Targeting Pain Science Education in Youth With Chronic Pain What Are the Sticking Points for Youth and Their Parents?

Abstract

Objectives:

An important part of providing pain science education is to first assess baseline knowledge and beliefs about pain, thereby identifying misconceptions and establishing individually-tailored learning objectives. The Concept of Pain Inventory (COPI) was developed to support this need. This study aimed to characterize the concept of pain in care-seeking youth and their parents, to examine its clinical and demographic correlates, and to identify conceptual gaps.

Materials and Methods:

Following an initial interdisciplinary evaluation, a cohort of 127 youth aged 8 to 18 years, and their parents, completed a series of questionnaires.

Results:

Parents had slightly higher COPI scores than youth did, reflecting parents’ greater alignment with contemporary pain science. The moderate positive association with older age among youth (r = .32) suggests that COPI is sensitive to cognitive development and life experiences. Youth and parent COPI responses were weakly associated (r = 0.24), highlighting the importance of targeting the concept of pain in both groups. For both parents and youth, ‘Learning about pain can help you feel less pain’ was the least endorsed concept. This conceptual ‘gap’ is a key point of intervention that could potentially lead to greater engagement with multidisciplinary pain treatment.

Discussion:

The COPI appears useful for identifying conceptual gaps or ‘sticking points’; this may be an important step to pre-emptively address misconceptions about pain through pain science education. Future research should determine the utility of COPI in assessing and treating youth seeking care for pain. The COPI may be a useful tool for tailoring pain science education to youth and their parents.

Approximately 30% of youth aged 8 to 18 years report pain lasting 3 months or longer.^1,2 Youth challenged by persistent pain may access multidisciplinary care to address the many biopsychosocial contributors to, as well as functional domains impacted by, their pain experience.^2,3 One important component of such care is pain science education (PSE),⁴ which aims to instill a basic understanding of the scientific basis for persisting pain, the biopsychosocial nature of pain, and the potentially large disconnect between pain and tissue damage.^5–7 A primary goal of PSE is to enable and empower individuals to make optimal decisions regarding their pain care. Research in adults shows the utility of PSE to facilitate participation in active self-management of pain.^8,9 Recent evidence in youth appears promising.^10–13

An important part of providing PSE is to first assess baseline knowledge and beliefs about pain, thereby identifying conceptual gaps in a contemporary scientific understanding of pain, identifying misconceptions about pain, and establishing individually-tailored learning objectives. The Concept of Pain Inventory (COPI) is a brief, 14-item tool, that was developed to support this need (Supplemental Digital Content 1, http://links.lww.com/CJP/A897). In an initial study involving an iterative development process with Australian children aged 8 to 12 years seeking care for pain,¹⁴ total COPI scores were inversely correlated with pain intensity and functional disability at baseline and at a 4-week follow-up. This provides preliminary evidence that a contemporary scientific understanding of pain is associated with clinical characteristics of pain in youth. However, that initial study comprised youth seeking care for both acute and persistent pain; COPI has not yet been investigated in a larger sample of youth seeking care for persistent pain. Moreover, the utility of COPI for older adolescents (aged over 13 y) is unknown. In addition, the utility of COPI for assessing parents’ concept of pain and how this relates to the child’s own concept of pain is also unknown. Given the well-established role of parents within the context of youth pain treatment,¹⁵ there may be additional benefit in assessing parents’ concept of pain and the relationship between the parent’s and youth’s concept of pain.

The current study aimed to address the research gaps. Specifically, we aimed to characterize the concept of pain in youth seeking care for persistent pain and their parents and to examine clinical and demographic correlates of the concept of pain in this population. In addition, we aimed to identify items with a particularly low or uncertain endorsement that could represent important conceptual gaps and misconceptions of pain, or ‘sticking points’ to be addressed in persisting pain care for youth.

MATERIALS AND METHODS

Participants

We included all youth and their parents who were seeking care for persisting pain at the Pediatric Pain Management Clinic (PPMC) at Stanford Children’s Hospital (Menlo Park). All youth who had completed an initial interdisciplinary evaluation between August 2018 and November 2020 were invited to participate in the present study. Each youth’s caregiver/parent was also invited to participate. Exclusion criteria included youth with cognitive impairments and youth and parents who are unable to read or communicate in English. Patients visiting the PPMC ranged in age from ~6 to 18 years, and patients who were between the ages of 8 and 17 at the time of their evaluation were included in the current study.

