A priori calculated required sample size not feasible after 3 years of recruitment.

The aim of this study is to examine the effect of using a humanoid robot during a painful puncture procedure (e.g., portal catheter punction, IV punctures, blood withdrawal, etc.) in children with chronic conditions (e.g., chronic immune deficiency, colitis ulcerosa, metabolic disorders, cystic fibrosis, etc.) on the pain and fear memory of child and parent in comparison to usual care. The use of a humanoid robot has proven to be effective on reducing healthy children's pain and distress towards vaccinations. Still, whether these benefits generalize to chronically ill children, to pain/fear memories and to the parents needs to be examined.

The current research proposal is the first one to examine the use of a humanoid robot to distract children with chronic conditions who undergo repeated painful procedures and, compared to usual care, its effectiveness in reducing negative pain and fear experience and reducing negative pain and fear memory biases.

Children aged 8-12y with chronic conditions receiving regular treatment by painful puncture procedures, and their parents will be recruited at the University Hospital Brussels and Ghent. Baseline assessments include child's anticipated pain, pain-related fear and self-efficacy by child's self-report and parent proxy report, child's attention control, attention bias, pressure hyperalgesia, energy-balance related behavior, child's and parent's catastrophizing about the child's pain, parental behavioral responses to their child's pain and parent's emotions towards the child's painful procedure. Stratified block randomization will be used to assign the children to the control group (usual care) or intervention group (robot distraction). Immediately after the puncture procedure, experienced pain, pain-related fear, self-efficacy and pain catastrophizing (state) will be assessed again by child and parent, as well as emotions of the parent. One week post-procedure the participants will be contacted by telephone for a short interview in order to assess child's and parent's pain and pain-related fear memory, as well as child's and parent's future pain, pain-related fear and self-efficacy expectancies using the same scales as administered before and immediately after the procedure, with in addition a free recall and prompted recall about contextual details about the day of the procedure. Recalled pain and pain-related fear ratings that are higher compared to initial reports are considered indicative of negative memory biases.

Objectives

The main objective of this pilot RCT is to examine if distraction by using a humanoid robot during a painful puncture procedure (I) contributes to less negative pain and pain-related fear memory biases (O) in comparison with the usual care (C) in children with chronic conditions between 8 and 12 years old and their parents (P). Furthermore, it wil be investigated if the use of a humanoid robot to distract these children during a puncture procedure contributes to lower levels of self-reported experienced pain-related fear and pain during the puncture and less negative expectations regarding future painful procedures. The investigators expect to see the same positive effects in the proxy reports of the parents. Also, the role of possible moderator variables such as child's self-efficacy, attention control, attention bias and energy-balance related behavior, child's and parent's pain catastrophizing about the child's pain, parent's emotions and responses to children's pain episodes will be investigated.

Methods

Participant recruitment

Patients will be recruited by the children's hospital of the University Hospital Brussels and Ghent. All children between 8-12y with chronic (non-primary pain) condition undergoing regular painful puncture procedures as part of their treatment, and their parents will be informed about the study and receive a consent form. The recruitment procedure will start in July 2019 and pursue until the calculated sample size is reached.

Given the observation that children up to 12 years of age are more likely than children over 12 years to believe that a robot possesses human characteristics, the study will focus on children between the ages of 8 and 12 years. The lower age limit is chosen because child self-report questionnaires have been validated from that age. Self-reports in children also require a variety of cognitive skills, which develop gradually and are only acquired at the age of 7 or 8 in most children. Children and their parents can only be included in the study if they meet the inclusion criteria and after written informed consent from the (co-participating) parent/guardian and child is obtained. All parents/guardians and children will receive verbal and written information on all aspects of the study prior to consent for participation.

