Patient Education and Counseling 46 (2002) 169173

Still falling: a community-wide infant walker injury prevention initiative

Gregory P. Connersa,b,*, Tener G. Veenemaa,c, Carol A. Kavanaghb,d,
John Riccie, Charles M. Callahanf
a

Department of Emergency Medicine, School of Medicine and Dentistry, University of Rochester, 601 Elmwood Avenue, Box 655, Rochester, NY 14642, USA
b
Department of Pediatrics, School of Medicine and Dentistry, University of Rochester, 601 Elmwood Avenue, Box 655, Rochester, NY 14642, USA
c
School of Nursing, University of Rochester, Rochester, NY 14642, USA
d
Department of Pediatrics, Rochester General Hospital, Rochester, NY 14642, USA
e
Monroe County Health Department, Rochester, NY 14642, USA
f
Department of Pediatrics, University of Massachusetts, Worcester, MA, USA

Abstract
Despite the well-known risk of injury associated with use of infant walkers, they remain popular, leading to large numbers of walker-related
injuries. A coalition of health care providers and educators, with the assistance of retailers and medical and human service agencies, undertook
an intensive multifaceted, community-wide intervention to educate the general and health care public regarding the dangers of infant walker
use and thereby reduce the number of walker-related injuries in our community. Following this intervention, 28% fewer children presented
annually at the two area pediatric emergency departments for walker-related falls down stairs than during the 30 months before the
intervention. The magnitude of this reduction attributable to the intervention, however, is uncertain, as national trends during the study period
revealed a similar decrease in walker-related injuries. Educational interventions alone may signicantly reduce but not eliminate walkerrelated injuries; national policy measures are likely also necessary. # 2002 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Infant walker; Falls; Injury prevention; Behavior change

1. Introduction
Although the dangers associated with use of infant
walkers have been recognized since 1982 [1,2], their use
remains widespread [317]. Consumers purchase about
1.2 million new infant walkers at a cost of about US$ 40
million each year (data courtesy National SAFEKIDS Campaign). Parental decision making regarding the use of infant
walkers is not clearly understood [18]. Parents may acquire
infant walkers with the belief that their use accelerates
development of independent walking skills, although there
is evidence to the contrary [19]. Imparting both mobility and
speed to a normally non-ambulatory infant, walker use has
been associated with falls down stairs, tip overs, collisions
with objects, burns, and poisonings [11,12,14,16,17]. In the
US, walker use is among the most common mechanisms of
injury for children under 18 months old [7,14], leading to
thousands of emergency department visits each year [20].
Although most walker-related injuries are relatively minor,

*
Corresponding author. Tel.: 1-585-273-4795; fax: 1-585-473-3516.
E-mail address: gregory_conners@urmc.rochester.edu (G.P. Conners).

fractures and serious head injuries are not uncommon

[1,7,12,15,16]; the deaths of at least two American children
per year are attributed to the use of infant walkers [11].
Because of the associated risk of injury, efforts have been
made to discourage infant walker use. The American
Medical Association has recommended that physicians
counsel parents regarding this risk [21]. The American
Academy of Pediatrics has recommended a ban on the sale
of infant walkers in the US, along with educational efforts
and community programs to dispose of existing walkers
[11].
Pediatricians from the Rochester, NY community were
among the rst to call attention to the hazards of infant
walkers nearly two decades ago [1]. However, the numbers
of children requiring evaluation in Rochester's pediatric
emergency departments (PedEDs) because of infant
walker-related falls down stairs has remained high. This
paper describes a community-wide educational intervention
designed to reduce the number of infant walker-related
injuries by increasing public and health care provider awareness of the hazards of infant walkers and an evaluation of its
effect on the numbers of children presenting to the PedEDs
because of walker-related falls down stairs.

