Abstract
Assessing parent readiness to support their child's post-concussion management requires valid and reliable measures. Therefore, the objective of this study was to develop and conduct preliminary tests of reliability and validity of survey measures of parent concussion management knowledge and self-efficacy. Additionally, we tested the hypothesis that among parents of youth who had sustained a concussion, higher scores on measures of knowledge and self-efficacy would predict greater likelihood of engaging in recommended concussion management behaviors during their child's recovery. Measure development occurred with reference to parenting behaviors included in the Center for Disease Control and Prevention's Pediatric Mild Traumatic Brain Injury (mTBI) Management Guidelines. A multi-stage mixed- methods approach was employed, including expert review, cognitive interviews with parents, quantitative item reduction, and tests of reliability and validity. All participants were English-speaking parents of school-aged children in the United States. A stepwise measure development process was followed, with different participant groups across steps (including opt-in web-based survey panels and in-person recruitment from the population of parents of pediatric patients seen in a large pediatric emergency department). In total, 774 parents participated in study activities. The final knowledge index had 10 items, and the final self-efficacy scale had 13 items across four subscales (emotional support, rehabilitation support, monitoring, and external engagement). Internal consistency reliability was 0.63 for the knowledge index and 0.79–0.91 for self-efficacy sub-scales, and validation tests were in the hypothesized directions. In a test of predictive validity, we observed that among parents of youth patients with recent concussion, higher self-efficacy scores at the time of discharge from the pediatric emergency department were positively correlated (r = 0.12) with greater likelihood of engaging in recommended support behaviors at 2-week follow-up. There was no association between concussion management knowledge at discharge and parenting behaviors at follow-up. Parents have the potential to play an important role in concussion management. The measures of knowledge and self-efficacy developed in this study can help identify parent needs and evaluate interventions aiming to support parenting post-concussion.
Keywords: concussion, health literacy, measure development, parent
Introduction
Every year, >1,000,000 youth in the United States sustain a concussion.1 Appropriate management is important for minimizing risk of subsequent injury and optimizing quality of life during recovery, including psychosocial functioning.2 Consensus guidelines recommend rest for 24–48 h, then moderating cognitive and physical demands in the school and home setting in response to symptoms, and avoiding activities with a risk of contact or collision until fully recovered.3 Post-concussion, many youth experience elevated symptomatology of anxiety and depression,4 which means that emotional support and monitoring of psychosocial functioning are also important aspects of supporting youth quality of life during this time period.5
Parents and guardians (hereafter referred to as parents) are key partners in implementing evidence-based post-concussion management in the home, school, and sport settings.6 However, many parents have difficulty supporting their child's recovery.7 One contributor may be their inadequate concussion management-relevant health literacy. Consistent with Sørensen's Integrated Model of Health Literacy, individuals who are more health literate are better able to access health information, understand this information, and act based on this information.8 Health literacy is impacted by structural, social, and situational factors;8 across a range of health issues, lower health literacy is associated with less use of healthcare services and worse health outcomes.9 Others have documented a lack of concussion knowledge (i.e., the “understanding” element of health literacy) among parents,10 with such deficits heightened among less affluent parents and those with fewer years of formal education.11,12 Parental self-efficacy related to concussion management (i.e., confidence in their ability to engage in recommended support behaviors, or the “act” element of health literacy) has not as yet been assessed.
Assessing parent readiness to provide evidence-based support to their child as the child recovers from a concussion requires valid and reliable measures. Such measures are critical for conducting needs assessments to inform intervention development, and for evaluating interventions that aim to increase parent readiness to provide post-concussion support. At present, there are no measures that assess parent self-efficacy. Existing indices that measure concussion knowledge were designed for a general audience rather than specifically with reference to parents' role in the provision of guideline-consistent support post-concussion, meaning that content is not matched to parenting behaviors. To address these gaps, the goal of the present study was to develop and conduct preliminary tests of reliability and validity of survey measures to assess parent knowledge and self-efficacy related to the provision of guideline-consistent support for their child during the child's recovery from a concussion. Additionally, we tested the hypothesis that among parents of youth who sustained a concussion, higher scores on measures of knowledge and self-efficacy would be associated with greater likelihood of engaging in recommended support behaviors during their child's recovery.
Methods
Study design
This study was designed to develop measures of parent knowledge and self-efficacy regarding their role in pediatric concussion management. Elements included (1) item development and refinement, (2) item reduction and preliminary tests of validity and reliability, and (3) longitudinal assessment of predictive validity. The [Seattle Children's Research Institute's] Institutional Review Board approved all research activities. All parent participants provided informed consent and were compensated $10 per survey. When parents were discussing the injury of an adolescent child (age 13–17), this child also provided assent.
