Decreasing disruptive behaviour during routine dental visits: a video modelling intervention for young children

Abstract

Objective: To evaluate the benefits of a collaborative partnership between paediatric dentists and behavioural health providers in which a practical video modelling intervention, with the aim to reduce disruptive behaviours in young children, is implemented. Methods: The video was created by a dentist using readily available technology and implemented in a busy practice setting. A clinical sample of 40 children, 3–6 years old, was recruited from a continuous sample of patients seen at the clinic. Participants were randomised into two groups and shown either the brief video model or a control video prior to a routine dental visit. All sessions were videotaped and independently scored by blinded observers. Behavioural data were recorded using 15-second partial-interval recording and included physical and vocal disruptions. Subjective measures of cooperation were also completed by observers and dental professionals. Results: Independent samples t-tests show that the treatment group had a significantly lower mean percentage of intervals in which disruptive behaviour was observed [t(38) = 2.94, P = 0.008] compared with the control group. Subjective rating scales revealed significantly higher ratings of cooperation for the treatment group from the dentist [t(38) = −5.19, P = 0.000], the dental assistant [t(38) = −4.01, P = 0.001] and the blinded coder [t(38) = −3.54, P = 0.002]. Significant relationships were found between the percentage of actual disruptive behaviour and subjective ratings of the dentist (r = −0.82, P < 0.01). Conclusions: Watching a brief dentist-created video model of expected procedures can reduce disruptive behaviour and increase cooperation for young children making their first visit to a busy medical setting.

INTRODUCTION

Disruptive behaviours during dental visits can include excessive movement, verbal disruptions and non-compliance. Mismanagement of these behaviours can limit children’s access to quality oral health care and hinder the success of the treatment for both the child and the paediatric dentist1., 2., 3., 4., 5.. This is of particular concern given that approximately 20% of children experience high levels of distress, fear and anxiety during dental procedures6., 7.. The intensity of these behaviours is even more pronounced in young children⁸, especially those with infrequent or no prior visits to a dentist. As a result, parents might forego some elective procedures (e.g. routine check-ups) because of a history of their child engaging in disruptive behaviour and non-compliance. Additionally, some dentists are reluctant to treat young children or to achieve compliance using restrictive means, such as physical restraint or sedation⁹.

In an attempt to manage disruptive behaviours without restraint or sedation, dentists have developed numerous approaches to behaviour management¹⁰. Unfortunately, dentists are not typically reimbursed for the time they spend learning or implementing behaviour management techniques11., 12.. Thus, dentists are more inclined to use behaviour management procedures that are time- and cost-effective and are most likely to use procedures that produce the best results with the least amount of time and effort¹³. It may also be difficult for a dentist to decide which behavioural management technique to use, given societal, parental and other providers’ influence. This confusion, in conjunction with the lack of use of appropriate strategies by parents, contributes to the persistence of behavioural problems within dental settings, often requiring additional consultation and support from external referral sources (e.g. behavioural health providers). Discussion of problems in accessing behavioural health services is outside the scope of this paper; however, limited access to such professionals within dental settings highlights the importance of interdisciplinary collaborative efforts. These types of collaborations promote the development of evidence-based strategies that can be used by dentists and dental assistants without requiring additional staff or ongoing behavioural health support.

Live demonstration and modelling of expected behaviours and sensations are recognised as an empirically supported approach for helping children develop coping skills and for reducing distress commonly experienced during medical procedures¹⁴. Modelling is thought to be valuable in preparing children for medical procedures by providing information about what procedures and sensations can be expected and by adding predictability and familiarity that can desensitise children to procedures that might otherwise produce distress. Limitations to this approach involve staffing and time constraints. One potentially more practical and cost-effective method involves video modelling. Video modelling is an approach in which an individual learns or strengthens an adaptive skill by watching a video of someone performing that skill during a targeted procedure or routine. The individual watches a video of a skill or behaviour that will be needed and then later imitates the behaviour of the model during the identified routine. Video modelling has considerable empirical support as an effective procedure for helping children and adolescents with autism spectrum disorder or other neurodevelopmental disabilities develop important skills in schools15., 16., work environments17., 18. and the home¹⁹; however, extension of this research into a neurotypical population is not as vast. Video self-modelling has a somewhat more extensive literature base within the typically developing population20., 21., 22.; however, time/staffing constraints and other logistical considerations (e.g. accessing patients before scheduled appointments) impede its use within active clinical settings. Thus, third-person video modelling is attractive as a means of managing disruptive behaviour in these situations because it does not require changes to or interruptions of the ongoing health-care routine and avoids the constraints associated with live or self-modelling.