Procedures

As part of standard care at the PPMC, before initial evaluations, all patients completed a battery of measures (eg, demographic and pain data, psychosocial functioning measures) administered electronically through the Pediatric Collaborative Health Outcomes Information Registry (Peds-CHOIR).¹⁶ During the initial evaluation at the PPMC, youth and families were evaluated jointly by a pain physician and pain psychologist, and depending on the presenting concern (eg, musculoskeletal pain), also by a physical therapist. At the end of the evaluation, families received discipline-specific verbal summaries of the findings, which included diagnoses and causes (when applicable) and recommendations and treatment plans. Following those summaries, families were introduced to a research coordinator who informed them of the study. Patients and parents who wanted to participate consented and enrolled at this point. Depending on whether families were evaluated in person or through telehealth, study-specific self-administered questionnaires were then completed on a secure/encrypted tablet in the clinic waiting area or through a secure link sent through email. In addition, the demographic data and selected questionnaires (ie, Pain Catastrophizing Scale (PCS), Fear of Pain Questionnaire for Children–short form (FOPQ-10)) that had been collected before the evaluation (through Peds-CHOIR) were extracted for inclusion in the present study. All study-specific measures were collected through REDcap. Participants were compensated with a US $10 gift card. This study was approved by the Institutional Review Board at Stanford University (Protocol # 46499).

Measures

Demographics

Demographic data (age, gender, race, and ethnicity) and patient-reported primary pain problem data were extracted from the Peds-CHOIR registry.

Concept of Pain

COPI is a tool assessing the concept of pain and has demonstrated acceptable internal consistency (Cronbach alpha = 0.78) and moderate test-retest reliability (intraclass correlation coefficient (3,1) = 0.55; 95% CI, 0.37–0.68) in children.¹⁴ Items are rated on a 5-point scale from ‘strongly disagree’ (0) to ‘strongly agree’ (4). Items are summed to derive a total score, with higher scores reflecting knowledge and beliefs more closely aligned with contemporary pain science education. Parents completed the same items as their youth, and the same scoring parameters were used for the parent data. The present study administered a version of COPI that included all 21 items originally generated¹⁴ and was conducted concurrent to that separate study interrogating the 21 COPI items further. Our intent was to administer all 21 items and should further testing identify redundant or misbehaving items, we would not analyze responses to those items. That further testing resulted in the removal of 7 items,¹⁴ so in the current study, we did not analyze responses from those 7 items.

Pain Intensity

Youth were asked to provide their “typical or usual level of pain,” and their “current level of pain” on two 11-point numeric rating scales, from “no pain” (0) to “most pain possible” (10).^17,18 Pain location was reported using us a list of common body locations.

Functional Impairment

Pain-related functional impairment was assessed using the Functional Disability Inventory (FDI),¹⁹ which assesses difficulty in physical and psychosocial functioning due to physical health. The instrument consists of 15 items concerning perceptions of activity limitations during the past 2 weeks. Each item is rated on a 5-point Likert scale ranging from 0=‘no trouble’ to 4=‘impossible.’ The total score ranges from 0 to 60, such that higher scores indicate greater functional impairment. Published clinical reference points indicate responses ranging from 0 to 12=no/minimal disability, 13 to 29=moderate disability, and 30 to 60=severe disability.²⁰ The FDI has previously demonstrated strong internal consistency (α=0.86), test-retest reliability, and validity.²¹ In this sample, the internal consistency of the FDI was good (α=0.89).

Fear of Pain and Avoidance of Activities

The Fear of Pain Questionnaire for Children–short form (FOPQ-SF)^22,23 was used to assess pain-related fear and avoidance of activities. The FOPQ-SF consists of 10 items rated on a 5-point Likert scale ranging from “strongly disagree” (0) to “strongly agree” (4). Items are summed to create a total score ranging from 0 to 40, where higher scores indicate greater fear of pain and avoidance of activities. Published clinical reference points available indicate responses ranging from 0 to 19=low fear avoidance, 20 to 25=moderate fear avoidance, and 26 to 40=high fear avoidance. The FOPQ-SF has previously demonstrated strong internal consistency (α=0.84), test-retest reliability, and sensitivity to change.^22,23 In this sample, the internal consistency of the FOPQ-SF was good (α=0.84).

