Longitudinal
Evaluation
of a Statewide
Network
of Hospital
Programs
to Improve
Child
Passenger
Safety
Richard
B. Colletti,
MD
From the University of Vermont, Burlington
ABSTRACT.
As a project begun in 1979 to improve childpassenger safety, a statewide network of hospital-based
car safety seat rental and education programs was imple-mented by a coalition of volunteers, nurses, physicians, and state agencies. A car safety seat program was estab-lished at every hospital delivering newborns. To evaluate the effectiveness of this network of programs, 1,846
new-borns (87% of hospital discharges) were observed at
discharge at ten hospitals during 33 observation studies
from 1979 to 1984. Mean correct car safety seat usage
increased from <21% in 1979 to 82% in 1984.
Concur-rently, use of the dangerous lap or arms position de-creased from 70% preprogram to 6% postprogram, and
incorrect car safety seat usage decreased from 28% to 3%. In addition 1,597 children from 0 to 3 years of age were
observed at an annual county fair from 1982 to 1984. Use
of either a vehicle seat belt alone or car safety seat with
harness and seat belt increased from 34% in 1982 to 67%
in 1984. Car safety seat and seat belt misuse occurred in
42% of infants <1 year of age and 20% of children 1 to 3
years of age. These findings suggest that a network of hospital-based car safety seat rental and education pro-grams is an effective means of improving the passenger safety of newborns, infants, and young children.
Pediat-rics 1986;77:523-529; accident prevention, health
educa-tion, motor vehicle injury, car safety seat, child passenger.
In the United States about 75% of vehicular
crashes result in property damage only, 25% result in injuries, and 1% result in death.’ About 2% of all deaths in the United States are due to motor vehicle accidents (70% are due to major cardiovas-cular diseases and malignant neoplasms).2 In 1979, there were 51,900 deaths related to the use of motor vehicles; 78% were drivers or passengers. Fatality rates (per 100 million passenger miles) are highest
Received for publication March 29, 1985; accepted June 10,
1985.
Reprint requests to (R.B.C.) University ofVermont, 1 5 Prospect St, Burlington, VT 05401.
PEDIATRICS (ISSN 0031 4005). Copyright © 1986 by the American Academy of Pediatrics.
for persons 16 to 29 years of age and for the elderly.’ Motor vehicle accidents are the leading cause of death in childhood, causing 15% of deaths of chil-dren 1 to 4 years of age and 25% of deaths of children 5 to 14 years of age. For children 0 to 12 years old, the motor vehicle occupant fatality rate
is highest in the first 6 months of life.4 Fatal infant
accidents are more likely when maternal education and age is low and maternal parity is high.5
Car safety seats are designed to provide optimal protection in a crash to children 0 to 4 years of age. With correct use, the infant is secured to the car
safety seat by a harness, and the car safety seat is
anchored to the car by the vehicle seat belt. When used correctly, car safety seats prevent 90% of deaths and 60% of injuries.6 Yet, in 1974 in Mary-land, Massachusetts, and Virginia, only 6% to 12% of children 0 to 9 years of age were properly secured in a car safety seat or vehicle seat belt.7 In 1977 in Tennessee and Kentucky, car safety seat usage for
children 0 to 3 years ofage was 8% to 11%.8 In 1980
in Maryland and Oklahoma, it was 7%,9 in Rhode Island 18%, and in Massachusetts 22%.’#{176}
As early as 1974, it was observed that infants were more likely than older children to be in car
safety seats.7 However, infants are also more likely
to be incorrectly secured. Thus, although 37% of infants <1 year of age were in car safety seats or vehicle seat belts, only 33% of them were correctly restrained.
TABLE 1. Studies of the Effectiveness of Techniques to Increase Car Safety Seat Usage
Location Year N Age Observed Observed “Correct” Usage
(%)
Reference No.