Item development and refinement
We aimed to measure two distinct constructs: (1) knowledge and (2) self-efficacy. We defined knowledge as retrieval of information relevant to performing recommended parental support behaviors during their child's concussion recovery. We defined self-efficacy as parent confidence in their ability to successfully engage in these behaviors. We reviewed the Center for Disease Control and Prevention's pediatric mild traumatic brain injury (mTBI) management guidelines5 and mapped guidance/instruction for parents into discrete parenting behaviors. This produced an initial list of 26 behaviors (see Tables S1 and S2). Preliminary self-efficacy items were generated to correspond to each of these behaviors, following guidance about constructing self-efficacy items (e.g., to reflect how challenging it would be to perform specific behaviors in real-world conditions).13,14 To generate preliminary knowledge items, we built upon prior expert consensus about the knowledge domains relevant for parents to engage in post-concussion support behaviors.6 We reviewed existing measures of concussion knowledge among adults,15–20 extracted all items, and grouped these by knowledge domain (e.g., symptoms, natural history). Where necessary, we generated new knowledge questions so that there was more than one item corresponding to each parenting behavior. Next, four clinicians with expertise in pediatric concussion management reviewed the preliminary pool of items for wording clarity, accuracy of responses, and consistency with consensus guidance. Minor modifications were made based on their feedback.
Next, we pre-tested knowledge items to efficiently reduce the number of items that would subsequently progress to cognitive interviews and psychometric testing. The full set of preliminary survey items was first administered to 240 participants recruited through Amazon Mechanical Turk (MTurk).21,22 MTurk is an opt-in survey panel commonly used in social science research to efficiently recruit a large number of participants. A link to a survey hosted on REDCap was posted to MTurk and was accessible only to panel members self-identifying as parents of school-aged children and adolescents (age 5–17). Participants completed all knowledge questions. Items were removed or modified if >80% of participants answered the question correctly (items were not removed based on a large percentage answering it incorrectly). Other reasons for removing items at this stage included question content being redundant with more evenly distributed items in the same domain or the question not being relevant to all participants. Exclusions are documented in Tables S1 and S2.
Ten parents subsequently reviewed the retained knowledge and self-efficacy questions in one-on-one cognitive interviews. Participants were recruited from a large pediatric emergency department (ED). Individuals were eligible if they were the parent or guardian of a least one child or adolescent 5–17 years of age; child injury history was not an inclusion criteria. Participants completed the Brief Health Literacy Screen, a three-item validated measure of health literacy,23,24 and reported their health insurance status. Interviews were conducted in English, in person by a trained research assistant, and lasted between 10 and 20 min. Participants were asked to read through the survey questions and were instructed to mark any items that they found confusing or unusual and to circle specific words or phrases that were problematic. For each item marked, interviewers asked follow-up questions, such as “What about this item is confusing or difficult to understand?” or “Are there ways this question could be asked instead?” After the first five interviews, we reviewed parent responses to prioritize changes. Subsequently, we completed five additional interviews with minimal feedback on the updated items. Items were finalized after this round of cognitive interviews.
Item reduction and psychometric testing
At this stage, we sought to finalize parsimonious measures with evidence of reliability and validity. Similar to the process described in the pre-testing step, a link to a web-based survey was posted to MTurk for parent completion. A total of 463 consenting participants completed 19 knowledge items, 19 self-efficacy items, and a battery of additional scales, which were included for purposes of testing the validity of the newly developed measures. These included the Parenting Sense of Competency scale (PSOC),25 the Brief Health Literacy Screening Tool,23 demographic characteristics (age, gender, race, educational level, and number and ages of children) and child history of concussion. Socially desirable responding was assessed using the Social Desirability Scale (SDS), a short form of the Marlowe Crowne Social Desirability Scale.26
We used item-level and index-level statistics to identify items for possible removal. Items were eliminated based on statistical performance, balanced with the desire to retain two items per domain when possible. For knowledge items, we assessed item discrimination, or how well each knowledge item discriminated between high scorers and low scorers. This was done in two ways. First, item-total correlations <0.3 were flagged for potential exclusion.27,28 Second, we calculated the difference in the proportion of respondents answering correctly among high scorers as compared with low scorers, defined as the highest and lowest one quarter of respondents respectively, for the total score on all knowledge items combined.29 Finally, item difficulty, defined as the percent of respondents answering correctly, was used to identify items less effective at discriminating between respondents, for being either too easy (> 80% correct), or too difficult (< 20% correct).27,30 Together, these results were used to identify potentially problematic knowledge items, informing decisions around item reduction. Ten knowledge items were retained. Self-efficacy response distributions were similarly inspected for any potential floor or ceiling effects. We additionally considered item-to-total correlations as well as item-to-item correlations to look for poorly performing items or highly correlated pairs of items (> 0.8), which could suggest redundancy. Thirteen self-efficacy items were retained.