Despite evidence supporting use of video modelling across settings, this process is not widely used in dentistry to prepare children for treatment. Indeed, video modelling was not included as a viable management alternative in professional guidelines for behaviour management until 2015¹⁰. This may be largely because video modelling has not been widely researched in preparing children for routine dental examinations. Only a small number of studies have examined filmed modelling alone as a preparation for dental care in children. While these studies found that filmed modelling can help reduce children’s distress, they found that the benefits may be limited to only certain children under certain conditions and there was no clear agreement on which conditions are best23., 24., 25., 26. (e.g. length of film in relation to age, repeated presentation vs. single presentation, pairing video with guided practice). The methodologies of these studies also were often limited by reliance on subjective, rather than direct and objective, measures of distress. Furthermore, the procedures used in the studies were not particularly amenable for use in everyday practice, creating numerous logistical difficulties that may have hampered dissemination²⁷. For instance, the videos were often lengthy and required 10–15 minutes of viewing time before dental treatment could begin. Previous studies also used videos created entirely by researchers, with little participation by end-users who might be interested in implementing them. Previous studies also provide little, if any, information on what to include in video modelling, making replication and dissemination less likely.

Fortunately, recent advances in portable technology have made it increasingly easy to make and implement video modelling in dental clinics. The ubiquitous availability of portable tablet devices and smartphones, most with their own high-quality video cameras, are reducing both financial and logistical barriers to video modelling and may ultimately enhance dissemination of this intervention. In addition, there is increasing evidence that end-users can participate more actively in creating and implementing video modelling interventions with guidance from investigators18., 19.. Therefore, by evaluating a dentist-created and implemented intervention in a busy practice setting (using readily-available video technology), we hope to advance discussion of behavioural and oral health-care collaborative efforts and research.

The purpose of this study was to evaluate the outcomes of a collaborative interdisciplinary project and the preliminary benefits of a practical video modelling intervention for reducing disruptive behaviours in young children. The study was designed to improve on past research by: (i) engaging a dentist more actively in video production; (ii) using accessible video-making technology; (iii) providing detailed information on how the intervention was designed; and (iv) including direct, as well as subjective, measures of disruptive behaviour in children.

METHODS

Participants and setting

The study was conducted with a clinical sample at a paediatric dental clinic associated with a large Midwestern academic medical centre. This interdisciplinary collaboration was achieved by leveraging previous partnerships between the Department of Psychology and College of Dentistry that, in the past, led to joint initiatives as well as to completion of research projects associated with the certification requirements of the General Practice Residency Program. The second author (a dentist) initiated the collaboration in view of the first author’s (a behavioural-health provider) past research and behavioural expertise, especially regarding the use of video modelling to teach adaptive and social behaviours. Both providers were involved with study conceptualisation; the second author created the video model, recruited participants and implemented all procedures with participants; and the first author oversaw the study design, behavioural definitions, study protocol, video coding and training, behavioural consultation, and video production consultation. The study protocol was approved by the Institutional Review Board (University of Nebraska Medical Center) and all research was conducted in full accordance with the World Medical Association Declaration of Helsinki. Data were collected for 3 months.

Forty children (3–6 years of age) participated in the study. Participants were recruited from the paediatric dental clinic as children who were making their first visit to the clinic (but not necessarily their first visit to a dentist). Participants were recruited from a continuous sample of patients seen at the clinic and they were selected based on their need for an initial preventative dental screening. Eligibility criteria were: (i) children between the ages of 3 and 6 years; and (ii) children making their first visit to the clinic. Exclusion criteria were: (i) children who had easily discernible limitations of physical or mental status; and (ii) children who did not speak English. The sample size was limited to 40 participants due to dentist transferring to a different clinic; however, post-hoc power analyses revealed sufficient power to suggest significant between-group comparisons (see the Results section).