Sample size calculations were performed with G*Power 3.1.9.2. The a priori sample size calculation was based on the treatment effects on children's pain-related fear memories in the study of Pickrell et al. (2007) and targeting a study power of (80%) with α=0.05. Allocation ratio (N2/N1) was defined as 1, resulting in 47 patients in the experimental group and 47 in the control group (n=94 for the total sample size). Accounting for a 10% dropout and keeping in mind that every sample size calculation has its limitations, the investigators propose studying at least 52 patients per group for this study. Hence, the total number of participants to be included will be equal to 104. Because the included diseases are all rare diseases, the calculated sample size remains as a target.

Assessments:

Using an online questionnaire battery (↑ study feasibility), a number of measures will be assessed in the participating children and their parent(s). Online questionnaires will set up with LimeSurvey.

Risk management: In case the online questionnaires are not filled in by the start of the procedure, the test day will be moved to the next puncture procedure.

Measurement time points: T0 = day before puncture procedure, T1 = immediately before puncture procedure, T2 = immediately after puncture procedure, T3 = one week after puncture procedure

Demographics and medical data

At T0 demographics (e.g. age, sex, ethnicity, socioeconomic status) and medical information will be collected from the patient's clinical charts and will be further supplemented by child and parent report.

Anthropometrics

At T1, basic anthropometrics (body height, body weight, waist circumference) will be gathered from each participating child. In addition, the body composition (i.e., body mass index, fat mass, fat free mass, muscle mass, intra- and extra cellular water, etc.) will be determined by a body composition analyzer (Tanita).

Pressure hyperalgesia: pressure pain threshold (PPT) - CHILD

At T1, pressure hyperalgesia of all participating children will be assessed by evaluating pressure pain thresholds (PPTs) at the dominant tibialis anterior and trapezius muscle by means of a digital pressure algometer (FPX 50, Wagner Instruments, Greenwich, CT, USA). To familiarize the subject with PPT measurements, one familiarization session will be done at the idominant upper leg. This technique to evaluate PPTs is a well-known and well validated method in the exploration of pathophysiological mechanisms involved in pain. High reliability levels of this technique are apparent from various studies. Also, pressure algometry has proven to have an excellent intra-rater agreement, a satisfactory inter-rater agreement and is well tolerated even by very young children.

Pain catastrophizing Scale for Children (PCS-C) - CHILD

At T0, children's catastrophic thinking about pain will be assessed with the Dutch version of the Pain Catastrophizing Scale for Children (PCS-C). The PCS-C consists of 13 items describing thoughts and feelings which the child may experience when they are in pain, in independent but strongly related dimensions of rumination, magnification and helplessness. Children are asked to rate how frequently they experience these thoughts and feelings on a 5-point Likert scale (0 = 'not at all', 4 = 'extremely'). The PCS-C results in a total score (ranging from 0 to 52). The PCS-C has been shown to be a reliable and valid tool for children of 9 to 15 years old and a strong internal consistency for pain catastrophizing (α=.90) is already demonstrated in the pediatric population.

Pain Catastrophizing Scale for Children (PCS-C state) - CHILD

Just before (T1) and immediately after (T2) the puncture procedure, the extent of child state (i.e. situation-specific) catastrophizing thoughts about the procedure will be assessed using an adaptation the PCS-C. Based on previous research, the investigators use a state version of the PCS-C, consisting of 1 adapted item from each of the 3 subscales. A mean score of these 3 items will be calculated ranging from 0 to 10. Research has shown a significant association between child state catastrophizing and child pain intensity.

Attentional Control Scale for Children (ACS-C) - CHILD

At T0, attention control will be assessed using the Attentional Control Scale for Children (ACS-C), which is a 20 item self-rating scale that assesses abilities to focus and shift attention. Ten statements assess the focusing component of attention (''When I concentrate myself, I do not notice what is happening in the room around me'') and the other ten statements assess the shifting of attentional resources (''When I am doing something, I can easily stop and switch to some other task''). A higher score on this scale indicates a better capacity of attentional control. Several studies report good internal consistency of the ACS-C.

