Article Text

Download PDFPDF

Facilitators and barriers to child restraint use in motor vehicles: a qualitative evidence synthesis
  1. Soumyadeep Bhaumik1,
  2. Kate Hunter2,3,
  3. Richard Matzopoulos4,
  4. Megan Prinsloo5,
  5. Rebecca Q Ivers3,6,
  6. Margaret Peden3,7,8
  1. 1 Injury Division, George Institute for Global Health, New Delhi, India
  2. 2 Injury Division, George Institute for Global Health, Camperdown, New South Wales, Australia
  3. 3 School of Public Health and Community Medicine, University of New South Wales, Sydney, New South Wales, Australia
  4. 4 School of Public Health & Family Medicine, University of Cape Town, Cape Town, Western Cape, South Africa
  5. 5 Burden of Disease Research Unit, South African Medical Research Council, Tygerberg, South Africa
  6. 6 Ivers, George Institute for Global Health, Camperdown, New South Wales, Australia
  7. 7 George Institute for Global Health UK, Oxford, Oxfordshire, UK
  8. 8 Nuffield Department of Womens and Reproductive Health, University of Oxford, Oxford, Oxfordshire, UK
  1. Correspondence to Margaret Peden, George Institute for Global Health UK, Oxford OX1 2BQ, UK; margie.peden{at}


Background Road traffic collisions contribute a significant burden of mortality and morbidity to children globally. The improper or non-use of child restraints can result in children sustaining significant injuries in the event of a collision. Systematic reviews on the effectiveness of various interventions to increase the use of child restraints already exist but to the best of our knowledge, there has been no qualitative evidence syntheses on the facilitators and barriers to child restraint usage. This review aims to fill that gap.

Methods We searched for qualitative studies, which focused on perceptions, values and experiences of children, parents/caregivers or any other relevant stakeholders on the use of restraints for children travelling in motor vehicles in PubMed, EMBASE and Global Health and screened reference lists of all included studies. We assessed the quality of included studies with the Critical Appraisal Skills Programme (CASP) checklist and used the PROGRESS Plus lens for an equity focused analysis.

Results We identified a total of 335 records from searching the databases and five records from other sources. After screening, we identified 17 studies that met our inclusion criteria. All but one study (which had children as participants) focused on the perceptions, attitudes and barriers of parents or caregivers. The included studies were from three high-income (n=14) and one upper-middle income (n=3) country. In addition, although many focused exclusively on participants from culturally and linguistically diverse minorities, the issue of equity was not well addressed. Five major themes emerged from the analysis. (1) perceived risk for injuries and perceived safety benefits of child restraint usage varies in different settings and between different types of caregivers; (2) practical issues around the use of child restraints is a major barrier to its uptake as a child safety measure; (3) restraint use is considered as a mechanism to discipline children rather than as a safety device by parents and as children became older they actively seek opportunities to negotiate the non-usage of restraints; (4) adoption and enforcement of laws shape perceptions and usage in all settings and (5) perceptions and norms of child safety differ among culturally and linguistically diverse groups.

Conclusion The results of this systematic review should be considered when designing interventions to promote the uptake of child restraints. However, there is a need to conduct qualitative research around the facilitators and barriers to child restraint usage in low-income and middle-income countries. Furthermore, there is a need for more evidence conducted in semiurban and rural areas and to involve fathers, policy-makers, implementers and enforcement agencies in such studies.

  • interventions
  • uptake/adherence
  • child survival
  • systematic review
  • child
View Full Text

Statistics from


Death, disability and economic burden due to road traffic injuries are a major public health problem globally and are expected to increase for at least the next two decades.1 Children contribute a significant proportion of the burden2 and so addressing child road traffic injury risk factors is crucial.

Increasing the use of child restraints is one of the five key behavioural factors being tracked globally that is aimed at reducing the burden of road injury and is the only one specific to children.2 In addition, Target 8 of the Global Voluntary performance Targets for Road Safety, which was adopted by United Nations Member States through UN GA resolution 72/271 in April 2018, specifically aims to increase the use of standard child restraint systems to 100% by 2030.3 As of 2016 only 33 countries, almost all of them high-income countries (HIC), had adopted best practice legislation in relation to the use of child restraints.2 Ensuring the use of child restraints has been a major challenge despite the legal requirements in these countries.4–9 While systematic reviews on the effectiveness of various interventions to promote the use of child restraints have already been conducted,10 there has not been a synthesis of evidence to understand the facilitators and barriers to the use of child restraints. The current study, therefore, aims to synthesise qualitative evidence on the barriers to and facilitators of the use of restraints for children travelling in motor vehicles. To the best of our knowledge, this is the first qualitative evidence synthesis on the topic and will provide important insights to understand the implementation of best practices and to improve the design of programmes to increase child restraint usage.


