Objective. To assess the effectiveness of an emergency department (ED)-based home safety intervention on caregivers’ behaviors and practices related to home safety.
Methods. We conducted a randomized, clinical trial of 96 consecutive caregivers of children who were younger than 5 years and presented to an urban pediatric ED for evaluation of an acute unintentional injury sustained in the home. After completing a structured home safety questionnaire via face-to-face interview, caregivers were randomly assigned to receive either comprehensive home safety education and free safety devices or focused, injury-specific ED discharge instructions. Participants were contacted by telephone 2 months after the initial ED visit for repeat administration of the safety questionnaire. The pretest and posttest questionnaires were scored such that the accrual of points correlated with reporting of safer practices. Scores were then normalized to a 100-point scale. The overall safety score reflected performance on the entire questionnaire, and the 8 category safety scores reflected performance in single areas of home injury prevention (fire, burn, poison, near-drowning, aspiration, cuts/piercings, falls, and safety device use). The main outcome was degree of improvement in safety practices as assessed by improvement in safety scores.
Results. The intervention group demonstrated a significantly higher average overall safety score at follow-up than the control group (73.3% ± 8.4% vs 66.8% ±11.1) and significant improvements in poison, cut/piercing, and burns category scores. Caregivers in the intervention group also demonstrated greater improvement in reported use of the distributed safety devices.
Conclusions. This educational and device disbursement intervention was effective in improving the home safety practices of caregivers of young children. Moreover, the ED was used effectively to disseminate home injury prevention information.
Each year, millions of US children are injured in their own homes.1 Injuries from falls, poisonings, fires, burns, lacerations, and suffocations are leading causes of childhood morbidity and mortality.2,3 Injuries that occur in the home environment account for 4.1 million emergency department (ED) visits annually for children younger than 18 years.4 Repeat occurrence after first injury episode is common for young children and their siblings,5–8 despite that many such injuries are preventable. As evidenced by the inclusion of childhood injury prevention in the Healthy People 2010 objectives, reduction of home injuries is a national priority.9
Effective injury prevention requires an awareness of the common threats to young children as well as the means to protect against them, yet, frequently, parents lack the knowledge of effective countermeasures to prevent injuries to children and largely underuse protective safety devices.10–12 Therefore, it can be surmised that reduction of childhood injuries in the home environment could be achieved through widespread education and adoption of safety measures. As such, the American Academy of Pediatrics and national injury experts cite injury prevention as a priority area for counseling during routine health maintenance visits for young children.13,14
Early investigations conducted in the primary care setting have shown positive results with regard to parent knowledge acquisition, hazard recognition, the use of distributed safety devices, and, in a few studies, reduction of injury incidence. In a systematic review of the literature through 1991, Bass et al15 showed that 18 of 20 primary care–based auto occupant and home safety interventions demonstrated at least 1 positive result. However, a more recent review restricted to office-based clinical trials of home injury prevention concluded that counseling has little impact on practices that are designed to childproof the home.16 Moreover, a recent well-designed trial was unable to show a reduction in injuries after a randomized disbursement of a home safety intervention in the primary care setting.17 In addition, there is evidence to suggest that time constraints and other competing demands limit physicians’ abilities to deliver injury prevention counseling during routine health maintenance visits, and few families recall receiving important prevention education.18–20
We believe that the ED may represent an underrecognized site for the dissemination of targeted injury prevention information. To date, only Woolf et al21 have explored the ED setting as a venue for a home injury prevention intervention. In their study, the effectiveness of a randomly distributed intervention that increased poisoning prevention knowledge and practices was demonstrated.
We hypothesized that an educational and safety device disbursement conducted as part of the ED encounter after an acute injury sustained in the home may be an effective intervention for educating parents regarding prevention of such injuries. The aim of this study was to assess the effectiveness of an ED-based home safety intervention on caregivers’ practices related to home safety.
We conducted a randomized, clinical trial comparing comprehensive home safety education and safety device distribution with focused, injury-specific ED discharge instructions. Those who were eligible for inclusion were caregivers of children who were younger than 5 years and presented to an urban pediatric ED for treatment of acute, unintentional injuries sustained in the home. In addition to serving as a tertiary ED for a tristate region, the ED serves the emergency needs of the local urban community, which is predominantly lower income black. The following injury mechanisms were included: falls, burns/fires, cuts/piercings, poisonings, foreign body aspirations or ingestions, and near-drownings. These injury mechanisms were selected because of their associated high incidence or potential for high morbidity and mortality.2,3 Only English-speaking caregivers and those who resided within the city limits were considered for inclusion. Caregivers of children with injuries that did not occur in the home, were inflicted, or were fatal were excluded. A “caregiver” was defined as an adult person, at least 16 years of age, who cares for the child at least 8 hours per day, 3 days per week.