Before or After or
Without In- With
Inter-tervention vention
Pittsburgh 1976-1977
1978-1979
955
132
Newborn
2-4 mo 2 mo
l5mo
6 8-11
21 20-28
29 50
50 56
12
13
Kansas City, MO 30 Newborn
imo
0 67
23 29
14
Chicago 1981 182 Newborn 8 11 15
Kalamazoo, MI 0-9 mo 9 34 16
Albuquerque 1980-1981 195 4 mo 9 38 17
Vermont 1979-1984 618
92 581
Newborn
<1 yr
1-3 yr
<17 82
49-75 59-68
Present study
Every Ride . . .A Safe Ride” program of the
Amen-ican Academy of Pediatrics.9”
The effectiveness of techniques to promote car
safety seat usage was evaluated by observing car safety seat usage in six published studies’2’7 (Table 1). At a Pittsburgh hospital in 1976 to 1977, Reisin-ger and Williams’2 found that literature, personal
discussion, a car safety seat free or convenient for
purchase, and demonstration of correct use did not
substantially increase correct car safety seat usage
by newborns at hospital discharge or at follow-up 2
to 4 months later. In 1978 to 1979, these authors’3
used personal counseling by a pediatrician in the
hospital and the office plus literature and
demon-stration. They observed increased usage at 2 months of age in the experimental group (50%) v a
comparison group (29%) but between 4 months and 15 months of age observed differences were slight
and not statistically significant.
In a Kansas City hospital, a personal demonstra-tion of correct car safety seat usage and a free loaner car safety seat increased correct usage by newborns at hospital discharge from 0% to 67%, but by 1 month of age usage in the experimental group (29%) was not substantially higher than that of the control group (23%).’ At a Chicago hospital in 1981, a postpartum program using a group dem-onstration, film, and literature did not substantially increase correct car safety seat usage at discharge.’5
In contrast, at a Kalamazoo hospital, a program
using personal instruction, film, pamphlet, and car safety seat rental appeared to increase observed
correct usage from 9% by general community
in-fants to 34% by program infants 0 to 9 months of
age.’6 At an Albuquerque clinic in 1980 to 1981, correct car safety seat usage by 4-month-old infants was only 9%17 After an experimental program in-cluding group demonstration, film, pamphlet, pre-natal discussion, and free car safety seat, usage increased to 38%.
These six intervention studies resulted in either
no increase, a modest increase, or a short-term increase in correct car safety seat usage. The studies are limited by the use of one-time intervention, restricted location or population, or short-term fol-low-up observations. We report on the longitudinal observations of the effectiveness of a statewide network of hospital-based car safety seat rental and education programs.
METHODS
In 1979, individuals and groups committed to child passenger safety formed a coalition called Vermont SEAT (Seatbelts Eliminate Automobile Tragedies). It coordinated the efforts of the state agency of transportation and departments of health and education and created a statewide network of volunteers, nurses, hospital administrators, and physicians. The initial goal was to have 90% of all newborns in Vermont be discharged from hospital
in car safety seats. To accomplish this, it was
de-cided to establish a car safety seat rental and edu-cation program at each hospital delivering babies.
As described in detail previously,’8 at a statewide conference, invited hospital representatives were asked to use volunteers to establish and operate a hospital-based car seat rental program. Each hos-pital representative was provided a copy of “EarlyRider,” an instructional manual printed and provided free by the National Highway Traffic Safety Administration. In the manual were detailed guidelines on how to establish a car safety seat rental program.
The “EarlyRider” manual also gave guidelines
for an education curriculum. It included chapters
systematic contact with each postpartum mother,
and a procedure in which the discharge person
assists the parent in placing the child in the car seat, fastening the harness, and securing the vehicle seat belt.
Hospitals were also asked to determine car safety seat usage rates by newborns before and after start-ing a rental program. Forms provided were to be
completed by the discharge nurses who observed
infants in the automobile as they departed from the
hospital. Earlier forms indicated only whether the infant was in a car safety seat, but later forms also indicated whether the infant was secured by a
han-ness or a vehicle seat belt.
To detemine what factors at each hospital might
contribute to the success or failure of its rental and
education program, each hospital completed an in-itial and annual questionnaire. Information was obtained about prenatal contacts, physician partic-ipation, postpartum ward activities, modeling and behavioral rehearsal techniques, discharge proce-dunes, car safety seat availability, program
visibil-ity, additional outreach, personnel, financial data, participation by the hospital administration,
insur-ance coverage, car safety seat storage, problems, major strengths, and major weaknesses. Each hos-pital was visited every 6 to 12 months to discuss program operations.
Although the immediate goal of Vermont SEAT
was to increase car safety seat usage by newborns, it was hoped that as infants grew to toddlers and
preschoolers their parents would continue to secure them in car safety seats and vehicle seat belts by habit as well as by intent. To estimate the potential long-term effects of the hospital-based programs, we observed car safety seat and vehicle seat belt
usage by children 0 to 3 years of age at an annual county fair each year for 3 years beginning in 1982.
