-
nication skills to tell parents about adverse effects, and an extra step is involved even for
older children. The pivotal role of parents as intermediaries between the prescribing
physician may be affected by parental linguistic, socioeconomic, or educational challenges
and cultural attitudes and belief.5A crucial step toward prevention of medication errors is
identifying risk factors.
Health care disparities represent a pervasive national problem, affecting health care
access, use, and clinical outcomes. Race and ethnicity have been shown to be intimately
tied to characteristics such as socioeconomic status (SES), cultural factors such as limited
English proficiency (LEP),5,6and the child’s health needs such as perceived health status
and chronic illnesses.7Black and Hispanic children are less likely than white children to
have a regular source of medical care. Furthermore, non-English speaking families are less
likely than English-speaking families to report a regular source of medical care for their
children.8-10Among hospitalized pediatric patients, those whose parents had LEP had a
To determine whether there are racial/ethnic, socioeconomic, parental linguistic, or parental educational dispar-
wo reports from the Institute of Medicine have called attention to the need to
improve patient safety and the substantial racial and ethnic disparities in health-
care.1,2Medications are the most common medical therapy and the most frequent
Adverse drug events
See editorial p 153, and
related articles, p 214
and p 219
From the Department of Pediatrics, Weill
Cornell Medical College and Komansky
Center for Children’s Health at New York
Presbyterian Hospital, New York, New
York (S.Z., R.K.); Department of Medicine,
Children’s Hospital Boston, Boston, Massa-
chusetts (D.G.); Institute for Healthcare
Improvement, Cambridge, Massachusetts
(D.G.); Harvard Medical School, Boston,
Massachusetts (D.G., D.B., R.K.); Division of
General Internal Medicine, Brigham and
Women’s Hospital, Boston, Massachusetts
(C.K., C.Y., D.B., R.K.); Information Systems,
Partners Healthcare System, Boston, Mas-
sachusetts (D.B.); and Department of Pub-
lic Health, Weill Cornell Medical College,
New York, New York (R.K.).
Supported by the Agency for Healthcare
research and Quality (P01-HS11534) and
the Commonwealth Fund. Dr Zandieh is
the recipient of APA (Ambulatory Pediatric
Association) young investigator award.
Submitted for publication Jun 26, 2007; last
revision received Aug 24, 2007; accepted
Sep 28, 2007.
Reprint requests: Dr Stephanie Zandieh,
Division of Pediatrics, Weill Cornell Medical
College, 525 E 68th St, Box 139, New
York, NY 10021. E-mail: stz2001@med.
0022-3476/$ - see front matter
Copyright © 2008 Mosby Inc. All rights
2-fold increase in risk for an unintended or potentially adverse
outcome.11Earlier literature has also found that individuals
who have LEP also face significant health literacy chal-
lenges.12Individuals with low functional health literacy often
are unable to read and correctly interpret instructions on
To date, relatively little research has examined whether
there are increased risks of adverse drug events for vulnerable
children in the ambulatory setting, where children receive
most medical care.16Therefore, we undertook this study to
examine the rates and types of preventable ADEs and deter-
mine whether there are racial/ethnic, socioeconomic, linguis-
tic, or educational disparities for children who experience an
ADE in the pediatric ambulatory care setting.
Medication errors can be categorized as errors with little
potential for harm (medication errors), errors with significant
potential for harm (near misses), and errors with actual harm
(preventable ADE).4,17One example of a preventable ADE
would be delay in giving a child a necessary antibiotic for a
serious infection, resulting in worsening symptoms requiring
a visit to the emergency department for intravenous antibiot-
ics. This complication could have been prevented had the
parents given the antibiotic appropriately. ADEs not associ-
ated with errors are considered non-preventable ADEs. An
example of a non-preventable ADE occurs when parents give
the appropriate doses of antibiotics for a bacterial infection
and their child develops diarrhea, which represents a well-
known adverse effect of antibiotics.
