Self-perceived importance and confidence of adolescents transitioning to adult care

Abstract

Purpose

Motivational interviewing (MI) techniques are used by health care teams to engage adolescents and young adults (AYAs) in health care self-management and pediatric to adult health care transition (HCT) planning efforts. The aim of this study was to assess the initial level of motivation of AYAs prior to receipt of HCT anticipatory guidance and to determine associations with demographic and health coverage factors.

Methods

This retrospective study included 5112 AYAs, aged 12–26 years, from four health systems. All AYAs completed the Got Transition readiness assessment that includes MI questions on importance and confidence related to the move to an adult provider.Independent variables included demographic and health coverage factors: age, sex, race, ethnicity, language, and insurance type. The statistical approach included summary statistics, chi-square tests of independence and log-likelihood ratio tests, and generalized linear models and contrasts.

Results

The study results demonstrate initial trends in importance and confidence scores for AYAs before they became part of a HCT planning process. Importance scores increased from 12-14 through 18–20 years of age, then decreased in the 21–26-year group. Confidence scores increased from the 12–14 through the 21–26-year group.Confidence scores were generally higher than importance scores and were accompanied by smaller standard deviations. Ethnicity and insurance type also demonstrated an association with MI scoring.

Discussion

This study provides baseline scores on two key MI importance and confidence questions that can facilitate clinician understanding of AYA engagement in discussing the changes needed to move to adult care and guides the clinician to start earlier than just before transfer that often occurs around age 21.

Highlights

• •Age, ethnicity, and insurance type were associated with importance and confidence.
• •Mean AYA scores were generally higher for confidence than importance.
• •Provides new insight into motivation levels of AYAs for health care transition.
• •Results can assist clinicians to optimally address transition readiness.
• •Demonstrates the need to initiate health care transition preparation earlier.

1. Introduction

Pediatric to adult health care transition (HCT) is a challenging shift for adolescents and young adults (AYAs). Challenges include changes from family to patient centered care, ending long-standing provider relations, and finding and adapting to new health care systems.¹ Systematic literature reviews have shown that a structured health care transition process that includes planning, transfer, and integration into an adult model of care statistically improves outcomes for youth with special health care needs (SHCN): adherence, patient experience, and health care utilization.2, 3 The 2021–2022 National Survey of Children’s Health results found that 22.1 % of adolescents, aged 12–17 years with SHCN, and 17.8 % without SHCN received services for planning transition to adult care⁴ indicating a widespread gap in HCT preparation.

National HCT best practices have been outlined in the American Academy of Pediatrics (AAP), American Academy of Family Physicians (AAFP), and American College of Physicians Clinical (ACP) Report, reaffirmed in 2023 ¹ (http://publications.aap.org/pediatrics/article-pdf/142/5/e20182587/1524035/peds_20182587.pdf) and include Got Transition’s Six Core Elements of Health Care Transition, an approach to guide clinicians and health care systems as they support patients in this transition to adult health care with or without transferring to an adult clinician. The Six Core Elements lay out an age-related roadmap that includes conducting annual transition readiness assessments and providing self-care skill-building starting at age 14 and continuing into young adulthood.¹ Got Transition’s customizable self-report transition readiness assessment includes two questions on HCT importance and confidence to initiate the process of HCT motivational interviewing (MI), specifically asking about the move to a clinician who cares for adults before age 22.⁵

In a systematic review and meta-analysis of randomized controlled trials, MI techniques have been shown to be an effective method to encourage individuals to make behavior changes and have demonstrated a statistically significant impact on a variety of medical and behavioral outcome measures that include confidence, intention to change, and engagement in treatment.⁶ A motivational interviewing based social network intervention to reduce substance use in emerging young adults demonstrated acceptability and potential feasibility.7, 8 Ball et al., 2011 found MI beneficial with adolescents aged 13–17 years with obesity when self-efficacy was low, or at later stages, after the occurrence of setbacks.⁹ Self-efficacy has been identified as the most important component of MI in a scoping review.¹⁰ A study that utilized self-determination theory with AYAs aged 17–23 years found that methods that support AYAs’ self-management competence (skills and confidence) and provider relatedness (support for autonomy) predicted transition readiness in those with SHCNs.¹¹ Promoting AYA confidence in managing their chronic condition and health self-efficacy, in addition to appropriate parent involvement and meeting the adult health care team prior to transfer, resulted in improved outcomes (participation, well-being, satisfaction, and condition specific measures) in a five year longitudinal study of AYAs with long-term medical conditions.¹²