All participants received the usual verbal ED discharge instructions related to the type of injury sustained by the child. In addition, for ethical reasons, we provided all participants with a brochure entitled “Home Safety Tips: How to Make Your Home Safer for You and Your Child.” This handout was 2 pages and contains general information about the prevention of common household injuries to young children. Illustrative figures accompany text divided into 7 sections: fire safety, burn hazards, choking hazards, drowning hazards, fall prevention, poisoning prevention, and laceration prevention. The handout contains information and recommendations on all of the safety practices assessed in the study. Participants who were randomized to the control group received the handout with verbal counseling limited to prevention of the type of injury sustained by the child. Participants who were randomized to receive the intervention were provided comprehensive home safety counseling via a scripted, verbal review of the entire handout as well as the distribution and explanation of the contents of a home safety kit provided free of charge (estimated retail cost: $32). The contents of the home safety kit were cabinet latches, drawer latches, electrical outlet covers, tub spout covers, nonslip bath decals, bath water thermometer, small parts tester (choking tube), poison control telephone number stickers, and literature related to fire and window safety.
The study was approved by the Institutional Review Board of the study site. Written, informed consent of all participants was obtained. Research assistants in the ED enrolled consecutive participants between the hours of 8 am and midnight, 7 days per week during September through December 2001. Potential participants were identified, and a preliminary screening for eligibility was performed on the basis of the patient’s age, address, injury mechanism, and location of event. After the child’s medical evaluation, all consenting caregivers completed a structured home safety questionnaire via face-to-face interview administered by trained study personnel. The questionnaire consisted of closed-ended, multiple-choice questions and was designed to last ∼10 minutes. A multidisciplinary panel of injury experts determined the survey content using the available home safety literature and the American Academy of Pediatrics Injury Prevention Program22 recommendations. Topics included but were not limited to the injury mechanisms listed as inclusion criteria above. The use of specific devices was assessed, including those distributed as part of the intervention (listed below) as well as smoke and carbon monoxide alarms, stair gates, and gun locking devices. The questionnaire was pilot-tested with a similar group of caregivers from the surrounding community and modified to address any identified limitations.
After completion of the initial survey, study staff randomly assigned participants to intervention or control groups using a series of numbered, opaque envelopes that had been prerandomized in computer-generated blocks of 10. The allocation sequence and the block size were unknown to the staff who enrolled and assigned participants.
To assess the effect of the intervention, participants were contacted by telephone 6 to 8 weeks after the initial ED visit for repeat administration of the safety questionnaire. Trained study personnel different from those who enrolled participants in the ED conducted the follow-up interviews and were unaware of group status. Contact was considered unsuccessful when the telephone number provided by the caregivers at study entry was unusable (wrong or disconnected number) or when 5 unsuccessful attempts were made.
Each participant who finished the study completed 2 questionnaires, the pre- and posttests. The questionnaires were scored such that points were accumulated for desirable responses, and the accrual of more points correlated with the reporting of safer practices. Scores were then normalized to a 100-point scale. For each questionnaire, an overall safety score (OSS) was computed on the basis of performance on the entire questionnaire (51 items). In addition, individual questionnaire items related to single areas of home injury prevention were aggregated into category safety scores (CSSs). Eight CSSs were generated to reflect safety practices in the following areas of home injury prevention: fires (14 items), burns (12 items), poisonings (6 items), submersion (4 items), aspirations (6 items), cuts (6 items), falls (7 items), and safety device use (9 items). For the OSS and each CSS, the change in score (posttest CSS − pretest CSS) was calculated.
Statistical analyses were performed using STATA 6.0 software (www.stata.com). Continuous variables were summarized using mean, median, and range, and discrete variables were described using frequencies. Differences in mechanism of injury and demographic characteristics between the caregivers who completed the study and those who were lost to follow-up were assessed using χ2 or Fisher exact tests for categorical variables and t test for continuous variables. Between-group differences in OSS and CSSs were tested using t test. Exploratory analyses using multiple linear regression were conducted to evaluate the relationships between OSS and caregiver age, education, and employment status. In addition, stepwise multiple linear regression was used to evaluate the relative contributions of the CSSs to OSS variance within the intervention group.