While there are several approaches to qualitative evidence synthesis, the current review broadly follows the principles and guidance laid down in a recent series of papers by the Cochrane Qualitative and Implementation Methods Group.11–16 The protocol for the study was registered prospectively in PROSPERO (CRD42018102497).

Theoretical approach

Our approach in this review is from the ‘critical realism’ standpoint with a pragmatic lens. Critical Realism is a philosophical standpoint which recognises that an observable behaviour—in this instance, the use of a child restraint—is influenced by unseen phenomena such as values and expectations. While child restraint use is observable, the use of child restraints is not independent of values, attitudes, prior experiences, capacities and motivations. As such, through this philosophical standpoint, and wide qualitative evidence synthesis, we endeavoured to understand the diverse and multifaceted external reality in all its complexity and depth through perceptions and interpretations reported in individual qualitative studies. A pragmatic lens allows us to include individual qualitative studies irrespective of their ontological stance.

Criteria for considering studies for this review

Type of studies

We included primary studies that have used qualitative approaches for both data collection and data analysis. We recognised focus groups, individual in-depth interviews, ethnographic interviews, participant observation, diaries and other narrative methods of data collection as valid tools for qualitative data collection. We also recognised narrative analysis, thematic analysis, grounded theory, phenomenological approaches and discourse analysis as valid methods for qualitative analyses. We excluded studies that used qualitative methods for data collection but did not analyse the data qualitatively. We included qualitative studies irrespective of whether or not they were conducted in conjunction with the implementation of interventions for the use of restraints for children travelling in motor vehicles. We included mixed-methods studies only when both the methods and results for the qualitative component were reported separately.


We included studies that focused on the perceptions, values and experiences of children; parents and/or caregivers and any other relevant stakeholder involved in formulating policies or programmes or implementing child restraint programmes.


We included studies from anywhere in the world and in any setting provided that they met our inclusion criteria.

Types of phenomena of interest

We included studies where the primary focus were perceptions, attitudes and experiences of parents/caregivers and children and other stakeholders towards restraints for children travelling in motor vehicles.

Searching for qualitative evidence

We searched PubMed, EMBASE (EMBASE Classic 1947 to 1973 and EMBASE 1974 to 2018 July 26) and Global Health for eligible studies using search strategies developed for this purpose (online supplementary appendix 1) and updated up to 27July 2018. We also searched the reference lists of all included studies found by electronic database search. The search was not restricted by language.

Selection of studies

In the first phase, one review author (SB) independently assessed eligibility of primary studies based on titles and abstracts. We then acquired the full text of all papers identified as potentially relevant by at least one review author for independent review by two authors (SB and MP). We did not contact study authors for any further information because of pragmatic reasons and timelines.

Data extraction

We extracted data from studies using an extraction form that was developed iteratively. We extracted data on context, participants, study design and methods in addition to what was required for the quality appraisal of included studies and for thematic analyses. We extracted information on participants using the PROGRESS Plus framework. The framework (in the form of an aide-memoire) is used to explicitly understand factors that lead to health inequity. It assesses place of residence, race/ethnicity/culture/language, occupation, gender/sex, religion, socioeconomic status, social capital plus additional factors that may indicate disadvantage.17

Assessment of the quality of the included qualitative studies

We appraised the quality of the included qualitative studies by using the Critical Appraisal Skills Programme (CASP) quality assessment tool for qualitative studies.18 The tool, which has been recommended for use by the Cochrane Qualitative and Implementation Methods Group, comprises 10 questions (applied in table 1) and has been previously used in several other qualitative evidence syntheses studies.

