- Email: [email protected]
Training Pediatric Residents to Provide Parent Education: A Randomized Controlled Trial
Training Pediatric Residents to Provide Parent Education: A Randomized Controlled Trial Erin McCormick, MSW, MPH; Suzanne E. U. Kerns, PhD; Heather McPhillips, MD, MPH; Jeffrey Wright, MD; Dimitri A. Christakis, MD, MPH; Frederick P. Rivara, MD, MPH From the Seattle Children’s Research Institute (Dr McCormick); University of Washington (Dr Kerns and Rivara); Seattle Children’s Hospital and University of Washington (Dr McPhillips); Seattle Children’s Research Institute (Dr Wright); and University of Washington (Dr Christakis) The authors declare that they have no conflict of interest. Address correspondence to Erin McCormick, MSW, MPH, Center for Child Health, Behavior and Development, Seattle Children’s Research Institute, 2001 Eighth Ave, Suite 400, Seattle, WA 98121 (e-mail: [email protected]). Received for publication November 7, 2013; accepted March 17, 2014.
ABSTRACT OBJECTIVE: We evaluated the effect of Primary Care Positive
RESULTS: Primary Care Triple P training had a positive, significant, and persistent impact on residents’ parenting consultation skills (mean increase on Parent Consultation Skills Checklist 48.11, 95% confidence interval [CI] 40.07, 57.36). Parents visiting intervention-trained residents demonstrated improved disciplinary practices compared to parents visiting control residents (mean change in Child Discipline Survey 0.322, 95% CI 0.02, 0.71), with stronger differential effects for parents with lower baseline skills (mean Child Discipline Survey change 0.822, 95% CI 0.48, 1.83). No differences were found for child behavior or parenting sense of confidence. CONCLUSIONS: Training residents in Primary Care Triple P can have a positive impact on consultation skills and parent disciplinary practices. This finding adds strength to the call for increased residency training in behavioral pediatrics. KEYWORDS: child; graduate medical education; parenting; pediatric resident; primary care; Triple P
Parenting Program (Triple P) training on pediatric residents and the families they serve to test 2 hypotheses: first, training would significantly improve resident skill in identifying and addressing discrete parenting and child behavior problems; and second, parents would report an improvement in their sense of selfefficacy, use of positive discipline strategies, and their child’s behavior. METHODS: Study participants included pediatric residents from 3 community clinics of a pediatric residency program, as well as English-speaking parents of children aged 18 months to 12 years without a diagnosed behavior disorder cared for by study residents. Residents were randomized to receive Primary Care Triple P training either at the beginning or end of the study period. The measured resident outcomes were selfassessed confidence and skills in giving parenting advice. The measured family outcomes were parent sense of self-efficacy, child externalizing behavior, and discipline strategies.
ACADEMIC PEDIATRICS 2014;14:353–360
WHAT’S NEW
problems, such evidence-based programs currently reach a limited proportion of parents.13–15 Public health approaches to preventing physical and mental health problems in children increasingly recognize the medical home as a crucial site of child abuse and neglect identification and early detection and treatment of mental, emotional, and behavioral disorders. Pediatric primary care providers are often the first contact point for children and families experiencing behavioral and parenting difficulties, and research consistently demonstrates that parents view their primary care providers as credible sources of parenting information.16–19 However, primary care providers vary in their degree of confidence and skill in managing mental, emotional, and behavioral concerns. In part, these variations may be due to a lack of specific instruction on brief, evidence-based behavioral interventions during residency.20,21 Pediatric residency offers an ideal time to teach and practice evidence-based behavioral and parenting interventions to ensure that these professional competencies develop.22
Training pediatric residents in the Positive Parenting Program (Triple P) improved provider skill and confidence in conducting parent consultations about child behavior concerns. Study finings support prior research on Triple P’s training effect and demonstrate promising family-level outcomes to explore in future research.
EFFECTIVE PARENTING IS a key determinant of healthy child development and strongly affects a child’s risk of developing emotional and behavioral disorders.1–3 Early emotional and behavioral symptoms as well as coercive and negative parenting can place children at increased risk for later behavior problems, child maltreatment, and other significant physical and emotional sequelae.4–8 Interruption and cessation of negative parenting styles can attenuate and even prevent progression to abuse and/ or serious mental health problems.9–12 Although there are evidence-based parenting programs with demonstrated success in improving parenting skills and child behavioral ACADEMIC PEDIATRICS Copyright ª 2014 by Academic Pediatric Association
353
Volume 14, Number 4 July–August 2014
354
MCCORMICK ET AL
The Positive Parenting Program (Triple P) is an evidence-based parenting program aimed at promoting positive parenting and preventing child abuse and neglect. Although Triple P has proven effective at reducing coercive parenting practices and disruptive behavior when delivered in community clinics by nurses and mental health professionals, there is limited evaluation on whether pediatricians can effectively deliver this model.23–26 This study has 2 objectives: to evaluate the impact of Triple P training on pediatric residents’ skill and confidence in delivering parenting information; and to understand the potential impact of residents’ Triple P training status on parents’ sense of self efficacy in parenting and discipline strategies as well as child behavior.
ACADEMIC PEDIATRICS
The sample of parents was selected on the basis of their affiliation with one of the residents in the clinical trial, and provision of Triple P services to an individual patient was not required for parent enrollment into the study. Therefore, this study examined a general impact of resident training on outcomes for families. It was not a direct test of the Triple P intervention effectiveness on families. Participating families may or may not have received Triple P during the course of this study, as parents were not asked directly about Triple P. The Seattle Children’s Research Institute institutional review board approved all study procedures, and the trial was registered at ClinicalTrials.gov (NCT01946958). PARTICIPANTS
METHODS STUDY SETTINGS The study was conducted in 3 distinct community clinics operated by University of Washington faculty, which serve as resident continuity clinic teaching sites. One was an inner-city pediatric clinic, another was a health department clinic, and the third was a clinic in a large multispecialty primary care center. DESIGN The randomized controlled trial included 2 study populations: pediatric residents and parents of children served by these residents. Consented residents were randomized using a computer randomization sequence. Eligible residents who consented to participation were randomized into one of two conditions: 1) intervention, which involved immediate training in Primary Care Triple P (PC Triple P), or 2) wait list control, which involved the usual provision of well-child services until subsequent training in PC Triple P. Intervention residents received PC Triple P training at the beginning of the study, and control residents received training 9 months later. Because residents were the unit of randomization, participating clinics contained both intervention and control group residents.
Figure 1. Flow chart of resident participation.
PEDIATRIC RESIDENTS Eligible residents from all years of residency (PL1, 2, and 3) attended continuity clinics at participating clinics. The residency program director introduced the project to eligible pediatric residents via e-mail. After the e-mail introduction, the research coordinator approached eligible residents to discuss participation. Figure 1 depicts the flow of residents through the trial as well as retention and attrition numbers at key study intervals. PARENTS AND CAREGIVERS After training the intervention group residents, parent recruitment began across both condition and in all participating clinics. Eligible parents had a well-child appointment the intervention group received PC Triple P training with a participating resident, spoke English, and had a child between the ages of 18 months and 12 years. Children who had received current DSM-IV diagnoses or who were receiving treatment for a behavioral problem were excluded from the trial because the Triple P training administered to residents was not intended for this subpopulation. Parents were approached and consented in person at their primary care clinic after their child’s well-child appointment. Parents were blinded to resident randomization status and the Triple P intervention. Study staff presented the
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
355
Figure 2. Flow chart of parent participation.
project as an exploration of how pediatricians address and manage parents’ concerns about child behavior and parenting strategies. Figure 2 depicts the flow of parent participants through the trial as well as retention and attrition at key study intervals. INTERVENTION: PRIMARY CARE TRIPLE P TRAINING AND ACCREDITATION Residents randomized to each group received standardized PC Triple P training (12 direct training hours over 1.5 days) followed by a half-day accreditation session 6 to 8 weeks after training. This training course prepares practitioners to deliver a brief parent education intervention to address commonly occurring developmental and behavioral difficulties in children aged 0 to 12 years. This level of Triple P can be delivered as part of a universal parenting support strategy or as an early intervention strategy for parents with children at risk of developing oppositional, disruptive, and/or aggressive behaviors. The PC Triple P training course is appropriate for practitioners who routinely offer advice and support to parents during time-limited appointments in the primary care setting. The course provides a comprehensive overview of common developmental and behavioral problems in children as well as a range of parenting strategies to promote child development. The course includes practical skills training in a range of parent interview skills necessary in the delivery of a brief parenting intervention. More specifically, the training course enhances provider skills and knowledge in the following areas: Early detection and identification of family risk and protective factors. Causes of child behavior problems. Core principles of positive parenting and effective behavior change. Specific positive parenting strategies for promoting children’s development, social competence, and selfcontrol. Effective parent consultation and skills training procedures. Identification of indicators suggesting more intensive intervention. Appropriate referral procedures.
