Article Text
PDF
Abstract
Background Although effective when used correctly, child restraint systems (CRS) are commonly misused. Caregivers must make accurate judgements about the quality of their CRS installations, but there is little research on the psychological, technological, or contextual factors that might influence these judgements.
Methods Seventy-five caregivers were observed installing a CRS into a vehicle and completed self-report surveys measuring risk appraisals, previous utilisation of CRS resources, task difficulty, and confidence that the CRS was installed correctly.
Results Approximately 30% of caregivers installed the CRS inaccurately and insecurely, but reported that it was correctly installed. Predictors of confidence were ease of use (β=0.47) and exposure to CRS resources (β=−0.34). Installation errors and CRS security were unrelated to caregivers’ confidence.
Conclusions An interdisciplinary approach is needed to understand factors influencing caregivers’ judgements about their installations, optimise channels to connect caregivers to CRS resources, and to design safety technologies in light of these findings.
Statistics from Altmetric.com
Ninety-six countries have laws requiring children ride restrained; in the USA these laws are enacted at the state level.1 Despite these laws and the recognised safety benefit of a correctly installed child restraint system (CRS), misuse of CRS remains high.2–,5 Installing a CRS correctly can be a challenging task requiring strength, agility, literacy, and comprehension of instructions.6–8 A recent nationally representative study found that 90% of caregivers reported being confident or very confident that they had correctly installed the CRS and properly harnessed the child into the CRS.9 In contrast, estimates of safety-compromising CRS misuse were as high as 72%.2
As noted by the National Highway Traffic Safety Administration, this dichotomy between over-confidence and actual practice is problematical because over-confident caregivers will probably not seek help in ensuring that their CRS is properly installed.9 Currently, we do not understand why caregivers believe they correctly installed the CRS when in fact they did not. Elucidating the predictors of these judgements can inform the development of better technologies and psycho-educational interventions designed to enable caregivers to install CRS correctly and identify when they need assistance.
CRS checkpoints, staffed by child passenger safety technicians (CPST), were developed to address installation problems. Although checkpoints can be a useful tool, research suggests they are not used as much as the misuse statistics dictate they should be.7 ,10 Reasons for this underutilisation are hypothesised to be related to caregivers’ overestimating the likelihood that they installed the CRS correctly combined with underestimating the injury risk to the child in the event of a crash.7 Therefore, caregivers’ inability to judge effectively whether CRS were installed correctly might have a detrimental effect on child occupant safety.
In addition to checkpoints, other formal resources for CRS information include policy statements, guidance from primary care physicians, websites, brochures, and CRS and vehicle manuals. Although less well understood, caregivers might get information through informal channels such as conversations with friends, relatives and co-parents. As a whole, we know little about how caregivers use CRS resources in the course of their day-to-day lives, especially informal resources, and how use of these resources might affect caregivers’ cognitions and behaviours regarding child passenger safety, including misuse. International research suggests that social deprivation, broadly defined as a lack of access to injury prevention or health promotion resources and infrastructures at the community level, can play a significant role in explaining sociodemographic disparities in the rates of unintentional injuries and deaths to children.11 Therefore, understanding caregivers’ access to and use of available resources is important.
To address these gaps, the goals of the current study were to examine whether exposure to CRS resources, risk appraisals about injury likelihood, indicators of correct CRS installation (ie, security and accuracy), and perceptions about the ease of installation were associated with caregivers’ confidence in the quality of their CRS installations.
Methods
Study design and sample
This study consisted of a one-time observational assessment in which 75 adult caregivers installed a forward-facing CRS in a vehicle and completed a psychosocial and sociodemographic survey as part of a larger study (n=125) examining the effect of CRS technologies on adults’ ability to install a CRS effectively among experienced (n=75) and inexperienced (n=50) adult installers. The 50 inexperienced participants were not included in the current analysis, as by definition they did not have exposure to CRS resources, a key component of the research objective.