Motivational interviewing has evolved to an overlapping four step process that involves engaging, focusing, evoking, and planning.¹³ The AAP/AAFP/ACP clinical report emphasizes the importance of AYA engagement in HCT. Importance and confidence are conceptually distinct steps in patient activation, and confidence has been identified as a validated domain to indicate engagement.14, 15 Confidence can be used as a proxy for patient engagement using a 0–10 scale to measure confidence and promote further communication with health care providers.¹⁶ Several studies have examined importance and confidence as they relate to HCT. Syverson et.al found that AYAs aged 16–22 years with SHCNs rated transition importance on a 1–10 scale significantly higher when a discussion related to health insurance occurred compared to AYAs who did not receive insurance related anticipatory guidance.¹⁷ A study of AYAs aged 15–22 years in Singapore found that transition importance and confidence scores were high at initial assessment with a mean score of 7 for both importance and confidence, using a 1–10 scale, and there was no significant difference between age groups.¹⁸ Initial high mean importance and confidence scores were also noted among AYAs aged 14–20 years in a quality improvement project (QIP) conducted in a federally qualified health center (FQHC).¹⁹ Another QIP found that AYAs with congenital heart disease and disabilities had significantly lower confidence scores than those without disabilities.²⁰

Despite extensive evidence of the benefits of MI, available studies published on the use of MI importance and confidence scores related to HCT have limited sample sizes, seldom examined those without SHCN, and have not analyzed differences by sociodemographic factors. Clinicians working in primary care clinics routinely see a broad mix of AYAs with and without special health care needs. The AAP/AAFP/ACP recommends that all AYAs with and without SHCNs receive transition planning.¹ This study addresses the following research questions: 1) What are the initial levels of self-reported importance and confidence related to HCT prior to receipt of HCT anticipatory guidance in a large sample of AYAs, aged 12–26 years with and without chronic conditions, from four large health systems; and 2) What associations exist between ratings of HCT importance and confidence and demographic and health characteristics that include age, sex, gender identity, race, ethnicity, language, and insurance type.

2. Methods

2.1. Study population, procedures, and materials

The study population was a convenience sample of 5112 AYAs aged 12–26 years from four health systems: Children’s Mercy Kansas City (Kansas City, MO), Cleveland Clinic Children’s (Cleveland, OH), Henry Ford Health (Detroit, MI), and Ravenswood Family Health Center, an FQHC (East Palo Alto, CA). The four health systems are part of the Got Transition HCT learning community. This community meets monthly to present quality improvement activities around system-wide changes in HCT. For this study, four of the ten participating health systems compiled and analyzed retrospective data from the first transition readiness assessment of AYAs in their system. All AYAs completed the Got Transition readiness assessment that includes the MI questions of importance and confidence related to the move to an adult clinician: 1) How important is it to you to prepare for/change to an adult doctor/clinician before age 22? and 2) How confident do you feel about your ability to prepare for/change to an adult doctor/clinician? (https://gottransition.org/6ce/?leaving-readiness-assessment-youth). This assessment was administered when the HCT process was initiated with each patient, prior to receipt of HCT specific anticipatory guidance. These times varied within each health system and continued through March 2020 at the start of COVID-19 the pandemic.