It is the most populous county in the state (Table 2, hospital 5), and usage rates in more rural areas may be lower. Observations in 1984 were made 1 month after a new state law went into effect re-quiring children 0 to 3 years of age to be properly
secured in car safety seats or vehicle seat belts.
Although car safety seats used incorrectly may provide protection in some crashes, there is serious concern that failure to fully comply with
instnuc-tions for correct usage compromises their effective-ness.’#{176}Dynamic crash testing indicates that opti-mal protection is obtained when, in addition to use of harness and vehicle seat belt, a tether is used if required and the infant-type can safety seat is facing rearward. A rear seat position is preferred if secured by a vehicle seat belt alone. Previous studies have defined “connect” car safety seat usage variably, including: harness (or shield), vehicle seat belt, and
tether if required7; vehicle seat belt fastened around car safety seat8”2’4; on harness, vehicle seat belt,
and rearward facing if required.’6’7 For the present
study we defined correct usage for newborns at hospital discharge as use of harness and vehicle seat belt. To determine whether the rearward facing position was used, we recently revised our data
collection form and observed newborns at one hos-pital (Table 2, hospital 5). In addition, at the 1984
county fair we recorded observations of car safety
seat use with harness and vehicle seat belt, tether if required, rearward facing if required, and neat seat position if in a vehicle seat belt only.
RESULTS
Hospital-based car safety seat rental and educa-tion programs began at three hospitals in 1980, seven hospitals in 1981, and two hospitals in 1982.
These 12 hospitals delivered 98% of newborns in Vermont each year. Two more hospitals started
programs in 1984.
Volunteers at each hospital used the
“Early-Rider” manual and their own creativity to inde-pendently organize their programs according to the
resources of each community. Start-up money to purchase car safety seats was usually obtained from
donations to the local hospital auxiliary fund or
sometimes from general hospital funds. Parents
rented the car safety seat for 9 to 12 months, for
TABLE 2. Car Safety Seat Usage and Educational Intervention at Ten Hospitals
Hospital No.
1 2 3 4 5 6 7 8 9 10
Car safety seat usage
(with harness and seat belt) (%) 93 92 92 85 81 79 75 73 72 69 Educational components*
Prenatal
x
x
X XContact each X X X X X X X
Pamphlets X X X X X X X X
Film
x
X X XDemonstration X X X X X X X X
Rehearsal X X X
. . .
I
90
80
70
60
50 40 30 20
10
0
0
I
0
$15 to $20, with a refund of $5 to $10 when car seats were returned. Initially, each program pur-chased 50 to 200 car safety seats. Because returned seats were rented again several times, each program
operated with a small profit and was self-sustaining. Rental income was used to purchase additional car
safety seats, replacement parts, and cleaning serv-ices. The hospital gift shop was the usual base of
operation, but at some hospitals the maternity ward on an administrator’s office was the site where car
safety seats were stored, loaned, and returned and
records kept.
An average of five volunteers shared the three
hours per week required to operate the rental
pro-grams. Some hospital auxiliaries used the program to attract new members. At some hospitals, as many
as 30 volunteers participated in the project. Some of the larger hospitals required up to a total of seven hours of volunteer work per week. At some hospitals paid employees such as nurses or clerks
performed part or all of the duties. Most hospital administrators supported these programs because
they were highly visible in the community and
enhanced public relations. In 1984, there were about
8,000 live births annually in Vermont and 3,000 rented infant car seats in circulation.18
From 1979 to 1984, 1,846 newborns (87% of hos-pital discharges) were observed at discharge during
33 observation studies at ten hospitals (Fig 1).
Mean correct car safety seat usage (with harness and vehicle seat belt) increased from <21% in 1979 to 31% in 1980, 35% in 1981, 59% in 1982, 75% in
1983, and 82% in 1984.
Preprogram usage was observed at four hospitals. At two hospitals before starting programs, total car
1OO
. .
$
& S.
I iO. I . I I . I
79
80
81
82
83
84
Fig 1. Usage of car safety seats (CSS) at discharge at
ten hospitals during 33 observation studies, 1979 to 1984.
.,
CSS with harness and belt; 0, CSS with or withoutharness and belt; arrows, time when hospital programs
started.
safety seat usage was 21% in 1979; after programs
were begun there, usage increased to >60% in 1980 to 1981. At two other hospitals, preprogram usage in 1980 to 1981 was still low (13%), but usage increased after programs were begun there.
