Modeling improvements in booster seat use: a discrete choice conjoint experiment

Factors associated with booster seat use among Indigenous peoples in Canada

OBJECTIVE

Booster seat use among the general population remains relatively low, despite their effectiveness in preventing injury among children when involved in motor vehicle collisions. Given the prevention of injuries that booster seats provide, understanding the factors that hinder or facilitate the use of these seats is critical, particularly in communities that are often overlooked when conducting general population studies. To date, no studies have examined the prevalence and predictors of booster seat use among Indigenous peoples in Canada. The purpose of this study was to examine the use of booster seat use among Indigenous peoples across Canada and the factors that impact their use.

METHODS

Data were collected from a survey of participants from First Nations communities and organizations serving Indigenous peoples nationwide. Hypotheses arising from known predictors of booster seat use across the general population were tested using logistic regression models.

RESULTS

The strongest predictor of booster seat use, even when all other study factors were accounted for, was the reduction of barriers related to the use of booster seats, such as a child's resistance to being placed or staying in the passive safety restraint or a parent, guardian, or other caregiver being unwilling to use or unsure of how to install/setup the booster seat.

CONCLUSION

Most Indigenous participants consistently used booster seats to safely secure children being transported in vehicles. However, this compliance rate is well below that of the general population. Accessibility and affordability of child safety restraints and/or children's refusal to use booster seats, as well as having more than 1 child to secure, were identified as mitigating factors. Access to and the affordability of booster seats, coupled with clear and understandable information on how to use them, are critical components to compliance. Raising awareness among Indigenous peoples communities regarding the importance of using booster seats is imperative. To achieve this, Indigenous peoples must lead discussions to ensure that child safety strategies not only are based on research and best practices but are culturally connected and community driven. Through meaningful collaboration, vehicle-related injuries and mortality among Indigenous children can be significantly reduced.

An introduction to "discrete choice experiments" for behavior analysts

In this paper, we introduce discrete choice experiments (DCEs) and provide foundational knowledge on the topic. DCEs are one of the most popular methods within econometrics to study the distribution of choices within a population. DCEs are particularly useful when studying the effects of categorical variables on choice. Procedurally, a DCE involves recruiting a large sample of individuals exposed to a set of choice arrays. The factors that are suspected to affect choice are varied systematically across the choice arrays. Most commonly, DCE data are analyzed with a multinomial logit statistical model with a goal of determining the relative utility of each relevant factor. We also discuss DCEs in comparison with behavioral choice models, such as those based on the matching law, and we show an example of a DCE to illustrate how a DCE can be used to understand choice with behavioral, social, and organizational factors.

A Cross-Sectional Geographic Information Systems Study of a Pediatric Emergency Department Child Restraint System Distribution Program

INTRODUCTION

A pediatric ED program sought to promote injury prevention through distribution of child restraint systems. Program funds are paid for child passenger safety technician certification of all personnel. Pediatric emergency nurses distributed child restraint systems at hospital discharge and dedicated technicians at fitting stations. Researchers described program characteristics, developed a baseline understanding of program outreach using geographic information systems, and evaluated adherence to manufacturer guidelines with a sensitivity analysis.

METHODS

This retrospective cross-sectional study used distribution forms linked to hospital records from 2013 to 2016. Testing for differences used nonparametric methods. Median values and interquartile ranges for weight and height of children were compared with manufacturer guidelines. Geographic information systems visualized recipients' street addresses and motor vehicle crashes on an underlying base map.

RESULTS

There were 312 child restraint systems distributed: of which 179 (57.4%) at the hospital, 126 (40.4%) at fitting stations, and 7 (2.2%) missing a location. Among those on Medicaid, 64.4% received a child restraint system at the hospital compared with 35.6% at fitting stations (χ² = 5.40, P < .02). Fitting stations had limited outreach to rural residents. Finally, results from the sensitivity analysis showed that devices were issued according to manufacturer guidelines.

DISCUSSION

Despite the workplace pressures of clinical care, pediatric emergency nurses delivered educational information and demonstrated hands-on installation at similar rates to dedicated technicians. Distribution of child restraint systems through the hospital reached a uniquely underserved population. Further research should investigate methods to improve fitting station outreach among Medicaid recipients.