All retained knowledge questions were scored as true or false. The knowledge index score was calculated as the total number of correct responses, with a possible range of 0–10 where higher scores indicated greater knowledge. All self-efficacy questions were scored on a five-point unipolar Likert-style scale, with response options of not at all hard (1), slightly hard (2), moderately hard (3), very hard (4), and extremely hard (5). For the self-efficacy scale, the score was calculated as the item-level average, with a possible range of 1–5. Scores were set to missing for participants missing responses to more than half the items (n = 1 participant).
We calculated McDonald's Omega to assess the internal consistency of the knowledge index and self-efficacy scale. Convergent validity was assessed by examining the correlation among knowledge, self-efficacy, and other theoretically similar constructs. We expected positive correlations among knowledge, self-efficacy, and scores on the Brief health literacy scale.23 We also expected positive correlations between self-efficacy and scores on the Parenting Sense of Competency.25 Discriminant validity was assessed by the extent to which scores on the knowledge and self-efficacy measures differentiated between groups theoretically expected to be dissimilar. We expected that knowledge and self-efficacy scores would be highest among those with the highest educational attainment level, and among parents who had a child who had previously sustained a concussion. Correlation between self-efficacy and Social Desirability Scale score was also assessed.
Longitudinal prediction of parent behavior
Longitudinal data were collected in a separate sample of parents of youth patients with concussion (n = 51). Participants were parents in a large pediatric ED whose child (5–17 years of age) had sustained a closed head injury in the previous 3 days and was being discharged. Families were who were waiting for discharge paperwork were provided with information about the study and invited to complete a baseline survey. They were then e-mailed or texted the follow-up survey 2 weeks later. Baseline surveys included the final 10 knowledge items and 13 self-efficacy items. Follow-up surveys queried whether the parents had engaged in each of the 13 different parenting behaviors corresponding to the 13 self-efficacy items (see Tables S1 and S2) during the previous 2 weeks. For example, if the self-efficacy item “How hard do you think it would be to talk to you child about how they are feeling emotionally,” the corresponding behavior item was: “Thinking about today and yesterday, did you talk to your child about how they are feeling emotionally?” Response options for each behavior were yes (3), somewhat (2), or no (1). For behaviors that were expected to be performed once during recovery (i.e., follow-up appointment, temporarily stop playing sports, receive clearance from a healthcare provider to return to sports, and let schoolteacher know about the injury) the prompt was “Please indicate if you have done the following since your child's injury.” Responses were summed to create an index in which higher scores indicated that more behaviors were performed during that 2-week period. Correlation among the knowledge index, self-efficacy scale, and behavior index were assessed by calculating Pearson's r.
Results
Item development and refinement
Participants in the pre-testing phase (n = 240) were on average 39 years old and two thirds self-identified as white. Of the 10 participants in cognitive interviews, 50% reported using public insurance, and 50% had a health literacy score of 15 (the highest possible score, indicating high health literacy). Participant demographics are reported in Table 1.
Table 1.
Sample Demographic and Other Characteristics of Separate Samples of Parents Across Study Phases
| Pre-testing (n = 240) | Psychometric testing (n = 463) | Longitudinal prediction (n = 61) | |
|---|---|---|---|
| Age (years), mean (SD) | 39.2 (11.5) (n = 237) | 38.4 (11.5) (n = 459) | 40.6 (5.46) (n = 60) |
| Gender | |||
| Female | 117 (48.8%) | 245 (52.9%) | 43 (70.5%) |
| Male | 123 (51.2%) | 215 (46.4%) | 18 (29.5%) |
| Non-binary | – | 1 (0.2%) | – |
| (missing) | – | 2 (0.4%) | – |
| Race and ethnicity | |||
| Hispanic/Latino | 12 (5.0%) | 19 (4.1%) | |
| Middle Eastern | – | – | |
| American Indian/Alaskan Native | 4 (1.7%) | 5 (1.1%) | |
| Native Hawaiian/Pacific Islander | – | 1 (0.2%) | – |
| Asian | 40 (16.7%) | 145 (31.3%) | |
| African American/Black | 21 (8.8%) | 25 (5.4%) | |
| White | 160 (66.7%) | 262 (56.6%) | |
| Other | 3 (1.3%) | 5 (1.1%) | |
| More than one race (missing) |
1 (0.2%) | ||
| Parent education | – | – | |
| High school or less | 31 (6.7%) | ||