0738-3991/02/$ see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 7 3 8 - 3 9 9 1 ( 0 1 ) 0 0 2 1 0 - 5

170

G.P. Conners et al. / Patient Education and Counseling 46 (2002) 169173

2. Methods
A coalition of Rochester area health care providers (``Too
Many Tumbling Tots Coalition'') was formed to address the
continuing problem of injuries related to the use of mobile
infant walkers. The goal of the coalition was to reduce the
number of infant walker-related injuries in Monroe County
(Rochester, NY). Methods for achieving this goal included
public and health care provider education regarding the
dangers of infant walker use, reducing the number of
walkers in use, and encouraging parents and caregivers to
use safe alternatives to infant walkers.
2.1. Intervention
The coalition began by soliciting support for the project
with local public service and health care agencies. These
groups were asked to give support and written endorsement
on project materials. County and city medical societies,
childcare organizations, emergency medical services, the
county health department injury prevention division, and
the regional burn and poison prevention centers agreed
to participate. Posters, brochures, and pamphlets were
designed and produced for parents, day care providers,
and other caregivers. Readibility (grade level < 6) was
checked with Microsoft-Word software and by professional
educators. The materials and posters, which included vivid
drawings of a child rolling down stairs in a walker out of
reach of his mother, emphasized that even under the best
conditions walker use is unsafe and stressed the use of infant
walker alternatives. Packets containing these materials were
sent to over 240 health care providers' ofces and clinics,
including pediatricians, family medicine and emergency
medicine physicians, and obstetricians. A cover letter briey
outlined the project and requested providers' participation,
both in distributing the educational materials to their patients
and in counseling them against the use of infant walkers.
Materials were also distributed at local shopping malls,
libraries, and post ofces. In addition, the Monroe County
Department of Health sent educational materials to parents
with newborn children's birth certicates. To supplement
these efforts, members of the coalition gave educational
presentations to several childcare agencies, volunteer agencies, and medical societies. Through these efforts, several
non-prot childcare agencies limited or curtailed walker
use; for example, the local Salvation Army agreed to discontinue the re-distribution of walkers through their stores.
In order to decrease the number of walkers in use, local
retailers were asked to voluntarily discontinue the sale of
infant walkers. Following informational meetings with the
managers of the three largest local retailers of baby products,
they agreed. These merchants also agreed to support a 1-day
``walker roundup'', at which, in return for a chance at one of
three donated coupons for US$ 100 of merchandise, families
were asked to turn in their infant walkers for destruction. A
press conference promoting the event and the anti-walker

message was held in a university hospital PedED; coalition

members and a mother and her child who had been injured in
an infant walker were featured. Several local radio and
television stations, newspapers, and a parent magazine
reported on the press conference. Additional materials promoting the ``walker roundup'' were posted at area malls,
retailers, libraries, and other public places. On the day of the
roundup, extensive local media coverage was again present,
resulting in extensive local and limited national coverage.
2.2. Evaluation
The combined setting for this study included the two
tertiary care PedEDs with trauma center capabilities serving
a county population of 716,000. Physicians (pediatric emergency medicine and surgical faculty and resident housestaff)
stafng both of these centers are afliated with the School of
Medicine and Dentistry, University of Rochester, NY.
The primary outcome measure was a comparison of the
mean annual numbers of children evaluated for walkerrelated falls down stairs at the two PedEDs for the
30 months before (1 January 1994 to 30 June 1996) to
the 12 months after (1 January 1997 to 31 December 1997)
the intervention. The 6-month intervention implementation
period (1 July 1996 to 31 December 1996) was excluded
from the analysis. Case ascertainment by retrospective chart
review involved a computerized search of all PedED visits
for children <18 months of age with ICD-9 external cause of
injury E-code 880.9 (designating a fall down stairs), followed by a manual chart review by the investigators to
determine the role of an infant walker. Coding practices
were stable throughout the study period and no attempts
were made to access other sources for data.
Mean annual numbers of PedED visits pre- and postintervention were compared using the F-test [22]. Using
two-tailed tests, differences were considered signicant for
P < 0:05. The University of Rochester Research Subjects
Review Board approved this study.
3. Results
During the 30 months prior to the intervention,
132 children (52.8 annually) were evaluated in the PedEDs
for falls down stairs while in an infant walker (see Table 1).
During the 12 months following the intervention, 38 children were evaluated for infant walker-related falls, a 28%
mean annual reduction (P < 0:001). Although the large
majority (90%) of these 160 children were treated and
released from the PedEDs, two children sustained traumatic
brain injuries requiring intensive care unit admission, while
one, described elsewhere [23], required surgery for abdominal injury.
During the same pre- and post-intervention time periods,
the frequency per year of all PedED visits for children
<18 months of age decreased 20.5% (7159 versus 5689,