The dental clinic was located on the second floor of a children’s hospital and consisted of nine dental chairs, including four private treatment rooms. An isolated private treatment room was used as the session room and was consistent with a typical dental treatment room, including an electric dental chair, an X-ray unit and a computer workstation. All sessions and dental procedures were completed by the second author and the same dental assistant.

Materials – video modelling production

The video used for the intervention group was filmed together by the first and second authors using a tablet (iPad 2) and all stock software (e.g. iMovie). The peer model used in the video was a Caucasian 5-year-old volunteer from the dental practice who was unfamiliar to all participants. The movie was filmed in the same clinic space in which study data were collected.

The video showed the dentist introducing himself and the model, showed them walking back to the treatment room, and then showed the model undergo brief exposure to each of the major components of a dental examination and cleaning, performed by both the dentist and a dental assistant, including:

• •Getting in the chair
• •Wearing a bib
• •Wearing a leaded apron
• •Accepting the camera placed in position
• •Accepting the X-ray slide and biting down
• •Reclining back with hands on stomach
• •Wearing sunglasses and light turned on
• •Accepting the evacuator and rinse
• •Accepting prophy application
• •Accepting flossing
• •Accepting use of explorer and mirror
• •Accepting application of fluoride.

Throughout exposure to each step, the dentist provided procedural and sensory information about what to expect (i.e. Tell-Show-Do) directly in the video; therefore, no editing was required later to include voice-over narration. The dentist also provided the model with specific praise for cooperative behaviour throughout the video (e.g. keeping still, having hands on stomach, following directions) and was shown to provide the model with a small reward at the end of the appointment. The dentist had the peer model practice the routine before filming and decided to reshoot the video several times to improve sound and performance by the model and the dental staff. No sound editing or special graphics software was used. Editing was limited to that which can be conducted using the stock software native to the tablet device and involved removing a few seconds from both the beginning and end of the final video so that it was 4 minutes in length.

The video shown to the control group was also 4 minutes in length and comprised a clip of a popular children’s cartoon. Data were taken by the participating dentist on how much of the video/cartoon the participant watched (e.g. <1 minute, 1–3 minutes, >3 minutes) and found that 100% of participants watched all or most of the video/cartoon (‘more than 3 minutes’).

Measurement

Demographic questionnaire

Each parent completed a background questionnaire asking about the participant’s age, gender, ethnicity, if he/she had been to a dentist before (if so, how many times) and whether the participant had ever received a diagnosis of developmental delay or another diagnosis resulting in early intervention or special school services.

Direct observational measures of behaviour

Behavioural data were documented using 15-second partial-interval recording with which the blinded coder (first author) marked whether a specific behaviour occurred during each interval. The coding system used for this study was a slight adaptation of a coding system used in previous research in which disruptive behavior was studied within dental settings13., 28.. Behaviours coded for each child included physical and vocal disruptions. A physical disruption was defined as 15 cm or more of continuous movement of the body or head or a series of continuous repetitive movements accumulating to six inches or more. A vocal disruption was defined as an unprompted vocalisation, including crying, screaming, moaning and/or complaining. Vocalisations and body movements in response to questions were not coded.

Data were also taken on whether specific behaviour-management strategies were used during the session. These included physical management by the professional (e.g. dentist or dental assistant physically restricting movement by holding or blocking part of child’s body); whether or not a parent was asked to come into the room because of disruptive behaviour; and whether the child required a more restrictive procedure (i.e. knee-to-knee examination). The data recorded also included the percentage of procedures (e.g. X-rays, cleaning, dental examination and fluoride treatment) that were not completed for each participant.

All sessions were videotaped so that two observers (the first author and an undergraduate student trained to conduct interobserver agreement) could independently measure child behaviour. These observers were blind to whether the participant was in the treatment group or the control group.