Table 1

Methodological limitations of included studies using the CASP qualitative study tool

Syntheses methodology

We used thematic synthesis as defined by Thomas and Harden for the purpose of qualitative evidence syntheses.19 Box 1 presents the detailed rationale for the choice of thematic syntheses as defined in the RETREAT framework.20

Box 1

Retreat approach for selecting qualitative evidence syntheses approaches20

Review question: What are the barriers and facilitators for the use of restraints for children travelling in motor vehicles?

Epistemology: Critical realism with the lens of pragmatism.

Time/timeframe: 6 months.

Resources: Funded project with access to database and availability of qualitative software.

Expertise: Generic qualitative research skills; specialist generic evidence syntheses skills; specialist subject expertise.

Audience and purpose: Primarily academics but also other stakeholders, including policy-makers (health, transport, urban planning), civil society representatives and the private sector.

Types of data: Scoping done indicates availability of some conceptually rich and adequately thick studies. We did not do a comprehensive search during the scoping phase.

Chosen method: Thematic syntheses as outlined by Thomas and Harden.

Rationale for choice: This systematic review does not seek to contribute to existing theory but aims to inform current practice around the implementation of interventions on restraints for children travelling in motor vehicles. The rapid nature of the review and the team expertise are other factors that guided the choice of method.

We followed the standard methods outlined by the thematic approach wherein the entire text labelled as ‘results’ or ‘findings’ in study reports was used verbatim for further analyses. Broadly, this consisted of the following steps:

  • Coding text and developing descriptive themes. The review author conducted line-by-line coding using NVIVO in a set of three articles and developed a hierarchical coding framework and then applied this to newer articles. After every third article, the coding framework was revised iteratively, based on the identification of newer concepts. The final coding framework developed as an output of the iterative process was then applied to all included studies. Repeat checks, constant comparison and discussion between the reviewers was undertaken to ensure consistency. Similarities and differences between codes were constantly compared with develop descriptive themes.

  • Development of analytical themes. In the final level of synthesis, we developed higher-level analytical themes by inferring barriers and facilitators for the use of restraints for children travelling in motor vehicles going beyond the interpretations being offered by individual primary studies.

Differences in protocol and the full qualitative evidence synthesis are described in online supplementary appendix 2.


Results of the search

We identified a total of 335 records from the databases and five records from other sources. After removing 90 duplicates, we screened 250 articles based on their titles and abstracts only. We considered 24 full-text papers for inclusion in this synthesis and found 17 studies that met our inclusion criteria (figure 1).

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart of the qualitative evidence syntheses.

Description of included studies

Characteristics of the included studies are summarised in table 2 and the details are narratively summarised in subsequent sections.

Table 2

Characteristics of included studies

Focus of included studies

Most of the studies intended to understand perceptions, attitudes and barriers to the use of child restraints in motor vehicles. The qualitative studies were standalone projects except for three studies, which were undertaken in conjunction with an intervention.21–23

Type of studies

The studies included were either standalone qualitative methodologies or mixed methods. The different types of studies included were as follows.

  • Focus group discussions only.21 22 24–32

  • In-depth interviews only.33

  • Focus groups and in-depth interviews.34

  • Mixed-method study designs23 35–37

Study participants

All but two of the included studies targeted parents or caregivers of children as participants. Only one study34 included children in a qualitative component in addition to parents or caregivers. One study included known violators of the child restraints law, including parents and caregivers26 and two studies focussed on pregnant women.31 35 Many studies included participants from culturally and linguistically diverse minorities25 30 37—two looked specifically at low-income communities.29 32

Study setting

All the qualitative studies included were from HIC (USA=9, Australia=4, Saudi Arabia=1) or upper-middle-income countries (China=3) (UMIC). Studies were mostly community-based.

Methodological quality of the included studies

The methodological quality of the included studies as assessed by the CASP tool is summarised in table 1. Most studies were clear about their research aims, had justified the use of a qualitative design and the overall design was appropriate. The recruitment strategy was not clear in four studies.24 28 31 35 Nine studies24–26 28–31 36 37 were adjudged to not have sufficiently rigorous data analysis methods (and one study was adjudged unclear34 on several counts including for methods of analysis), reflexivity not being reported and data saturation not being reached or not reported. Three studies22 24 26 were adjudged to not have clear statement of findings since they did not present quotes of the participants to back-up their results and/or did not describe basic characteristics of participants.