| Some college or associates degree | 113 (24.4%) | ||
| Bachelor's degree or higher | 313 (67.6%) | ||
| (missing) | 6 (1.3%) | ||
| How many children do you have? | – | ||
| 1 | 85 (35.4%) | 195 (42.1%) | |
| 2 | 94 (39.2%) | 178 (38.4%) | |
| 3 | 24 (10.0%) | 52 (11.2%) | |
| 4 | 5 (2.1%) | 24 (5.2%) | |
| 5 | 5 (2.1%) | 5 (1.1%) | |
| 6 or more | 5 (2.1%) | 2 (0.4%) | |
| (missing) | 22 (9.2%) | 7 (1.5%) | |
| Have you ever had a child diagnosed with a concussion by a medical professional? | – | ||
| No | 139 (57.9%) | 137 (29.6%) | |
| Yes | 101 (42.1%) | 325 (70.2%) | |
| (missing) | 1 (0.2%) | ||
| Have you ever been diagnosed with a concussion by a medical professional? | – | – | |
| Yes | 183 (39.5%) | ||
| No | 278 (60.0%) | ||
| (missing) | 2 (0.4%) | ||
| Brief Health Literacy Scale score, mean (SD) | 15.4 (4.1) |
16.0 (3.6) |
|
| Parenting Competency Scale score, mean (SD) | – |
67.41 (11.66) |
– |
| Age of child | – | – | |
| 5 years | 15 (24.6%) | ||
| 6 years | 10 (16.4%) | ||
| 7 years | 7 (11.5%) | ||
| 8 years | 9 (14.75%) | ||
| 9 years | 6 (9.84%) | ||
| 10 years | 2 (3.28%) | ||
| 11 years | 3 (4.92%) | ||
| 12 years | 3 (4.92%) | ||
| 13 years | 2 (3.28%) | ||
| 14 years | 3 (4.92%) | ||
| 15 years | – | ||
| 16 years | 1 (1.64%) | ||
| Number of children of specific ages. | – | ||
| <5 years | 108 (45.0%) | 186 (40.2%) | |
| 5 years | 22 (9.2%) | 59 (12.7%) | |
| 6 years | 20 (8.3%) | 38 (8.2%) | |
| 7 years | 20 (8.3%) | 40 (8.6%) | |
| 8 years | 17 (7.1%) | 44 (9.5%) | |
| 9 years | 14 (5.8%) | 35 (7.6%) | |
| 10 years | 17 (7.1%) | 27 (5.8%) | |
| 11 years | 13 (5.4%) | 31 (6.7%) | |
| 12 years | 20 (8.3%) | 32 (6.9%) | |
| 13 years | 6 (2.5%) | 27 (5.8%) | |
| 14 years | 18 (7.5%) | 18 (3.9%) | |
| 15 years | 11 (4.6%) | 23 (5.0%) | |
| 16 years | 14 (5.8%) | 26 (5.6%) | |
| 17 years | 12 (5.0%) | 22 (4.8%) | |
| 18+ years | 56 (23.3%) | 98 (21.2%) |
Twenty-two knowledge items were omitted and nine were modified during the item refinement phase (see Tables S1 and S2 for itemized modifications). Modifications were made to item wording to improve clarity based on feedback from cognitive interviews. These included adding more context (e.g., trouble concentrating on schoolwork), providing specific examples for phrases such as “how they are feeling emotionally,” and combining questions that seemed repetitive.
Item reduction and psychometric testing
Participants in this phase (n = 463) were on average 38 years old, and 57% self-identified as white. Additional demographic characteristics are presented in Table 1. Data missingness for both knowledge and self-efficacy items was low (< 2% for all items).
Knowledge item distributions ranged from 14% to 90% correct. Seven knowledge items were dropped for poor item difficulty and/or poor item discrimination, because of lower item-to-total correlations, or smaller differences in high as compared with low scorers. Possible floor/ceiling effects among self-efficacy items were considered when more respondents selected the top category than any other category; this ranged from 29% to 49% of respondents. Five items were dropped for potential ceiling effects and/or being less strongly correlated overall and within domain. One item was dropped because of conceptual overlap with other items.
Knowledge index score for the 10 retained true/false items ranged from 1 to 9 out of 10 total possible points, with a mean score of 4.8 points (standard deviation [SD] 1.6). Mean self-efficacy score was 3.6 (SD = 0.85, range 1–5). Item-level responses are summarized in Table 2 (knowledge) and Table 3 (self-efficacy).
Table 2.
Proportion of Parents in Psychometric Testing Sample With Correct Response to True-False Knowledge Questions (n = 463)
| Question | Proportion correct | Standard deviation |
|---|---|---|
| 1. Concussion symptoms usually last more than one month | 0.28 | 0.45 |
| 2. It is common for symptoms to get worse the first few days after concussion | 0.14 | 0.35 |
| 3. After having a concussion, a child should be completely symptom free before returning to school | 0.34 | 0.47 |
| 4. Trouble concentrating on schoolwork can be a symptom of concussion | 0.81 | 0.39 |
| 5. A child should not exercise at all during the first week after a concussion | 0.22 | 0.41 |
| 6. If a child has been recovering from a concussion and gets a headache when they try jogging, they should avoid jogging for two weeks | 0.23 | 0.42 |
| 7. Feeling depressed and anxious one month after concussion is a normal part of recovering and does not require treatment | 0.57 | 0.50 |
| 8. Anxiety can be a symptom of concussion | 0.75 | 0.43 |
| 9. Getting more sleep can help with concussion symptoms | 0.71 | 0.45 |
| 10. Trouble sleeping can be a symptom of concussion | 0.77 | 0.42 |
Table 3.