Pain Catastrophizing—Child

The Pain Catastrophizing Scale for Children (PCS-C)²⁴ was used to assess negative thinking associated with pain. It is comprised of 13 items rated on a 5-point Likert scale ranging from “not at all” (0) to “extremely”(4). The total score ranges from 0 to 52, where higher scores indicate higher levels of catastrophic thinking. The following clinical reference points have been established for the PCS-C: 0 to 14 = low, 15 to 25 = moderate, and ≥ 26 = high catastrophic thinking),²⁵ and have demonstrated strong internal consistency (α = 0.87).²⁴ In this sample, the internal consistency of the PCS-C was high (α = 0.91).

Pediatric Pain Screening Tool

The Pediatric Pain Screening Tool (PPST)²⁶ is a 9-item self-report screening tool used to identify prognostic factors (eg, interrupted sleep, mood concerns) associated with adverse outcomes among youth with chronic pain. The PPST contains a Physical (4 items) and Psychosocial (5 items) subscale. Patients respond “disagree” (0) or “agree” (1) to statements (eg, “It is difficult for me to fall asleep and stay asleep at night”). Items are summed to yield a total score ranging from 0 to 9. Scores on the PPST help identify patients at low, medium, and high risk for adverse outcomes, with allocation to high risk based on responses to psychosocial items (score >3 on those items).

Pain Catastrophizing—Parent

The Pain Catastrophizing Scale for Parents (PCS-P)²⁷ assesses parents’ negative cognitions associated with their child’s pain. It is comprised of 13 items rated on a 5-point Likert scale from “strongly disagree” (0) to “strongly agree” (4). Items are summed to create a total score ranging from 0 to 52, with higher scores indicating higher levels of catastrophic thinking. In this sample, the internal consistency of the PCS-P was high (α=0.92).

Parent Pain-Related Distress

The Parent Risk and Impact Screening Measure (PRISM)²⁸ is a 12-item screening to assess parent physical, emotional, and behavioral responses to their child’s pain. Parents respond “disagree” (0) or “agree” (1) to written prompts (eg, “I believe my child’s pain problem is out of control”), yielding a total score ranging from 0 to 12. Parents are classified into high-or low-risk based on the total score (high risk=>6). In this sample, the internal consistency of the PRISM was good (α=0.84).

Statistical Analyses

A minimum target sample size of 105 respondents was established based on the principle of 5 to 10 individuals per inventory item²⁹ and the 21-item pool.

Data were double-entered and analyzed using SPSS v22.0 (SPSS Inc) and Rstudio (Rstudio Team). Descriptive statistics were conducted to examine underlying assumptions of normality (skew and kurtosis, considered significant if <−2 or >2) and to characterize youth and parent concept of pain. “Sticking points” were items with low rates of agreement. We compared the distribution of youth scores to previously published data.¹⁴ To examine associations, Pearson product-moment-correlation (r, interpreted: weak=0.10–0.29, moderate=0.30 to 0.49, and strong=>0.5) and Kendall rank coefficient (tau, interpreted: weak= 0.10–0.19, moderate=0.20–0.29, and strong=>0.3) analyses were conducted.

To assess the agreement between youth and parent’s concept of pain at baseline, the Bland-Altman method³⁰ was used. The average of youth and parent total COPI scores were plotted against the differences between these scores. The mean and SD of the mean difference were calculated, as well as the true value of the mean, by using 95% CIs to further assess the existence of systematic bias.

RESULTS

Participants

During the study period, 182 youth aged 8 to 17 years underwent an initial evaluation in the PPMC. Of these, 127 (70%) provided consent and completed all measures and were included in the study. Youth predominately identified as white (38%) and female (71%) with a mean (SD) age of 13.9 (2.5) years and median (IQR) pain duration of 28.5 (10 to 70.5) months. Youth reported a variety of pain locations, including head, shoulder/neck, back, and abdomen; 68% reported that they experienced pain daily. The mean (SD) typical pain intensity rating was 5.7 (1.9). The mean (SD; range) doctor visits in the preceding year, related to the child’s pain, was 4.4 (3.1, 1 to 20).

In addition, 1 parent/caregiver per patient consented to participate and completed all questionnaires. From herein, parent/caregiver participants will be referred to as “parents.” Most parents were the patient’s mother (91%); 6% were the patient’s father, and 3% were a different type of caregiver. Mean (SD, range) parent age was 45.5 (6.7, 32 to 65) years; 49% reported being employed full-time; 52% reported having earned at least a bachelor’s degree. Regarding pain, 55% of parents reported experiencing chronic pain themselves in their history. Table 1 contains demographics and baseline clinical variables for youth and their parents.