At completion of the training course, practitioners had the ability to deliver counseling via 1 to 4 sessions to families they serve.27 After the initial 1.5-day training, practitioners were encouraged to practice their new skills until the accreditation session (6 to 8 weeks after the initial training). Accreditation is a necessary component of the competency-based training course and is required for official recognition of program proficiency. At accreditation, practitioners have an opportunity to demonstrate, via roleplay, their proficiency in the two competency areas targeted for PC Triple P accreditation: 1) discussion of a parenting plan, and 2) discussion and modification of a parenting plan that has been implemented. During accreditation, participants receive coaching and feedback on their performance of the targeted competences. At this session, practitioners submit an accreditation quiz, testing comprehension of the Triple P model and intervention components. Practitioners must pass the accreditation quiz and skill demonstrations to be a Triple P–accredited practitioner. Study staff collaborated with Triple P America to hold trainings, and Triple P America provided doctoral-level Triple P trainers for the trainings and accreditation sessions. RESIDENT-REPORT MEASURES PRACTITIONER SKILL IN ADDRESSING BEHAVIORAL CONCERNS WITH PARENTS The Parent Consultation Skills Checklist (PCSC) is a standardized instrument that assesses a practitioner’s proficiency in 4 core domains: assessment, active skills training, dealing with process issues, and clinical application of positive parenting strategies.28 The tool also assesses confidence in important parent consultation skills. The measure consists of 18 items, which are rated on a 7-point scale ranging from 1 (not at all confident) to 7 (very confident). Factor analysis has indicated that the checklist measures a single construct (a ¼ 0.97) with all items loading significantly on the scale.15 The checklist has been shown to have good internal consistency (a ¼ .96–.97).29 Maximum score for the
356
MCCORMICK ET AL
PCSC is 154, with higher scores indicating greater competency. PARENT-REPORT MEASURES CHILD BEHAVIOR We measured child behavior with the externalizing subscale of the Child Behavior Checklist (CBCL).30–32 The CBCL/1.5–5 is appropriate for children between 1.5 and 5 years of age, and the CBCL/6–18 is appropriate for children from 6 to 18 years of age. The externalizing scales measure problems involving conflicts with other people and parent expectations for their child. The CBCL/ 1.5–5 externalizing scale consists of 24 items, while the CBCL/6–18 scale contains 35 items. Discriminant analyses of the CBCL/6–18 indicate that 85% of children are correctly classified as referred or nonreferred. To standardize responses across age ranges, we calculated T scores for all participants. PARENTAL SENSE OF SELF-EFFICACY We measured parental self-efficacy with the Parenting Sense of Competence Scale (PSOC).33 The PSOC is a 16-item tool assessing parents’ views of their competence on two dimensions: satisfaction with their parenting role and feeling of efficacy as a parent. The total score, satisfaction score, and efficacy score show a satisfactory level of internal consistency (0.79, 0.75, and 0.76, respectively). The maximum score for this measure is 96, with higher scores indicating greater confidence a parent feels in his or her parenting abilities. USE OF POSITIVE DISCIPLINE STRATEGIES We selected 9 items from the International Society for the Prevention of Child Abuse and Neglect’s Child Abuse Screening Tool–Parent Version (ICAST-P) to identify parental discipline strategies and beliefs.34 It has face validity and reasonable internal consistency, and it has been demonstrated to be acceptable to parents in the countries where it was piloted. Higher scores indicate that more positive discipline techniques were used. ASSESSMENT TIMING The PCSC served as a self-report of resident selfefficacy in parent consultation. Because participating residents received the PC Triple P training at specific (and different) times depending on their study status (intervention or control), we examined skill differences at key time points in the trial. At time T1, both the intervention and control group residents completed their baseline measure of skills using the PCSC. The intervention group repeated the PCSC immediately after training (T2). Approximately 9 months after the intervention group training (T3), the intervention group completed the PCSC again to measure the long term effect of the training. When intervention residents completed the follow-up PCSC, control residents received training in PC Triple P. Control residents completed the PCSC at the beginning
ACADEMIC PEDIATRICS
of the training (also T3). The control group repeated the PCSC at the end of the training (T4). Parents had the option to complete study assessments (CBCL, PSOC, and Child Discipline Survey) at the time of recruitment in the clinic or at home, returning the measures via mail. Three months after their consent appointment, study staff contacted parents by telephone. At this time, parents completed the CBCL, PSOC, and ICAST-P by telephone with study staff. DATA ANALYSIS Data were key entered and analyzed by SPSS (IBM, Armonk, NY). We examined mean differences in scores from one time to another, adjusting for baseline differences. We also conducted a subanalysis for each of the parent measures on the group of parents whose scores in that measure were in the bottom half of the distribution.
RESULTS DEMOGRAPHICS Fifty-three pediatric residents participated in the randomized control trial, with 25 assigned to the intervention group and 28 to the control group (Table 1). The majority of residents were female (79%), and the vast majority (96%) were not parents. Residents in the two groups were distributed across the three years of training. Sixty percent of residents were assigned to one continuity clinic, the largest clinic in our training system. A total of 101 families participated in the trial. Parents were evenly distributed between intervention and control residents (Table 2). The mean age for participating children was 4.8 years (standard deviation 3.25), and male and female children were represented equally. Parent participants were predominantly mothers. The majority of families had less than a 1-year relationship with the resident. Of the families that reported their household income, 51% reported annual household incomes lower than $30,000. Over half of families identified as an ethnic minority or other nonwhite. Age and language eligibility requirements limited family participation at each clinic. The result of these limitations was a high proportion of recruited families from a single clinic. Table 1. Characteristics of Pediatric Resident Study Participants* Characteristic Female sex Guardian status ¼ parent Continuity clinic assignment Clinic 1 Clinic 2 Clinic 3 Residency year Intern Second year Third year
Intervention, n (%) (n ¼ 25)
Control, n (%) (n ¼ 28)
19 (76) 1 (4)
23 (82) 1 (4)
4 (16) 17 (68) 4 (16)
6 (21) 15 (54) 7 (25)
10 (40) 8 (32) 7 (28)
11 (39) 11 (39) 6 (21)
*There were no significant differences between groups.
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
Table 2. Characteristics of Family Study Participants* Characteristic Pediatric clinic† Clinic 1 Clinic 2 Clinic 3 How long family knew provider <1 y 1–2 y $3 y Child’s age, y Child’s sex Female Relationship to child Mother Parent’s race White African American/black Other Highest level of education High school or below Some college College degree or higher Household income Less than $29,999 $30,000–$49,999 $50,000 and above
Intervention (n ¼ 51)
Control (n ¼ 50)
12 (24) 37 (73) 2 (4)
17 (34) 21 (42) 12 (24)
39 (77) 10 (20) 2 (4) 4.7 (3.1)
27 (59) 15 (33) 4 (9) 4.9 (3.4)
24 (47)
24 (48)
41 (80)
40 (80)
23 (45) 18 (35) 20 (20)
15 (30) 12 (24) 23 (46)
8 (17) 21 (44) 19 (40)
18 (36) 16 (32) 16 (32)
19 (46) 13 (32) 9 (22)
25 (55) 10 (22) 10 (22)
*Some respondents did not answer every self-report question. Data are presented as n (%) or mean standard deviation. †A significant difference was found in recruitment clinic site.