G.P. Conners et al. / Patient Education and Counseling 46 (2002) 169173

171

Table 1
Children of age <18 months evaluated in two pediatric emergency departments before and after an intervention to reduce baby walker injuries
Before intervention

After intervention

P-value

Time period (month)

Total patients (age < 18 months)
Non-walker-related patients (age < 18 months)

30
17897 (7159 per year)
17765 (7106 per year)

12
5689
5651

<0.001
<0.001

Falls down stairs

Walker-related
Non-walker-related

132 (52.8 per year)

126 (50.4 per year)

38
48

<0.001
0.36

Total

258 (103.2 per year)

86

<0.001

P < 0:001). The annual number of PedED visits for children

<18 months of age for reasons other than walker injuries
decreased 20.5% (7106 versus 5651, P < 0:001). Similarly,
when annual PedED visits for all falls down stairs for
children <18 months of age were compared, there was a
16.7% annual reduction (103.2 versus 86, P < 0:001). However, visits per year for falls down stairs unrelated to walkers
remained unchanged pre- and post-intervention (50.4 versus
48, P 0:36).
4. Discussion
Despite the well-known risk of injury from the use of
infant walkers, walker use remains popular and associated
injuries are common, occurring in 1240% of users
[2,7,11,24]. In one of the earliest such studies, Kavanagh
and Banco reported in 1982 that 31% of the Rochester, NYarea children they studied who used walkers had suffered
walker-associated injuries [1]. Despite the compelling message of these local data and national anti-walker efforts
[11,21], infant walker-associated injuries have persisted in
the Rochester community. Prior to our intervention, one
child per week required evaluation in one of the two
Rochester PedEDs.
The American Academy of Pediatrics recommends media
campaigns ``to educate parents about the hazards and lack of
benets of walkers'' and community programs to destroy
existing walkers [11]. This intervention met these recommendations, as have similar projects in Seattle, WA [25] and
Winnipeg, Manitoba (Tenenbein, personal communication).
The number of walkers destroyed at the ``walker roundup''
was small (N 32) compared with the number in use.
However, our educational intervention likely led to destruction of walkers by parents separate from the walker roundup;
several families have reported to us, in a non-systematic
way, that, although they did not attend the ``walker
roundup'' they destroyed their child's walkers on learning
of the danger of injury. Following the intervention, there was
a signicant decrease in the number of walker-related falls
down stairs presenting to the PedEDs.
Coincident with the decrease in PedED visits for falls
down stairs associated with walker use, all PedED visits for
children <18 months of age decreased. The pressures of

managed care in decreasing PedED utilization may partially

explain the decrease for all visits; it is unlikely, however, that
the drop in census fully accounts for the decrease in PedED
visits due to falls down stairs associated with walkers. The
mechanism of injury alone, that being a fall down a ight of
stairs, would likely prompt most pediatricians and emergency medical technicians to direct their young patients to a
PedED with trauma center capability. The data from this
study demonstrate that almost all of the decrease in PedED
visits for falls down stairs for young children was due to the
decrease in falls associated with walkers, while falls down
stairs unrelated to walkers were unchanged, pre- and postintervention. We thus conclude that the measured reduction
in the number of infant walker-related falls down stairs
evaluated in the PedEDs likely reects a genuine decrease
in their number.
4.1. Limitations
Based on our historical comparison, infant walker-related
injuries decreased signicantly during the study period.
Because of the nature of our intervention, we were unable
to identify a relevant untreated concurrent ``control''
PedED. Inferring causality based on a comparison solely
with a historical control, however, is problematic in a
changing environment. Without this contemporary comparison group, we cannot conclude that our intervention was the
sole cause of the decline in walker-related injuries. Other
factors that could have accounted for this decline may
include other educational walker-related efforts, replacement of mobile walkers with stationary walker-like products, and the effects of the US Consumer Product Safety
Commission's voluntary guidelines for safer walker design
[26]. That the relevant clinical environment was changing is
suggested by the National Electronic Injury Surveillance
System's estimates of annual numbers of infant walkerrelated injuries of children treated in US Hospital Emergency Departments during the study period: 24,682 in 1994,
20,477 in 1995, 17,047 in 1996, and 16,009 in 1997 [20].
Although these estimates must be interpreted with caution
[3,20], they suggest a nation-wide 35% decrease in the
number of infant walker-related injuries between 1994
and 1997. A comparison of the 1997 estimate with an
average of the 1994, 1995, and half the 1996 estimate,