Equity lens using PROGRESS Plus

We applied an equity lens on a posthoc basis using the PROGRESS Plus framework to better understand the equity issues around facilitators and barriers for child restraint use17 and this is shown in table 3. However, for many studies, the reporting was not detailed enough to enable a better understanding of equity. Almost all the studies were conducted in big cities in HIC and seven studies specifically focused on culturally and linguistically diverse groups.23 25 29 30 32 36 37 Studies typically did not intend to explore differences between mothers and fathers. Occupation, a key factor which might influence risk perception and risk-taking behaviour as well as socioeconomic status was not reported in 11 studies.23–27 29–31 33–35 Religion was mentioned only in one study;31 socioeconomic conditions were not reported in six studies.22 25 26 30 31 34 Social capital was reported only in one study.37 Even when they have reported equity factors, studies did not look specifically on its effect, except when they looked at culturally and linguistically diverse communities.

Table 3

Equity characteristics of participants in included studies using the PROGRESS Plus framework

Syntheses of findings

Five key themes emerged from the thematic analysis of the included studies. (1) Perceived risk for injuries and perceived safety benefit of child restraint usage varies in different settings and between different types of caregivers. (2) Practical issues around the use of child restraints was a major barrier to its uptake as a child safety measure. (3) Restraint use was considered a mechanism to discipline children rather than a safety device by parents and as children became older, they actively seek opportunities to negotiate non-usage of restraint. (4) Adoption and enforcement of laws shaped perceptions and usage in all settings. (5) Perceptions and norms of child safety differed among culturally and linguistically diverse groups.

Perceived risk for injuries and perceived safety benefit of child restraint usage varies in different setting and between different types of caregivers

A mixture of perceived risk of injuries and the perceived safety benefit of using child restraints to prevent road traffic injuries influenced uptake of restraints and this varied both across different contexts and between different types of care givers.

Many studies21 24 27–31 33–35 included participants who acknowledged that it was ‘dangerous for children to travel unrestrained’30 owing to a high risk of injuries. Participants also reported that the risk of injuries was dependent on the duration of rides, speed, driver experience (being ‘safe drivers’) and ‘safe areas’.24 26 28 29 33 34 36 Authors reported that among non-users of child restraints, child restraint use was determined by risk perception with authors concluding that a ‘short distance and slow speed in the city would not cause serious injury even in a crash; CSS is not that important’. This was evidenced by the quote from parents ‘We don’t ask my child to wear the seat belt because most of the time we are driving in the downtown area’.33 Mothers often mentioned that fathers, grandparents and other caregivers had lower risk perceptions and allowed children to be transported unrestrained which made enforcement difficult. In one study,25 ‘mothers felt it was important for more educational efforts targeted specifically at fathers, who often were working and therefore not present during potential teachable moments such as doctor visits, or school and community events’. Often fathers played an important role in decision-making on buying restraints, and convincing them was often difficult due to a lack of knowledge ‘I heard it and want to buy it, but I fail to persuade my husband because I don’t know much about it’ (non-restraint user mother).33 Participants in the study from Saudi Arabia mentioned that, ‘Allah is in control of everything that happens, and it is all in His hands. Whether or not I use a car seat will not matter’.31 Similar thoughts were expressed by Arab-speaking male participants in another study.30

The perceived safety benefits of child restraints was low in many settings.22 29 30 33 Concerns were raised about the balance between ‘being belted vs the risks of being trapped in the car’.29 35 Participants often expressed doubts on the preventability of particular types of restraints or belt-positioning. Booster seats and forward-facing restraints were mentioned to be untrustworthy by participants; ‘I don’t think they [belt-positioning booster seats] are safe—they are more dangerous than a car seat [forward-facing restraint]’.27 Cultural notions on what constitutes safety were also reflected when participants mentioned that holding children on laps provided better safety in case of a crash—‘It is a natural human response that people would hold children tightly when a car brakes. If a child is restrained in a safety seat, I feel I could not protect my child when a car suddenly brakes’.35

Practical issues around the use of child restraints was a major barrier to its uptake as a child safety measure