Interobserver agreement

Interobserver agreement was conducted on direct observational measures. The secondary video coder was an undergraduate student trained by the first author on operational definitions and data collection. Training was provided by the first author using multiple video examples of dental visits and until the secondary coder attained at least 95% agreement with the first author across three consecutive practice sessions. During actual assessment of the outcomes of the study, interobserver agreement was assessed for at least 30% of all sessions. Agreement was scored if both coders recorded the occurrence or non-occurrence of the same behaviour during an interval. Percentage of agreement was calculated by dividing the number of agreements by the total number of agreements plus disagreements and multiplying by 100 for each behaviour. Mean agreement was 93% (range: 87%–100%) for verbal disruptions and 94% (range: 84%–100%) for physical disruptions. Mean agreement was 92% (range: 89%–100%) for physical management procedures used, 99% (range: 99%–100%) for parent presence, and 98% (range: 94%–100%) for knee-to-knee procedures.

Subjective rating scales of behaviour

After each session, both the dentist and the dental assistant were asked to complete subjective ratings of the child’s level of cooperation and disruptive behaviour. As the dentist and dental assistant were in charge of showing the video to participants, they were not blinded when rating behaviour. However, each of the blind observers also completed identical questionnaires after watching the video of each session. The Likert-type scale used in the present study has been used previously in dental research²⁹ and consists of six ratings (1–6), ranging from ‘extremely cooperative’ (score 1) to ‘extremely uncooperative’ (score 6). All raters were provided with operational definitions of each rating. The professionals were asked to complete the scales independently.

Procedure

The participating children’s parent or guardian was approached before the appointment and prior to the child receiving any treatment, to inform them of the research and to secure parental permission. Written informed consent was obtained from all parents/guardians for all children prior to initiation of study procedures. If consent was obtained, the child and his/her parent were escorted to a small conference room adjacent to the examination room.

Each child participant was randomly assigned, via coin flip, to either the video modelling group or the control group. With the exception of watching either the video model or cartoon, all procedures were identical across both groups. Parents were only present while the child watched the video. After watching the video, each child was escorted to the examination room while the parents were asked to wait in the designated waiting area. The same examination room was reserved for the duration of the study and was isolated from the other treatment rooms and examination areas. Upon entering the examination room, the participant was asked to sit in the dental chair. Each participant then underwent a dental recall appointment which comprised occlusal and bitewing X-rays, prophylaxis and flossing performed by the dental assistant and intra- and extra-oral examinations performed by the dentist. The intention was to expose all participants to each of these procedures; however, not every procedure was completed for every participant when problem behaviour occurred. The dentist and assistant were not given any specific prompts or feedback on their interaction with each participant during the appointment. Instead, the dentist and assistant were asked to run the session in a typical manner while at least attempting to initiate all procedures in the same order for all participants. Dental professionals were encouraged not to reference the video during clinical procedures in to order to keep clinical setting and behavioural management standard across groups. In this clinic, dental professionals were attempting to implement many practice parameters recommended by the American Academy of Pediatric Dentistry regarding behavioural management (e.g. communicative guidance, Tell-Show-Do, descriptive praise, protective stabilisation); however, more advanced interventions (e.g. distraction, pre-visit imagery, differential reinforcement) were not observed or reported to be in use at the time of the study.

RESULTS

Descriptive statistics

Descriptive statistics for all participants in each group, including age, gender, number of prior visits and ethnicity, are provided in Table 1. Overall, 40 children (19 in the control group and 21 in the treatment group) participated in the study. There were no significant differences between groups on the baseline descriptive characteristics. All participants were eligible for Medicaid. Sixty percent (n = 24) of participants had been to the dentist before (at a different clinic) for routine dental care.

Table 1.

Baseline characteristics of participants

Variable	Control (n = 19)	Treatment (n = 21)
Age (years)	4.5 ± 1.7	4.7 ± 1.0
Male gender	14 (74)	10 (52)
Number of prior visits	1.2 ± 1.3	1.3 ± 1.7
Ethnicity
African American	5 (26)	7 (33)
Asian American	3 (16)	0 (0)
Caucasian	4 (21)	2 (10)
Hispanic	7 (37)	12 (57)

Data are given as mean ± SD or n (%).