172

G.P. Conners et al. / Patient Education and Counseling 46 (2002) 169173

paralleling our study methodology, reects a 25% decrease

nationally, similar to our nding of a 28% decrease. Separating the effects of our intervention from those that would
have occurred otherwise as part of a national trend is thus
difcult, especially when the possible role of our intervention (and similar interventions elsewhere) as contributors to
this national trend is considered.
This study focused on infants presenting to PedEDs
due to falls down steps. We did not study the effect of
this intervention on the numbers of children evaluated for
walker-related injuries by other physicians and other local
emergency departments. Similarly, the effect of the intervention on other walker-related injuries, including samelevel falls, thermal and electrical burns, and toxic ingestions
was not measured; in other studies, these typically comprise
about 10% of injuries associated with walker use leading
to medical evaluation [12]. Thus, the true number of walkerrelated injuries in our area is probably higher, and it is
possible that our community-wide intervention may have
been more effective than we report.
Finally, the retrospective nature of this study exposes the
case denition to measurement bias. It is possible that
following the intervention, parents and physicians were
more likely to be aware, recall, and record in the medical
chart the role of infant walkers in fall-related injuries.
Interpreted in this fashion, measurement bias would have
diminished the true effectiveness of this intervention. Alternatively, parents may have been reluctant to admit the role a
walker played in their child's injury and some parents may
have avoided bringing children to PedEDs that had participated in well-publicized anti-walker campaigns. The direction of this bias would exaggerate the true effectiveness of
the intervention.
5. Conclusions
Following a well-publicized community-wide multifaceted educational intervention to promote parent and physician awareness of the dangers of infant walkers, PedED
visits for walker-related falls down stairs declined by 28%.
In light of a concurrent decline in walker-related injuries
nationally, we are unable to determine the magnitude of
this decline solely attributable to the intervention. Even if
entirely attributable to the intervention, the magnitude of
this reduction, which is consistent with those of other
successful similar interventions [26], suggests that educational efforts of this nature may signicantly reduce, but not
eliminate, the targeted injury.
It remains to be seen if manufacturers' voluntary reengineering of infant walkers sold in the US will have
any effect on decreasing the injury rate. These standards
require new walkers to be either too wide to t through a
standard-size doorway or to have a braking device that stops
at the edge of a step. These safety-enhancing revisions to the
American Society for Testing and Materials (ASTM) stan-

dard F977-96, entitled ``Standard Consumer Safety Performance Specication for Infant Walkers'' [27] were enacted
in 1996, and apply to walkers manufactured after 30 June
1997; infant walkers manufactured before that date remain
widely available. Even if redesigned walkers prove effective
in decreasing the number of falls, the wide availability and
extended durability of older infant walkers may delay by
years any signicant decline in injury rate. A more aggressive injury control policy should include enforcement of a
ban on sales and a recall of the many walkers already in use
[9,13]. Such a policy may be required to eliminate the
majority of these common, preventable, sometimes serious
injuries to infants.
5.1. Practice implications
As per the recommendations of the American Academy of
Pediatrics [11] and others [13], we performed an intensive,
community-based educational intervention designed to
reduce the large number of infant walker injuries in our
area. Although we are unable to fully determine the magnitude of the 28% measured reduction in PedED visits that is
attributable to the intervention, it is clear that this intervention fell short of eliminating walker-related injuries.
Although longer-term or even more intensive community
interventions might be more effective, it seems likely that
governmental policy changes are needed to profoundly
reduce numbers of walker injuries. The effects of the current
US voluntary standard remain uncertain. Even changes in
national standards alone may not prove as effective as
hoped; in Canada, walker-related injuries persisted despite
an effective national ban on new walker sales [6].
Acknowledgements
This project was supported in part by grants from the
Kessler Foundation and the Children's Hospital at Strong. It
was presented in part at the American Academy of Pediatrics
Annual Meeting, San Francisco, CA, 1621 October 1998; it
was also presented in part at the ``Health Policy and Health
Services Research: Nursing's Essential Contribution'' conference of the American Academy of Nursing and Columbia
University, New York, NY, 1214 May 1999. Thanks to
Regina Graziani for her administrative support of this
project.
References
[1] Kavanagh CA, Banco L. The infant walker: a previously unrecognized health hazard. Am J Dis Child 1982;136:2056.
[2] Fazen III LE, Felizberto PA. Baby walker injuries. Pediatrics
1982;70:1069.
[3] Weiss HB. Limitations of child surveillance data from the CPSC's
National Electronic Injury Surveillance System: the case of baby
walker related data. Inj Prev 1996;2:616.