Population and Setting
The study population was a prospective cohort of pa-
tients ?21 years old who were seen from July 2002 to April
2003, in Boston, as previously described.18Six diverse practice
sites were sampled: 2 associated with a teaching hospital, 2
urban health clinics, and 2 suburban practices. Children ?21
years old who had an office visit during the study period,
received at least 1 prescription, and spoke English, Spanish,
or Cambodian were eligible. Patients seen multiple times
during the study period were included only once. All prescrip-
tions for oral contraceptives and possible treatment for sexu-
ally transmitted diseases were excluded because of concerns
about patient privacy. Prescriptions for equipment and for-
mula were also excluded. Furthermore, we excluded patients
when they opted out of the study, did not meet study criteria,
were lost to follow-up, or had incomplete data. This study
received human subjects’ approval by Weill Cornell Medical
College of Cornell University and Brigham and Women’s
We collected data from practice data sources (review of
duplicate prescriptions, chart review, and administrative data)
and telephone surveys. At the time of the patient’s clinic visit,
families were asked to participate in the study. Ten to 14 days
after that visit, the patient’s parents (or main caregivers for the
child) were telephoned and asked to read the labels of all
prescription bottles and report about any new symptoms that
might be related to a drug. When any symptoms were re-
ported, further structured questions were asked about timing
and any actions the parent took (eg, stopping the medication,
telling their child’s pediatrician). Then, approximately 3
months after the index visit, a research nurse examined the
child’s medical chart to confirm ADEs or medication errors
detected on telephone interview, discover previously undetec-
ted ADEs, and collect additional patient data regarding co-
morbidities, disease severity, medication regimens, and pre-
The primary outcome measure was the presence of a
preventable ADE, defined as actual harm from medication
use. All ADEs identified by a trained research nurse were
presented to a pair of physician reviewers who independently
categorized these events into medication errors, near misses,
and ADEs. ADEs were identified and classified by using an
established and validated protocol used in several earlier stud-
through discussion and consensus. Each ADE was further
defined as either preventable (associated with a medication
error) or non-preventable (not associated with a medication
error), and given a severity rating. The kappa statistics for
inter-rater reliability were 0.89 for classification of event, 0.75
for severity of event, and 0.95 for preventability of event.
Our primary predictor variables were gathered from
telephone surveys. These included: race/ethnicity, reported
annual family income, parental educational attainment, and
parental self-reported English proficiency. Race/ethnicity was
obtained with the question, “Which of the following describes
your child’s race?” (white, black, Hispanic, or other).21An-
nual reported family income was obtained with the question,
“What is your average yearly income?” (?$30,000, $30,000-
49,999, $50,000-79,999, and ?$80,000). Parental educa-
tional attainment was obtained with the question, “What is
the highest level of education you have completed?” (less than
high school, high school graduate, some college, and college
graduate/post-graduate). Self-reported English proficiency
was determined with the question, “How well do you think
you speak English?” (very well, well, poorly, and not at all).5,22
English proficiency was defined as “proficient” when the
response was very well, and “LEP” combined responses well,
poorly, and not at all.
Independent variables of interest were gathered from
telephone survey and chart review. These included the child’s
sociodemographic characteristics (age, sex, health status, pres-
ence of a chronic disease, and Medicaid status), the family’s
226Zandieh et alThe Journal of Pediatrics • February 2008
poverty status (above or below the 2003 Department of
Health and Human Services poverty guidelines), health care
access (practice site and continuity of care), and medication
regimen complexity (number of prescriptions at index visit).
Health status was obtained with the question, “In general how
would you rate your child’s health at the present time?”
(excellent, very good, good, poor, and fair).23Presence of a
chronic disease was determined with the question, “Does your
child have a chronic or long-term health condition (a condi-
tion lasting longer than 3 months)?” (yes/no). Continuity of
care was determined through chart review, in which the
duration of continuous care at the office practice was re-
corded. Poverty status was calculated by taking a family’s
annual income and number of family members, comparing
them with the 2003 poverty guidelines issued by the Depart-
ment of Health and Human Services,24and then classifying
families in 2 groups, either less than the Federal Poverty Level
or higher than the Federal Poverty Level.