A convenience sample was recruited from primary care practices affiliated with a large urban children's hospital in the northeastern USA. In addition, fliers were posted in various community locations, including childcare centres, college campuses and workplaces. Data collection occurred from June to August 2012. ‘Experienced’ participants had to be regular caregivers to at least one child under 4 years of age, transported this child at least two times a week, and have installed any type of CRS a minimum of five times within the previous 6 months. Participants received US$25 for their time and effort. Study visits occurred in a hospital parking lot or at participants’ homes depending on their preference. Participants were given the choice of completing the installations using either their personal vehicle or the study vehicle, a contemporary standard minivan with rear bench seating. All but one participant used the van.
Materials
A commercially available forward-facing CRS with a five-point harness system was used for the observational installation assessment. The CRS is designed for children who weigh between 11 and 38 kg and stand 76–145 cm tall. Attachment to the vehicle can be achieved with either the vehicle seat belt or lower anchors and tethers for children (LATCH), a dedicated attachment method for CRS installation. Although the CRS could also be used as a belt-positioning booster seat for children weighing at least 18–54 kg and standing 107–165 cm tall, the CRS was not evaluated in booster mode. For the purpose of this study, installation instructions and graphics were initially created by the CRS manufacturer and revised by the study team to fit on one printed page following usability guidelines.12 Final instructions consisted of six steps with complementary diagrams. The Flesch–Kincaid reading level of the instructions was 7.3 (7th grade reading level), slightly higher than the 5–6th grade reading level recommended by the National Institutes of Health for health communication materials,13 but less than most CRS instructions that range from 7th to 12th grade and have an average reading level of 9th or 10th grade.6 ,12
Procedures
Observational assessment
Participants were instructed to install the CRS in the second row of the van, use whichever installation method (seat belt vs LATCH) felt more comfortable, not to use the top tether, take as much time as needed to read the instructions, and reference the instructions as many times as they wished during the installation. The installation began when the participant said he or she was ready to begin and concluded either when the participant said they were finished or when 30 min had elapsed.
CRS installations were assessed for accuracy and security. The CRS installation was deemed accurate if none of the following nine errors occurred: twisted seat belt; twisted LATCH anchor straps; non-approved position for use of LATCH anchors; attachment of LATCH anchor strap to non-anchor point in the vehicle; use of both LATCH and seatbelt; incorrect seatbelt routing; incorrect routing of LATCH anchor belt; inability to attach the CRS to the vehicle, or other error (eg, did not buckle seat belt). A secure installation was defined as lateral and forward movement less than 2.54 cm (1 inch). To obtain this measurement, masking tape was used to mark the initial location of the CRS on the vehicle seat and then the CRS was pulled in each direction at the belt or LATCH anchor points. The resulting amount of forward and lateral movement was assessed at the seat bight (where the seat cushion and seat back come together) using a standard measuring tape and was recorded on the observation sheet. All observational assessments were performed by a CPST.
Surveys
Sociodemographic information, risk appraisals and use of CRS resources were measured before the observational assessment. Risk appraisals about crash and injury likelihood were measured using item constructions typically used as part of the Theory of reasoned action/planned behavior:14 ‘I am likely to get in a crash in the next 1 month when I'm driving in the car by myself’, ‘I am likely to get in a crash in the next 1 month when I'm driving in the car with my child(ren)’, and ‘If you were to get into a crash when driving with your children, how confident are you that your child would be injury-free?’ Response choices were on a five-point scale: strongly disagree (1) to strongly agree (5) for the first two items and not at all confident (1) to very confident (5) for the third item.