Mean Response to Self-Efficacy Items by Parents in Psychometric Testing Samplea (n = 463)
| Question | Mean | Standard deviation |
|---|---|---|
| Emotional support (“How hard would it be for you to talk with your child about…” | ||
| 1. How they are feeling emotionally (e.g., sadness, anxiety) | 3.57 | 1.10 |
| 2. How the time it takes to recover from a brain injury can be unpredictable | 3.47 | 1.22 |
| 3. Their concerns about their recovery from their brain injury | 3.50 | 1.21 |
| Monitoring (“How hard would it be for you to track…”) | ||
| 4. How your child's symptoms from their brain injury are or are not changing | 3.47 | 1.16 |
| 5. How your child's symptoms change when they start doing more physical activity | 3.45 | 1.15 |
| 6. How your child is feeling and doing at school | 3.38 | 1.17 |
| Rehabilitation support (“How hard would it be for you to…”) | ||
| 7. Help or encourage your child to do more physical activity as symptoms improve | 3.74 | 1.17 |
| 8. Allow child to return to normal electronic media use (e.g., cell phone, video games) as symptoms improve | 3.59 | 1.16 |
| 9. Help child adjust school work based on how they're feeling | 3.67 | 1.10 |
| 10. Help child adjust their bedtime or sleep environment to make sure they get enough sleep | 3.57 | 1.09 |
| External engagement (“How hard would it be for you to…”) | ||
| 11. Stop your child from playing sports until they receive clearance from a doctor | 3.32 | 1.14 |
| 12. Let their schoolteacher know about their brain injury | 3.98 | 1.20 |
| 13. Schedule a follow-up appointment for your child with their regular doctor or clinic | 4.02 | 1.17 |
All items scored on a five-point scale (range of 1–5) and reverse coded so that higher values reflect greater self- efficacy
McDonald's Omega for the knowledge index was 0.63, suggesting moderate internal consistency reliability. Self-efficacy was 0.91 for the full scale, 0.88 for self-efficacy related to the provision of emotional support, 0.87 for self-efficacy related to rehabilitation-related support, 0.79 for self-efficacy related to external engagement, and 0.83 for self-efficacy related to monitoring their child's recovery.
We observed a moderate positive correlation between knowledge index score and score on the Brief Health Literacy Scale (r = 0.515). A similar correlation was observed between self-efficacy and health literacy (r = 0.47). Self-efficacy scores were moderately correlated with scores on the Parenting Sense of Competency Scale (r = 0.51). Mean knowledge score among those with a self-reported history of having a concussion or having a child with a concussion was 5.0, compared with 4.6 among those with no type of concussion history (p = 0.01). Mean self-efficacy score was 3.5 for those with versus 3.7 for those without a history of concussion (either self or child; p = 0.03). Knowledge score was higher (5.2) for those without a bachelor's degree than for those with a bachelor's or graduate degree (4.6; p < 0.01). Self-efficacy score was also higher for those without a bachelor's degree than for those with a bachelor's or graduate degree (4.0 vs. 3.4; p < 0.01). There was a weak correlation between self-efficacy and scores on the Social Desirability Scale (r = 0.065).
Longitudinal prediction of parent behavior
Fifty-one parents of youth patients with concussion completed baseline and 2-week follow-up measures (Table 1). At follow-up, all parents engaged in at least one of the 13 measured behaviors; however, no parent engaged in all the behaviors (Table 4). The behaviors that parents were the most likely to engage in (responding “yes”) were: stopping child from temporarily playing sports (73%), letting their child's teacher know about their brain injury (71%), and allowing their child to return to normal electronic media use as symptoms improve (64%). The behaviors that parents were the least likely to engage in were talking to their child about how the time it takes to recover can be unpredictable (20%), talking to their child about their concerns about recovery (22%), and having a follow-up appointment (37%).
Table 4.