We analyzed all patients who participated in the 10-day
survey and had a chart review, particularly because parental
report is a critical method for detection of preventable ADEs.
Five percent of patients who had the initial survey did not
have a chart review because of the lack of the medical chart.
We report rates of ADEs and preventable ADEs per 100
patients. Descriptive statistics were used to examine child and
parental characteristics associated with the occurrence of
ADEs and preventable ADEs. The associations of indepen-
dent variables with each outcome variable were assessed with
univariate analysis with the ?2or Fisher exact (categorical
data) or Student t (continuous data) tests (P ? .05 was
considered significant). SAS software (SAS Institute, Cary,
NC) was used for statistical analysis.
Next, we assessed for correlation between our predictor
variables and found all to be statistically correlated. However,
the highest correlation was found between annual income and
parental education (r ? .67, P ? .001).To reduce collinearity,
in multivariate analysis, we included only parental education
and not annual income in our final model.
Our multivariate analyses compared children who ex-
perienced a preventable ADE with children without any
ADE. Through this analysis, we identified among all children
receiving a medication prescription risk factors (patient/fam-
ily, medication, healthcare system) associated with a prevent-
able ADE. Logistic regression with a forward stepwise ap-
proach was performed; variables with a P value ?.05 and
variables that were part of our a priori hypothesis were re-
tained. The final model included these independent variables:
race/ethnicity, parental educational attainment, parental En-
glish proficiency, continuity of care, practice type, and num-
ber of prescriptions. We also performed a similar analysis that
assumed an over-dispersion of preventable ADEs per person
relative to a Poisson distribution and found no significant
difference in results, thus we report only our logistic analyses.
During the study period, 21,209 visits were made by
13,919 patients, 3838 of whom received a prescription. Of
these, we studied 1689 patients who had both completed the
10-day survey and had a chart review (Table I).
Table I. Demographics of patients eligible for initial survey
Non-respondents of initial
survey (n ? 1801)
Respondents of initial survey, 10 day follow-up,
and chart review (n ? 1689)
912 (51%)835 (50%) .53
Because of rounding, some categories may not total 100%.
There were 92 participants who had a 10-day follow-up, but not a chart review, and therefore were not included in this analysis.
Risk Factors in Preventable Adverse Drug Events in Pediatric Outpatients227
Characteristics of the study group are marked in
Table II. These 1689 patients were given 2155 prescriptions,
for an average of 1.3 prescriptions per patient. The number of
prescriptions at the index visit ranged from 1 to 7. Children
who had a chronic illness were more likely to have ?3
medications prescribed at their visit than children without a
chronic illness (7% in children with a chronic illness and 3%
in children without, P ? .05).
Our study population was 49% white, 15% black, 21%
Hispanic, and 14% other ethnic groups (eg, Native Ameri-
cans, Asians, and Native Pacific Islanders; Table III). Com-
paring the racial/ethnic mixture of our study population for
SES measures we found that Hispanic parents were more
likely to have lower educational attainment, lower annual
income, and 2-times as likely to have LEP.
We found 283 ADEs occurred in 242 children (14%),
of which 57 were preventable ADEs (rate, 3%; 95% CI,
35-4%) in 55 children and 226 were non-preventable ADEs
(rate, 14%; 95% CI, 11%-15%) in 186 children. Forty-nine
children experienced only a preventable ADE, and 6 children
had both a preventable ADE and a non-preventable ADE.