Attention Bias Assessment (dot-probe task) - CHILD

At T1, all participating children will be asked to complete a dot-probe task, which is a well-investigated reaction time task designed to measure selective attention to threat. In this dot-probe task, a threatening (i.e., pain) and neutral stimulus are presented simultaneously (i.e., a painful child face paired with a neutral child face) at two different spatial locations on a screen for a short time. Because both stimuli are in different spatial locations, they compete for attention. After the appearance of these stimuli, a dot probe emerges at the location of the threatening/pain stimulus (congruent presentation) or at the location of the neutral stimulus (incongruent presentation). The assignment of attention is measured by the time needed to respond to the dot probe. It is reasoned that responding to the probe will be faster when the child's attention is already allocated to location where the probe appears. Most probe detection studies found indeed that anxious individuals respond faster to congruent trials than to incongruent trials (congruency effect). This task is implicit, does not require instruction, and subjects need no or minimal training to perform the test successfully. To enhance personal salience of the dot-probe task, children will be informed that pictures they will see during the dot-probe task are derived from other children who underwent a puncture procedure.

Energy balance-related behavior (ENERGY CSS child) - CHILD

At T0, a selection of questions regarding physical activity and sedentary behavior of the ENERGY Cross-Sectional Survey (ENERGY CSS child) used in the ENERGY-project will be assessed by the participating children. This ENERGY-child questionnaire was developed in order to assess energy balance-related behaviors of the child as well as personal, family and school-environmental determinants related to these behaviors. The questionnaire has been shown to have good test-retest reliability and moderate to good construct validity for the large majority of items.

Numerical Rating Scale for anticipated pain and pain-related fear (NRS-11) - CHILD&PROXY

Before (T1), immediately after (T2) and 1 week after (T3) the puncture procedure, child's pain-related fear and pain with regard to the procedure will be assessed using an 11-point 1-item scale with the anchors of 0 (not scared) and 10 (extremely scared), and 0 (no pain) and 10 (extremely painful). Research has shown that the NRS-11 and FPS-R are strongly associated (r = 0.78,p < 0.001) and that the score distribution of the NRS-11 is comparable with that of the FPS-R. Miró et al. (2009) showed good convergent, discriminant and criterion validity for the NRS-11.

Faces Pain Scale-Revised (FPS-R) - CHILD&PROXY

Before (T1), immediately after (T2) and 1 week after (T3) the puncture procedure, child's pain with regard to the puncture procedure will be assessed using The Faces Pain Scale-Revised (FPS-R). This is a revised version of the original scale, which was developed by Bieri et al. (1990) for the self-assessment of pain severity by children. The revised scale consists of 6 sex-neutral line drawings illustrating an increasing level of pain intensity from the most left face to the most right face. The participant responds by indicating which of the 6 faces corresponds to his or her level of pain. It is easy to administer and requires no equipment except for the photocopied faces. Its adequate psychometric properties have already emerged from several studies.

Children's fear scale (CFS) - CHILD&PROXY

Before (T1), immediately after (T2) and 1 week after (T3) the puncture procedure, child's pain-related fear with regard to the puncture procedure will be assessed using the Children's Fear Scale (CFS). This scale is based on the Faces Anxiety Scale, developed by McKinley et al. (2003). This one-item scale consists of a row of 5 sex-neutral faces varying from a not afraid (neutral) face on the left to a face that shows extreme fear on the right. The participant responds by indicating which of the 5 faces corresponds to his or her level of pain-related fear. It is easy to administer and requires no equipment except for the photocopied faces. The scale has been shown to have a good interrater reliability and test-retest reliability.

Self-efficacy (1 item) - CHILD&PROXY

Before (T1), immediately after (T2) and 1 week after (T3) the puncture procedure, the child and the parent will score the same question about how he/she or his/her child is dealing/dealt with the puncture procedure on a standard 11-point numeric rating scale from 0 (not good at all) to 10 (extremely good).