The two questions of importance and confidence were the primary dependent variables. These questions represent the first stage in the process of MI in which patient engagement is encouraged by eliciting their perceptions prior to implementation of HCT processes and further MI. Independent variables included demographic and health coverage factors: age, sex, gender identity, race, ethnicity, language, and insurance. Patient ages were grouped as 12–14, 15–17, 18–20, and 21–26 years. The ages were grouped to approximate stages of adolescence (early, middle, and late) and young adulthood while maintaining sufficient numbers in each group for comparison. Gender identity was excluded due to missing data. Although the study included youth with and without SHCN, it did not identify or differentiate responses of youth with chronic conditions due to challenges in determining what would be considered a chronic condition across the four institutions and the broad range of patient diagnoses and complexity among AYAs with SHCNs. Questions of importance and confidence regarding changing to an adult clinician were assessed using a 0–10 Likert scale. For a limited subset of patients cared for by family practice clinicians in the FQHC (<2 % of the total study population), the importance and confidence questions were related to managing their own health care since transition to adult care would not include a clinician change (https://gottransition.org/6ce/?staying-readiness-assessment-youth). Got Transition’s readiness assessment also includes customizable questions on health and health care readiness that were not analyzed for this study.

Methods to retrieve data included manual chart review at Ravenswood Family Health Center, where completed readiness assessments were scanned and stored in the electronic health record (EHR). All demographic and health characteristics and importance and confidence scores were manually pulled at Ravenswood by the study lead and were stored securely on an Excel file. A code book was set up in the Stanford Research Electronic Data Capture (REDCap) that included all possible response options for each demographic factor. Children’s Mercy Kansas City, Cleveland Clinic Children’s, and Henry Ford Health pulled EHR reports that matched the REDCap codebook. Data reports were viewed in a secure system at each institution. The study lead at each institution performed a manual review of charts to verify the accuracy of pulled data.

Study data were collected and managed using REDCap electronic data capture tools hosted at Stanford.21, 22 REDCap (Research Electronic Data Capture) is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources.

A data use agreement was completed through Stanford University Office of Research Administration. De-identified pulled chart data were uploaded to and managed through Stanford REDCap by the study lead at each institution. De-identified data for all institutions were then downloaded to a CSV file by the lead author. This study was approved by the IRBs for Stanford University (IRB of record), Children’s Mercy Kansas City, Cleveland Clinic Children’s, and Henry Ford Health. The study was also approved by the Clinical Quality Management Committee of Ravenswood Family Health Center.

2.2. Statistical approach

The methods included a multifaceted statistical approach that includes summary statistics, chi-square tests of independence and log-likelihood ratio tests (G²), and generalized linear model (GLM) contrasts to investigate patterns in adolescents’ motivation towards transitioning to adult care. Summary statistics of importance and confidence responses are presented by demographic and health coverage groupings including age, sex, race, ethnicity, language, and insurance.

P-values from chi-square tests of independence were used to see if relationships existed between the demographic variables and importance and confidence scores. Generalized linear models were used to investigate potentially significant relationships between importance and confidence scores (treated as continuous variables) and the demographic and insurance groups. Because of imbalanced samples and variances, likelihood ratio tests were implemented to measure distances between the observed data and the null expectation - and based on the chi-squared distribution - to gauge if the expected number of observations in each category fits the theoretical expectation. This is used as a goodness-of-fit with GLM models to indicate practical significance/clinical relevance of the outcome. All results were computed using the R language and environment for Statistical Computing²³ and the dplyr package.²⁴

3. Results

Data consist of n = 5112 individuals aged 12–26 years from the four institutions: Children’s Mercy Kansas City (n = 3439), Cleveland Clinic Children’s (n = 117), Henry Ford Health (n = 922), and Ravenswood (n = 634). Table 1 shows demographic and insurance characteristics of the study population. When their respective health systems initiated HCT planning, 11 % were 12–14years, 56 % were between 15 and 17 years, 27 % were 18–20, and 6 % were 21–26 years. Two thirds were female, 62 % were White, 14 % were Black, and for ethnicity, 19 % identified as Hispanic or Latino (Hispanic/Latino). English was the predominant language. More than half were privately insured followed by about a third who were publicly insured.

Table 1.

Study demographic factors.