Proportion of Parents Engaging in Parenting Behaviors 2 Weeks After Discharge from the Pediatric Emergency Department Post-Concussion and Correlation Between Pre-Discharge Self-Efficacy and Parenting Behaviors (n = 51)
| |
Proportion |
|
|
|||
|---|---|---|---|---|---|---|
| Behavior | No | Somewhat | Yes | Standard deviation | Correlation with corresponding self-efficacy item | |
| Emotional support (“Did you talk to your child about…”) | ||||||
| 1. How they are feeling emotionally (e.g., sadness, anxiety) | 0.22 | 0.35 | 0.43 | 0.78 | 0.04 | |
| 2. How the time it takes to recover from a brain injury can be different from person to person | 0.53 | 0.27 | 0.20 | 0.79 | 0.16 | |
| 3. Their concerns about their recovery from their brain injury | 0.53 | 0.25 | 0.22 | 0.81 | 0.25 | |
| Monitoring (“Did you track…”) | ||||||
| 4. How your child's symptoms from their brain injury are or are not changing | 0.37 | 0.27 | 0.35 | 0.86 | -0.11 | |
| 5. How your child's symptoms change when they start doing more physical activity | 0.43 | 0.24 | 0.33 | 0.88 | 0.08 | |
| 6. How your child is feeling and doing at school | 0.35 | 0.18 | 0.47 | 0.91 | -0.10 | |
| Rehabilitation support (“Did you…”) | ||||||
| 7. Help or encourage your child to do more physical activity as symptoms improve | 0.18 | 0.38 | 0.44 | 0.75 | 0.06 | |
| 8. Allow child to return to normal electronic media use (e.g., cell phone, video games) as symptoms improve | 0.10 | 0.26 | 0.64 | 0.68 | 0.05 | |
| 9. Help child adjust schoolwork based on how they're feeling | 0.25 | 0.24 | 0.51 | 0.84 | 0.30 | |
| 10. Help child adjust their bedtime or sleep environment to make sure they get enough sleep | 0.12 | 0.25 | 0.63 | 0.70 | -0.11 | |
| External engagement (“Did you…”) | ||||||
| 11. Stop your child from temporarily playing sports | 0.27 | – | 0.73 | 0.90 | 0.01 | |
| 12. Let their schoolteacher know about their brain injury | 0.29 | – | 0.71 | 0.91 | 0.16 | |
| 13. Have a follow-up appointment for your child with their regular doctor or clinic | 0.63 | – | 0.37 | 0.98 | -0.08 | |
There was a positive correlation between the overall self-efficacy score at baseline and number of concussion support behaviors parents reported engaging in at 2-week follow-up (Pearson's r = 0.12). Self-efficacy items were mapped onto corresponding parenting behaviors, and post-hoc analyses assessed item-level correlations between self-efficacy responses and behaviors. Correlations were highest for self-efficacy questions relating to talking to their child about their concerns about the child's recovery from their brain injury (ρ = 0.25) and helping their child adjust schoolwork based on how they're feeling (ρ = 0.30).
Knowledge index scores ranged from 3 to 9 out of 10, with a mean score of 6.1 (SD = 1.4). Knowledge at baseline was not correlated with concussion management behaviors at 2- week follow-up (Pearson's r = 0.052).
Discussion
We developed measures of parent knowledge and self-efficacy related to the provision of guideline-consistent support during their child's concussion recovery using a systematic process that engaged parents and expert clinicians. These measures provide an important starting point for needs-assessment research. To that end, among parents of youth patients being discharged from the ED, we observed that those who had greater self-efficacy were more likely to engage in recommended emotional support, rehabilitation support, monitoring, and external engagement behaviors during the first 2 weeks of their child's recovery. Correlation between knowledge and subsequent parenting behaviors was not observed.
These results suggest that parent self-efficacy may be one appropriate target for interventions aiming to support parenting post-concussion. However, we caution that correlation between self-efficacy and parenting behaviors was relatively small. This is likely because the ability of parents to engage in each behavior is constrained by their context. For example, the ability of parents to schedule a follow-up appointment for their child may be influenced by their proximity to and the availability of a healthcare provider, whether their child has a medical home (e.g., a place where they can routinely access medical care), their ability to take time off work, the nature of their health insurance, and the burden of transportation. Consistent with Sørensen and colleagues' guidance on supporting individuals with low health literacy, both individual and systems-level interventions should be considered to increase parent support for their child post-concussion.8 At an individual level, behavioral interventions can be designed to help increase parenting capacity to engage in targeted behaviors by increasing their skills and confidence in overcoming barriers. There are a number of evidence-based strategies that can help modify these individual-level assets, such as guided practice, mastery experiences, goal setting, cue altering, and establishing contingent reinforcement and rewards.31 However, change is also needed at a systems level to reduce burdens on parents as they navigate health systems and support their child during injury recovery.8
The lack of prospective association between knowledge and parenting behaviors underscores the limitations of approaches to parent education focused solely on knowledge translation. Across theories of health behavior, knowledge is almost never the sole determinant of whether an individual engages in a recommended health behavior.32 An important first step is reviewing the theoretical frameworks underlying parent discharge education post-concussion: is it focused solely on ensuring that parents are informed about what they are supposed to do? Or is it also targeting parent self-efficacy or other potentially theoretically indicated barriers to action (e.g., motivation, perceived norms, contextual support, cues to action). Ensuring that parents can appropriately support their child as the child recovers from a concussion requires approaches to post-discharge support targeting behavioral determinants other than just knowledge.