RESIDENT OUTCOMES Analyses across time demonstrate that Triple P training had a positive, significant, and persistent impact on the skills, competence, and confidence of pediatric residents, as measured by the PCSC (Fig. 3). At baseline (T1), no significant differences existed between the control and intervention residents’ scores on the baseline measure (mean difference 4.90, 95% confidence interval [CI] 5.52, 15.34). At the intervention group posttraining (T2), the intervention group reported significantly higher levels of confidence and skill in conducting parent consultations (mean difference 48.11, 95% CI 40.07, 57.36) than their control group counterparts. Fifty-three residents were included in baseline and posttraining analyses. At ninemonth follow-up (T3), the intervention group maintained a significantly higher mean score than at baseline, with only a slight drop from the postintervention training. At follow-up, the difference between the intervention and control group remained significant (mean difference 39.03, 95% CI 28.75, 50.91). At this time point, the control
Figure 3. Resident PCSC mean scores over time in trial.
357
group received training. At the postintervention training (T4), the difference in self-reported skill and confidence between intervention and control group was no longer statistically significant (mean difference 0.07, 95% CI 8.11, 9.49). Thirty-nine residents (74%) were included in follow-up and control group training analyses. Over the course of the study, we lost residents to graduation, out-of-state rotations, and time conflicts. PARENT AND CHILD OUTCOMES We analyzed the change in parents’ sense of selfefficacy, discipline strategies, and child behavior over time using the parent-completed assessments from baseline and follow-up. Our analysis of the parent and child data showed mixed results. We attempted follow-up with all parent participants; however, only 56% of parents completed follow-up at parent T2. At baseline, there were no significant differences in any outcome measure between groups (Table 3). At follow-up, there were no significant differences between groups or within groups. However, in the adjusted analyses, parents who had received care from an intervention-trained resident showed greater improvements on the Child Discipline Survey at follow-up, indicating improved disciplinary practices in the parents exposed to Triple P–trained residents (Table 4). The difference (on the ICAST-P measure, the discipline proxy) between groups was revealed in the regression models, with the greatest effect found in families that scored at or below the 50th percentile on the ICAST-P scale at baseline (b ¼ 0.816, 95% CI 0.48, 1.83) (Table 4). The effect was also stronger in families that had a longer relationship with their pediatrician (b ¼ 0.693, 95% CI 0.10, 1.70) (Table 5). No intergroup differences were found for child behavior or parenting sense of confidence.
DISCUSSION Training pediatric residents to deliver parenting education using Triple P improved provider self-assessed skill and confidence; it also positively affected parental disciplinary behaviors in families that received care from intervention-trained residents. The impact of interventiontrained residents on parent’s sense of self-efficacy, discipline strategies, and child behavior was mixed, with the greatest group effect occurring on parent-reported discipline strategies, especially among parents who reported the most problems at baseline. Parents exposed to Triple P–trained residents demonstrated a higher rate of positive discipline strategies at follow-up. However, there appeared to be little impact of the intervention on the other parent measures. One possible explanation for the reported improvement in discipline strategies is that discipline is a frequently discussed psychosocial concern for parents. Burklow et al found that misbehavior was the most common psychosocial concerns parents presented with to their pediatrician, and discipline was the more frequently discussed strategy to address the behavioral concern.17 Managing child
358
MCCORMICK ET AL
ACADEMIC PEDIATRICS
Table 3. Parent Report Outcomes Over Time* Outcome Measure Parent’s sense of confidence Intervention Control Use of positive discipline strategies Intervention Control Child behavior problems Intervention Control
Baseline (T1), Mean SD
Follow-up (T2), Mean SD
72.94 10.42 75.44 10.46
70.89 11.18 76.10 14.72
7.20 .92 7.22 .84
7.81 .40 7.65 .67
48.64 10.35 47.20 8.75
49.45 8.93 52.85 7.82
95% CI at T1†
95% CI at T2†
1.62, 6.62
1.76, 12.17
0.32, 0.37
0.44, 0.13
5.35, 2.46
1.51, 8.31
SD ¼ standard deviation; CI ¼ confidence interval. *Intervention group experienced no significant differences between baseline and follow-up in all outcomes. †No significant differences exist between groups at either baseline or follow-up.
misbehavior might be the entry point to a longer-term discussion about promoting positive interaction between parent and child. It should also be noted that behavior problems worsened in the control group over time, a change not seen in the intervention group, suggesting an effect on behavior as well. The study had a number of limitations, including a small sample size as well as recruitment and implementation level barriers. The cost of Triple P training, coupled with the available funding, permitted only a limited number of training sessions. The Triple P model is not a train-thetrainer model. Rather, it requires a certified trainer contracted by Triple P America and mandates an upper limit of 20 people per training. Residents’ complicated work schedules, made even more complicated by work-hour restrictions, makes patient continuity difficult. Furthermore, residents have lower patient volumes than practicing pediatricians. Additionally, clinic and study staff encountered a moderate rate of patients who failed to keep appointments. We lost parents during follow-up as a result of an inability to contact families, despite multiple attempts (disconnected numbers, unanswered or unreturned calls, etc). We had a high retention rate at follow-up for those we were able to reach by phone. Because the family sample size was small, and because families were enrolled into the study regardless of receipt of Triple P, the possible direct impact of the Triple P intervention may be diluted in this sample. However, the impact of resident training on parenting outcomes in general may point to a generalization effect of the training that is
robust—that is, knowledge gained through the training provided a framework for parenting support that extended beyond the use of a specific model. We also faced challenges in fully supporting trained residents without compromising the control group. Because residents were the unit of randomization instead of the clinic, clinics were staffed by both control and intervention residents. Triple P materials were not displayed in participating clinics, and program support was not integrated into existing group supervision structures at clinics. In addition, residents did not always have access to Triple P– trained preceptors. Although several clinic preceptors attended the Triple P training, clinics had gaps in Triple P staff knowledge and skills. Finally, not all residents completed the entire training. Although all interventiontrained residents attended the initial training days, less than half (44%) attended accreditation sessions. All of these limitations may have constricted use of Triple P with families and attenuated the level of support provided to residents for enhancing their use of the model. Further research is needed to understand the potential of this early intervention strategy on families and should focus on creating environments in which implementation of the Triple P program is encouraged and comprehensively reinforced. Our findings are consistent with a prior study evaluating the effect of training general practitioners in PC Triple P.24 Results of this study show promising outcomes of parent consultations and greater use of parent consultation skills. These training effects are directly related to competencies
Table 4. Multivariate Analysis of the Relationship of Parent-Report Outcomes to Resident and Family Characteristics* b (95% CI) for: Variable
PSOC
CDS
CBCL
Intervention Clinic Length of relationship with provider Child’s sex Child’s age Respondent’s relationship to child Race Education Income
0.007 (5.52, 5.91) 0.012 (5.06, 5.62) 0.107 (4.03, 9.24) 0.394 (15.75, 4.42) 0.141 (0.44, 1.63) 0.027 (1.97, 1.57) 0.140 (1.25, 4.34) 0.164 (1.12, 4.63) 0.147 (2.94, 0.969)
0.322 (0.02, 0.71) 0.012 (0.38, 0.31) 0.235 (0.16, 0.64) 0.289 (0.65, 0.03) 0.062 (0.05, 0.07) 0.232 (0.04, 0.18) 0.152 (0.10, 0.24) 0.186 (0.09, 0.26) 0.036 (0.12, 0.10)
2.67 (8.19, 2.84) 0.396 (5.14, 5.94) 0.137 (6.50, 6.78) 2.57 (3.29, 8.43) 0.371 (1.41, 0.67) 0.259 (1.99, 1.47) 0.434 (3.23, 2.37) 1.17 (1.65, 3.99) 0.754 (2.67, 1.16)
CI ¼ confidence interval; PSOC ¼ Parent Sense of Competence Scale; CDS ¼ Child Discipline Survey; CBCL ¼ Child Behavior Checklist. *Each outcome was adjusted for baseline score.