Ob-served increases in car safety seat usage coincided
with the starting of hospital-based programs.
The best correlate of increased car safety seat usage appears to be duration of hospital program operation (Fig 2). Mean correct usage increased from <17% preprogram to 40% the first year of program operation, 62% the second year, 79% the third year, and 89% the fourth and fifth years. Concurrently, use of the dangerous lap or arms position (in which the infant is held on the lap or
in the arms of a passenger) declined dramatically
from 70% preprogram to 38% the first year of program operation, 25% the second year, 12% the third year, and 6% the fourth and fifth years. Sim-ilarly, failure to use harness or vehicle seat belt when using the car safety seat also decined from
28% the first year of program operation to 1 1 % the
second year, 8% the third year, and 3% the fourth and fifth years.
To determine whether infant car safety seats were rearward facing, as is recommended, an addi-tional 1-month observation study was performed in December 1984 at a hospital with a rental and education program in operation for 5 years. Of 91 discharges, 86 newborns were observed, 85 were in car safety seats, and 84 (98%) were in harness, seat belt, and facing rearward.
Educational intervention varied considerably among hospitals. Most hospitals contacted each mother to inform her about car safety seats and the rental program, distributed pamphlets, and
dem-PreProgam
Post
Progam
(years)
Fig 2. Mean usage of car safety seat (CSS) total (0),
Css with harness and belt
(#{149}),
and lap or arms positionAge of Child
%50
40
-<1
1
2
3
onstrated correct car safety seat usage (Table 2). A minority of hospitals provided information pre-natally on had parents rehearse connect car safety seat usage. One hospital used six techniques, two used five, four used three, and three used two. The most recent correct can safety seat usage rates at these ten hospitals ranged between 69% and 93%.
There was no correlation by hospital between the usage nate and the type or number of educational techniques, the level of maternal education, or fre-quency of maternal use of Medicaid.
At an annual county fair, 1,597 children 0 to 3
years of age were observed from 1982 to 1984 (Fig 3). Use ofeither a vehicle seat belt only or car safety
seat with harness and belt increased from 34% in
1982 to 67% in 1984. For infants <1 year of age, it increased from 49% in 1982 to 75% in 1984; for children 1 year of age from 48% to 82%; 2 years of age from 35% to 68%; and 3 years of age from 23% to 53%. In 1984, use of vehicle seat belt only for children 1 year of age was 4%, 2 years of age 16%, and 3 years of age 31%.
Although 86% of infants <1 year of age were in
car safety seats, 42% were used incorrectly (Table
3). Thus, by the strictest criteria, only 49% were
correctly secured. Most errors in this age group
100
90
80
70
601-
NB
30
20
10
0L
I I I82 83 84
Fig 3. Usage of either car safety seat (CSS) (with
har-ness and belt) or vehicle seat belt alone by child
passen-gers <1 year old (0), 1 year old
(a),
2 years and 3years (0) at an annual county fair, compared with mean CSS usage (with harness and belt) by newborns at
hos-pital discharge (#{149}).
TABLE 3. Incorrect Usage of Car Safety Seats and
Vehicle Seat Belts at a County Fair in 1984*
<1 Yr 1-3 Yr
Observed 77 (100) 401 (100) Correct usage 45
(
58) 319(
80)Incorrect usaget 32 ( 42) 82 ( 20)
No harness 9
(
12) 41(
10)Nobelt 5( 6) 10( 2)
No tether (if required) 6
(
8) 27(
7)Not facing rear (if required) 24
(
31) 2 ( 1)In front seat in seat belt 0 ( 0) 25 ( 6)
* Results are numbers
(%)
of children observed.t
One or more errors.were failure to use a harness or belt or not facing
rearward when required. Although 74% of children
1 to 3 years of age were in either a vehicle seat belt
alone or car safety seat, 20% of these restraints were used incorrectly (Table 3). Thus, by the
strict-est criteria, only 59% of observed children 1 to 3
years of age were correctly restrained.
At hospital discharge, based on duplicate
obser-vations of 12 newborns, interobserver reliability was 97%. At the county fair in 1984, based on 74
duplicate observations, interobserver reliability was 97%.
DISCUSSION
This study demonstrates the feasibility and
lon-gitudinal effectiveness of a statewide network of hospital-based car safety seat rental and education
programs. Within 5 years, every hospital in the state delivering newborns had a car safety seat
program. Correct can safety seat usage with harness and seat belt by newborns at hospital discharge increased from <21% in 1979 to 82% in 1984. Increased car safety seat usage coincided with the start of hospital-based programs; use of the
danger-ous lap or arms position decined, as did incorrect
usage of car safety seats. These observations sug-gest that as a result of hospital-based can seat rental
and education programs the parents of newborns were able to obtain car safety seats and learned to
use them correctly.