G.P. Conners et al. / Patient Education and Counseling 46 (2002) 169173

[4] Petridou E, Simou E, Skondras C, Pistevos G, Lagos P, Papoutsakis
G. Hazards of baby walkers in a European context. Inj Prev
1996;2:11820.
[5] Stoffman JM, Bass MJ, Fox AM. Head injuries related to the use of
baby walkers. Can Med Assoc J 1984;131:5735.
[6] Morrison CD, Stanwick RS, Tenenbein M. Infant walker injuries
persist in Canada after sales have ceased. Pediatr Emerg Care
1996;12:1802.
[7] Mayr J, Gaisl M, Purtscher K, Noeres H, Schimpl G, Fasching G.
Baby walkers: an underestimated hazard for our children? Eur J
Pediatr 1994;153:5314.
[8] Coats TJ, Allen M. Baby walker related injuries: a continuing
problem. Arch Emerg Med 1991;8:525.
[9] Gleadhill DN, Robson WJ, Cudmore RE, Turnock RR. Baby
walkerstime to take a stand? Arch Dis Child 1987;62:4914.
[10] Sendut IH, Tan KK, Rivara F. Baby walker associated scalding
injuries seen at University Hospital Kuala Lumpur. Med J Malaysia
1995;50:1923.
[11] American Academy of Pediatrics. Committee on injury and poison
prevention. Pediatrics 1995;95:77880.
[12] Rieder MJ, Schwartz C, Newman J. Patterns of walker use and
walker injury. Pediatrics 1986;78:48893.
[13] Trinkoff A, Parks PL. Prevention strategies for infant walker-related
injuries. Pub Health Rep 1993;108:7848.
[14] Partington MD, Swanson JA, Meyer FB. Head injury and the use of
baby walkers: a continuing problem. Ann Emerg Med 1991;20:6524.
[15] Chiavello CT, Christoph RA, Bond GR. Infant walker-related
injuries: a prospective study of severity and incidence. Pediatrics
1994;93:9746.

173

[16] Smith GA, Bowman MJ, Luria JW, Shields BJ. Baby walker-related
injuries continue despite warning labels and public education.
Pediatrics 1997;100:e1.
[17] Mroz LS, Krenzelok EP. Examining the contribution of infant
walkers to childhood poisoning. Vet Hum Toxicol 2000;42:3940.
[18] Bar-on ME, Boyle RM, Endriss EK. Parental decisions to use infant
walkers. Inj Prev 1998;4:299301.
[19] Seigel AC, Burton RV. Effects of baby walkers on motor and mental
development in human infants. J Dev Behav Pediatr 1999;20:355
61.
[20] US Consumer Products Safety Commission. National Electronic
Injury Surveillance System product summary reports for 19941997.
[21] Board of Trustees. American Medical Association. Use of infant
walkers. Am J Dis Child 1991;145:9334.
[22] Brownlee KA. Statistical theory and methodology. In: Science and
engineering, 2nd ed. New York: Wiley, 1960. p. 1835.
[23] Conners GP, Weber CE, Emmens RW. Intussusception following a
baby walker injury. J Emerg Med 1999;17:26971.
[24] Thein MM, Lee J, Tay V, Ling SL. Infant walker use, injuries, and
motor development. Inj Prev 1997;3:636.
[25] Swanson MW, Rivara FP, Gomness JM. Baby walkers: hazardous
vehicles for infants, Wash Pub Health 1994;12:213.
[26] Sadan N, Raz A, Wolach B. Impact of community educational
programmes on foreign body aspiration in Israel. Eur J Pediatr
1995;154:85962.
[27] American Society for Testing and Materials. Standard Consumer
Safety Performance Specification for Infant Walkers (ASTM F97796). Philadelphia, PA: American Society for Testing and Materials,
1996.