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
359
Table 5. Multivariate Analysis of Relationship of Families Scoring Below 50th Percentile on Parent Report Outcome Measures to Resident and Family Characteristics* b (95% CI) for: Variable
PSOC
CDS
CBCL
Intervention Clinic Length of relationship with provider Child’s sex Child’s age Respondent’s relationship to child Race Education Income
0.268 (16.08, 4.10) 0.438 (7.91, 23.17) 0.073 (26.63, 30.05) 0.890 (36.57, 3.24) 0.262 (3.87, 1.18) 0.372 (2.98, 6.38) 0.345 (7.86, 14.91) 0.766 (0.374, 15.28) 0.379 (6.97, 2.24)
0.816 (0.48, 1.83) 0.189 (0.10, 0.48) 0.693 (0.10, 1.70) 0.833 (1.79, 0.39) 0.200 (0.19, 0.11) 0.262 (0.30, 0.85) 0.412 (0.08, 0.55) 0.465 (0.76, 0.26) 0.128 (0.22, 0.32)
4.62 (11.16, 1.93) 3.65 (10.14, 2.84) 0.586 (9.34, 8.17) 1.83 (4.44, 8.09) .561 (1.18, 2.31) 1.153 (1.04, 3.34) 0.384 (3.51, 2.75) 1.63 (4.01, 0.76) 0.243 (2.07, 1.59)
CI ¼ confidence interval; PSOC ¼ Parent Sense of Competence Scale; CDS ¼ Child Discipline Survey; CBCL ¼ Child Behavior Checklist. *Each outcome was adjusted for baseline score.
as outlined by the American Academy of Pediatrics.31 This study has implications for the clinical training and practice of pediatricians. Studies such as this can provide residency programs with an approach to improve competency in behavioral pediatrics, including identification of and early intervention with child behavior concerns.35 Residency is a stage of graduate medical training that allows for advanced training and for the development of automatic and integrated skills. Because residents receive greater clinical support and manage fewer patients than professionals, they are uniquely positioned to receive and practice behavioral health interventions. However, residents’ schedules and short-term relationships with patients may provide unique challenges to integrating new skills. Residents likely need extra support to increase comfort in initiating parenting discussions and eliciting information about the child’s behavior and parenting strategies. Incorporating training during residency would ensure a competent and confident workforce. The burden of responsibility on primary care doctors continues to grow even as appointments shorten, reimbursement rates fall, and resources diminish. Given these realities, we need to better prepare our pediatric primary care providers to identify and manage early signs of psychosocial problems in the families they serve.
ACKNOWLEDGMENTS Funding for this study was provided by the Doris Duke Charitable Foundation. We wish to thank the clinics, pediatric residents, and parents who participated in this study.
REFERENCES 1. Barth RP. Preventing child abuse and neglect with parent training: evidence and opportunities. Future Child. 2009;19:95–118. 2. Collins WA, Maccoby EE, Steinberg L, et al. Contemporary research on parenting: the case for nature and nature. Am Psychol. 2000;55: 218–232. 3. Sanders M. Triple P–Positive Parenting Program as a public health approach to strengthening parenting. J Fam Psychol. 2008;22: 506–517. 4. Garner AS, Shonkoff JP, Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood, Adoption, and Dependent Care, Section on Developmental and Behavioral Pediatrics. Early childhood adversity, toxic stress, and the role of the pedi-
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
19.
atrician: translating developmental science into lifelong health. Pediatrics. 2012;129:e224–e231. Lereya ST, Samara M, Wolke D. Parenting behavior and the risk of becoming a victim and a bully/victim: a meta-analysis study. Child Abuse Negl. 2013;37:1091–1108. Anda RF, Felitti VJ, Bremmer JD, et al. The enduring effects of abuse and related adverse experiences in childhood: a convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatr Clin Neurosci. 2006;256:174–186. Center on the Developing Child at Harvard University. The foundations of lifelong health are built early in childhood. 2010. Available at: http://www.developingchild.harvard.edu. Accessed October 22, 2010. Hillis SD, Anda RF, Dube SR, et al. The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics. 2004;113:320–327. Briggs-Gowan MJ, Owens PL, Schwab-Stone ME, et al. Persistence of psychiatric disorders in pediatric settings. J Am Acad Child Adolesc Psychiatry. 2003;42:1360–1369. Hurlburt MS, Nguyen K, Reid J, et al. Efficacy of the Incredible Years group parent program with families in Head Start who self-reported a history of child maltreatment. Child Abuse Negl. 2013;37:531–543. Kaminski JW, Perou R, Visser SN, et al. Behavioral and socioemotional outcomes through age 5 years of the legacy for children public health approach to improving developmental outcomes among children born into poverty. Am J Public Health. 2013;103:1058–1066. Institute of Medicine. Preventing Mental, Emotional, and Behavioral Disorders Among Young People: Progress and Possibilities. Washington, DC: National Academies Press; 2009. Lundahl B, Risser HJ, Lovejoy MC. A meta-analysis of parent training: moderators and follow-up effects. Clin Psychol Rev. 2006; 26:86–104. Nixon RDV. Treatment of behavior problems in preschoolers: a review of parent training programs. Clin Psychol Rev. 2002;22:525–546. Sanders M, Murphy-Brennan M, McAuliffe C. The development, evaluation of a training programme for general practitioners in evidence-based parent consultation skills. Int J Ment Health Promot. 2003;5:13–20. Earls MF, Hay SS. Setting the stage for success: implementation of developmental and behavioral screening and surveillance in primary care practice—the North Carolina Assuring Better Child Health and Development project. Pediatrics. 2006;118:e183–e188. Burklow KA, Vaughn LM, Valerius KS, Schultz JR. Parental expectations regarding discussions on psychosocial topics during pediatric office visits. Clin Pediatr. 2001;40:555–562. Zero to Three. Parenting Infants and Toddlers Today: Research Findings. Washington, DC: Hart Research Associates. Available at: http:// www.zerotothree.org/about-us/funded-projects/parenting-resources/ final_survey_report_3-11-2010.pdf; 2009. Accessed April 5, 2014. Coker TR, Shaikh Y, Chung PJ. Parent-reported quality of preventive care for children at-risk for developmental delay. Acad Pediatr. 2012; 12:384–390.
360
MCCORMICK ET AL
20. Grant E, Macnab A, Wambera K. The effectiveness of pediatric residency education in preparing graduates to manage neurological and neurobehavioral issues in practice. Acad Med. 2007;82:304–309. 21. Foy JM, Kelleher KJ, Laraque D. American Academy of Pediatrics Task Force on Mental Health. Enhancing pediatric mental health care: strategies for preparing a primary care practice. Pediatrics. 2010;125(suppl 3):S87–S108. 22. Flaherty E, Stirling J, Committee on Child Abuse and Neglect. Clinical report: the pediatrician’s role in child maltreatment prevention. Pediatrics. 2010;126:833–841. 23. Turner KMT, Sanders M. Help when it’s needed first: a controlled evaluation of preventative behavioral family intervention in a primary care setting. Behav Ther. 2006;37:131–142. 24. Sanders M, Tully L, Turner K, Maher C. Training GPs in parent consultation skills. Aust Fam Physician. 2003;32:763–768. 25. De Graaf I, Speetjens P, Smit F, et al. Effectiveness of the Triple P Positive Parenting Program on behavioral problems in children: a metaanalysis. Behav Modif. 2008;23:714–735. 26. Nowak C, Heinrichs N. A comprehensive meta-analysis of Triple P– Positive Parenting Program using hierarchical linear modeling: effectiveness and moderating variables. Clin Child Fam Psychol Rev. 2008; 11:114–144. 27. Participant Notes for Primary Care Triple P Provider Training. Milton, Australia: Triple P International Pty Ltd; 2010.
ACADEMIC PEDIATRICS 28. Turner KMT, Sanders MR. Parent Consultation Skills Checklist. Brisbane, Australia: Parenting and Family Support Centre, University of Queensland; 1996. 29. Turner KMT, Nicholson JM, Sanders MR. The role of practitioner self efficacy, training and workplace factors on the implementation of an evidence-based parenting intervention in primary care. J Prim Prev. 2011;32:95–112. 30. Achenbach TM, Rescorla LA. Manual for the ASEBA Preschool Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Child, Youth and Families; 2000. 31. Achenbach TM, Rescorla LA. Manual for the ASEBA School-Age Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth and Families; 2001. 32. Jensen PS, Watanabe HK, Richters JE, et al. Scales, diagnoses, and child psychopathology: II. Comparing the CBCL and the DISC against external validators. J Abnorm Child Psychol. 1996;24: 151–168. 33. Johnston C, Mash EJ. A measure of parenting satisfaction and efficacry. J Clin Child Psychol. 1989;18:167–175. 34. Runyan DK, Dunne MP, Zolotor AJ, et al. The development and piloting of the ISPCAN Child Abuse Screening Tool–Parent version (ICAST-P). Child Abuse Negl. 2009;33:826–832. 35. Committee on Psychosocial Aspects of Child and Family Health and Task Force on Mental Health. The future of pediatrics: mental health competencies for pediatric primary care. Pediatrics. 2009;124: 410–421.