Sample size was calculated on the basis of pilot data and assumed scenarios for the within-group correlation. On the basis of a mean (standard deviation) pretest OSS (on a 100-point scale) of 59.1 (13.6), a 2-tailed hypothesis test, an α level of .05, and correlations ranging from 0.1 to 0.6, we estimated that a difference of 10% could be detected by enrolling between 14 and 33 participants per group with a power of 80%.
Between September 6, 2001, and December 10, 2001, a total of 355 children who were injured by the specified mechanisms and were younger than 5 years were screened by study staff. Of these, exclusions were made for the following reasons: injury occurred at location other than primary residence (98), nonurban residence (45), suspected child abuse/neglect (7), primary caregiver absent (4), previous study enrollment (3), and not English speaking (2). Of the 196 remaining eligible participants, 39 were not approached because multiple eligible caregivers were in the ED simultaneously. Of those approached, 136 (87%) gave consent for enrollment. Randomization assigned 69 caregivers to the control group and 67 to the intervention group. All received the allocated intervention, although 1 of the caregivers in the intervention group left the safety kit behind after discharge from the ED. Two participants from the control group were later found to be ineligible and excluded because of nonurban residence. Of the sample, 48 (72%) caregivers in the control group and 50 (73%) caregivers in the intervention group were successfully contacted for repeat survey administration. Two caregivers (1 intervention and 1 control) declined participation in the posttest survey. In compliance with their requests, posttest data were not collected. For the purposes of analysis, these participants were treated as losses to follow-up. The caregiver who left the safety kit behind was included in accordance with an intention-to-treat analysis (Fig 1). Randomization resulted in comparable treatment groups with respect to caregiver and child demographics, caregiver educational level, homeownership status, and child injury mechanism. More caregivers in the control group were employed as compared with the intervention group (Table 1). At entry, the caregivers’ home safety practices were similar as indicated by equivalent OSSs and CSSs achieved on the pretest safety questionnaire (Table 2).
Ninety-six of the caregivers were contacted successfully for repeat administration of the survey. The median length of follow-up time was 68 days (range: 39-146). The length of follow-up did not differ by group assignment, and there was no relationship between follow-up time and posttest OSS (Pearson correlation coefficient: 0.15; P < .3). As a group, the caregivers who were lost to follow-up did not differ from those who were contacted successfully with respect to group assignment (P = .91), pretest OSS (P < .28), caregiver age (P < .47), caregiver relationship (P < .42), child age (P < .15), or injury mechanism (P < .11).
The intervention group demonstrated a significantly higher average OSS at follow-up than the control group (73.3% ± 8.4% vs 66.8% ±11.1; P < .002). For the CSSs, the intervention group showed significant improvement in poison safety score (P < .02), cut/piercing safety score (P < .001), and burn safety score (P < .03). There were no significant differences between the groups in score improvement for questions related to fall prevention (P < .79), water safety (P < .33), aspiration prevention (P < .12), and fire prevention (P < .61). Caregivers in the intervention group were more likely to improve reported use of the distributed safety devices than caregivers in the control group (65.4% ± 20.5 vs 44.3% ± 22.3; P < .001). Reported safety device use increased 55% from baseline in the intervention group (Table 2).
Relationship of Caregiver Characteristics to OSS
We conducted multiple linear regression analyses restricted to the participants in the intervention group to determine the associations between education, caregiver age, and employment status with improvement in safety score. We found that none of these characteristics predicted compliance as defined by improvement in OSS (caregiver age: P < .8; education: P < .7; employment: P < .2). Furthermore, the 3 variables explained only 4% of the variance (R2 = .04).
Relative Contributions of Improvement in CSSs to Improvement in OSS
A stepwise linear regression model of the 8 CSSs against the change in OSS within the intervention group demonstrated that 57% of the variance (R2 = .57) in OSS was explained by the change in device use score alone. This suggests that the benefit achieved by the intervention group was mostly attributable to the increased use of the distributed safety devices.