It also appears that parents continue to use car
safety seats and vehicle seat belts as their newborns
grow. At one location in Vermont in 1984, 86% of
observed infants <1 year of age were in car safety seats; 75% were in car safety seats with harness
and seat belt. A slightly lower percentage of chil-dren 1 to 3 years of age were secured; 68% were either in a vehicle seat belt alone or a car safety
seat with harness and seat belt.
As has been observed previously,7 however, with
a vehicle seat belt alone or car safety seat with
harness and seat belt, only 68% of 2-year-old and
53% of 3-year-old children were so secured. We have also observed that car safety seats may fre-quently be used incorrectly. Although more than 90% of car safety seats were used correctly at hos-pital discharge and 80% were used correctly by infants 1 to 3 years of age, 42% of infants <1 year
of age in car safety seats were by the strictest criteria incorrectly secured. Nonetheless, in the last
3 years of this study, correct usage of car safety
seats and vehicle seat belts has increased
substan-tially for all ages observed.
There are advantages to a car safety seat program
within a hospital. Car safety seats can be
inexpen-sively and conveniently provided at the time of
need. A hospital-based education program can tar-get its effort at a captive audience which is, during the postpartum period, motivated to learn, willing to change its behavior, and vulnerable to
persua-sion.
The observed increase in car safety seat usage
may be attributable to methods not used in other studies. Educational intervention, although varia-ble from hospital to hospital, was generally multi-faceted in origin. Education was provided by nurses, physicians, and volunteers with the support of
hos-pital administration. The educators were known, credible, and trusted figures in the community.
Parents knew that there was a statewide effort to
improve child passenger safety. During 3 years of
the study, there was ample newspaper and
televi-sion coverage of legislative lobbying and debate
about mandatory can safety seat legislation. Child
passenger safety was further promoted statewide by state agencies with public information and edu-cation programs and at state meetings of pediatni-cians, nurses, hospital auxilians, and hospital ad-ministnators. Car safety seats became socially
ac-ceptable and the basis of a new standard of child
passenger safety in communities in Vermont.
The level of maternal education in Vermont2#{176} is
higher than the national average.2’ In Vermont,
86% of mothers giving birth had completed high
school compared with only 78% for all races nation-ally and 81 % for whites nationally. Because level of maternal education is an important determinant of health behavior, including use of car safety seat,’6 parents in Vermont may be somewhat more respon-sive to educational intervention. A Vermont
states-man once said, “Of all the governors of the United
States, I think I have the best opportunity to ob-serve the general good which may be effected by cooperation among groups” (The New York Times,
Jan 11, 1985, p AlO). Although not unique to
Ver-mont, the spirit of cooperation, coalition, and
com-munication among communities may have contnib-uted to the effectiveness of the programs.
It is unlikely that the observed increased usage of car safety seats is merely secondary to the na-tional attention given to child passenger safety. Legislatures in all 50 states have passed laws (most
between 1982 and 1984) requiring infants and
young children to be secured in car safety seats and seat belts when traveling in motor vehicles.22 When the effect of a law has been studied, increased
connect car safety seat usage has been observed. In Tennessee, usage increased from 8% to 29% after
2.5 years,8 and in Rhode Island, it increased from
22% to 35% after 4 months.’#{176} In Minnesota, it increased from 44% to 63% for infants <1 year of
age and from 15% to 25% for children 1 to 3 years
of age, 2.5 years after its law had been in effect, in
1984.23 These findings suggest that, although it may
reduce injuries24 and fatalities,25 legislation alone
will not alter the behavior of the majority of
par-ents. A law requiring car safety seat and seat belt usage in Vermont went into effect on July 1, 1984,
after all 33 observation studies at ten hospitals had
been completed. It will be interesting to determine
whether this law will further increase car safety
seat usage.
In summary, the Vermont program appears to have succeeded because the program was based in
hospitals, education was multifaceted, the target
population was highly educable, and the program
was statewide, facilitating the transformation of
communities’ notions of socially acceptible ways for children to travel in motor vehicles. For the pro-motion of child passenger safety, we recommend an
intervention mode in which a statewide or regional network of hospital-based car safety seat rental and education programs serves as a foundation for
ed-ucating parents. Increased car safety seat and
ye-hide seat belt usage is most likely to result from a comprehensive strategy combining education, availability, and enforced legislation.