None of the preventable ADEs was life threatening or fatal,
14% were serious, and 86% were significant. An example of a
serious preventable ADE was a 9-year-old child with strep-
tococcal pharyngitis for whom amoxicillin was prescribed
whose parent did not complete the course of medicine, which
resulted in a return visit for persistent symptoms. Forty pre-
ventable ADEs (70%) occurred during parental administration
of medication to their child, and 15 preventable ADEs (26%)
occurred at ordering. The most common drugs involved in
preventable ADEs were amoxicillin or amoxicillin-clavulanate
(26%), inhaled steroids (11%), topical anti-fungal (7%), anti-
histamines (7%), and inhaled bronchodilators (5%).
Detection Method of ADEs
Of the 283 ADEs discovered, 218 were identified by
surveying patients and 51 ADEs were discovered by reviewing
charts, and 14 ADEs were identified by both means. The
distinction in the identifying source of the ADE has impor-
tant implications for the results. For example, we found that
black children, when compared with white children, had a
1.8-fold (95% CI, 0.74-3) higher odds of an ADE through
chart review, but a 0.58-fold lower (95% CI, 0.36-0.94) odds
of having an ADE reported by parental survey. Many other
vulnerable subgroups of children showed the same phenom-
ena. Objective chart review found higher rates of ADEs for
Hispanic children (odds ratio [OR], 1.4; 95% CI, 0.73-2.7;
compared with white children) and for children of parents
with LEP (OR, 1.5; 95% CI, 0.7-2.2; compared with parents
who were English proficient); parents in these subgroups may
have been reticent to report ADEs on the telephone survey
(Hispanic versus white children: OR, 0.64; 95% CI, 0.42-
0.97; LEP versus English proficient: OR, 0.54; 95% CI,
Children with Preventable ADEs Compared with
Children without ADEs
In the univariate analysis, we found that children of
parents who reported they spoke English poorly were twice as
likely to have a preventable ADE (OR, 2.3; 95% CI, 1.01-
5.34) than children of parents who spoke English very well
(Table IV). Similarly, children with less continuity of care
(?1 year) were more likely to have a preventable ADE than
children with more continuity of care (OR, 1.83; 95% CI,
1.01-3.34). In multivariate analysis, only children with mul-
tiple prescriptions were at increased risk of having a prevent-
able ADE (OR, 1.46; 95% CI, 1.01-2.11).
Table II. Characteristics of study sample
CharacteristicsTotal (n ? 1689)
Health status good/fair/poor
Educational attainment ?12 years
50,000 ? 80,000
30,000 ? 50,000
Above Federal Poverty Level†
Not at all
Urban health center
Continuity of care ?1 year
Number of prescriptions
†Poverty status was calculated by taking the family’s annual income and number of
family members, comparing them with the 2003 poverty guidelines issued by the
Department of Health and Human Services, and then classifying families in 2 groups,
either at the Federal Poverty Level or above the Federal Poverty Level.
228 Zandieh et alThe Journal of Pediatrics • February 2008
Relatively little research has described the risk factors
for ADEs in the pediatric outpatient setting. In this study of
6 primary care practices, we found that 14% of patients
experienced an ADE, of which approximately a quarter were
preventable. Children with complex medication regimens
were at increased risk of having a preventable ADE. We did
not find any disparities in preventable ADEs in pediatric
As expected, children with chronic illnesses had more
medications prescribed, which in turn increased their risk of
experiencing a preventable ADE. We do not believe that
these findings are caused by prescribing of less familiar med-
ications, because most drugs resulting in preventable ADEs
were associated with common medications, such as antibiot-
ics. Perhaps more likely is that providers did not sufficiently or
accurately communicate medication administration informa-
tion to parents, because most preventable ADEs occurred
during home administration. Improved communication
among healthcare providers and patients potentially could
prevent ADEs in the outpatient setting, as in the inpatient
We used a well-known and effective method for detec-
tion of ADEs.20,25-28Detection of voluntarily reported ADEs
depends on who is responsible for reporting the data. In the
outpatient setting, patients tend to be better reporters than
health professionals.17,29Gandhi et al showed in the adult
ambulatory care setting that patient surveys identified most
ADEs.20These results are comparable with ours in that we
identified 81% of ADEs through survey and 19% by chart
review. The differences between patients and health profes-
sionals reporting ADEs may be caused by patients being more
prone to report symptoms in hopes of feeling better, whereas
health professionals may be afraid of the potential ramifica-
tions (eg, fear of malpractice risk or damage to their reputa-
tion).30In an outpatient setting, interviewing patients directly
is an important source of information about medication inci-
An apparent ascertainment bias was found in our re-
sponse rates: more ADEs were identified through chart re-
view than parental report in black and Hispanic parents. This
is a surprising result because, as just stated, overall more
ADEs tend to be found through parental report. We suspect
that the low odds of reported ADEs by survey reflect the
conservative reporting style of black and Hispanic parents
rather than an actual reduction in the occurrence of ADEs.