Participant’s Characteristics	Project Participants
	N = 5112	%
Age
12–14 years	574	11
15–17 years	2859	56
18–20 years	1363	27
21–26 years	316	6
Sex
Female	3043	60
Male	2064	40
Race
Hispanic	792	16
Black or African American	709	14
Native Hawaiian or Pacific Islander	34	<1
White	3175	62
Asian	110	2
American Indian/Alaska Native	28	<1
Multiracial or more than one race	128	3
Other	58	1
Declined/not available/unknown/not reported	67	1
Ethnicity
Hispanic	945	19
Non-Hispanic	4103	80
Declined/Other	58	1
Language
English	4955	97
Spanish	135	3
Insurance
Public/Medicaid	1808	35
Private/Commercial	2900	57
Uninsured/Self pay	127	2
Other	276	5

The proportion of AYA respondents’ mean scores on importance and confidence by each participating site and demographic and insurance characteristics show that confidence scores are generally higher than importance scores and are accompanied by smaller standard deviations, with some notable differences by study site, race, ethnicity, and language (Table 2).

Table 2.

Proportion of AYA respondents’ mean scores on importance and confidence by participating site and demographic and insurance characteristics.

	Importance		Confidence
Demographics	n	Mean (SD)	n	Mean (SD)
Study Sites (N = 5112)	5048		5044
Cleveland Clinic	117	6.17 (3.34)	117	7.74 (2.38)
Children’s Mercy	3434	6.90 (2.99)	3431	7.68 (2.42)
Henry Ford	922	5.52 (3.51)	922	7.01 (2.96)
Ravenswood FHC	575	7.63 (2.51)	574	7.08 (2.40)
Age	5048		5044
12–14	551	6.18 (3.30)	549	6.82 (2.85)
15–17	2820	6.81 (3.02)	2819	7.50 (2.47)
18–20	1361	6.87 (3.10)	1360	7.67 (2.49)
21–26	316	6.17 (3.45)	316	7.84 (2.64)
Sex	5045		5041
Female	3008	6.82 (3.10)	3008	7.48 (2.59)
Male	2037	6.57 (3.12)	2033	7.52 (2.47)
Race	4974		4970
Hispanic	747	7.56 (2.55)	749	7.19 (2.41)
Black	703	6.61 (3.43)	703	7.19 (2.89)
Native Hawaiian PI or Pacific Islander	29	7.93 (2.72)	29	7.10 (2.69)
White	3171	6.55 (3.14)	3166	7.67 (2.46)
Asian	110	6.25 (2.76)	109	7.23 (2.37)
Multiracial (6)	128	7.46 (2.72)	128	7.44 (2.47)
Other	58	6.79 (3.00)	58	7.26 (2.49)
American Indian/Alaska Native	28	6.32 (3.27)	28	8.25 (2.46)
Ethnicity	4986		4982
Hispanic	896	7.40 (2.65)	897	7.19 (2.43)
Non-Hispanic	4090	6.58 (3.18)	4085	7.57 (2.55)
Language	4915		4907
English	4915	6.69 (3.11)	4907	7.50 (2.54)
Spanish	130	7.92 (2.65)	134	7.42 (2.69)
I Insurance	4771		4767
Public	1748	7.04 (2.99)	1743	7.20 (2.63)
Private	2896	6.59 (3.13)	2898	7.69 (2.44)
Uninsured	127	7.11 (3.00)	126	7.54 (2.57)

n = sample; mean = arithmetic mean; SD = standard deviation.

Chi-square tests of independence showed overall low p-values which indicate there are strong relationships between the demographic factors and importance and confidence scores. P-values for importance and confidence for all demographic variables are significant except for sex and language for confidence, indicating that neither of those demographic variables are related to confidence score. Notably, the age demographic variable showed strong associations with importance (X2 = 313, P = .000, G2 = 32.41) and confidence (X2 = 330, P = .000, G2 = 51.33) scores. Also, ethnicity shows strong association with importance (X2 = 543, P = .000, G2 = 56.46) and confidence (X2 = 124, P = .000, G2 = 19.33). Insurance also shows an association with importance (X2 = 460, P = .000, G2 = 47.75 and confidence (X2 = 285, P = .000, G2 = 44.44).

Results demonstrate trends in scores by age prior to initiation of health care transition anticipatory guidance and services, when AYAs were establishing with a HCT program (Fig. 1). Importance scores increased from 12-14 through 18–20 years of age, then decreased in the 21–26-year group. Confidence scores increased from the 12–14 through the 21–26-year age groups.

Fig. 1.

Mean importance and confidence scores by age group.