TABLE 1.

Demographics and Baseline Clinical Variables of Youth Attending an Initial Pain Evaluation (n = 127)

Variable	n (%) (For Categorical Variables)	Mean (SD) (For Continuous Variables)
Age (y) Sex	—	13.9 (2.5)
Male	33 (26)	—
Female	90 (71)	—
Prefer not to answer	2 (2)	—
Missing Race (‘tick all that apply’)	2 (2)	—
Caucasian	48 (38)	—
Hispanic or Latino	17 (13)	—
Asian	6(5)	—
Black or African-American	3 (2)	—
American Indian or Alaskan Native	3 (2)	—
Multiracial	1 (1)	—
Other	2 (2)	—
Missing	48 (38)	—
Primary language spoken at home
English	118 (93)	—
Spanish	5 (4)	—
Missing	4(3)	—
Previous pain science education–yes	58 (46)	—
Typical pain (scores 0–10)-Child	—	5.7 (1.9)
Current pain (scores 0–10)-Child	—	4.5 (2.6)
Fear of Pain Questionnaire total (scores 0–40)	—	22.5 (6.4)
Pain-related fear (scores 0–16)	—	10.3 (3.8)
Fear-avoidance of activities (scores 0–24)	—	12.3 (3.9)
Functional disability (scores 0–60)	—	21.7 (10.7)
Pain catastrophizing (scores 0–52)-child	—	28.0 (10.0)
Pain duration (mo)		Median (IQR) = 28.5 (10–70.5)
Pain location (can endorse one or more)
Head	58 (46)	—
Face/jaw	14 (13)	—
Shoulder/neck	47 (37)	—
Chest	29 (23)	—
Arm/hand	34 (27)	—
Abdominal	48 (38)	—
Back	56 (44)	—
Leg/foot	58 (46)	—
Other	25 (20)	—
Parent history of recurrent or persistent pain	69 (55)	—
Parent current recurrent or persistent pain	48 (38)	—
Pain catastrophizing (scores 0–52)–Parent	—	22.5 (9.5)
Pediatric pain risk factors (scores 0–9)	—	8.9 (2.5)
Parent Risk and Impact Screening Measure (PRISM scores 0–12)		6.3 (3.4)

Note: Characteristics may not sum to 100% due to the effect of rounding.

All measures are summed scored such that higher scores indicate more a more clinically vulnerable presentation (eg, higher fear of pain, higher functional disability).

On average, youth reported moderate pain-related fear avoidance, high catastrophic thinking, moderate functional disability, and a high risk for adverse outcomes (Table 1). High variability in youth responses was observed for functional disability and catastrophic thinking. On average, parents reported moderate levels of catastrophic thinking and would be categorized as high risk for pain-related distress through the PRISM.

Youths’ and Parents’ Concept of Pain

Internal Consistency and Understandability

The internal consistency for the current sample was just below acceptable in youth (Cronbach alpha=0.69) and good in parents (Cronbach alpha=0.88). See Table 2 for descriptive characteristics (mean, SD, skew, and kurtosis) of items for the COPI responses of youths and their parents. Regarding independent completion of the COPI, 87% of youth responded “no” to the question: “Did you have to get your parent to help you with these questions about pain?” A moderate correlation with age (tau=0.21, P=0.007) revealed older participants more frequently reported independent completion of COPI than younger participants.

TABLE 2.

Descriptive Statistics for Items for the COPI Responses of Youth and their Parents