Does Controlling for Scale Heterogeneity Better Explain Respondents' Preference Segmentation in Discrete Choice Experiments? A Case Study of US Health Insurance Demand

Analyses of preference evidence frequently confuse heterogeneity in the effects of attribute parameters (i.e., taste coefficients) and the scale parameter (i.e., variance). Standard latent class models often produce unreasonable classes with high variance and disordered coefficients because of confounding estimates of effect and scale heterogeneity. In this study, we estimated a scale-adjusted latent class model in which scale classes (heteroskedasticity) were identified using respondents' randomness in choice behavior on the internet panel (e.g., time to completion and time of day). Hence, the model distinctly explained the taste/preference variation among classes associated with individual socioeconomic characters, in which scales are adjusted. Using data from a discrete-choice experiment on US health insurance demand among single employees, the results demonstrated how incorporating behavioral data enhances the interpretation of heterogeneous effects. Once scale heterogeneity was controlled, we found substantial heterogeneity with 4 taste classes. Two of the taste classes were highly premium sensitive (economy class), coming mostly from the low-income group, and the class associated with better educational backgrounds preferred to have a better quality of coverage of health insurance plans. The third class was a highly quality-sensitive class, with a higher SES background and lower self-stated health condition. The last class was identified as stayers, who were not premium or quality sensitive. This case study demonstrates that one size does not fit all in the analysis of preference heterogeneity. The novel use of behavioral data in the latent class analysis is generalizable to a wide range of health preference studies.

A comparison of parent and child perspectives about barriers to and facilitators of bicycle helmet and booster seat use

To date, little work has compared similarities and differences between parent and young people's perceptions of barriers to and facilitators of bicycle helmet and booster seat usage. Our goal was to conduct such a comparison in order to inform future safety campaigns. Eleven focus groups with a total of 68 parents and 11 focus groups with a total of 76 young people were conducted. Recruitment was conducted and focus groups were held in diverse neighbourhoods to facilitate participation by families from a variety of cultural, linguistic, and socioeconomic backgrounds. Overall, parents and their children agreed on 50% of the barriers identified for bicycle helmet use and approximately 40% of the barriers for booster seats. Barriers common to both types of equipment for parents and children included comfort, style and design, and fear of teasing. Common facilitators included perceived safety, and comfort. While there was considerable overlap between the perspectives of parents and young people, there were also differences, underscoring the importance of addressing both perspectives. The barriers and facilitators identified were modifiable to a large extent; based on these, recommendations for future injury prevention campaigns were outlined.

Situational Use of Child Restraint Systems and Carpooling Behaviors in Parents and Caregivers

Suboptimal compliance with child restraint system (CRS) recommendations can increase risk for injury or death in a motor vehicle crash. The purpose of this study was to examine scenarios associated with incomplete CRS use and non-use in children ages 4⁻10 years. We used a cross-sectional online survey with a convenience sample of parent/caregivers from the United States, age ≥18 years, with a child age 4⁻10 years in their home, who could read and spoke English, and drove child ≥6 times in previous three months. We used descriptive statistics and Mann-Whitney U to describe and compare the distribution of responses to situational use of CRSs among car seat users and booster seat users. We also used descriptive statistics and the Mann-Whitney U to describe and compare the distribution of responses to carpooling items among booster seat users and non-booster seat users. There were significant differences among those who reported most often using booster seats (n = 282) and car seats (n = 127) in situations involving rental cars, driving just around the corner, car too crowded to fit the CRS, not enough CRSs in the vehicle, the CRS is missing from the car, or the child is in someone else's car without a CRS (p < 0.05). Among those who reported most often using booster seats and who carpooled other children (n = 159), 71.7% (n = 114) always used a booster seat for their own child. When carpooling other children, booster seat users were significantly more likely to use booster seats for other children ages 4⁻10 than the non-booster seat users (p < 0.01). Continued education and programs surrounding CRS use is critical, particularly for children who should be in booster seats.

Promoting booster seat use for young children: A school-based intervention pilot study

Purpose

Misuse and/or lack of booster seat use are often associated with high rates of injury and death among school-aged children. This pilot study examined the efficacy and the potential effectiveness of a booster seat intervention in the classroom.

Methods

Two elementary schools participated (randomly assigned as one intervention school and one control school). At the intervention school, a certified car seat specialist and a police officer held an interactive booster seat session. The height and age for each child were recorded. Children received a certificate indicating whether they met the requirements for booster seat use and a postcard with car seat restraint specifications. Children in the control school received a brochure on car seat safety. Pre- and post-intervention self-reports were collected and booster seat use was observed.

Results

Observational findings showed a decline in booster seat use at the control school and an increase in use at the intervention school. Self-reports of booster seat use indicated a decline at both schools; however, cell sizes were too small to permit statistical analyses.