ABSTRACT OBJECTIVE: We evaluated the effect of Primary Care Positive
RESULTS: Primary Care Triple P training had a positive, significant, and persistent impact on residents’ parenting consultation skills (mean increase on Parent Consultation Skills Checklist 48.11, 95% confidence interval [CI] 40.07, 57.36). Parents visiting intervention-trained residents demonstrated improved disciplinary practices compared to parents visiting control residents (mean change in Child Discipline Survey 0.322, 95% CI 0.02, 0.71), with stronger differential effects for parents with lower baseline skills (mean Child Discipline Survey change 0.822, 95% CI 0.48, 1.83). No differences were found for child behavior or parenting sense of confidence. CONCLUSIONS: Training residents in Primary Care Triple P can have a positive impact on consultation skills and parent disciplinary practices. This finding adds strength to the call for increased residency training in behavioral pediatrics. KEYWORDS: child; graduate medical education; parenting; pediatric resident; primary care; Triple P
Parenting Program (Triple P) training on pediatric residents and the families they serve to test 2 hypotheses: first, training would significantly improve resident skill in identifying and addressing discrete parenting and child behavior problems; and second, parents would report an improvement in their sense of selfefficacy, use of positive discipline strategies, and their child’s behavior. METHODS: Study participants included pediatric residents from 3 community clinics of a pediatric residency program, as well as English-speaking parents of children aged 18 months to 12 years without a diagnosed behavior disorder cared for by study residents. Residents were randomized to receive Primary Care Triple P training either at the beginning or end of the study period. The measured resident outcomes were selfassessed confidence and skills in giving parenting advice. The measured family outcomes were parent sense of self-efficacy, child externalizing behavior, and discipline strategies.
ACADEMIC PEDIATRICS 2014;14:353–360
WHAT’S NEW
problems, such evidence-based programs currently reach a limited proportion of parents.13–15 Public health approaches to preventing physical and mental health problems in children increasingly recognize the medical home as a crucial site of child abuse and neglect identification and early detection and treatment of mental, emotional, and behavioral disorders. Pediatric primary care providers are often the first contact point for children and families experiencing behavioral and parenting difficulties, and research consistently demonstrates that parents view their primary care providers as credible sources of parenting information.16–19 However, primary care providers vary in their degree of confidence and skill in managing mental, emotional, and behavioral concerns. In part, these variations may be due to a lack of specific instruction on brief, evidence-based behavioral interventions during residency.20,21 Pediatric residency offers an ideal time to teach and practice evidence-based behavioral and parenting interventions to ensure that these professional competencies develop.22
Training pediatric residents in the Positive Parenting Program (Triple P) improved provider skill and confidence in conducting parent consultations about child behavior concerns. Study finings support prior research on Triple P’s training effect and demonstrate promising family-level outcomes to explore in future research.
EFFECTIVE PARENTING IS a key determinant of healthy child development and strongly affects a child’s risk of developing emotional and behavioral disorders.1–3 Early emotional and behavioral symptoms as well as coercive and negative parenting can place children at increased risk for later behavior problems, child maltreatment, and other significant physical and emotional sequelae.4–8 Interruption and cessation of negative parenting styles can attenuate and even prevent progression to abuse and/ or serious mental health problems.9–12 Although there are evidence-based parenting programs with demonstrated success in improving parenting skills and child behavioral ACADEMIC PEDIATRICS Copyright ª 2014 by Academic Pediatric Association
353
Volume 14, Number 4 July–August 2014
354
MCCORMICK ET AL
The Positive Parenting Program (Triple P) is an evidence-based parenting program aimed at promoting positive parenting and preventing child abuse and neglect. Although Triple P has proven effective at reducing coercive parenting practices and disruptive behavior when delivered in community clinics by nurses and mental health professionals, there is limited evaluation on whether pediatricians can effectively deliver this model.23–26 This study has 2 objectives: to evaluate the impact of Triple P training on pediatric residents’ skill and confidence in delivering parenting information; and to understand the potential impact of residents’ Triple P training status on parents’ sense of self efficacy in parenting and discipline strategies as well as child behavior.
ACADEMIC PEDIATRICS
The sample of parents was selected on the basis of their affiliation with one of the residents in the clinical trial, and provision of Triple P services to an individual patient was not required for parent enrollment into the study. Therefore, this study examined a general impact of resident training on outcomes for families. It was not a direct test of the Triple P intervention effectiveness on families. Participating families may or may not have received Triple P during the course of this study, as parents were not asked directly about Triple P. The Seattle Children’s Research Institute institutional review board approved all study procedures, and the trial was registered at ClinicalTrials.gov (NCT01946958). PARTICIPANTS
METHODS STUDY SETTINGS The study was conducted in 3 distinct community clinics operated by University of Washington faculty, which serve as resident continuity clinic teaching sites. One was an inner-city pediatric clinic, another was a health department clinic, and the third was a clinic in a large multispecialty primary care center. DESIGN The randomized controlled trial included 2 study populations: pediatric residents and parents of children served by these residents. Consented residents were randomized using a computer randomization sequence. Eligible residents who consented to participation were randomized into one of two conditions: 1) intervention, which involved immediate training in Primary Care Triple P (PC Triple P), or 2) wait list control, which involved the usual provision of well-child services until subsequent training in PC Triple P. Intervention residents received PC Triple P training at the beginning of the study, and control residents received training 9 months later. Because residents were the unit of randomization, participating clinics contained both intervention and control group residents.
Figure 1. Flow chart of resident participation.
PEDIATRIC RESIDENTS Eligible residents from all years of residency (PL1, 2, and 3) attended continuity clinics at participating clinics. The residency program director introduced the project to eligible pediatric residents via e-mail. After the e-mail introduction, the research coordinator approached eligible residents to discuss participation. Figure 1 depicts the flow of residents through the trial as well as retention and attrition numbers at key study intervals. PARENTS AND CAREGIVERS After training the intervention group residents, parent recruitment began across both condition and in all participating clinics. Eligible parents had a well-child appointment the intervention group received PC Triple P training with a participating resident, spoke English, and had a child between the ages of 18 months and 12 years. Children who had received current DSM-IV diagnoses or who were receiving treatment for a behavioral problem were excluded from the trial because the Triple P training administered to residents was not intended for this subpopulation. Parents were approached and consented in person at their primary care clinic after their child’s well-child appointment. Parents were blinded to resident randomization status and the Triple P intervention. Study staff presented the
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
355
Figure 2. Flow chart of parent participation.
project as an exploration of how pediatricians address and manage parents’ concerns about child behavior and parenting strategies. Figure 2 depicts the flow of parent participants through the trial as well as retention and attrition at key study intervals. INTERVENTION: PRIMARY CARE TRIPLE P TRAINING AND ACCREDITATION Residents randomized to each group received standardized PC Triple P training (12 direct training hours over 1.5 days) followed by a half-day accreditation session 6 to 8 weeks after training. This training course prepares practitioners to deliver a brief parent education intervention to address commonly occurring developmental and behavioral difficulties in children aged 0 to 12 years. This level of Triple P can be delivered as part of a universal parenting support strategy or as an early intervention strategy for parents with children at risk of developing oppositional, disruptive, and/or aggressive behaviors. The PC Triple P training course is appropriate for practitioners who routinely offer advice and support to parents during time-limited appointments in the primary care setting. The course provides a comprehensive overview of common developmental and behavioral problems in children as well as a range of parenting strategies to promote child development. The course includes practical skills training in a range of parent interview skills necessary in the delivery of a brief parenting intervention. More specifically, the training course enhances provider skills and knowledge in the following areas: Early detection and identification of family risk and protective factors. Causes of child behavior problems. Core principles of positive parenting and effective behavior change. Specific positive parenting strategies for promoting children’s development, social competence, and selfcontrol. Effective parent consultation and skills training procedures. Identification of indicators suggesting more intensive intervention. Appropriate referral procedures.