	Youth				Parent
	Mean	SD	Skewness	Kurtosis	Mean	SD	Skewness	Kurtosis
1. Feeling stressed can make you feel more pain.	3.03	0.78	−0.67	0.40	3.10	0.79	−1.08	2.03*
2. Feeling sad can make you feel more pain.	2.31	1.10	−0.38	−0.64	2.83	0.85	−1.39	2.75*
3. Being distracted can make you feel less pain.	3.05	0.80	−0.66	0.20	3.09	0.78	−1.29	3.27*
4. Doing something you enjoy can make you feel less pain.	2.75	0.98	−0.98	0.71	2.97	0.79	−1.24	2.86*
5. Pain is a warning that the body needs to be protected.	2.50	0.87	−0.23	−0.30	2.74	0.86	−1.01	1.46
6. Feeling pain for a long time can make the brain more sensitive to warning messages.	2.60	0.86	−0.20	−0.55	2.80	0.93	−0.79	0.67
7. You can feel a lot of pain even when an injury is small.	3.06	0.77	−1.05	2.08*	3.16	0.72	−1.29	3.70*
8. Learning about pain can help you to feel less pain.	1.94	1.01	−0.35	−0.78	2.60	0.86	−0.67	0.80
9. You can have an injury and feel no pain.	2.39	1.06	−0.68	−0.29	2.91	0.84	−1.16	1.97
10. The brain can make pain better or worse.	2.65	0.89	−0.58	0.28	2.89	0.74	−0.91	1.92
11. You can feel a little bit of pain even when an injury is big.	2.78	0.94	−1.10	1.25	2.99	0.88	−1.35	2.45*
12. You can feel pain even after an injury heals.	2.94	0.82	−0.83	1.00	3.05	0.76	−1.09	2.39*
13. Pain usually feels better if you move your body a little bit more each day.	2.24	0.93	−0.56	−0.08	2.70	0.90	−0.64	0.31
14. The brain processes lots of details before you feel pain.	2.26	0.79	0.09	0.04	2.72	0.82	−0.84	1.57

^*indicates a value of < −2 or > 2.

Response Patterns

Visual inspection of responses to individual items (Fig. 1) indicated similar response patterns for both groups with overall high levels of agreement (green) with COPI items for comparisons between both (1) youth and parents and (2) youth and the reference cohort of Australian children published elsewhere.¹⁴ The median (IQR) total COPI score for youth was 37 (33 to 40), and the median (IQR) total parent COPI score was 40 (38 to 44). Youth provided more “disagree” and “unsure” responses than their parents did. The most common response for parents was “agree.” In assessing the relationship between parent’s and youth’s concept of pain, the Pearson product-moment-correlation revealed a positive relationship between parent and child total COPI scores r=0.24, P=0.006, such that as the parent’s total COPI scores increased, so too did child’s total COPI scores.

FIGURE 1.

A, The distribution of responses to individual COPI items for youth and parents. B, The distribution of responses to individual COPI items for youth compared with the child data taken from Pate et al.¹⁴

Sticking Points

Some items had low rates of agreement and therefore appeared to be ‘sticking points’ among youths’ concept of pain: 65% responded with ‘unsure’ or disagreement to item 8 ‘Learning about your pain can make you feel better.’; 65% responded with ‘unsure’ or disagreement to item 14 ‘The brain processes a lot of details before you feel pain’; 56% responded with ‘unsure’ or disagreement to item 13 ‘Pain usually feels better if you move your body a little bit more each day’; 50% responded with ‘unsure’ or disagreement to item 2 ‘Feeling sad can make you feel more pain.’ Among parents, no item elicited less than 50% agreement; however, some questions did garner more varied responses than others. Examining the association between child and parent responses to the sticking points listed above, there were significant relationships regarding item 8 (tau=0.18, P=0.021) and item 14 (tau=0.22, P=0.006); however, no significant relationship was observed for item 13 (tau=0.10, P=0.218) and item 2 (tau=0.11, P=0.140). Examining the 127 individual-level responses, no child selected the same response for all COPI items. In contrast, 5 parents selected the same response for all COPI items. Sixty-two percent of youth responded ‘unsure’ to 0–3 of the 14 COPI items; 63% responded ‘unsure’ to 0–2 of the 14 COPI items.

Associations with Demographic and Clinical Characteristics

Given the positive correlation between youth’s COPI total score and older age (r=0.32, P<0.001), age was controlled for in partial correlation analyses with all clinical variables. For youth, no association was found between COPI total scores and: current pain intensity (r=0.14, P=0.13), average pain intensity (r=0.09, P=0.38), functional disability (r=0.11, P=0.21), pain catastrophizing (r=0.00, P=0.17), fear of pain (r=0.14, P=0.17 ), avoidance of activities (r=0.09, P=0.38 ), and risk-level on the PPST (r=0.04, P=0.73).

For parents, no significant associations were found between COPI total score and youth’s functional disability (r=−0.03, P=0.71), the youth fear of pain (r=0.01, P=0.94), and avoidance of activities (r=0.01, P=0.96), and parent pain catastrophizing (r=−0.01, P=0.96). There were significant associations between parent COPI total score and the youth’s pain catastrophizing (r=−0.30, P=0.002)²⁰ and PRISM total (r=0.21, P=0.02), PRISM Parent Behaviors (r=0.21, P=0.02), and PRISM Family Impact (r=0.22, P=0.01) scores.