Thus, the overall odds for ADEs that we reported may
underestimate the true difference between black and Hispanic
children compared with white children. Furthermore, these
differences may reflect that black and Hispanic parents are not
recognizing that new symptoms in their children may be
caused by a medication or perhaps because they worry about
the consequences of reporting their own errors (eg, not com-
pleting a 10-day course of antibiotics).32Conversely, these
findings may suggest that more educated white parents over-
report medical symptoms.
We also found that children of parents with LEP were
significantly less likely to report that their child had an ADE,
which may also make it more difficult for these parents to
communicate about ADEs with health care providers. Parents
with LEP may not be able to communicate adequately the
possible medication incidents in a telephone interview. Al-
though LEP does not represent the same thing as limited
health literacy, both may lead to harm. Patients with limited
health literacy not only struggle with written communication,
but also with oral communication.33
Our study had several limitations. First, although we
included 6 diverse pediatric practices that were hospital- and
community-based, the results may not be generalizable to
other settings. Second, we found an ascertainment bias be-
cause black and Hispanic parents were relatively less likely to
report an ADE. This may have affected the number of
Table III. Characteristics of sample by race/ethnicity
n ? 815
n/total n (%)
n ? 256
n/total n (%)
n ? 343
n/total n (%)
n ? 239
n/total n (%)
50,000 to ?80,000
30,000 to ?50,000
Percentages may not add to 100 because of rounding.
*English proficient: parent report of speaking English very well.
Risk Factors in Preventable Adverse Drug Events in Pediatric Outpatients 229
adverse events identified. Third, the respondent to the survey
may not have been the same person who spoke to the physi-
cian or was the primary caretaker for the child, therefore the
accuracy of parental recall associated to either the recognition
or reporting of an adverse drug event is unknown. Lastly, and
most important, although this is the largest study to date
examining ADEs in the pediatric ambulatory care setting, our
small sample size limits our findings.
Our findings have 2 important policy implications. For
policymakers and providers who are interested in improving
patient safety, better methods are needed to identify prevent-
able ADEs, especially among black and Hispanic parents with
LEP. Second, identification of parents’ health literacy and
appropriate tailoring of medication-related information are
required. Improving the effectiveness of physician-patient
communication should ultimately lead to a decrease in pre-
system. Washington, DC: Institute of Medicine, National Academy of Sciences; 1999.
Nelson AR. Unequal treatment: report of the Institute of Medicine on racial and
ethnic disparities in healthcare. Ann Thorac Surg 2003;76:S1377-81.
Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al.
Incidence of adverse events and negligence in hospitalized patients. Results of the
Harvard Medical Practice Study I. N Engl J Med 1991;324:370-6.
Kaushal R, Bates DW, Landrigan C, McKenna KJ, Clapp MD, Federico F, et al.
Medication errors and adverse drug events in pediatric inpatients. JAMA 2001;285:
Flores G, Abreu M, Tomany-Korman SC. Limited English proficiency, primary
language at home, and disparities in children’s health care: how language barriers are
measured matters. Public Health Rep 2005;120:418-30.