To better examine the variability of importance and confidence scores by age that could be lost by looking at the mean scores, importance and confidence scores were divided by low (0−3), medium (4−6), and high (7−10) for the four age groups (Table 3). The age groups with the higher proportions of low scores on both importance and confidence were the youngest and, surprisingly, the oldest age groups. Roughly a quarter in each age group scored in the medium range for both questions. More than half in each age group scored high in importance (ranging from 52.8 % to 61.3 %) and confidence (ranging from 59.7 % to 73.1 %).

Table 3.

Proportion of AYAs with low, medium, and high importance and confidence scores by age group.

	Importance		Confidence		Importance Scores %			Confidence Scores %
Age	N 5048	Mean (SD)	N 5044	Mean (SD)	0–3 Low (n = 768)	4–6 Med (n = 1281)	7–10 High (n-2999)	0–3 Low (n = 373)	4–6 medium (n = 1138)	7–10 High (n = 3533)
12–14	551	6.18 (3.30)	549	6.82 (2.85)	20.1 % (111)	27.0 % (149)	52.8 % (291)	11.7 % (64)	28.6 % (157)	59.7 % (328)
15–17	2820	6.81 (3.02)	2819	7.50 (2.47)	14.0 % (394)	25.5 (719)	60.5 % (1707)	6.7 % (190)	22.8 % (644)	70.4 % (1985)
18–20	1361	6.87 (3.10)	1360	7.67 (2.49)	14.5 % (197)	24.2 % (330)	61.3 % (834)	6.8 % (93)	20.4 % (278)	72.7 % (989)
21–26	316	6.17 (3.45)	316	7.84 (2.64)	20.9 % (66)	26.3 % (83)	52.8 % (167)	8.2 %(26)	17.7 % (59)	73.1 % (231)

n = sample; Mean = arithmetic mean; SD = standard deviation.

Table 4 shows the relevant group contrast statistics for GLM models with p-values < 0.05. Statistically significant results for the GLM contrasts in importance and confidence scores between age groups within the independent variable groupings are presented. For importance scores, the 12–14-year-old group and the 21–26-year-old group had the lowest mean scores (Table 2), and each of these youngest and oldest groups demonstrated statistically significant differences (Table 4) in scores compared to the 15–17- and 18–20-year-old groups (mean differences.63–.71, p values.000–.0003). Statistically significant group mean differences were noted between the 12–14-year age group and all other age groups for confidence (mean differences.69–1.02, p = .000).

Table 4.

Generalized linear model contrast results.

		Importance				Confidence
Variable	Comparison	Difference of Group Means	Standard Error	Dfa	Adjusted P Valueb	Difference of Group Means	Standard Error	df	Adjusted P Value
Age (years)	(12−14) - (15−17)	−0.631c	0.144	5044	0.000	−0.687	0.118	5040	0.000
	(12−14) - (18−20)	−0.695	0.157	5044	0.000	−0.848	0.128	5040	0.000
	(12−14) - (21−26)	0.010	0.219	5044	1.000	−1.023	0.178	5040	0.000
	(15−17) - (18−20)	−0.064	0.102	5044	0.925	0.161	0.083	5040	0.218
	(15−17) - (21−26)	0.641	0.184	5044	0.003	−0.335	0.150	5040	0.114
	(18−20) - (21−26)	0.705	0.194	5044	0.002	−0.175	0.158	5040	0.686
Sex	Female-Male	0.257	0.089	5043	0.004	−0.046	0.073	5039	0.524
Race	Hispanic/Latino-White	1.013	0.125	5043	0.000	−0.472	0.103	5028	0.000
	Black/African American-White	0.058	0.129	5032	1.000	−0.478	0.105	5028	0.000
	Hispanic/Latino-Black/African American	0.955	0.162	5032	0.000	0.006	0.133	5028	1.000
	Hispanic/Latino-Asian	1.310	0.315	5032	0.001	−0.036	0.259	5028	1.000
Ethnicity	Hispanic/Latino-Non-Hispanic/Latino	0.816	0.114	5041	0.000	−0.386	0.093	5037	0.000
Language	English-Spanish	−1.227	0.276	5043	0.000	0.078	0.222	5037	0.725
Insurance	Public-Private	0.455	0.094	5043	0.000	−0.488	0.077	5039	0.000