At completion of the training course, practitioners had the ability to deliver counseling via 1 to 4 sessions to families they serve.27 After the initial 1.5-day training, practitioners were encouraged to practice their new skills until the accreditation session (6 to 8 weeks after the initial training). Accreditation is a necessary component of the competency-based training course and is required for official recognition of program proficiency. At accreditation, practitioners have an opportunity to demonstrate, via roleplay, their proficiency in the two competency areas targeted for PC Triple P accreditation: 1) discussion of a parenting plan, and 2) discussion and modification of a parenting plan that has been implemented. During accreditation, participants receive coaching and feedback on their performance of the targeted competences. At this session, practitioners submit an accreditation quiz, testing comprehension of the Triple P model and intervention components. Practitioners must pass the accreditation quiz and skill demonstrations to be a Triple P–accredited practitioner. Study staff collaborated with Triple P America to hold trainings, and Triple P America provided doctoral-level Triple P trainers for the trainings and accreditation sessions. RESIDENT-REPORT MEASURES PRACTITIONER SKILL IN ADDRESSING BEHAVIORAL CONCERNS WITH PARENTS The Parent Consultation Skills Checklist (PCSC) is a standardized instrument that assesses a practitioner’s proficiency in 4 core domains: assessment, active skills training, dealing with process issues, and clinical application of positive parenting strategies.28 The tool also assesses confidence in important parent consultation skills. The measure consists of 18 items, which are rated on a 7-point scale ranging from 1 (not at all confident) to 7 (very confident). Factor analysis has indicated that the checklist measures a single construct (a ¼ 0.97) with all items loading significantly on the scale.15 The checklist has been shown to have good internal consistency (a ¼ .96–.97).29 Maximum score for the
356
MCCORMICK ET AL
PCSC is 154, with higher scores indicating greater competency. PARENT-REPORT MEASURES CHILD BEHAVIOR We measured child behavior with the externalizing subscale of the Child Behavior Checklist (CBCL).30–32 The CBCL/1.5–5 is appropriate for children between 1.5 and 5 years of age, and the CBCL/6–18 is appropriate for children from 6 to 18 years of age. The externalizing scales measure problems involving conflicts with other people and parent expectations for their child. The CBCL/ 1.5–5 externalizing scale consists of 24 items, while the CBCL/6–18 scale contains 35 items. Discriminant analyses of the CBCL/6–18 indicate that 85% of children are correctly classified as referred or nonreferred. To standardize responses across age ranges, we calculated T scores for all participants. PARENTAL SENSE OF SELF-EFFICACY We measured parental self-efficacy with the Parenting Sense of Competence Scale (PSOC).33 The PSOC is a 16-item tool assessing parents’ views of their competence on two dimensions: satisfaction with their parenting role and feeling of efficacy as a parent. The total score, satisfaction score, and efficacy score show a satisfactory level of internal consistency (0.79, 0.75, and 0.76, respectively). The maximum score for this measure is 96, with higher scores indicating greater confidence a parent feels in his or her parenting abilities. USE OF POSITIVE DISCIPLINE STRATEGIES We selected 9 items from the International Society for the Prevention of Child Abuse and Neglect’s Child Abuse Screening Tool–Parent Version (ICAST-P) to identify parental discipline strategies and beliefs.34 It has face validity and reasonable internal consistency, and it has been demonstrated to be acceptable to parents in the countries where it was piloted. Higher scores indicate that more positive discipline techniques were used. ASSESSMENT TIMING The PCSC served as a self-report of resident selfefficacy in parent consultation. Because participating residents received the PC Triple P training at specific (and different) times depending on their study status (intervention or control), we examined skill differences at key time points in the trial. At time T1, both the intervention and control group residents completed their baseline measure of skills using the PCSC. The intervention group repeated the PCSC immediately after training (T2). Approximately 9 months after the intervention group training (T3), the intervention group completed the PCSC again to measure the long term effect of the training. When intervention residents completed the follow-up PCSC, control residents received training in PC Triple P. Control residents completed the PCSC at the beginning
ACADEMIC PEDIATRICS
of the training (also T3). The control group repeated the PCSC at the end of the training (T4). Parents had the option to complete study assessments (CBCL, PSOC, and Child Discipline Survey) at the time of recruitment in the clinic or at home, returning the measures via mail. Three months after their consent appointment, study staff contacted parents by telephone. At this time, parents completed the CBCL, PSOC, and ICAST-P by telephone with study staff. DATA ANALYSIS Data were key entered and analyzed by SPSS (IBM, Armonk, NY). We examined mean differences in scores from one time to another, adjusting for baseline differences. We also conducted a subanalysis for each of the parent measures on the group of parents whose scores in that measure were in the bottom half of the distribution.
RESULTS DEMOGRAPHICS Fifty-three pediatric residents participated in the randomized control trial, with 25 assigned to the intervention group and 28 to the control group (Table 1). The majority of residents were female (79%), and the vast majority (96%) were not parents. Residents in the two groups were distributed across the three years of training. Sixty percent of residents were assigned to one continuity clinic, the largest clinic in our training system. A total of 101 families participated in the trial. Parents were evenly distributed between intervention and control residents (Table 2). The mean age for participating children was 4.8 years (standard deviation 3.25), and male and female children were represented equally. Parent participants were predominantly mothers. The majority of families had less than a 1-year relationship with the resident. Of the families that reported their household income, 51% reported annual household incomes lower than $30,000. Over half of families identified as an ethnic minority or other nonwhite. Age and language eligibility requirements limited family participation at each clinic. The result of these limitations was a high proportion of recruited families from a single clinic. Table 1. Characteristics of Pediatric Resident Study Participants* Characteristic Female sex Guardian status ¼ parent Continuity clinic assignment Clinic 1 Clinic 2 Clinic 3 Residency year Intern Second year Third year
Intervention, n (%) (n ¼ 25)
Control, n (%) (n ¼ 28)
19 (76) 1 (4)
23 (82) 1 (4)
4 (16) 17 (68) 4 (16)
6 (21) 15 (54) 7 (25)
10 (40) 8 (32) 7 (28)
11 (39) 11 (39) 6 (21)
*There were no significant differences between groups.
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
Table 2. Characteristics of Family Study Participants* Characteristic Pediatric clinic† Clinic 1 Clinic 2 Clinic 3 How long family knew provider <1 y 1–2 y $3 y Child’s age, y Child’s sex Female Relationship to child Mother Parent’s race White African American/black Other Highest level of education High school or below Some college College degree or higher Household income Less than $29,999 $30,000–$49,999 $50,000 and above
Intervention (n ¼ 51)
Control (n ¼ 50)
12 (24) 37 (73) 2 (4)
17 (34) 21 (42) 12 (24)
39 (77) 10 (20) 2 (4) 4.7 (3.1)
27 (59) 15 (33) 4 (9) 4.9 (3.4)
24 (47)
24 (48)
41 (80)
40 (80)
23 (45) 18 (35) 20 (20)
15 (30) 12 (24) 23 (46)
8 (17) 21 (44) 19 (40)
18 (36) 16 (32) 16 (32)
19 (46) 13 (32) 9 (22)
25 (55) 10 (22) 10 (22)
*Some respondents did not answer every self-report question. Data are presented as n (%) or mean standard deviation. †A significant difference was found in recruitment clinic site.
RESIDENT OUTCOMES Analyses across time demonstrate that Triple P training had a positive, significant, and persistent impact on the skills, competence, and confidence of pediatric residents, as measured by the PCSC (Fig. 3). At baseline (T1), no significant differences existed between the control and intervention residents’ scores on the baseline measure (mean difference 4.90, 95% confidence interval [CI] 5.52, 15.34). At the intervention group posttraining (T2), the intervention group reported significantly higher levels of confidence and skill in conducting parent consultations (mean difference 48.11, 95% CI 40.07, 57.36) than their control group counterparts. Fifty-three residents were included in baseline and posttraining analyses. At ninemonth follow-up (T3), the intervention group maintained a significantly higher mean score than at baseline, with only a slight drop from the postintervention training. At follow-up, the difference between the intervention and control group remained significant (mean difference 39.03, 95% CI 28.75, 50.91). At this time point, the control
Figure 3. Resident PCSC mean scores over time in trial.