Exposure to CRS resources was measured by two items that asked parents where they obtained information about child passenger safety and how they learned to install a CRS. In response to: ‘Where do you get information about child passenger safety?’ caregivers selected from the following list: primary care provider, friend, relative, brochure/book, class, online and other. In response to: ‘How did you learn how to install a child safety seat?’ caregivers selected from: CRS manual, vehicle manual, CPST, friend, relative, brochure, class, online and other. Four variables were created using these two items: quantity of CRS resources (ie, total number of resources identified); passive formal exposure (sum of exposure to passive formal sources of information; eg, brochure, online); active formal exposure (sum of exposure to active sources of information; eg, CPST, primary care provider); and informal exposure (sum of exposure to informal sources of information; eg, friend, relative). We conceptualised active exposure as resulting from an interaction between the caregiver and another person and passive exposure as the unidirectional consumption of information in which the caregiver read or listened to information without a social interaction. Formal resources were conceptualised as resources that were developed with the intention of the communication of CRS-related information potentially by a member of an agency or organisation, and informal resources as information provided by laypersons who may know the caregiver personally or socially.
After each CRS installation, caregivers rated the ease of installation, very difficult (1) to very easy (5), and their confidence that the CRS was installed correctly, not at all confident to (1) to very confident (5). Paper surveys were reviewed for completeness and entered into a Research electronic data capture (REDCap) software system,15 a secure, web-based application designed to support data capture for research studies. The study was approved by the institutional review board of the Children's Hospital of Philadelphia. Participants received US$25 in remuneration for their time and effort.
Results
Sixty-eight caregivers were parents to at least one child 4 years or younger; the remaining seven caregivers were a grandparent, aunt/uncle, foster parent or step-parent. Eighty-seven per cent of the caregivers were women. Fifty-six per cent were non-hispanic black, 23% were non-hispanic white, and 21% were of another race/ethnicity. Caregivers ranged in age from 18 to 55 years old; M (SD)=31.63 (8.15). Sixty-four per cent had less than a 4-year degree. Forty-five per cent were married, 32% did not have a partner (never married, single or divorced), and 17% were living with a partner at the time of the study. Seventy-five per cent drove a car in the past week. All participants completed the installation in under 30 min; M (SD)=177.6 (12.3) seconds, just under 3 min.
Installation security and accuracy
Fifty-three per cent of caregivers achieved an accurate installation (ie, no accuracy errors) and 16% achieved a secure installation (ie, less than 1 inch or 2.54 cm of movement). On average there was: M (SE): 92.4 mm (12.3) of lateral movement and M (SE): 35.8 mm (2.8) of forward movement. Caregivers who achieved a secure installation had CRS that demonstrated: M (SE): 16.09 mm (3.53) of lateral movement and M (SE): 16.51 mm (3.54) of forward movement compared to caregivers with insecure installations whose CRS demonstrated: a M (SE): 107.63 mm (13.86) of lateral movement and M (SE): 39.62 mm (3.00) of forward movement. Seventy-seven per cent of caregivers installed the CRS using the seat belt, 14% used LATCH, and 8% used both (which was noted as an error); percentages do not add to 100% due to rounding. A more detailed accounting of the frequency of specific errors related to accuracy and security committed can be found in table 1. Thirty-two per cent of the caregivers had both an inaccurate and insecure installation and a confidence rating of ‘3’ (scale midpoint) or greater. Caregivers who completed the installation accurately reported greater perceived ease of installation: M (SD)=3.18 (1.15) compared to caregivers who did not install the CRS accurately: M (SD)=2.63 (1.06); F (1,73)=4.52, ɳ2=0.06, p=0.04.
Frequency of accuracy* and security errors
Exposure to CPS resources
Caregivers were asked to identify whether they had previous exposure to 15 different CRS and child passenger safety information resources. Caregivers reported using 20% of the total measured resources: M (SD)=2.90 (1.27), observed range=1–6. Of the four formal active resources caregivers reported using an average of: M (SD)=1.13 (0.77) resources, observed range=1–3, corresponding to a quarter of resources in this category. Seventeen per cent of the seven formal passive resources were utilised: M (SD)=1.21 (0.90), observed range=0–4. Finally, utilisation of the four informal resources: M (SD)=0.58 (0.90) was least common at 14%. We did not find an association between the total quantity of resources or exposure to a particular category of resources and the accuracy and security of CRS installations; using active resources was inversely associated with using passive resources, see tables 2 and 3.