Participants in all studies mentioned several practical barriers in using child restraints. Cost was almost a universal cause of concern which limited use.21 24 25 27 29 30 34 36 Educational sessions were thought to be useful when accompanied by discounts or giveaways of restraints. Rebates were thought to be an impractical method because of the difficulty in obtaining rebates and the time delay. Discounts were also reported to be inadequate.24 A few studies reported the need for them to be provided completely free.23 37 Interestingly, costs were not reported to be a barrier in three studies from China. The difference in studies from China might be due to the fact that most participants in these studies had jobs or were earning at or above the average wage in the city (one of these studies did not report on socioeconomic or job status but authors reported cost was not a barrier).22 33 35

The issue of cost was also related to other practical constraints like family size. Restraints were thought to be like, ‘sardines in the back seat… (The car seats) take up too much space. In particular, full-back booster seats allow for only two children in a mid-sized car, with no room for extra passengers’.28 This was particularly problematic in cultures which do not have nuclear family set-ups or had large family sizes and in communities where parenting responsibilities were often shared or where providing or taking lifts was common.21 25 26 28–31 One study from Australia mentioned the importance of ‘learning from the authorised fitter while others felt paying for an authorised restraint fitter to fit the car seat was a waste of money’.36

The practicalities around restraints as children grow up and the need for frequent transitioning after a particular age or when they have attained a particular weight or height was identified as a major practical issue for use by participants in seven studies.23 24 27 28 30 34 36 Some of these studies highlighted the lack of knowledge about transition as well as different viewpoints among parents about the suitability of such legislation for older children.

The comfort of children was also raised as an important consideration in many studies.28 30 31 33 Comfort often outweighed safety concerns leading to non-usage: ‘when the kid’s on the restraint actually crying a lot, crying like crazy, and at that moment if you cannot stop him crying, especially on a long journey’.30 The only study that included children as participants reported that most of their comments were about comfort and that they ‘liked ones that were large enough for them and disliked ones that seemed or felt ‘too skinny’. Comfort seems to be a significant issue for children’.34 As such, comfort of children seems to be an important practical enabler to increased use of restraints. Few studies had parents mentioning that booster seats provided a better view of the outside to their children, thus improving compliance.24 28 Other issues that limited the use of child restraints were lack of availability, incompatibility between cars and booster seat designs, lack of shoulder belts in the rear seats of older vehicles and the time consuming and complicated nature of buckling children.24 25 27 28 30 34

Restraint use is considered a mechanism to discipline children rather than a safety device by parents and as children grow older, they actively seek opportunities to negotiate non-usage of restraints

Many parents saw the use of restraints as more of an issue of disciplining their children. This meant that parents often relaxed rules for restraints as need be. For example, several parents had commented that they had an ‘obedient child, and therefore the parent could let the child use a lap belt’.24 They also ‘gave in to children who resisted CSS use, often citing a need to keep the child quiet to avoid distracting the driver’.26 Studies also reported that parents occasionally ‘would allow a child to travel in the front seat as a special treat or because the trip was perceived as a short one… (with a) desire to support children in their maturation and perceiving the move to the front seat as part of this: ‘there’s a sense that your kid is growing up and you kind of want to—it feels good when they’re taking their little steps like that and I guess subconsciously you weaken’’.27

As children grew older (reported around 3–5 years in different studies), peer pressure and the desire for children to be perceived as a ‘big kid’ was a major factor preventing restraint usage. This was particularly problematic because, ‘few older siblings or friends used booster seats’24 and as such parents often found it ‘harder to resist as children grow older27’ due to social norms and pressure, ‘many said that after a certain age (four or 5 years), children say that they feel like ‘babies’ if they sit in a car seat’.28

How parents negotiated with children making these demands seemed to play an important role in child restraint use. Some parents could draw a distinction between safety (non-negotiable) and bath time and eating habits (negotiable), whereas others failed to do so and allowed children to negotiate on the use of restraints. Older children mostly learnt to unbuckle themselves too, which probably limited negotiability.

Adoption and enforcement of laws shaped perceptions and usage in all settings

The adoption and enforcement of laws mandating usage of restraints in children was identified as a very important factor influencing usage as well as influencing perceptions almost universally. This was true even when there was low awareness about the actual specifics of the law or there were gaps in the actual law (particularly around transition ages). Having a clear set of laws and guidelines which had been properly communicated leading to a high level of awareness was thought to be very important.