357
group received training. At the postintervention training (T4), the difference in self-reported skill and confidence between intervention and control group was no longer statistically significant (mean difference 0.07, 95% CI 8.11, 9.49). Thirty-nine residents (74%) were included in follow-up and control group training analyses. Over the course of the study, we lost residents to graduation, out-of-state rotations, and time conflicts. PARENT AND CHILD OUTCOMES We analyzed the change in parents’ sense of selfefficacy, discipline strategies, and child behavior over time using the parent-completed assessments from baseline and follow-up. Our analysis of the parent and child data showed mixed results. We attempted follow-up with all parent participants; however, only 56% of parents completed follow-up at parent T2. At baseline, there were no significant differences in any outcome measure between groups (Table 3). At follow-up, there were no significant differences between groups or within groups. However, in the adjusted analyses, parents who had received care from an intervention-trained resident showed greater improvements on the Child Discipline Survey at follow-up, indicating improved disciplinary practices in the parents exposed to Triple P–trained residents (Table 4). The difference (on the ICAST-P measure, the discipline proxy) between groups was revealed in the regression models, with the greatest effect found in families that scored at or below the 50th percentile on the ICAST-P scale at baseline (b ¼ 0.816, 95% CI 0.48, 1.83) (Table 4). The effect was also stronger in families that had a longer relationship with their pediatrician (b ¼ 0.693, 95% CI 0.10, 1.70) (Table 5). No intergroup differences were found for child behavior or parenting sense of confidence.
DISCUSSION Training pediatric residents to deliver parenting education using Triple P improved provider self-assessed skill and confidence; it also positively affected parental disciplinary behaviors in families that received care from intervention-trained residents. The impact of interventiontrained residents on parent’s sense of self-efficacy, discipline strategies, and child behavior was mixed, with the greatest group effect occurring on parent-reported discipline strategies, especially among parents who reported the most problems at baseline. Parents exposed to Triple P–trained residents demonstrated a higher rate of positive discipline strategies at follow-up. However, there appeared to be little impact of the intervention on the other parent measures. One possible explanation for the reported improvement in discipline strategies is that discipline is a frequently discussed psychosocial concern for parents. Burklow et al found that misbehavior was the most common psychosocial concerns parents presented with to their pediatrician, and discipline was the more frequently discussed strategy to address the behavioral concern.17 Managing child
358
MCCORMICK ET AL
ACADEMIC PEDIATRICS
Table 3. Parent Report Outcomes Over Time* Outcome Measure Parent’s sense of confidence Intervention Control Use of positive discipline strategies Intervention Control Child behavior problems Intervention Control
Baseline (T1), Mean SD
Follow-up (T2), Mean SD
72.94 10.42 75.44 10.46
70.89 11.18 76.10 14.72
7.20 .92 7.22 .84
7.81 .40 7.65 .67
48.64 10.35 47.20 8.75
49.45 8.93 52.85 7.82
95% CI at T1†
95% CI at T2†
1.62, 6.62
1.76, 12.17
0.32, 0.37
0.44, 0.13
5.35, 2.46
1.51, 8.31
SD ¼ standard deviation; CI ¼ confidence interval. *Intervention group experienced no significant differences between baseline and follow-up in all outcomes. †No significant differences exist between groups at either baseline or follow-up.
misbehavior might be the entry point to a longer-term discussion about promoting positive interaction between parent and child. It should also be noted that behavior problems worsened in the control group over time, a change not seen in the intervention group, suggesting an effect on behavior as well. The study had a number of limitations, including a small sample size as well as recruitment and implementation level barriers. The cost of Triple P training, coupled with the available funding, permitted only a limited number of training sessions. The Triple P model is not a train-thetrainer model. Rather, it requires a certified trainer contracted by Triple P America and mandates an upper limit of 20 people per training. Residents’ complicated work schedules, made even more complicated by work-hour restrictions, makes patient continuity difficult. Furthermore, residents have lower patient volumes than practicing pediatricians. Additionally, clinic and study staff encountered a moderate rate of patients who failed to keep appointments. We lost parents during follow-up as a result of an inability to contact families, despite multiple attempts (disconnected numbers, unanswered or unreturned calls, etc). We had a high retention rate at follow-up for those we were able to reach by phone. Because the family sample size was small, and because families were enrolled into the study regardless of receipt of Triple P, the possible direct impact of the Triple P intervention may be diluted in this sample. However, the impact of resident training on parenting outcomes in general may point to a generalization effect of the training that is
robust—that is, knowledge gained through the training provided a framework for parenting support that extended beyond the use of a specific model. We also faced challenges in fully supporting trained residents without compromising the control group. Because residents were the unit of randomization instead of the clinic, clinics were staffed by both control and intervention residents. Triple P materials were not displayed in participating clinics, and program support was not integrated into existing group supervision structures at clinics. In addition, residents did not always have access to Triple P– trained preceptors. Although several clinic preceptors attended the Triple P training, clinics had gaps in Triple P staff knowledge and skills. Finally, not all residents completed the entire training. Although all interventiontrained residents attended the initial training days, less than half (44%) attended accreditation sessions. All of these limitations may have constricted use of Triple P with families and attenuated the level of support provided to residents for enhancing their use of the model. Further research is needed to understand the potential of this early intervention strategy on families and should focus on creating environments in which implementation of the Triple P program is encouraged and comprehensively reinforced. Our findings are consistent with a prior study evaluating the effect of training general practitioners in PC Triple P.24 Results of this study show promising outcomes of parent consultations and greater use of parent consultation skills. These training effects are directly related to competencies
Table 4. Multivariate Analysis of the Relationship of Parent-Report Outcomes to Resident and Family Characteristics* b (95% CI) for: Variable
PSOC
CDS
CBCL
Intervention Clinic Length of relationship with provider Child’s sex Child’s age Respondent’s relationship to child Race Education Income
0.007 (5.52, 5.91) 0.012 (5.06, 5.62) 0.107 (4.03, 9.24) 0.394 (15.75, 4.42) 0.141 (0.44, 1.63) 0.027 (1.97, 1.57) 0.140 (1.25, 4.34) 0.164 (1.12, 4.63) 0.147 (2.94, 0.969)
0.322 (0.02, 0.71) 0.012 (0.38, 0.31) 0.235 (0.16, 0.64) 0.289 (0.65, 0.03) 0.062 (0.05, 0.07) 0.232 (0.04, 0.18) 0.152 (0.10, 0.24) 0.186 (0.09, 0.26) 0.036 (0.12, 0.10)
2.67 (8.19, 2.84) 0.396 (5.14, 5.94) 0.137 (6.50, 6.78) 2.57 (3.29, 8.43) 0.371 (1.41, 0.67) 0.259 (1.99, 1.47) 0.434 (3.23, 2.37) 1.17 (1.65, 3.99) 0.754 (2.67, 1.16)
CI ¼ confidence interval; PSOC ¼ Parent Sense of Competence Scale; CDS ¼ Child Discipline Survey; CBCL ¼ Child Behavior Checklist. *Each outcome was adjusted for baseline score.