Rogers AJ, Delgado CA, Simon HK. The effect of limited English proficiency on
admission rates from a pediatric ED: stratification by triage acuity. Am J Emerg Med
Stevens GD, Shi L. Racial and ethnic disparities in the primary care experiences
of children: a review of the literature. Med Care Res Rev 2003;60:3-30.
Flores G, Olson L, Tomany-Korman SC. Racial and ethnic disparities in early
childhood health and health care. Pediatrics 2005;115:e183-93.
Flores G, Bauchner H, Feinstein AR, Nguyen US. The impact of ethnicity, family
income, and parental education on children’s health and use of health services. Am J
Public Health 1999;89:1066-71.
Weinick RM, Krauss NA. Racial/ethnic differences in children’s access to care.
Am J Public Health 2000;90:1771-4.
Cohen AL, Rivara F, Marcuse EK, McPhillips H, Davis R. Are language barriers
associated with serious medical events in hospitalized pediatric patients? Pediatrics
Nielsen-Bohlman L, Panzer A, Kindig D. Health literacy: a prescription to end
confusion. Washington, DC: Institute of Medicine; 2004.
Williams MV, Parker RM, Baker DW, Parikh NS, Pitkin K, Coates WC, et al.
Inadequate functional health literacy among patients at two public hospitals. JAMA
Davis TC, Wolf MS, Bass PF III, Middlebrooks M, Kennen E, Baker DW, et al.
Low literacy impairs comprehension of prescription drug warning labels. J Gen Intern
Davis TC, Wolf MS, Bass PF III, Thompson JA, Tilson HH, Neuberger M,
et al. Literacy and misunderstanding prescription drug labels. Ann Intern Med 2006;
Dovey S, Weitzman M, Fryer G, Green L, Yawn B, Lanier D, et al. The ecology
of medical care for children in the United States. Pediatrics 2003;111(5 Pt 1):1024-9.
Morimoto T, Gandhi TK, Seger AC, Hsieh TC, Bates DW. Adverse drug events
and medication errors: detection and classification methods. Qual Saf Health Care
Kaushal R, Goldmann DA, Keohane CA, Hale AS, Christino M, Honour M,
et al. Adverse drug events and near-misses in pediatric outpatients. Ambul Pediatr. In
Fortescue EB, Kaushal R, Landrigan CP, McKenna KJ, Clapp MD, Federico F,
et al. Prioritizing strategies for preventing medication errors and adverse drug events in
pediatric inpatients. Pediatrics 2003;111(4 Pt 1):722-9.
Gandhi TK, Weingart SN, Borus J, Seger AC, Peterson J, Burdick E, et al.
Adverse drug events in ambulatory care. N Engl J Med 2003;348:1556-64.
Office of Management and Budget. Revisions to the Standards for the Classifi-
cation of Federal Data on Race and Ethnicity. Washington, DC: US Census Bureau.
Available at: http://www.census.gov/population/www/socdemo/race/Ombdir15.html.
Accessed September 13, 2006.
Flores G. The impact of medical interpreter services on the quality of health care:
a systematic review. Med Care Res Rev 2005;62:255-99.
McDowell I, Newell C. Measuring Health: A guide to rating scales and ques-
tionnaires. New York: Oxford University Press; 1996.
2003 Poverty Guidelines. Department of Health and Human Services. Available
at: http://aspe.hhs.gov/poverty/03fedreg.htm. Accessed July 7, 2006.
Bates DW, Boyle DL, Vander Vliet MB, Schneider J, Leape L. Relationship
between medication errors and adverse drug events. J Gen Intern Med
Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al. Incidence
of adverse drug events and potential adverse drug events. Implications for prevention.
Gurwitz JH, Field TS, Harrold LR, Rothschild J, Debellis K, Seger AC, et al.
Incidence and preventability of adverse drug events among older persons in the ambu-
latory setting. JAMA 2003;289:1107-16.