Conclusion

Anecdotally researchers found the sessions were easy to conduct and were well received by the children and could be easily integrated into programming in schools. Classroom sessions may have the potential to positively influence booster seat use among 6- to 8-year-olds.

Using Latent Class Analysis to Model Preference Heterogeneity in Health: A Systematic Review

BACKGROUND

Latent class analysis (LCA) has been increasingly used to explore preference heterogeneity, but the literature has not been systematically explored and hence best practices are not understood.

OBJECTIVE

We sought to document all applications of LCA in the stated-preference literature in health and to inform future studies by identifying current norms in published applications.

METHODS

We conducted a systematic review of the MEDLINE, EMBASE, EconLit, Web of Science, and PsycINFO databases. We included stated-preference studies that used LCA to explore preference heterogeneity in healthcare or public health. Two co-authors independently evaluated titles, abstracts, and full-text articles. Abstracted key outcomes included segmentation methods, preference elicitation methods, number of attributes and levels, sample size, model selection criteria, number of classes reported, and hypotheses tests. Study data quality and validity were assessed with the Purpose, Respondents, Explanation, Findings, and Significance (PREFS) quality checklist.

RESULTS

We identified 2560 titles, 99 of which met the inclusion criteria for the review. Two-thirds of the studies focused on the preferences of patients and the general population. In total, 80% of the studies used discrete choice experiments. Studies used between three and 20 attributes, most commonly four to six. Sample size in LCAs ranged from 47 to 2068, with one-third between 100 and 300. Over 90% of the studies used latent class logit models for segmentation. Bayesian information criterion (BIC), Akaike information criterion (AIC), and log-likelihood (LL) were commonly used for model selection, and class size and interpretability were also considered in some studies. About 80% of studies reported two to three classes. The number of classes reported was not correlated with any study characteristics or study population characteristics (p > 0.05). Only 30% of the studies reported using statistical tests to detect significant variations in preferences between classes. Less than half of the studies reported that individual characteristics were included in the segmentation models, and 30% reported that post-estimation analyses were conducted to examine class characteristics. While a higher percentage of studies discussed clinical implications of the segmentation results, an increasing number of studies proposed policy recommendations based on segmentation results since 2010.

CONCLUSIONS

LCA is increasingly used to study preference heterogeneity in health and support decision-making. However, there is little consensus on best practices as its application in health is relatively new. With an increasing demand to study preference heterogeneity, guidance is needed to improve the quality of applications of segmentation methods in health to support policy development and clinical practice.

Preference phenotypes to facilitate shared decision-making in rheumatoid arthritis

OBJECTIVE

Implementing treat-to-target (TTT) strategies requires that patients with rheumatoid arthritis (RA) and their rheumatologists decide on how best to escalate care when indicated. The objective of this study was to develop preference phenotypes to facilitate shared decision-making at the point of care for patients failing methotrexate monotherapy.

METHODS

We developed a conjoint analysis survey to measure the preferences of patient with RA for triple therapy, biologics and Janus kinase (JAK) inhibitors. The survey included seven attributes: administration, onset, bothersome side effects, serious infection, very rare side effects, amount of information and cost. Each choice set (n=12) included three hypothetical profiles. Preference phenotypes were identified by applying latent class analysis to the conjoint data.

RESULTS

1273 participants completed the survey. A five-group solution was chosen based on progressively lower values of the Akaike and Bayesian information criteria. Members of the largest group (group 3: 38.4%) were most strongly impacted by the cost of the medication. The next largest group (group 1: 25.8%) was most strongly influenced by the risk of bothersome side effects. Members of group 2 (11.2%) were also risk averse, but were most concerned with the risk of very rare side effects. Group 4 (6.6%) strongly preferred oral over parenteral medications. Members of group 5 (18.0%) were most strongly and equally influenced by onset of action and the risk of serious infections.

CONCLUSIONS

Treatment preferences of patients with RA can be measured and represented by distinct phenotypes. Our results underscore the variability in patients' values and the importance of using a shared decision-making approach to implement TTT.

An ecological model to factors associated with booster seat use: A population based study

Belt-positioning booster seat use (BPB) is an effective technology to prevent severe child injury in cases of car crash. However, in many countries, age-appropriate car restraint use for children aged 4-7 years old remains the lowest among all age groups. The aim of this study was to identify the main determinants of BPB use through a comprehensive approach. An ecological model was used to analyze individual, parent-child relationships, and neighborhood characteristics. Parents of children enrolled in the first and second grades completed a self-reported questionnaire (n=745). The data were subjected to multilevel modeling. The first level examined individual and parent-child relationship variables; in addition the second level tested between neighborhood variance. According to parental self- reports, 56.6% of their children had used a BPB on each car trip during the previous month. The results indicated that the determinants positively related to BPB use were individual and parental; namely, the number of children in the family, the parents' car seat belt use, parental knowledge of children's car safety principles, and a highly authoritative parenting style. Children's temperaments and parental supervision were not associated with BPB use. At the neighborhood level, a small difference was found between neighborhoods for BPB users compared to non-users.

