Preliminary development of a measure of parental behavioral responses to everyday pains in young children: the PREP

Supplemental Digital Content is Available in the Text. The PREP is a relatively short (10 item) measure assessing self-reported observable parent responses to their child's everyday pains.


Introduction
Everyday pains (EDPs; ie, minor bumps and scrapes incurred during everyday activities) are experienced frequently, particularly early in development when children begin walking and attending school. 13,23,32Everyday pains provide young children with opportunities for learning about pain during critical developmental periods.This learning could be influenced by parents' responses to painful events.Nevertheless, research examining EDPs is scarce, and no self-report measures of parent responses to young children's EDPs exist.Among infant, child, and adolescent samples, parent behaviors that direct attention toward pain (eg, verbal reassurance) or that reinforce pain behaviors (eg, protectiveness) are linked to increased child pain, distress, and disability [2][3][4]7,9,22,24,32 and have been conceptualized within an operant learning framework. 6,14,19 Howeer, it cannot be assumed that parent responses have the same effect in clinical and naturalistic settings 25,33 and in children with and without chronic conditions.To date, only a handful of studies have examined EDPs in children aged 1 to 7 years.,26,32 More recently, parents have used electronic diaries to self-report their child's pain and their own distress.However, this field is limited by the lack of self-report measures of caregiver responses to EDPs in young children's.
Informed by parent behaviors observed in a study of parent responses to young children's EDPs, 23 we developed a preliminary self-report measure of parent behavioral responses to EDPs during the toddler and preschool years (the PREP) and examined its relationship with child age, sex, and parent and child distress.

Participants
Parents (N 5 294) participated in an online survey which took approximately 30 minutes to complete.Participants missing .3items were excluded, leaving a final sample of N 5 290 parents, which is consistent with best practice recommendations for exploratory factor analyses (EFAs). 5,18Participants were primarily mothers (93%) of children aged 18 to 60 months (Table 1).Inclusion criteria were (1) have .1 child between 18-60 months, (2) able to read and understand English, and (3) have Internet access.The Research Ethics Board approved this study.

Demographics
Parents reported their age, racialized identity/ethnicity, relationship to child, marital status, and educational level.They reported on their child's age, sex, and racialized identity/ethnicity.

Everyday pains in young children
The Parent Responses to Everyday Pains (the PREP) questionnaire assessed parents' self-reported responses to their child's EDP experiences.Everyday pain experiences were defined as instances when a child's body comes into contact with another person or object and results in either: (1) a distress (eg, crying), anger, or protective (eg, help-seeking) reaction from the child or (2) the parent judges that the child experienced at least momentary, minor discomfort.Items were based on the Dalhousie Everyday Pains Scale-Revised, a behavioural checklist used in an observational study of parent responses to toddler's EDPs. 13Using a 5-point Likert scale ranging from 1 ("never") to 5 ("usually"), parents were asked to indicate how often they engage in 46 responses.Possible responses were separated into 28 verbal statements (eg, "Tell him/her to stop crying") and 18 observable actions (eg, "Distract child with a toy") (Supplementary Material, http://links.lww.com/PR9/A229).Given that the empirical study from which this measure was based 23 coded only parents' observable actions and not verbal statements (due to the busy nature of the play setting), only items reflecting observable responses to children's pain were included in the analyses.

Parent and child distress
Parents were asked to indicate how distressed they and their child are, on average, following the child's EDP experiences on a 6-point Likert scale ranging from 0 ("low distress") to 5 ("high distress").

Procedure
Parents were recruited to participate in the study through online and newspaper advertisements, posters, and email invitations through hospitals, community centers, and social media.Parents provided consent to participate and were entered into a draw for 1 of 3 $100 gift certificates.

Data analysis
Analyses were conducted using SPSS (v22).An EFA was performed on the 18 observable PREP items.Similar to previous pediatric pain studies, 12,24 iterated principal axis factoring using a nonorthogonal (oblique) rotation method was used as this method allows factors to be correlated, resulting in a more accurate, reproducible solution. 11Eigenvalues $1, 10 .50% of  PAIN Reports ® variance explained, 17,29,31 and the initial scree plot were used to select a factor structure.
To be retained in the model, individual items were required to have a primary loading above the recommended threshold of 0.32 31 and items that cross-loaded on the same factor had to have loadings of at least twice the value of the other. 12We systematically removed items one at a time and reran analyses each time, applying these criteria at each stage.Models were run until a clean solution was obtained.Final clean solutions were examined and compared.The superiority of models was judged based on interpretability and overall variance accounted for.Factor solutions were required to have at least 3 items loading onto each subscale. 10ivariate correlations were conducted between the final subscales and child age, sex, and child and parent distress.