Pain Catastrophizing Scale for Parents (PCS-P) - PARENT

At T0, parent's catastrophic thinking about their child's pain will be assessed with the Dutch version of the Pain Catastrophizing Scale for Parents (PCS-P). The PCS-P consists of 13 items describing different thoughts and feelings that parents may experience in relation to their child's pain, in independent but strongly related dimensions of rumination, magnification and helplessness. Parents are asked to rate how frequently they experience these thoughts and feelings on a 5-point Likert scale (0 = 'not at all', 4 = 'extremely'). The developers reported strong internal consistency (α=.93) and validity as demonstrated by associations with parent distress and child disability.

Pain Catastrophizing Scale for Parents (PCS-P state) - PARENT

Before (T1) and immediately after (T2) the puncture procedure, the extent of parental state (i.e. situation-specific) catastrophizing thoughts about the procedure will be assessed using an adaptation of PCS-P. Based on previous research, the investigators use a PCS-P state, consisting of 1 adapted item from each of the 3 subscales. A mean score of these 3 items will be calculated ranging from 0 to 10.

Inventory of parent/caregiver responses to the children's pain experience (IRPEDNA) - PARENT

To measure the responses of the parent to their child's pain episodes at T0, a shortened of the inventory of parent/caregiver responses to the children's pain experience (IRPEDNA) will be used. This scale is a self-administered questionnaire with three subscales: (1) solicitousness, (2) discouragement, and (3) promotion of well-behaviors and coping. The shortened version of 10 items only includes solicitousness and promotion of well-behaviors. All items reflect actions and behaviors that the parents may enact after seeing the children's pain behaviors. A good internal consistency of the 3 subscales has been demonstrated (coefficient alphas of 0.87, 0.83 and 0.87 respectively) and the subscales also show good criterion-related validity.

Emotions of the parents regarding their child undergoing a painful procedure - PARENT

Before (T1) and immediately after (T2) the puncture procedure, a series of emotion adjectives is used to assess the emotions of the parents in response to the child's painful procedure. This method is based on the work of Batson et al. (1987). All emotion adjectives will be rated on an 11-point scale ranging from 0 (not at all) to 10 (extremely). This method is short and easy-to-complete, which lends itself well to adapt to assess distress in response to a specific situation, in this case parental distress as a reaction to their children's painful puncture procedure. Based on Batson et al. (1987) and with the aim of reflecting self-oriented emotional responses or distress, 4 adjectives will be presented (worried, upset, anxious and sad). The mean score of these 4 adjectives will be calculated, whereby higher scores represent higher levels of parental distress. This method has proven to be valid and reliable.

Total time of procedure

The total time of duration of the procedure will be registered for each participant using a stopwatch, starting at the point at which the door closes after arrival of each person concerned in the room and ending when the puncture procedure is totally finished (everything is fixated on the body with a bandage). In this way, a difference in the duration of the procedure can possibly be established between the usual care group and the robot intervention group.

Kalinox® registration

Whether or not and how much Kalinox® (mixture of nitrous oxide (50 %) and oxygen (50 %)) is administered via a respirator during the puncture procedure is registered for each participant.

Randomization procedure and intervention:

Randomization

Stratified block randomization (balanced by sex (M/F), age (8-9 year/10-12 years) and site (University Hopsital Brussels/Ghent) will be used to assign patients to the control group or experimental group. Randomization will be executed by an independent researcher who will not be involved in treatment provision, assessments or in statistical analyses. This independent researcher will prepare a list with patient numbers/codes and the group allocation that results from this randomization procedure. Group allocation will take place after baseline assessments have been performed. Both groups will have the same people present in the operating room: 1 child, 1 parent, 1 or 2 nurses, 1 researcher.