No significant between-group differences on baseline descriptive characteristics were found.

Direct observational measures

Table 2 depicts independent samples t-tests completed to examine significant differences between groups on the objective behavioural measures. The results show that the treatment group had a significantly lower mean percentage of intervals in which disruptive behaviour was observed for Physical Disruptions [t(38) = 2.72, P = 0.014], Vocal Disruptions [t(38) = 2.36, P = 0.029] and Combined Disruptions (intervals in which either verbal or physical disruptions occurred) [t(38) = 2.94, P = 0.008] compared with the control group. Given that most often physical and vocal disruptive behaviour co-occurred, a post-hoc power analysis was performed with the Combined Disruptions variable to show the between-groups comparison effect size. This effect size was observed to be large (d = 0.94).

Table 2.

Between-group examination of direct behavioural measures

Variable	Control (n = 19)	Treatment (n = 21)
Physical Disruptions	7.5 ± 10.9	0.7 ± 1.6*
Vocal Disruptions	7.9 ± 13.2	0.6 ± 2.9*
Combined Disruptions	11.4 ± 15.4	0.8 ± 2.9*

Data are given as mean % of intervals ± SD.

^*Significance at P < 0.05 (two-tailed).

Summary of the observational data shows that 47% of participants in the control group and only 5% of children in the treatment group required some type of physical management employed by the dentist or dental assistant (e.g. restraint). Furthermore, in 21% of control group participants the procedures were incomplete compared with only 5% of the treatment group.

Subjective rating scales

Table 3 shows independent-samples t-tests for the results of the subjective rating scales across professionals and the primary video coder. The results revealed significantly higher ratings of cooperation for the treatment group from the dentist [t(38) = −5.19, P < 0.001], the dental assistant [t(38) = −4.01, P = 0.001] and the video coder [t(38) = −3.54, P = 0.002].

Table 3.

Between-group analysis of subjective rating scales

Variable	Control (n = 19)	Treatment (n = 21)
Dentist rating of behaviour	3.8 ± 1.7	5.9 ± 0.3*
Assistant rating of behaviour	4.1 ± 1.7	5.7 ± 0.5*
Blind observer rating of behaviour	4.4 ± 1.6	5.7 ± 0.6*

Data are given as mean ± SD

^*Significance at P < 0.01 (two-tailed).

Correlations amongst objective and subjective measures

Correlations were used to compare the objective measures of behaviour with the subjective ratings of behaviour. Significant relationships were found between the percentage of actual disruptive behaviour (combined) and the corresponding subjective ratings of the dentist (r = −0.82, P < 0.01), the dental assistant (r = −0.81, P < 0.01) and the primary video coder (r = −0.92, P < 0.01). This demonstrates that, overall, if a participant was rated as highly cooperative by the professional or video coder, these ratings were highly correlated with actual direct measures of behaviour.

DISCUSSION

The purpose of this study was to evaluate the effects of a relatively low-intensity dentist-implemented intervention on the disruptive behaviour and overall compliance of young children with little to no experience at a dentist. Extending the results of past research implementing video modelling with children of typical development and with disabilities21., 23. the results of the current study indicate that the intervention was an effective and practical method to address these target behaviours. Participants in the treatment group displayed significantly less disruptive behaviour and were rated as significantly more cooperative than their control counterparts. These results were consistent across both direct and subjective measures.

Summary of the observational data also suggests other noteworthy points. For instance, nearly half of the participants in the control group required some type of physical management to be employed by the dentist (e.g. restraint). This suggests that children who did not watch the video, and who were more disruptive during the appointment, required the dentist and/or dental assistant to change their standard procedure in some way, whether it involved holding the child’s face/head or implementing a more restrictive procedure such as a knee-to-knee. It is unknown as to whether or not physical management influenced the level of disruptive behaviour during appointments; nevertheless, interval analysis revealed that physical management by dental professional only occurred after a child displayed disruptive behaviour. Given past research showing dentists’ reluctance to use physical management or treat young children⁹, if having children watch a short video model decreases the likelihood of dental professionals having to use physical management, this probably increases the social validity of the intervention. Furthermore, it was not possible, in many of the control group participants, to complete all procedures. This suggests that having children watch a short video model might increase the likelihood that all procedures are completed (by decreasing the incidence of problem behaviour), consequently increasing the child’s access to quality oral health care.