Correlations among caregivers’ confidence, installation security, risk appraisals, and exposure to CRS resources (Spearman's r)
Association between exposure to CRS resources and CRS installation accuracy (ANOVA)
Risk appraisals
Caregivers perceived slightly less crash risk when driving with their child(ren) compared to when they drove alone, with child(ren): med (IQR)=1 (1,2) versus driving alone: med (IQR)=2 (1,3), Z=2.05, p=0.040. Beliefs that their child would be injury free in the event of crash were slightly above the scale mid-point (scale range 1–5): M (SD)=3.73 (0.92); higher values indicate greater confidence that the child would not be injured in the event of a crash. Stronger beliefs that their child would be injury free in the event of a crash were positively associated with confidence that the CRS was correctly installed: Spearman's r=0.26, p=0.022. A marginally significant inverse association was found between caregivers’ perceptions that children would be injury free in the event of a crash and perceived crash likelihood when driving with children in the car: Spearman's r=−0.22, p=0.054. As shown in table 2, greater beliefs that their child would be injury free in the event of the crash were associated with greater use of informal CRS resources: Spearman's r=0.24, p=0.034.
Predictors of installation confidence
Installation confidence was normally distributed across the study sample: M (SD)=3.20 (1.08). Hierarchical linear regression was used to identify predictors of caregivers’ confidence that the installation was completed correctly. Sociodemographic variables were entered in step 1, accuracy and security were entered in step 2 and step 3 respectively, ease of installation was entered in step 4, exposure to the total number of CRS resources was entered in step 5, and risk appraisals regarding injury likelihood was entered in step 6.
Caregivers’ confidence in their installation was not associated with the amount of accuracy or security of the installation: ΔR2=0.00, p=0.871 and ΔR2=0.00, p=0.455, respectively. Perceived ease of installation was positively associated with confidence, such that for every one unit increase in perceived ease of installation (1–5 Likert scale, higher values indicate greater ease), caregivers’ installation confidence increased by 0.47 scale points; ΔR2=0.22, p=0.000. For every one additional CRS resource, caregivers’ confidence decreased by −0.34 scale points; ΔR2=0.09, p=0.003. In total, the model accounted for 30% of variance in perceived confidence: F (9,65)=4.50, p=0.000 (see table 4).
Hierarchical linear multiple regression of predictors of caregiver confidence in a correct installation (n=75)
Discussion
Our results illustrate that CRS can be difficult to install accurately and securely for the average caregiver—just over half installed the CRS accurately and only one in six did so securely. Furthermore, in a population of caregivers with real-world experience installing CRS, those that installed the CRS incorrectly generally did not self-identify. Instead their judgements were more influenced by ease of use and previous exposure to CRS resources. This evidence suggests that caregivers are unlikely to self-identify that they installed a CRS incorrectly and need help.
We did find that exposure to a greater number of CRS resources was associated with less confidence. It is possible that exposure to a greater number of CRS resources educated caregivers about the difficulty that parents can have installing a CRS, but not how to do so correctly, and subsequently influenced their judgements about their CRS installations, or that caregivers with lower confidence regarding performing CRS installations sought out a greater number of resources. Our study design does not permit an assessment of either causal pathway; future experimental studies are needed to disentangle the direction of effects.
We also found that greater use of active forms of CRS information was associated with less use of passive resources. Access to active forms of formal CRS information (eg, CPST, or primary care provider) might be limited for some caregivers, especially those in socially deprived communities, and they compensate by using passive formal resources that are easier to obtain (eg, brochures and websites). This association could also be explained by individual differences related to how caregivers like to obtain information, passively or actively. Importantly, the caregivers might have used resources other than the ones we asked them about and we do not know anything about the accuracy of information provided by any of the sources we asked caregivers about in this study. Some sources might even have provided conflicting information. Exploratory research can help to determine what other CRS sources are available to and are utilised by caregivers and the accuracy of information by source type.