Agreement.

The Bland-Altman plot (Fig. 2) shows the difference between youth and parent COPI scores against their average. The discrepancy in the vertical direction shows the extent of the agreement between youth and parent scores, and the pattern indicates the nature of the discrepancy. Given the plotted points are mostly distributed below the zero line, systematic bias is observed between youth and parent scores (parents scored higher on average). The mean difference between youth and parent scores was −4 points out of a possible total score of 56. The SD of this difference, 8 points, indicates that most plotted points showed a difference of less than 16 points (2 SD’s). The pattern shows a horizontal cloud of plotted points: to formally assess for any potential proportional bias, a linear regression (of difference against mean) was run, resulting in a slope estimate (unstandardized B) of −0.395, P=0.137). This lack of significance indicates there is no proportional bias; that is, no greater difference between youth and parent scores for those with higher or lower scores on the COPI.

FIGURE 2.

Bland-Altman plot of the difference in COPI scores between youth and parents (y axis) versus mean of COPI scores of youth and parents (x-axis) (data from n = 127 youth and their parents).

DISCUSSION

A sample of 127 youth and their parents presenting to a tertiary pain clinic in the United States completed the COPI alongside demographic and clinical self-report measures. We found a moderate positive association with older age among youth (r=0.32). Parents had slightly higher COPI scores than youth did, reflecting parents’ greater alignment with contemporary pain science. Youth and parent COPI responses were weakly associated (r=0.24), highlighting the importance of targeting the concept of pain in both groups. For both parents and youth, ‘Learning about pain can help you feel less pain’ was the least endorsed concept. This conceptual ‘gap’ is a key point of intervention that, if addressed, could potentially lead to greater engagement and buy-in to multidisciplinary pain treatment.

This study corroborates and extends a previous validation study comprising a younger sample seeking care for both acute and chronic pain.²⁴ Total COPI scores in the current study were similar to those observed in the original validation study.¹⁴ Given that both samples had sought care for pain, this finding was anticipated. The current sample, however, had a wider age range, was older, predominantly female, and reported higher scores of pain intensity, avoidance and fear of pain, pain catastrophizing, and functional disability.¹⁴ In both studies, COPI scores were unrelated to pain catastrophizing and pain-related fear. The distribution of total COPI scores was also similar to a recent study of the COPI-Adult.³¹

Youth reported moderate functional impairment and fear avoidance, and high pain catastrophizing.^20,25 Total COPI scores were positively associated with age, such that each year of age resulted in 0.73 points higher total COPI scores. This supports the validity of the COPI because it is reasonable that scientific pain knowledge would increase alongside cognitive development and/or life experiences.³² Given these findings, we controlled for age in subsequent analyses and suggested the same should be done in future research.

No significant associations were found between COPI total scores and pain intensity or disability. This contrasts with initial validation findings,¹⁴ which reported higher COPI total scores inversely correlated with pain intensity and functional disability. The lack of association raises 2 interesting possibilities. Perhaps the COPI total score lacks content validity and does not capture aspects of the concept of pain that are most relevant to pain and disability in youth. Or, perhaps the COPI has adequate content validity, as previously reported,¹⁴ and the concept of pain is neither an important determinant of pain and disability, nor predictive of pain and disability, in youth. Regardless, the clear clinical implication is to expect conceptual variability in youth seeking pain care, regardless of their clinical presentation. This finding does not imply that a shift in the concept of pain does not promote a reduction in pain and disability, which has been demonstrated in adults,³³ with preliminary evidence that a shift in the concept of pain is an important determinant of long-term clinical benefit.³⁴

Parent total COPI scores were associated with several domains of parent and family functioning in unanticipated directions (ie, higher parent COPI responses were associated with greater risk on PRISM scores and greater impact of pain on family). This unexpected relationship highlights the importance of discussing treatment processes alongside PSE. For example, parents may have a concept of pain well-aligned with contemporary pain science but may not have the support or skills to apply the implications of their concept of pain to their child and to promote functional improvement. In contrast, parents who demonstrated a higher total COPI score had children who reported lower pain catastrophizing, highlighting the importance of parent intervention in supporting pain treatments for youth. There was no divergence or convergence between youth and parent scores according to mean score, as evidenced by the Bland-Altman plot of youth-parent difference in COPI versus youth-parent combined average COPI (Fig. 2). This finding supports individualized targeting of education for youth and their parents, regardless of their baseline scores.