Results
The initial EFA resulted in 5 factors with initial eigenvalues above 1.0% and 63.4% variance explained.The scree plot suggested a 3-factor solution; therefore, 5-factor, 4-factor, and 3-factor models were tested by forcing each respective solution.In the initial EFA, the 4-factor and 3-factor models accounted for 57.5% and 50.4% of the variance, respectively.
After systematically removing poorly loading and highly crossloading items for each model, final clean solutions were obtained.The 5-factor solution accounted for 55.6% of the total model variance; however, 2 of the factors included only 2 items.The 4factor model explained 56.6% of the total model variance; however, one of the subscales was indicative of an absence of soothing behaviors, which was deemed not as interpretable as the 3-factor model.The 3-factor model was deemed to be superior to the 4-factor and 5-factor models based on both interpretability and total variance explained.The final solution included 10 items across 3 factors: Distract (4 items), Physical Soothe (3 items), and Extra Attention (3 items).The final 3-factor solution explained 60% of the model variance with each factor explaining 33.2%, 19.6%, and 7.2% of the variance, respectively.The final model had primary loadings of .0.50 (Table 2).
Internal consistency was high for the Distract (0.87) and Physical Soothe (0.83) subscales and acceptable for the Extra Attention subscale (0.69).Table 3 presents mean subscale and item scores.

Discussion
This study describes the development and preliminary testing of a 10-item self-report measure of parental responses to EDPs during the toddler and preschool years.While some items (Distract, Extra Attention) are consistent with responses assessed on measures for use with older youth, 24 others (Physical Soothe) are more characteristic of earlier developmental periods.Parents of younger children in our previous study were more likely to engage in more physical soothing in response to their children's EDPs. 23All PREP subscales were related to child distress in the context of EDPs; however, relationships were small and require further investigation.While higher Physical Soothe and Extra Attention scores were related to greater parent distress to  children's EDPs, distracting responses were not.These findings may be conceptualized within Goubert's pain empathy model, 16 which posits that parents who experience self-oriented distress on observing their child in pain may be less able to engage in responses to child pain that can foster adaptive coping.Distracting responses have been linked to less child pain and distress in laboratory-based and clinical pain settings among infants to older adolescents. 1,8,21,28Physical soothing and providing extra attention may be driven by heightened parental distress and a self-oriented response; however, further research is needed.This study has limitations.The PREP includes only observable parent responses to children's EDPs in a sample of primarily mothers.Parents' verbal behaviors (eg, reassurance) have been linked to child pain outcomes in other acute pain settings 7,22 ; further measure development work examining verbal responses is warranted.The utility of the PREP as a nonverbal behavioral measure of parent responses might be limited to younger developmental periods.Effect sizes of the relationships between the PREP subscales and child and parent distress were small; therefore, practical implications cannot be definitively drawn.Research is needed with larger samples to further validate the PREP using confirmatory factor analysis and relationships to other measures of parent responses to child pain.Future research using the PREP with ecological momentary assessment methods could be used to assess variation in parental distress and responses to EDPs across contexts.
* Possible scores ranged from 0 to 5 with higher scores indicating more distress or pain.

Table 2
Exploratory factor analysis factor loading matrix for parent behavioral responses to everyday pains in the toddler and preschool years items for the final 3-factor model.The following items were removed: show no reaction/try to show no reaction; leave the room; pay no attention until he/she goes back to playing; rub the hurt area better; kiss the "boo-boo" or hurt area better; stay closer to him/ her; keep close watch on his/her activities so that he/she does not get hurt again; gently tap or swat his/her bottom.PREP, parent behavioral responses to everyday pains in the toddler and preschool years.

Table 3
Mean (SD) parent behavioral responses to everyday pains in the toddler and preschool years scores by subscale (N 5 290).
Possible scores ranged from 1 to 5; higher scores indicate more frequent engagement in pain response.* n 5 289 responses.