Kohn L, Corrigan J, Donaldson M. To err is human: building a safer health
Table IV. Unadjusted and adjusted odds ratio of
patients with preventable adverse drug events
OR (95% CI)
OR (95% CI)
Urban health center
Number of prescriptions
College, post graduate
Less than high school
Not at all
Duration of care
Model compared 49 children with the presence of a preventable ADEs to 1640 children
*Adjusted for all variables listed except annual income, which is highly correlated with
†P ? .05.
230Zandieh et alThe Journal of Pediatrics • February 2008
incidence of adverse drug events in two large academic long-term care facilities. Am J
Jha AK, Kuperman GJ, Teich JM, Leape L, Shea B, Rittenberg E, et al.
Identifying adverse drug events: development of a computer-based monitor and com-
parison with chart review and stimulated voluntary report. J Am Med Inform Assoc
Leape LL. Reporting of adverse events. N Engl J Med 2002;347:1633-8.
Gurwitz JH, Field TS, Judge J, Rochon P, Harrold LR, Cadoret C, et al. The
Patient-reported medication symptoms in primary care. Arch Intern Med 2005;165:
Stebbing C, Wong IC, Kaushal R, Jaffe A. The role of communication in
paediatric drug safety. Arch Dis Child 2007;92:440-5.
Schillinger D. Health literacy and error reduction: a fatal aspiration of a syringe
cap. Review of reviewed item. Available at: http://webmm.ahrq.gov/case.aspx?caseID?
53&searchStr?lethal?cap. Accessed June 22, 2006.
Weingart SN, Gandhi TK, Seger AC, Seger DL, Borus J, Burdick E, et al.
50 Years Ago in The Journal of Pediatrics
ACNE: A NEW APPROACH TO AN OLD PROBLEM
Baird JW. J Pediatr 1958;52:152-7
Baird argued that acne is the “scourge of youth” and should not be allowed to run its course but rather should be
treated. The mainstay of the new approach recommended was diet, including avoiding an excess of carbohydrates
(especially candy and carbonated cola beverages), chocolate, nuts, peanut butter, pork, and fried food. The authors
recommended that milk and milk products should not be used in excess but rather the diet should be rich in lean meats,
fruits, vegetables, and vitamins. Local treatment with a cream with drying and keratolytic powers was recommended to
be accompanied when necessary by the manual expression of comedones by a physician “using one of the mechanical
devices available for this purpose.” Treatment of severe cases could be supplemented with x-ray, short courses of oral
antibiotics, Staphylococcus vaccines, and estrogen hormones in girls over 18 years if acne exacerbations were associated with
Today, the realization that acne can be emotionally and physically scarring and that treatment is generally effective
and indicated is well accepted by the medical community and general population. However, much has changed in what
is considered acceptable treatment of acne over the last 50 years and most of what appeared to be promising at the time
this article was written is not longer indicated. Current mainstays of acne treatment do not include diet, the manual
expression of comedones, irradiation, or short courses of antibiotics. Rather, topical retinoids are now the primary
treatment although other topical agents (benzoyl peroxide, topical antibiotics, and azelaic acid) can also be used. Systemic
antibiotics (especially tetracycline and its derivatives) for at least 6 to 8 weeks are indicated if response to topicals has been
inadequate, if patients have moderate to severe acne, and/or a propensity for scarring. Trials of hormone therapy may be
appropriate or, in severe, nonresponsive cases, of isotretinoin. Surgical approaches are largely limited to intralesional
injection of low-dose glucocorticoids for persistent, painful lesions. (Dermabrasion and laser peel are used after the active
process has quieted.) Trials of phototherapy and vaccines offer some promise.
Bonita Stanton, MD
Professor and Schotanus Family Endowed Chair of Pediatrics
Carman and Ann Adams Department of Pediatrics
Children’s Hospital of Michigan
Wayne State University School of Medicine
Risk Factors in Preventable Adverse Drug Events in Pediatric Outpatients231