This study demonstrates the effectiveness of an educational and device disbursement intervention in improving the home safety practices of caregivers of young children who are at risk for injury in the home. Moreover, it is the first of its kind to use the ED setting for the dissemination of comprehensive home injury prevention information. Our results indicate that caregivers who were randomized to receive the intervention improved significantly in their overall practices regarding home safety, particularly within the areas of poisoning, cut, and burn prevention. In addition, caregivers who were provided with free safety equipment and instructions for proper use reported greater use of these devices as compared with caregivers in the control group.
Previous intervention studies have not shown consistently positive results in prevention of injuries sustained in the home. Several investigators have found that the disbursement of educational materials and/or inexpensive safety equipment increases parental safety knowledge and device use.15,23–25 Other results, however, have been contradictory. Dershewitz et al25 showed that primary care–based education had a limited effect on improving home safety and that parents who were given cabinet locking devices were no more likely to use them than control subjects. Similarly, Kendrick et al17 found no reduction in the number of injuries sustained after a randomized disbursement of safety advice and equipment. In a systematic review of office-based home safety clinical trials, DiGuiseppi and Roberts16 concluded that these studies have not been effective in influencing parents to adopt home safety measures. Some have suggested that a limiting factor may be the broad scope of education given through the interventions, as parents may be unable to absorb large amounts of diverse and frequently complicated information.13,16,18,26 The implication is that interventions might be more effective if they are simpler and more targeted.
The present study, which consisted of a broad childproofing intervention, was effective. Its novelty lies in its setting in an environment not traditionally used for home injury prevention education: the ED. This approach may have contributed to the success of the intervention by facilitating intervention when caregivers may be more receptive to preventive education, ie, during a “teachable moment.” As described in the early 1950s by Havighurst,27 the “teachable moment” is a key element of successful educating. It describes the phenomenon whereby learning is possible only when the time is right. Applied to health education, the “teachable moment” label has been used to describe naturally occurring health events that can motivate individuals to adopt risk-reducing behaviors.28 Behavioral theory provides additional insight into the potential advantage of the “teachable moment.” The Health Belief Model of behavior change contends that an individual’s motivation to adopt preventive health behaviors is influenced by perceptions of personal vulnerability and disease seriousness and commonly requires a triggering event.29–31 Therefore, the ED visit for an acute injury may represent the event necessary to nullify the misperceptions of invulnerability and to catalyze behavior change. For the promotion of countermeasures to promote home safety, the ED visit for a child’s acute injury may be a parent’s “teachable moment.”
Anticipatory guidance for injury prevention has been recognized as an integral part of the medical care provided for children. National injury experts and the American Academy of Pediatrics recommend specifically that parents of young children be advised about prevention of injuries sustained in the home,13,14 yet studies have shown that primary care physicians spend little time in injury prevention counseling, and few families receive important safety anticipatory guidance in the office-based setting.18–20 A commonly cited barrier to incorporating injury prevention counseling into routine well-child care is insufficient time.20,26 This suggests that additional opportunities for prevention counseling be sought. The ED visit for acute injury sustained in the home seems an opportune time to introduce or reemphasize home injury prevention strategies.
There are some limitations to consider in the interpretation of our results. First, the survey was based on self-reported safety practices that may have introduced reporting bias. Any potential reporting bias may have influenced the responses to items that assessed safety practices, resulting in our overestimation of the effectiveness of the intervention, especially with regard to safety device use. However, a recent study demonstrated a fairly high degree of consistency between self-reported home safety practices and those observed directly.32 Second, we used an intermediate measure for the true outcome of interest in that we measured improvement in safety practices rather than the less readily measured reduction of home injury events. However, our study has provided initial insight into the feasibility and effectiveness of an ED-based intervention using this intermediate outcome. Future investigations of ED-based home injury prevention interventions should aspire to measure reduction of injury as an outcome. Finally, this study was performed in a pediatric academic center ED. Time and staffing constraints in other EDs may limit the feasibility of reproducing this type of safety intervention. However, it should be noted that in this study, trained lay personnel provided all of the safety education in the ED. No additional work burden was placed on medical or nursing staff.
Pleas from advocates for a host of ED-based prevention programs abound amid national trends of increasing ED volume, acuity, and wait times.4 A brief injury prevention intervention targeted to high-risk individuals should have minimal negative impact on an already overburdened ED. Moreover, conducting an intervention in this environment may provide maximal benefit to families with demonstrated need. Although the ED cannot and should not supplant the role of the primary care physician’s office in providing anticipatory guidance, our study shows that the ED visit can be used effectively to disseminate home injury prevention information.