ACKNOWLEDGMENTS
This work was supported, in part, by Innovative
Proj-ect grant 1-A06 of the National Highway Traffic Safety
Administration. Project investigators were Esther
Tell-stone, RN, and Barbara Mayo. We thank Barbara
Schoolcraft, Michele Formen, John Harvey, the Vermont Agency of Transportation and Departments of Health and Education, and many other volunteers and support-ers.
REFERENCES
2. Wegman ME: Annual summary of vital statistics-1983.
Pediatrics 1983;74:981-990
3. Christoffel KK, Tanz R: Motor vehicle injury in childhood.
Pediatr Rev 1983;4:247-254
4. Baker SP: Motor vehicle occupant deaths in young children.
Pediatrics 1979;64:860-861
5. Wicklund K, Moss 5, Frost F: Effects of maternal education, age, and parity on fatal infant accidents. Am J Public Health
1984;74:1150-1152
6. Scherz RG: Fatal motor vehicle accidents of child passengers
from birth thrbugh 4 years of age in Washington State.
Pediatrics 1981;68:572-575
7. Williams AF: Observed child restraint use in automobiles.
Am J Dis Child 1976;130:1311-1317
8. Williams AF: The Tennessee child restraint law in its third year. Am J Public Health 1981;71:163-165
9. Eriksen MP, Gielen AC: The application of health education
principles to automobile child restraint programs. Health Educ Q1983;10:30-55
10. Williams AF, Wells JK: Evaluation of the Rhode Island child restraint law. Am J Public Health 1981;71:742-743
1 1. Colletti RB: Hospital-based rental programs to increase car
seat usage. Pediatrics1983;71:771-773
12. Reisinger KS, Williams AF: Evaluation of programs de-signed to increase the protection of infants in cars. Pediatrics
1978;62:280-287
13. Reisinger KS, Williams AF, Wells JK, et a!: Effect of
pedia-tricians’ counseling on infant restraint use. Pediatrics
1981;67:201-206
14. Christophersen ER, Sullivan MA: Increasing the protection
of newborn infants in cars. Pediatrics 1982;70:21-25
15. Goebel JB, Copps TJ, Sulayman RF: Infant car seat usage.
J Obstet Gynecol Neonatal Nurs 1984;13:33-35
16. Hletko PJ, Hletko JD, Shelness AM, et al: Demographic
predictors of infant car seat use. Am J Dis Child
1983;137:1061-1063
17. Berger LR, Saunders S. Armitage K, et al: Promoting the
use of car safety devices for infants: An intensive health
education approach. Pediatrics 1984;74:16-19
18. Colletti RB: Statewide hospital-based program to improve
child passenger safety. Health Educ Q1984;11:207-213 19. Shelness A, Jewett J: Observed misuse of child restraints,
in Society of Automative Engineers Child Injury and
Re-straint Conference Proceedings. San Diego, 1983, pp 207-216
20. 1982 Annual Report of Vital Statistics in Vermont. Burling-ton, Department of Health, Division of Health Statistics, 1982
21. National Center for Health Statistics: Advance report of
final natality statistics, 1982. Monthly Vital Statistics Re-port, US Department of Health and Human Services, vol 33, No. 6 (suppi) Sept 28, 1984
22. Sanders RS, Dan BB: Bless the seats and the children: The
physician and the legislative process. JAMA
1984;252:2613-2614
23. 1984 Observational Study and Driver Survey. St Paul,
Mm-nesota Department of Public Safety, Occupant Restraint
Program, 1984
24. Agran PF, Wehrle PF: Injury reduction by mandatory child
passenger safety laws. Am J Public Health 1985;75:128-129
25. Decker MD, Dewey MJ, Hutcheson RH, et al: The use and
efficacy of child retraint devices. JAMA 1984;252:2571-2575
ERRATUM
There is a new typographical error in the “corrected” theophylline dosage printed on page 813 in the November 1985 (vol 76, No. 5) issue of Pediatrics
concerning theophylline dosage in The Harriet Lane Handbook.
The correct dosage for infants 6-52 weeks old should be 0.2 x age in weeks
+ 5.0 (not 1.2 x age in weeks + 5.0).
Cynthia H. Cole, MD
University of Vermont College of
Medicine