Do parental decision-making patterns predict compliance with use of child booster seats?

Booster seat use for 4-9 year olds remains the lowest of all age groups in many countries. The objective of this study is to examine whether parents' decision-making patterns, as measured by the Melbourne Decision Making Questionnaire, relate to car booster seat use. Israeli parents of 4-7 years old children (n = 398) answered a questionnaire about car safety and decision-making habits. Ninety per cent of parents reported having a booster seat; 70.5% reported consistent booster seat use in general and on short drives during the last month (booster seat use compliance index). Greater compliance index was positively related to a vigilant decision-making pattern, passenger compliance with rear seat belts and families with fewer children. Lower booster seat use compliance index was associated with buck-passing decision-making pattern. Health professionals and policy-makers should take into account parents' habitual decision-making patterns when designing interventions for car booster seat compliance.

Boosting safety behaviour: Descriptive norms encourage child booster seat usage amongst low involvement parents

Campaigns advocating behavioural changes often employ social norms as a motivating technique, favouring injunctive norms (what is typically approved or disapproved) over descriptive norms (what is typically done). Here, we investigate an upside to including descriptive norms in health and safety appeals. Because descriptive norms are easy to process and understand, they should provide a heuristic to guide behaviour in those individuals who lack the interest or motivation to reflect on the advocated behaviour more deeply. When those descriptive norms are positive - suggesting that what is done is consistent with what ought to be done - including them in campaigns should be particularly beneficial at influencing this low-involvement segment. We test this proposition via research examining booster seat use amongst parents with children of booster seat age, and find that incorporating positive descriptive norms into a related campaign is particularly impactful for parents who report low involvement in the topic of booster seat safety. Descriptive norms are easy to state and easy to understand, and our research suggests that these norms resonate with low involvement individuals. As a result, we recommend incorporating descriptive norms when possible into health and safety campaigns.

Modeling the hospital safety partnership preferences of patients and their families: a discrete choice conjoint experiment

BACKGROUND

Patients and their families play an important role in efforts to improve health service safety.

OBJECTIVE

The objective of this study is to understand the safety partnership preferences of patients and their families.

METHOD

We used a discrete choice conjoint experiment to model the safety partnership preferences of 1,084 patients or those such as parents acting on their behalf. Participants made choices between hypothetical safety partnerships composed by experimentally varying 15 four-level partnership design attributes.

RESULTS

Participants preferred an approach to safety based on partnerships between patients and staff rather than a model delegating responsibility for safety to hospital staff. They valued the opportunity to participate in point of service safety partnerships, such as identity and medication double checks, that might afford an immediate risk reduction. Latent class analysis yielded two segments. Actively engaged participants (73.3%) comprised outpatients with higher education, who anticipated more benefits to safety partnerships, were more confident in their ability to contribute, and were more intent on participating. They were more likely to prefer a personal engagement strategy, valued scientific evidence, preferred a more active approach to safety education, and advocated disclosure of errors. The passively engaged segment (26.7%) anticipated fewer benefits, were less confident in their ability to contribute, and were less intent on participating. They were more likely to prefer an engagement strategy based on signage. They preferred that staff explain why they thought patients should help make care safer and decide whether errors were disclosed. Inpatients, those with immigrant backgrounds, and those with less education were more likely to be in this segment.

CONCLUSION

Health services need to communicate information regarding risks, ask about partnership preferences, create opportunities respecting individual differences, and ensure a positive response when patients raise safety concerns.

Understanding Academic Clinicians' Decision Making for the Treatment of Childhood Obesity

BACKGROUND

Although most clinicians agree that obesity is a major problem, treatment rates remain low. We conducted this discrete choice experiment (DCE) to understand academic clinicians' decisions in treating childhood obesity.

METHODS

A total of 198 academic pediatric surgeons, pediatricians, family physicians, and allied health professionals were recruited from 15 teaching hospitals across Canada to participate in this DCE. Participants completed 15 tasks choosing between three obesity treatment scenarios to identify the scenario in which they would most likely treat pediatric obesity.