Enforcement of the law and perception of likelihood of enforcement were identified as important factors for compliance. For example, sometimes respondents ‘did not use the child seat restraint at night, thinking that a police officer could not see into the car in the dark’.26

Presence of the law and its enforcement shaped parents’ perceptions and was also often used by parents to explain to children or negotiate with them. Children were often told that ‘Mommy will get in trouble28’ and thus it was ‘easier for the kids to obey the police’.30 In a study from the USA, the African-American community was dismayed that the laws did not take equity into perspective—‘It’s always about money. If it is mandatory, it should be affordable’.29

Perceptions and norms of child safety differed among culturally and linguistically diverse groups

The use of an equity lens allowed us to explore some issues around equity but due to limitations in the availability of primary data, this could not be fully evaluated around all PROGRESS Plus domains. While all studies were from HIC or UMIC context, many studies focused specifically on immigrants or culturally and linguistically diverse groups; one study36 was based in a regional area of a high-income country with parents of low socioeconomic status where approximately a third of participants were parents of Aboriginal children. Child safety in general was perceived very differently across groups implying that ‘one-size fits all type of approach’ might not be feasible when implementing programmes. As for example, immigrants mentioned ‘seatbelts and child safety restraints were not commonly used in their native countries, which may have contributed to the lack of awareness and low perception of risk among Latinos’.25 Immigrants perceived booster seats to be a reflection of an ‘overprotective culture’ or not being in alignment with their own culture in which ‘we love our kids. So we want to hug them… hold them in our laps in cars’.30 In fact, in the study specifically on Somali refugees and immigrants in USA, ‘child passenger safety, was never mentioned without prompting from the interviewer. Due to the abundance of other problems, safety issues are not top priority in the Somali community’.37 The study from Saudi Arabia had mothers who felt, ‘this may be American culture, but in our culture, they will laugh at us if we place the child’ (in a restraint).31 A study from USA28 which explored differences in norms between White mothers and African-American and Latino mothers found, in addition to differences in usage, that the African-American and Latino mothers mentioned that the comfort of children when sleeping in a booster seat was a key reason for non-usage. African-American and Latina mothers also felt that their children did not like to sit in a child safety seat and this was an issue compared with white mothers who said this was not so much of a problem.


Our narrative synthesis aimed at identifying the factors reported to influence the use of child restraints. The review was prospectively registered. We searched multiple databases and did not purposively sample within the studies we found.

We sought to understand the factors influencing the use of restraints from a wide section of stakeholders, including children but most of our studies focused on parental perceptions only. Only one study involved a few children in a single phase of the study and none of the studies included policy makers, traffic policemen or others involved in formulating or implementing laws or policies around child restraints. Even among parents, mostly the mothers were involved and not fathers or other male family members, who often play important roles. In certain cultural and linguistic groups, men are more often the drivers of motor vehicles. For example, in Saudi Arabia women have only been legally allowed to drive from 2018. As such, even subgroup analyses (such as type of participant, country income level and type of qualitative study, that is, associated with an intervention or standalone study) that we had planned a priori could not be conducted due to the paucity of data. We recommend future qualitative studies to include children and other participants to understand their perspectives. This would be particularly important for older children who actively negotiate for restraint exceptions with their parents.

The use of a PROGRESS Plus framework17 as an equity lens in the current study is novel. PROGRESS Plus has been used in quantitative systematic reviews and meta-analysis previously to understand how an intervention is affected by equity factors. However, we are not aware of its use in qualitative systematic reviews. Although data on equity issues in the primary qualitative studies were scarce thereby limiting interpretation, we contend that it is a feasible and essential element in qualitative evidence synthesis in order to explore how equity affects different social phenomena and its interpretations. More methodological work on this is essential. This posthoc approach was not planned during protocol phase but enabled us to understand the role of these equity parameters around this topic and demonstrate gaps in research in terms of equity. While differences based on gender and cultural and linguistically diverse backgrounds were evident, applying an equity lens through PROGRESS Plus broadens the scope to investigate factors that may influence child restraint use. This includes consideration of occupation, education, socioeconomic status, social capital and personal circumstances (such as disability), relationship status (eg, single/separated parents). Future qualitative studies should explore such factors to better understand child restraint usage. Most of the evidence we found was from high-income countries and from big cities and this should be considered when making judgements on generalisability of the results of the studies to low-income and middle-income countries or semiurban areas, where different issues might affect the use of child restraints.