Overall, caregivers were fairly confident that their child would be injury free in the event of a crash; probably a realistic estimation on the individual level. These observations were positively associated with their confidence that they completed the CRS installation correctly in the installation task and inversely associated with perceptions about crash likelihood (ie, the more likely participants believed a crash might happen, the less confident they were that their child would be injury free in that crash), although these associations were small. If respondents reported on the quality of the installation of the CRS(s) they use in their daily life, as opposed to the CRS used for the purpose of the study, we might have observed stronger associations among perceived crash and injury risk. We did observe that risk appraisals related to injury and crash were associated, but we did not ask parents to consider the specific role that the CRS might play in keeping their child safe and if they were concerned about how inaccurate CRS installations might compromise their child’s safety. These questions represent important avenues for future research.
This study is not without limitations. All but one of the participants did not use their own vehicle to install the CRS. Although using the same vehicle standardised assessment procedures across participants, lack of familiarity with the vehicle could have affected participants’ perceptions about ease of use, confidence, and installation quality. As we did not use the standard instructions provided with the CRS, and because adults were asked to install a potentially unfamiliar CRS in a potentially unfamiliar vehicle, it is not recommended to compare the proportions of errors obtained in this study with others that assess real-world use. Given known issues with the readability of CRS instructions, we wanted to make it easier for participants to get the information they needed to install the CRS in the context of our study. Finally, we treated each type of error equally and did not consider that potential impact of the error on child injury. Future studies can examine these issues in more detail.
Conclusion
The results of the present study indicate that caregivers do not make accurate judgements related to the quality of their CRS installations in that they overestimate the security and accuracy of the installation. More research is needed on how to design CRS that are easier to install correctly, understand what information caregivers are using to determine if installations are correct, and if there are modifiable individual differences that can be targeted in order to make these perceptions more accurate. Finally, we need to understand how installations are performed and judged by installers in real-world contexts in a variety of cultures and socio-political contexts in order to target CRS misuse effectively internationally.
What is already known on the subject
-
CRS can be very effective when used properly. It is critical for caregivers to make accurate judgements regarding the quality of their CRS installations.
-
Interventions such as CRS check-ups require that caregivers self-identify that they need assistance installing a CRS.
What this study adds
-
This study identifies predictors of caregivers’ judgements of CRS installation quality.
-
Caregivers’ judgements of CRS installation quality are not significantly related to indicators of improper installation.
Acknowledgments
The authors would like to thank the network of primary care clinicians, their patients and families for their contribution to this project and clinical research facilitated through the Pediatric Research Consortium (PeRC) at the Children’s Hospital of Philadelphia. In addition, they wish to thank Dennis Durbin, for his input during the early stages of this project as well as his critical review and feedback on the manuscript, and Kristina Puzino and Danielle Grams for their assistance with data collection.
Footnotes
-
Contributors All authors approved the final version of the manuscript. In addition each author made the following contributions: JHM contributed to study conception and design, data acquisition, data analysis and interpretation, and drafting and revising the manuscript. AEC, MRZ and LMC contributed to data acquisition and interpretation, and made critical revisions to the manuscript. SS and KBA contributed to study conception and design, data interpretation, and made critical revisions to the manuscript.
-
Funding This work was supported by the Centers for Disease Control and Prevention grant: Easy to use electric lower anchor tether winch for child safety seats (grant number 2R44CE00180-02A1) awarded to SS and by the National Science Foundation (grant number NSF EEC-1062166).
-
Competing interests None.
-
Ethics approval This protocol was reviewed by the institutional review board of the Children's Hospital of Philadelphia and deemed ‘exempt’.
-
Provenance and peer review Not commissioned; externally peer reviewed.