While effective, procedures of past video modelling research within dental contexts were not particularly amenable for use in everyday practice, possibly limiting dissemination and hampering widespread use. Thus, a number of additional advantages exist for use of the current intervention in private practice settings. First, the dentist was actively engaged in production of the video, employing readily-available and user-friendly media. Based on dentist and dental assistant reports, this intervention could easily be implemented in other paediatric dentistry clinics, with little-to-no training. Staff would simply need access to any device with filming capability and planned procedures for patients to view videos before appointments. Pre-appointment viewing could be achieved through in-clinic viewing using monitors or tablets, having videos available on the practice’s website or adding links to the video to text- or email-based appointment reminders. Second, the professionals did not require ongoing consultation from behavioural health providers or training on specialised behaviour management techniques. The only method outside the realm of ‘standard practice’ involved allowing the participants to watch a short video. In practice, the time required to show individuals the video could easily be offset by making the video available to families and children in the waiting room. Additionally, dental professionals were encouraged not to reference the video during clinical procedures for standardisation purposes; however, the intervention would probably be improved if the video is shown multiple times, shown in steps interspersed with reinforcement for compliance or frequently referenced throughout clinical procedures.

Regarding the general accuracy and validity of dentist and assistant behavioural ratings, the results of this study also suggest that subjective rating scales are probably appropriate for making accurate, data-based clinical decisions, without necessarily having to rely on exhaustive behavioural measures. Dental professionals could use such rating scales to rate patients quickly and accurately after an appointment, to communicate the possible need for a higher level of behavioural management. These children could then be selected to participate in more intensive interventions or be referred to a behavioural health provider.

This study did have some limitations. First, although the main findings were statistically significant, findings would be more robust and generalisable with a larger sample size. Increasing the sample size would allow more comparisons across groups, such as age-group comparisons. Also, as many of the children had previous experience at the dental office, it would help parse out the positive or negative influence of prior exposures. Furthermore, given the wide range of developmental characteristics within the 3- to 6-year age range, increasing the sample size would provide a better ability to understand the impact of developmental variables. Second, there were low levels of disruptive behaviour overall, suggesting that participants as a whole were generally well behaved. These data are not necessarily representative of past studies which suggest that a larger proportion of children exhibit distress and disruptive behaviour at the dentist1., 4.. The reasons for generally low levels of problem behaviour are unknown but may be related to dentist interactions or structure of the appointment. Future research should investigate the influence of ongoing interactional patterns between professionals and children during distress-provoking situations.

Despite these limitations, this study suggests that video modelling can lead to children who are calmer and easier to manage, and in conjunction with increases in user-friendly technology, does not require changes to or interruptions of the ongoing health-care routine. Also, with decreases in disruptive behaviour, children are rated as more cooperative by their dentist who are thus able to complete a higher percentage of procedures. This, in turn, increases young children’s access to quality oral health care and may encourage dental professionals to make video modelling part of their standard practice. Another implication of this study is that interdisciplinary research can successfully be conducted in a busy medical setting, increasing the likelihood of widespread implementation of practical and cost/time-efficient behavioural management procedures.

Acknowledgements

This initiative was funded as a pilot project of the Munroe-Meyer Institute for Genetics and Rehabilitation at the University of Nebraska Medical Center. This project also was completed with support from the Katherine Dodd Faculty Scholars Program in the Department of Pediatrics at Vanderbilt University Medical Center and support from the Vanderbilt Kennedy Center Eunice Kennedy Shriver National Institute of Child Health and Human Development Intellectual and Developmental Disabilities Center U54 HD08321. The opinions and views expressed herein are those of the authors.

Competing interests

The authors have no conflicts of interest to disclose.