As in primary qualitative research, we reflected on how our backgrounds and positions might have influenced the conduct of this study. As a team, we decided on the choice of review topic through extensive deliberation over e-mail. We had initially planned to do a quantitative systematic review and meta-analyses but the availability of such studies in literature and feedback on the need for understanding the facilitators and barriers from team members in South Africa and India led to a refocussing of the review to target factors influencing restraint use. In our review, only a single author (SB) conducted all the data extraction and we acknowledge this as a limitation, although we verified extraction to ensure accuracy. As a team we had researchers and clinicians from the UK, South Africa, Australia and India, with expertise and experience in injury epidemiology, child injury and qualitative research. While it was not possible to involve the entire team in the analyses, formulation of themes was initially done by SB, refined with KH and then the final derivation of themes, interpretation of the results and critical feedback was done collectively by the entire team. We conducted a thematic synthesis of the included studies to investigate factors that impact child restraint use, which was in keeping with the epistemology of those studies as per the RETREAT criteria.20 We have used standard methods to assess the quality of the included studies and saturation of themes was achieved. The quality of included qualitative studies varied and many did not report adequately on analysis methods or discussed issues around data saturation, positionality and reflexivity and some did not present quotes. We recommend future qualitative studies in this domain to use standardised reporting checklists like Consolidated criteria for Reporting Qualitative research or Standards for Reporting Qualitative Research.38 39

Results from our study demonstrate the wide range of issues that affect uptake of child restraints. As such, it is imperative for future child restraint research and programmes aimed at increasing child restraint use, focus both on effectiveness and embed the qualitative factors identified in our review. Implications from this review are broad. They highlight the need to develop targeted interventions for culturally and linguistically diverse groups (rather than just tailoring interventions), this may be particularly relevant in HIC. Further, caregivers within the same family have different perceptions (of injury risk and the value of using a child restraint) and there is a need to take that into consideration when developing programmes and messaging. Similarly, messaging should also consider the variations of perception between different parents and caregivers, this includes some parents’ perception that child restraint use is a form of discipline. Differentiating between negotiable (eg, sleeping late) and non-negotiable (safety) behaviour is key to promote child restraints as children grow up. As such, child-restraint usage programmes should focus on aspects of effective parenting too. Such perceptions could be factored into the development of effective parenting programmes to include negotiating with children. Laws appear to be universally useful in shaping perceptions as well as promoting usage. However, considering equity is important during formulation of laws. A robust and consistent law enforcement is also key to promote usage of restraints. Finally, the impracticalities reported by parents in the included studies around using the restraints (installing them in the vehicle and properly securing the child) demonstrate the need for better designs of child restraints, which could lead to both more children being restrained and fewer errors in child restraint use.


Many factors affect the use of car restraints for children which should be considered when planning and developing messaging and programmes to increase child restraint use. In addition, equity issues should also be targeted when planning interventions to promote uptake. There is also a need to conduct equity focussed qualitative research in low-income and middle-income countries in semiurban and rural settings, involving fathers and policy-makers, service providers and enforcement agencies to understand issues around the usage of child restraints in those settings in the future.

What is already known on the subject

  • Child restraint usage will save children’s lives in the event of a crash.

  • Many countries have child restraint laws but parents still do not use them. It is not clear what the barriers and/or facilitating factors are in this regard.

What this study adds

  • To our knowledge, this is the first qualitative evidence synthesis on the topic.

  • It provides important insights into understanding reasons behind compliance or non-compliance of child restraint usage.

  • It will inform the design of programmes locally to increase the usage of child restraints.

  • It will help countries achieve Target 8 of the Global Voluntary performance Targets for Road Safety which specifically aims to increase the use of standard child restraint systems to 100% by 2030.


View Abstract


  • Twitter @rebeccaivers, @margiepeden

  • Contributors MP conceived the study and obtained the funding. SB designed, conducted, analysed and interpreted the data with input from KH, RM, RQI and MP. SB drafted the first manuscript with input from KH and MP. The final text was approved by all authors.

  • Funding This systematic review was funded through an Oxford University Internal HEFCE GCRF grant (ref no. 005105). KH was funded through a NSW Health Early Mid-Career Research Fellowship.

  • Competing interests KH authored one included study and therefore was not involved in data extraction or quality assessment.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.