Limitations
Measures were developed using an English-speaking and predominantly white population in the United States. Parents who experience structural and social determinants of health inequities, such as racism and poverty, and parents for whom English is not their first language, may face barriers to concussion management self-efficacy and behavior that were not adequately captured in the present study. Further research is needed to test the reliability and validity of these measures in more diverse populations. We note that the sample used to test predictive validity was small, and may have been underpowered to detect associations between parent cognitions at discharge and subsequent parenting behaviors. Future work in larger prospective samples should re-test these associations.
Conclusion
Parents have the potential to play an important role in concussion management. The measures of knowledge and self-efficacy developed in this study can help identify parent needs and evaluate interventions aiming to support parenting post-concussion.
Supplementary Material
Transparency, Rigor, and Reproducibility Summary
The analysis plan was not formally pre-registered. Data from this study will be made available (as allowable according to institutional regulatory standards) by e-mailing the corresponding author. Analytic code used to conduct the analyses may be available by e-mailing the corresponding author. The full content of the manuscript is available on request by contacting the corresponding author.
Authors' Contributions
Emily Kroshus was responsible for conceptualization, methodology, investigation, writing and editing–original draft, and funding acquisition. Mary Kathleen Steiner was responsible for conceptualization, resources, data curation, writing–review and editing, visualization, and project administration. Sarah Lowry was responsible for conceptualization, methodology, validation, formal analysis, writing–original draft, and visualization. Casey Lion was responsible for conceptualization, and writing–review and editing. Eileen Klein was responsible for conceptualization, writing- review and editing, supervision. Bonnie Strelitz was responsible for investigation, resources, and writing–review and editing. Sara Chrisman was responsible for conceptualization, and writing–review and editing. Frederick Rivara was responsible for conceptualization, writing–review and editing, and supervision.
Funding Information
This study received funding from grant R21NS111065 National Institutes of Health/National Institute of Neurologic Disorders and Stroke.
Author Disclosure Statement
No competing financial interests exist.
Supplementary Material
References
- 1. Bryan MA, Rowhani-Rahbar A, Comstock RD, et al. Sports- and recreation-related concussions in US youth. Pediatrics 2016;138(1):e20154635; doi: 10.1542/peds.2015-4635 [DOI] [PubMed] [Google Scholar]
- 2. Polster DR, Giza CC, Babikian T. Health-related quality of life after concussion: how can we improve management of care? JAMA Pediatr 2016;170(12):e162985–e162985 [DOI] [PubMed] [Google Scholar]
- 3. Schneider KJ, Leddy JJ, Guskiewicz KM, et al. Rest and treatment/rehabilitation following sport-related concussion: a systematic review. Br J Sports Med 2017;51(12):930–934; doi: 10.1136/bjsports-2016-097475 [DOI] [PubMed] [Google Scholar]
- 4. Gornall A, Takagi M, Morawakage T, et al. Mental health after paediatric concussion: a systematic review and meta-analysis. Br J Sports Med 2021;55(18):1048–1058; doi: 10.1136/bjsports-2020-103548 [DOI] [PubMed] [Google Scholar]
- 5. Lumba-Brown A, Yeates KO, Sarmiento K, et al. Centers for Disease Control and Prevention Guideline on the Diagnosis and Management of Mild Traumatic Brain Injury Among Children. JAMA Pediatr 2018;172(11):e182853–e182853; doi: 10.1001/jamapediatrics.2018.2853 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kroshus E, Chrisman SPD, Harmon KG, et al. What do parents need to know about concussion? Developing consensus using the Delphi Method. Clin J Sport Med 2018; 31(2):139-144; doi: 10.1097/JSM.0000000000000702 [DOI] [PubMed] [Google Scholar]
- 7. Lyons VH, Moore M, Guiney R, et al. Strategies to address unmet needs and facilitate Return to Learn guideline adoption following concussion. J Sch Health 2017;87(6):416–426; doi: 10.1111/josh.12510 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Sørensen K, Van den Broucke S, Fullam J, et al. Health literacy and public health: a systematic review and integration of definitions and models. BMC Public Health 2012;12(1):80; doi: 10.1186/1471-2458-12-80 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Berkman ND, Sheridan SL, Donahue KE, et al. Low health literacy and health outcomes: an updated systematic review. Ann Intern Med 2011;155(2):97–107; doi: 10.7326/0003-4819-155-2-201107190-00005 [DOI] [PubMed] [Google Scholar]