Chi-square (X2) and deviance (G2 likelihood ratio test; minus two times the difference in log-likelihoods) were used for predicting importance and confidence scores via the GLMs, with high test statistics and low p-values indicative of good model fit. For both X2 and G2, p-values < 0.05 indicate good model fits and potentially clinically useful models. All fits were statistically significant except two models for Confidence score: X2 for Sex (x2 = 3, p = 0.083) and X2 for Language (X2 = 1, p = 0.317) both of which also had low G2 scores for Confidence (G2 = 0.41 for Sex and G2 = 0.12 for Language), comparatively small to all other G2 scores ranging between 11.27 and 70.24.

^aDf = degrees of freedom.

^bStatistical significance for p values <.05.

^cThe value in the Difference of Group Means column applies to the Comparison column. For example, in the first row for Age (years), the Difference of Group Means value 0.631 signifies that the 12–14-year-old age group had a mean score 0.631 lower for importance compared to the 15–17-year-old age group.

Additional statistically significant results for mean scores can be found in Table 4. For ethnicity, responses for Hispanic/Latino AYAs demonstrated a statistically significant difference in scores that were higher for importance (.82 mean difference, p = .000) and lower for confidence (.39 mean difference, p = .000) compared to non-Hispanic/non-Latino AYAs. For insurance type, AYAs with public/Medicaid insurance demonstrated statistically significant higher scores for importance (.46 mean difference, p = .000) and lower scores for confidence (.49, p = .000 mean difference) when compared to AYAs with private/commercial insurance. For sex, a small statistically significant difference was noted with females scoring higher for importance than males (.26 mean difference, p = .004)). For race, Hispanic/Latino AYAs scored higher for importance than White AYAs (1.01 mean difference, p = .000), higher than Black/African American AYAs (.96 mean difference, p = .000), and higher than Asian AYAs (1.31 mean difference, p = .001). For confidence, significant differences were noted with White AYAs scoring higher than Hispanic/Latino AYAs (.47 mean difference, p = .000) and higher than Black/African American AYAs (.48 mean difference p = .000).

4. Discussion

4.1. Motivational interviewing and HCT anticipatory guidance

This is the first study to examine initial importance and confidence responses of a large AYA population including those with and without chronic conditions across four diverse health care delivery systems and to compare scores to demographic and insurance factors. Key findings of this study were that 1) mean AYA scores were generally higher for confidence than importance; and 2) age, ethnicity, and type of insurance were associated with ratings for importance and confidence.

Motivational interviewing is an integral part of HCT anticipatory guidance and support. Having a baseline understanding of changes that occur with normal developmental processes as adolescents progress from early to middle and late adolescence and into young adulthood can be helpful to clinicians in both primary and specialty care who are engaging with AYAs in HCT education and counseling. Responses prior to HCT interventions indicated a gradual increase in initial importance scores with age until the 21–26-year group when importance scores declined, similar to the 12–14-year group, and a gradual increase in initial confidence scores with increasing age (Fig. 1). The decrease in self-ratings of importance to prepare for/change to an adult clinician in the 21–26-year age group emphasizes a need to address health care transition optimally at 15–20-years, when initial self-ratings of transition importance are highest. This finding is consistent with professional recommendations.¹ Introducing HCT at age 21or later is problematic for several reasons – not only are young adults less engaged, but they are also more likely to age out of their pediatric practices, leaving clinicians a very short time frame before transfer. In contrast to these recommendations and findings, most recent data of the National Survey of Children’s Health demonstrated in 2021–2022 that 22.1 % of youth with SCHN and 17.8 % without SHCN received HCT preparation before the age of 17 years.⁴ This highlights the need for HCT process improvements at individual practice and larger system levels to facilitate HCT anticipatory guidance for AYAs with and without SHCNs at early to middle adolescence.