Clinical Implications

Some items emerged as clear sticking points that may warrant targeted intervention. Among youth’s responses, the item most disagreed with was ‘Learning about pain can help you to feel less pain,’ which is interesting because this is a common starting point for pain- education. Clear messaging from the integrated team should include information about why pain science education is recommended and address barriers or resistance to multimodal treatment. In addition, if this item is disagreed with, or strongly disagreed with, clinicians can pivot their treatment plan to first address misconceptions of pain and the role of pain science education. In a cohort of recovered adult persistent pain patients, 70% indicated that the concept ‘Learning about pain is an important therapy in itself’ was very important for recovery.³⁵ This perspective was also clear in a wider, iterative investigation of consumer perspectives on what recovered patients perceive to be important components of PSE, with ‘learning about pain is important’ identified as a key learning objective of intervention (Leake et al, 2022, unpublished data under review). Other sticking points that emerged reflect elements of PSE that challenge the longstanding biomedical narrative of pain (e.g., ‘Feeling sad can make you feel more pain,’ ‘Pain usually feels better if you move your body a little bit more each day,’ and ‘The brain processes a lot of details before you feel pain’). Given this, COPI offers important clinical information for gaining an understanding of the patients’ framework of pain, including biomedical versus biopsychosocial perspectives and potential intervention targets. Directly discussing individual responses alongside information about shifting from a biomedical to a biopsychosocial framework of pain may be an effective intervention for challenging existing concepts of pain.

For parents, the item most disagreed with was ‘Pain usually feels better if you move your body a little bit more each day.’ This is clinically interesting because our finding concurs with other research showing that parent beliefs around protective behaviors are a barrier to behavior change and positive outcomes.¹⁵ Clinicians often support parents in scaffolding expectations around functional improvement due to previously held beliefs that expecting movement or physical activity was too demanding for youth given their pain. Although parents reported higher agreement with COPI items than youth did, similar items emerged as sticking points for both groups, and addressing item responses with parents may also challenge a biomedical conceptualization of pain and help address barriers to treatment progress. The correlation between youth and parent COPI scores points toward the importance of considering both youth’s and parents’ concepts of pain when evaluating and targeting pain science education. Moreover, examining concordance between parent and child variables may also be clinically valuable because they relate to outcomes.³⁶

Strengths and Limitations

The cultural appropriateness of COPI was not tested, potentially limiting generalizability. The final 14 COPI items¹⁴ were not administered as a standalone set of 14 items, but as part of a pool of 21 items, with only the final 14 items analyzed. Therefore, we cannot exclude the possibility that responses to the 7 items that were not extracted for analysis may have influenced responses to the items retained. Replicating these results using only the 14-item version would be helpful. Further, participants were assessed after their initial multidisciplinary pain evaluation, and thus immediate recall of key phrases may have confounded the validity of some COPI responses because we did not control for education provided during the evaluation appointments. Testing at multiple time points during treatment may allow the identification of important mediators by which changes in pain understanding relate, if at all, to clinical variables. Finally, the study protocol was not preregistered, which is now a recommended practice in the pain field.³⁷ We recognize this limits the transparency of the current work.

Future Research

More work is required to determine the utility of COPI in assessing and treating youth seeking pain care, with several important research questions at the forefront: Will outcomes improve if youth and their parents develop concepts of pain that align with current pain science? Will aligning youth and parent concepts of pain facilitate collaboration between them? Does someone’s concept of pain change over time following PSE, and are people aware of these changes or do they happen subconsciously? Does someone’s concept of pain, or particular COPI item responses, predict future clinical outcomes (addressing the question of how conceptual change relates to behavioral change)? How do youths’ and parents’ concept of pain translate to an understanding of their own pain experience (e.g., ‘I can feel pain even after an injury heals’). To address these questions, researchers and clinicians may use the COPI in a systematic way across time with their patients and may use COPI to both inform discussions around the parents’ and youth’s concept of pain. Given the very personal history patients and families have with pain, contextualizing responses qualitatively may also offer important insights to tailor treatment planning and clinical intervention.

CONCLUSIONS

The concept of the pain of youth aged 8 to 18 years and their parents presenting to a tertiary pain clinic varies at an individual and item level. COPI appears to be useful for identifying common ‘sticking points’ for tailoring PSE to individuals and their parents.