Financial support (only) was provided by Robert Wood Johnson Foundation.
- Received January 30, 2003.
- Accepted July 14, 2003.
- Reprint requests to (J.C.P.) Division of Emergency Medicine, Children’s Hospital of Philadelphia, 34th St and Civic Center Blvd, Philadelphia, PA 19104. E-mail:
- ↵National Safe Kids Campaign web site. Available at: www.safekids.org/tier2_rl.cfm?folder_id=174. Accessed January 21, 2003
- ↵Centers for Disease Control and Prevention web site. Available at: www.cdc.gov/ncipc/factsheets/childh.htm. Accessed January 21, 2003
- ↵Agran PF, Anderson F, Winn D, Trent R, Walton-Haynes L, Thayer S. Rates of pediatric injuries by 3-month intervals for children 0 to 3 years of age. Pediatrics.2003;111(6) . Available at: pediatrics.aappublications.org/cgi/content/full/111/6/e683
- ↵McCaig LF, Ly N. National Hospital Ambulatory Medical Care Survey; 2000. Available at: www.cdc.gov/nchs/data/ad/ad326.pdf. Accessed January 21, 2003
- ↵Sellar C, Ferguson JA, Goldacre MJ. Occurrence and repetition of hospital admissions for accidents in preschool children. BMJ.1991;302 :16– 19
- Bijur PE, Golding J, Haslum M. Persistence of occurrence of injury: can injuries of preschool children predict injuries of school-aged children? Pediatrics.1988;82 :707– 712
- Johnston BD, Grossman DC, Connell FA, Koepsell TD. High-risk periods for childhood injury among siblings. Pediatrics.2000;105 :562– 568
- ↵Litovitz TL, Flagler SL, Manoguerra AS, et al. Recurrent poisonings among pediatric poisoning victims. Med Tox.1989;4 :381– 386
- ↵US Department of Health and Human Services. Healthy People 2010: Understanding and Improving Health. 2nd ed. Washington, DC: US Government Printing Office; 2000
- ↵Cohen LR, Runyan CV, Downs SM, Bowling JM. Pediatric injury prevention counseling priorities. Pediatrics.1997;99 :704– 710
- ↵Committee on Injury and Poison Prevention. Office-based counseling for injury prevention. Pediatrics.1994;94 :566– 567
- ↵Bass JL, Christoffel KK, Widome M, et al. Childhood injury prevention counseling in primary care settings: critical review of the literature. Pediatrics.1993;92 :544– 550
- ↵Kendrick D, Marsch P, Fielding K, Miller P. Preventing injuries in children: cluster randomized controlled trial in primary care. BMJ.1999;318 :980– 983
- Quinlan KP, Sacks JJ, Kresnow M. Exposure to and compliance with pediatric injury prevention counseling–United States, 1994. Pediatrics.1998;102(5) . Available at: pediatrics.aappublications.org/cgi/content/full/102/5/e55
- ↵Cohen LR, Runyan CW. Barriers to pediatric injury prevention counseling. Inj Prev.1995;5 :36– 40
- ↵Woolf AD, Lewander W, Fillippone G, et al. Prevention of childhood poisoning: efficacy of an educational program carried out in an emergency clinic. Pediatrics.1987;80 :359– 363
- ↵The Injury Prevention Program (TIPP): Age related safety sheets. American Academy of Pediatrics. Available at: www.aap.org/family/2to4yrs.htm. Accessed January 21, 2003
- ↵Clamp M, Kendrick D. A randomized controlled trial of general practitioner safety advice for families with children under 5 years. BMJ.1998;316 :1576– 1579
- Kelly B, Sein C, McCarthy PL. Safety education in a pediatric primary care setting. Pediatrics.1987;79 :818– 823
- ↵Havighurst RJ. Human Development and Education. New York, NY; David McKay & Co; 1952
- ↵McBride CM, Emmons KM, Lipkus IM. Understanding the potential of teachable moments: the case of smoking cessation. Health Educ Res.2003;18 :156– 170
- ↵Watson M, Kendrick D, Coupland C. Validation of a home safety questionnaire used in a randomised controlled trial. Inj Prev.2003;9 :180– 183
- Copyright © 2004 by the American Academy of Pediatrics