- 10. Weerdenburg K, Schneeweiss S, Koo E, et al. Concussion and its management: what do parents know? Paediatr Child Health 2016;21(3):e22–e26; doi: 10.1093/pch/21.3.e22 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Bloodgood B, Inokuchi D, Shawver W, et al. Exploration of awareness, knowledge, and perceptions of traumatic brain injury among American youth athletes and their parents. J Adolesc Health 2013;53(1):34–39; doi: 10.1016/j.jadohealth.2013.01.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Cusimano MD, Zhang S, Topolovec-Vranic J, et al. Factors affecting the concussion knowledge of athletes, parents, coaches, and medical professionals. SAGE Open Med 2017;5:2050312117694794; doi: 10.1177/2050312117694794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Maibach E, Murphy DA. Self-efficacy in health promotion research and practice: conceptualization and measurement. Health Educ Res 1995;10(1):37–50; doi: 10.1093/her/10.1.37 [DOI] [Google Scholar]
- 14. Bandura A. Human agency in social cognitive theory. Am Psychol 1989;44(9):1175–1184; doi: 10.1037/0003-066X.44.9.1175 [DOI] [PubMed] [Google Scholar]
- 15. Thomas DG, Bradley L, Servi A, et al. Parental knowledge and recall of concussion discharge instructions. J Emerg Nurs 2018;44(1):52–56; doi: 10.1016/j.jen.2017.04.012 [DOI] [PubMed] [Google Scholar]
- 16. Rieger B, Lewandowski L, Potts H, et al. Parent knowledge and perceptions of concussion related to youth football. Cureus 2018;10(3):e2268; doi: 10.7759/cureus.2268 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. King C, Coughlan E. Blowing the whistle on concussion knowledge and education in youth sport referees. Open Access J Sports Med 2021;12:109–117; doi: 10.2147/OAJSM.S324191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Merz ZC, Van Patten R, Lace J. Current public knowledge pertaining to traumatic brain injury: influence of demographic factors, social trends, and sport concussion experience on the understanding of traumatic brain injury sequelae. Arch Clin Neuropsychol 2017;32(2):155–167; doi: 10.1093/arclin/acw092 [DOI] [PubMed] [Google Scholar]
- 19. Rosenbaum AM, Arnett PA. The development of a survey to examine knowledge about and attitudes toward concussion in high-school students. J Clin Exp Neuropsychol 2010;32(1):44–55; doi: 10.1080/13803390902806535 [DOI] [PubMed] [Google Scholar]
- 20. Owens JA, Spirito A, McGuinn M. The Children's Sleep Habits Questionnaire (CSHQ): psychometric properties of a survey instrument for school-aged children. Sleep 2000;23(8):1–9; doi: 10.1093/sleep/23.8.1d [DOI] [PubMed] [Google Scholar]
- 21. Walters K, Christakis DA, Wright DR. Are Mechanical Turk worker samples representative of health status and health behaviors in the U.S.? PLoS One 2018;13(6); doi: 10.1371/journal.pone.0198835 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Chandler J, Rosenzweig C, Moss AJ, et al. Online panels in social science research: expanding sampling methods beyond Mechanical Turk. Behav Res Methods 2019;51(5):2022–2038; doi: 10.3758/s13428-019-01273-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Haun J, Luther S, Dodd V, et al. Measurement variation across health literacy assessments: implications for assessment selection in research and practice. J Health Commun 2012;17(Supp. 3):141–159; doi: 10.1080/10810730.2012.712615 [DOI] [PubMed] [Google Scholar]
- 24. Wallston KA, Cawthon C, McNaughton CD, et al. Psychometric properties of the brief health literacy screen in clinical practice. J Gen Intern Med 2014;29(1):119–126; doi: 10.1007/s11606-013-2568-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Gibaud-Wallston, Wandersman LP. Parenting Sense of Competence Scale. Can J Behav Sci 1978. Available from: https://psycnet.apa.org/doiLanding?doi=10.1037%2Ft01311-000
- 26. Reynolds WM. Development of reliable and valid short forms of the marlowe-crowne social desirability scale. J Clin Psychol 1982;38(1):119–125; doi: [DOI] [Google Scholar]
- 27. Parmenter K, Wardle J. Evaluation and design of nutrition knowledge measures. J Nutr Educ 2000;32(5):269–277; doi: 10.1016/S0022-3182(00)70575-9 [DOI] [Google Scholar]
- 28. Fishbein M, Ajzen I. Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. 1975. Available from: https://trid.trb.org/view/1150648
- 29. Cappelleri JC, Jason Lundy J, Hays RD. Overview of classical test theory and item response theory for the quantitative assessment of items in developing patient-reported outcomes measures. Clin Ther 2014;36(5):648–662; doi: 10.1016/j.clinthera.2014.04.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Streiner DL, Norman GR, Cairney J.. Health Measurement Scales: A Practical Guide to Their Development and Use. Oxford University Press: Oxford, United Kingdom; 2015. [Google Scholar]
- 31. Kok G, Gottlieb NH, Peters G-JY, et al. A taxonomy of behaviour change methods: an Intervention Mapping approach. Health Psychol Rev 2016;10(3):297–312; doi: 10.1080/17437199.2015.1077155 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Glanz K, Rimer BK, Viswanath K.. Health Behavior: Theory, Research, and Practice. John Wiley & Sons: San Francisco, CA; 2015. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.