ACADEMIC PEDIATRICS
PEDIATRIC RESIDENTS AND PARENT EDUCATION
359
Table 5. Multivariate Analysis of Relationship of Families Scoring Below 50th Percentile on Parent Report Outcome Measures to Resident and Family Characteristics* b (95% CI) for: Variable
PSOC
CDS
CBCL
Intervention Clinic Length of relationship with provider Child’s sex Child’s age Respondent’s relationship to child Race Education Income
0.268 (16.08, 4.10) 0.438 (7.91, 23.17) 0.073 (26.63, 30.05) 0.890 (36.57, 3.24) 0.262 (3.87, 1.18) 0.372 (2.98, 6.38) 0.345 (7.86, 14.91) 0.766 (0.374, 15.28) 0.379 (6.97, 2.24)
0.816 (0.48, 1.83) 0.189 (0.10, 0.48) 0.693 (0.10, 1.70) 0.833 (1.79, 0.39) 0.200 (0.19, 0.11) 0.262 (0.30, 0.85) 0.412 (0.08, 0.55) 0.465 (0.76, 0.26) 0.128 (0.22, 0.32)
4.62 (11.16, 1.93) 3.65 (10.14, 2.84) 0.586 (9.34, 8.17) 1.83 (4.44, 8.09) .561 (1.18, 2.31) 1.153 (1.04, 3.34) 0.384 (3.51, 2.75) 1.63 (4.01, 0.76) 0.243 (2.07, 1.59)
CI ¼ confidence interval; PSOC ¼ Parent Sense of Competence Scale; CDS ¼ Child Discipline Survey; CBCL ¼ Child Behavior Checklist. *Each outcome was adjusted for baseline score.
as outlined by the American Academy of Pediatrics.31 This study has implications for the clinical training and practice of pediatricians. Studies such as this can provide residency programs with an approach to improve competency in behavioral pediatrics, including identification of and early intervention with child behavior concerns.35 Residency is a stage of graduate medical training that allows for advanced training and for the development of automatic and integrated skills. Because residents receive greater clinical support and manage fewer patients than professionals, they are uniquely positioned to receive and practice behavioral health interventions. However, residents’ schedules and short-term relationships with patients may provide unique challenges to integrating new skills. Residents likely need extra support to increase comfort in initiating parenting discussions and eliciting information about the child’s behavior and parenting strategies. Incorporating training during residency would ensure a competent and confident workforce. The burden of responsibility on primary care doctors continues to grow even as appointments shorten, reimbursement rates fall, and resources diminish. Given these realities, we need to better prepare our pediatric primary care providers to identify and manage early signs of psychosocial problems in the families they serve.
ACKNOWLEDGMENTS Funding for this study was provided by the Doris Duke Charitable Foundation. We wish to thank the clinics, pediatric residents, and parents who participated in this study.
REFERENCES 1. Barth RP. Preventing child abuse and neglect with parent training: evidence and opportunities. Future Child. 2009;19:95–118. 2. Collins WA, Maccoby EE, Steinberg L, et al. Contemporary research on parenting: the case for nature and nature. Am Psychol. 2000;55: 218–232. 3. Sanders M. Triple P–Positive Parenting Program as a public health approach to strengthening parenting. J Fam Psychol. 2008;22: 506–517. 4. Garner AS, Shonkoff JP, Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood, Adoption, and Dependent Care, Section on Developmental and Behavioral Pediatrics. Early childhood adversity, toxic stress, and the role of the pedi-
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
19.
atrician: translating developmental science into lifelong health. Pediatrics. 2012;129:e224–e231. Lereya ST, Samara M, Wolke D. Parenting behavior and the risk of becoming a victim and a bully/victim: a meta-analysis study. Child Abuse Negl. 2013;37:1091–1108. Anda RF, Felitti VJ, Bremmer JD, et al. The enduring effects of abuse and related adverse experiences in childhood: a convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatr Clin Neurosci. 2006;256:174–186. Center on the Developing Child at Harvard University. The foundations of lifelong health are built early in childhood. 2010. Available at: http://www.developingchild.harvard.edu. Accessed October 22, 2010. Hillis SD, Anda RF, Dube SR, et al. The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics. 2004;113:320–327. Briggs-Gowan MJ, Owens PL, Schwab-Stone ME, et al. Persistence of psychiatric disorders in pediatric settings. J Am Acad Child Adolesc Psychiatry. 2003;42:1360–1369. Hurlburt MS, Nguyen K, Reid J, et al. Efficacy of the Incredible Years group parent program with families in Head Start who self-reported a history of child maltreatment. Child Abuse Negl. 2013;37:531–543. Kaminski JW, Perou R, Visser SN, et al. Behavioral and socioemotional outcomes through age 5 years of the legacy for children public health approach to improving developmental outcomes among children born into poverty. Am J Public Health. 2013;103:1058–1066. Institute of Medicine. Preventing Mental, Emotional, and Behavioral Disorders Among Young People: Progress and Possibilities. Washington, DC: National Academies Press; 2009. Lundahl B, Risser HJ, Lovejoy MC. A meta-analysis of parent training: moderators and follow-up effects. Clin Psychol Rev. 2006; 26:86–104. Nixon RDV. Treatment of behavior problems in preschoolers: a review of parent training programs. Clin Psychol Rev. 2002;22:525–546. Sanders M, Murphy-Brennan M, McAuliffe C. The development, evaluation of a training programme for general practitioners in evidence-based parent consultation skills. Int J Ment Health Promot. 2003;5:13–20. Earls MF, Hay SS. Setting the stage for success: implementation of developmental and behavioral screening and surveillance in primary care practice—the North Carolina Assuring Better Child Health and Development project. Pediatrics. 2006;118:e183–e188. Burklow KA, Vaughn LM, Valerius KS, Schultz JR. Parental expectations regarding discussions on psychosocial topics during pediatric office visits. Clin Pediatr. 2001;40:555–562. Zero to Three. Parenting Infants and Toddlers Today: Research Findings. Washington, DC: Hart Research Associates. Available at: http:// www.zerotothree.org/about-us/funded-projects/parenting-resources/ final_survey_report_3-11-2010.pdf; 2009. Accessed April 5, 2014. Coker TR, Shaikh Y, Chung PJ. Parent-reported quality of preventive care for children at-risk for developmental delay. Acad Pediatr. 2012; 12:384–390.
360
MCCORMICK ET AL
20. Grant E, Macnab A, Wambera K. The effectiveness of pediatric residency education in preparing graduates to manage neurological and neurobehavioral issues in practice. Acad Med. 2007;82:304–309. 21. Foy JM, Kelleher KJ, Laraque D. American Academy of Pediatrics Task Force on Mental Health. Enhancing pediatric mental health care: strategies for preparing a primary care practice. Pediatrics. 2010;125(suppl 3):S87–S108. 22. Flaherty E, Stirling J, Committee on Child Abuse and Neglect. Clinical report: the pediatrician’s role in child maltreatment prevention. Pediatrics. 2010;126:833–841. 23. Turner KMT, Sanders M. Help when it’s needed first: a controlled evaluation of preventative behavioral family intervention in a primary care setting. Behav Ther. 2006;37:131–142. 24. Sanders M, Tully L, Turner K, Maher C. Training GPs in parent consultation skills. Aust Fam Physician. 2003;32:763–768. 25. De Graaf I, Speetjens P, Smit F, et al. Effectiveness of the Triple P Positive Parenting Program on behavioral problems in children: a metaanalysis. Behav Modif. 2008;23:714–735. 26. Nowak C, Heinrichs N. A comprehensive meta-analysis of Triple P– Positive Parenting Program using hierarchical linear modeling: effectiveness and moderating variables. Clin Child Fam Psychol Rev. 2008; 11:114–144. 27. Participant Notes for Primary Care Triple P Provider Training. Milton, Australia: Triple P International Pty Ltd; 2010.
ACADEMIC PEDIATRICS 28. Turner KMT, Sanders MR. Parent Consultation Skills Checklist. Brisbane, Australia: Parenting and Family Support Centre, University of Queensland; 1996. 29. Turner KMT, Nicholson JM, Sanders MR. The role of practitioner self efficacy, training and workplace factors on the implementation of an evidence-based parenting intervention in primary care. J Prim Prev. 2011;32:95–112. 30. Achenbach TM, Rescorla LA. Manual for the ASEBA Preschool Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Child, Youth and Families; 2000. 31. Achenbach TM, Rescorla LA. Manual for the ASEBA School-Age Forms and Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth and Families; 2001. 32. Jensen PS, Watanabe HK, Richters JE, et al. Scales, diagnoses, and child psychopathology: II. Comparing the CBCL and the DISC against external validators. J Abnorm Child Psychol. 1996;24: 151–168. 33. Johnston C, Mash EJ. A measure of parenting satisfaction and efficacry. J Clin Child Psychol. 1989;18:167–175. 34. Runyan DK, Dunne MP, Zolotor AJ, et al. The development and piloting of the ISPCAN Child Abuse Screening Tool–Parent version (ICAST-P). Child Abuse Negl. 2009;33:826–832. 35. Committee on Psychosocial Aspects of Child and Family Health and Task Force on Mental Health. The future of pediatrics: mental health competencies for pediatric primary care. Pediatrics. 2009;124: 410–421.