RESULTS

Latent class analysis revealed two classes with early intervention and late intervention preferences. Participants in the early intervention group (30%) were sensitive to variations in patient and family support. They would likely intervene if patients were obese, with normal lipid levels, were prediabetic, had high blood pressure, and when obesity was lifestyle associated. Late intervention clinicians (70%) were more likely to intervene if patients were morbidly obese, had abnormal lipid levels, required insulin for diabetes, had very high blood pressure, or when obesity impacted the patient's mental health. Simulations predicted that increasing colleague support for intervention, providing expert consultation, and mobilizing multidisciplinary support would increase the likelihood of treating pediatric obesity earlier from 16.1% to 81.5%.

CONCLUSIONS

This DCE was implemented to understand the factors clinicians use in making decisions. Most academic clinicians choose to intervene late in the clinical course when more-severe obesity-related morbidities are present. Increased support from colleagues, expert consultation, and multidisciplinary support are likely to lead to earlier treatment of obesity among academic clinicians caring for children.

Understanding preferences for disease-modifying drugs in osteoarthritis

OBJECTIVE

Numerous disease-modifying drugs for osteoarthritis (DMOADs) are under investigation. However, patients' preferences for drugs to prevent progression of OA are not known. The objective of this study was to quantify patient preferences for potential DMOADs.

METHODS

We administered a conjoint analysis survey to 304 patients attending outpatient general medicine and specialty clinics. All patients seated in the waiting rooms were asked if they would participate in a survey to elicit opinions about arthritis treatments. We performed simulations to estimate preferences for 4 options to prevent worsening of knee OA: best case (pill, highest benefit, lowest risk, lowest cost), worst case (infusion, lowest benefit, highest risk, highest cost), moderate subcutaneous injection (injection, mid-level benefit, mid-level risk, mid-level cost), and moderate infusion (same as subcutaneous injection except administered by infusion).

RESULTS

Subjects' median age was 57 years; 55% were women and 76% were white. Segmentation analyses revealed 4 patterns of preferences. A minority (5%) did not want to perform subcutaneous injections and would only consider DMOADs under the best-case scenario. Approximately 20% were risk sensitive and were willing to take DMOADs under the best-case scenario, but would start rejecting these medications as risk increased. A significant number rejected DMOADs under all conditions (16.4%); however, the largest segment (59.2%) had a strong preference for DMOADs across all scenarios.

CONCLUSION

Our results suggest that a significant percentage of a nonselected outpatient population might be willing to accept at least a moderate degree of risk in order to prevent worsening knee OA.

Buckle up safely (shoalhaven): a process and impact evaluation of a pragmatic, multifaceted preschool-based pilot program to increase correct use of age-appropriate child restraints

OBJECTIVE

To conduct a process and impact evaluation of a multifaceted education-based pilot program targeting correct use of age-appropriate restraints in a regional setting with a high proportion of Aboriginal and Torres Strait Islander families.

METHODS

The program was delivered in 2010 in 3 early learning centers where 31 percent of the children were of Aboriginal and Torres Strait Islander descent. Each component of the program was assessed for message consistency and uptake. To measure program effectiveness, participating children were matched 1:1 by age, language spoken at home, and annual household income with 71 children from the control arm of a contemporaneous trial. The outcome measure in the control and program centers (a 4-category ordinal scale of restraint use) was compared using ordinal logistic regression accounting for age of the parent.

RESULTS

Process evaluation found that though program components were delivered with a consistency of message, uptake was affected by turnover of all staff at one center and by parents experiencing difficulty in paying for subsidized restraints at each of the centers. Impact evaluation found that children from the centers receiving the program had nearly twice the odds of being in a better restraint category than children matched from the control group (adjusted odds ratio [ORadj] = 2.06, 95% confidence interval [CI], 1.09-3.90).

CONCLUSIONS

This was a pragmatic study reflecting the real-life issues of implementing a program in preschools where 57 percent of families had a low income and turnover of staff was high. Despite these issues, impact evaluation showed that the integrated educational program showed promise in increasing correct use of age-appropriate restraints. The findings from this pilot study support the use of an integrated educational program that includes access to subsidized restraints to promote best practice child restraint use among communities that include a high proportion of Aboriginal and Torres Strait Islander families in New South Wales. Future trials in similar settings should consider offering more support in centers with high turnover of staff and offering alternative methods of payment when families experience financial difficulties in purchasing the subsidized restraints. If proven in larger trials, this approach could reduce death and injuries in child passengers in this vulnerable group.