Intervention:

Usual care

All the performed punctures are in any case part of the standard/usual care. So, no extra puncture procedures are scheduled for this study. Within the usual care, no one of the attendees in the room will have specific interaction with the child during this procedure, with the exception of normal/necessary interaction by the nurse and/or parent. The child will sit down on the treatment table with the legs stretched out in front of him and the puncture-side arm in a 90-90 position next to the head or stretched out and lying down along the body, with the nurse at the puncture side and the parent at the other side standing next to him. The researcher is also present in the room, within the child's field of vision. At the end of the procedure, the nurse tells the child that he/she did very well.

Robot-assisted puncture procedure

During the experimental intervention, the child, a nurse, one of the parents/guardians, a researcher and the humanoid robot NAO (H25 Academic Edition, Aldebaran Robotics, Paris, France) will be present in the same room. The child will sit down in the same position as with the usual care. Next to the patient a humanoid robot of three-foot tall will sit on eye level on a slanted reading table, at the non-puncture side. The nurse will carry out the puncture procedure as performed as usual. The robot is programmed to distract the child during the entire procedure (before, during and after the puncture) by playing a game with the child based on his/her interests. The child can therefore choose between a number of games in different themes. In the end the robot tells the child that he/she did very well. During the whole intervention, the robot will be re-activated for each phase only when the child and the nurse are ready.

Remarks:

Whether or not and how much Kalinox® (mixture of nitrous oxide (50 %) and oxygen (50 %)) is administered via a respirator during the procedure is registered for each participant. Breathing the gas mixture causes the child to feel a bit drowsy, but the child's awareness and communication possibilities are preserved.

After all child and parent questions have been filled out, a closed envelope will be handed to the parent. This envelope contains the questionnaires that the parent and child will be asked to answer within 1 week after the procedure in a telephone interview. This telephone interview is aimed to assess child and parent (proxy) pain/pain-related fear memory, future pain/pain-related fear expectancies and child and parent self-efficacy expectancies towards future similar procedures. Also a free memory recall and a prompted recall about contextual details is conducted in the children. Parents and children are instructed to leave this envelope closed until they are asked to open it in the telephone interview.

Protocol deviations:

Initially, this study was intended to be a randomized clinical trial (N=104). Because of the rare nature of the eligible chronic conditions in children, the a priori determined sample size remained as a target. However, due to 1) a prolonged shutdown of research activities in 2020 due to restrictions inherent to the COVID-19 pandemic, 2) the very low number of eligible children at the participating hospitals between June 2019 and January 2022, and 3) limited funding availability for this research project, this study was modified to be a pilot and feasibility trial. Consequently, only a few initial pre-specified feasibility criteria were set in the official trial registration (e.g., needle-procedure duration). However, prior to any data analysis, consensus amongst the research team was reached to set some additional feasibility outcomes, including study procedure variables (e.g., recruitment rate, dropout rate, follow-up rate), medical procedure-related characteristics (e.g., procedure duration), intervention fidelity, and intervention acceptability, which were all consistently and clearly described and assessed during the study. The target sample size and statistical methods were adjusted to enable evaluation of intervention feasibility as well as efficacy. Accordingly, we aimed to recruit at least 10 participants per group to be able to explore intervention feasibility and preliminary assess the impact of robot-led distraction upon pain-related memory bias.

Inclusion Criteria:

All participants:

• Chronic disease
• Willing to participate and comply with pre-determined interventions
• 8-12y (children)
• Child and parent(s)/guardian cohabited for the past 5 years or, in case of divided custody, for at least half of the child's lifetime
• Currently being treated (not necessarily hospitalized) in the University Hospital Brussels

Exclusion Criteria:

• Developmental disabilities (e.g.: autism spectrum disorder, Attention Deficit (Hyperactivity) Disorder)
• Psychiatric disorders (e.g.: anxiety disorder, personality disorder)
• Significant vision or hearing impairment
• Not able to read and speak fluent Dutch (parent and child)