This study also demonstrated significant differences in importance and confidence responses with respect to ethnicity. The finding that Hispanic/Latino ethnicity was associated with higher scores for importance and lower scores for confidence may be indicative of the financial, language, and insurance challenges experienced by Hispanic/Latino AYAs disproportionately in the United States (US). Low health literacy has been associated with poverty, limited English language proficiency, and a lack of consistent health insurance among Hispanic adults in CA.²⁵ A study utilizing the U.S. Census Bureau’s Supplemental Poverty Measure data demonstrated that poverty rates were highest in first-generation and second-generation Hispanic children with two foreign born parents, primarily due to low family employment, with differences in parental education being the greatest contributor to intergenerational disparities in poverty. ²⁶ Hispanic/Latino AYAs in the US represent a broad diversity of socioeconomic and documentation statuses, immigration histories, and cultural and linguistic backgrounds.²⁷ This diversity could lead to variations in AYAs’ perspectives of HCT between and within Hispanic/Latino groups.

Similar to results for Hispanic/Latino AYAs, scores for insurance type demonstrated that AYAs with public insurance had higher scores for importance and lower scores for confidence compared to AYAs with private insurance. As previously discussed, research demonstrated that AYAs scored higher for importance if a discussion of insurance had occurred,¹⁷ demonstrating the impact of insurance factors on AYAs receiving HCT anticipatory guidance. Although the Affordable Care Act allowed for AYAs to be covered through 25 years on the parent’s insurance, the parents of many low income AYAs do not have access to employer-sponsored health insurance, and many are uninsured. Lack of stability with health insurance after turning 18 years and challenges of obtaining insurance in mixed immigration families could raise importance responses for AYAs on public insurance.

Another important take away from this study is the small, but important, population of AYAs that score low (0−3) on importance and/or confidence. Identifying this population can assist clinicians with where to begin HCT discussions. This study provides baseline data that can help with interpretation of scores for different age groups. White et al. (2018) recommend an initial focus on the lowest number, especially when there is a discrepancy between the importance and confidence scores.²⁸ If both are low, recommendations are to focus first on the importance score to gain the youth’s engagement in the process.²⁸ Importance has been identified as an initial step to patient activation in which the patient identifies that their role is important, which is followed by having the confidence to manage their health care.¹⁴ Sharing scoring observations with the AYAs can be helpful to gain an understanding of other issues. For example, an AYA might not want to leave the pediatric practice, so they might score low for importance and high for confidence.²⁸. Also, challenges including psychosocial stressors such as anxiety have been shown to affect AYA scoring.¹⁹

4.2. Limitations

Limitations of this retrospective chart review include varied time periods in which each institution initiated their HCT activities. The two-item constructs examined are a piece of much larger transition assessment that is part of HCT planning. Responses to these early motivational interviewing questions help to set the stage for anticipatory guidance. Got Transition’s Transition Readiness Assessment questions that follow the motivational interviewing importance and confidence questions were not part of this study since sites differed in the questions asked. This study focused on the pooled data from the four institutions. Although ratios of institutions relative to Children’s Mercy Kansas City (largest number of participants) are appropriate for predictive value, there was a difference in the number of participants at each site (Table 1).

Challenges were encountered with respect to racial categories, some of which highlight the difficulties in obtaining race-related data, particularly in a retrospective chart review. One of the institutions used an EHR that included Hispanic/Latino as a racial category. This example demonstrates that AYAs can experience challenges selecting a racial category if they need to select “Black” or White”, which has been well-documented with Hispanic/Latino patients.29, 30 Updated race/ethnicity standards for the United States Census 2024 will use one combined question for race and ethnicity and will allow for a single response, such as Hispanic or Latino, or multiple responses as needed (https://www.census.gov/newsroom/blogs/random-samplings/2024/04/updates-race-ethnicity-standards.html).³⁰ These challenges, in combination with differences in group sizes, made race comparisons inconclusive.

4.3. Recommendations for future studies

Potential confounding variables related to health literacy and social determinants of health such as economic stability, education access and quality, and social and community context should be considered in future studies looking at HCT and importance and confidence scoring.

Recommendations for future studies are to investigate relationships between self-efficacy, engagement, and responses to motivational interviewing importance and confidence questions. Studies on engagement have been done primarily with adults. Sebastian et al., 2014 described HCT readiness assessment as a potential way to measure engagement.³¹ Exploring relationships between importance and confidence questions, age, and the overall scores of the skills readiness assessment is recommended in a future prospective study, which would necessitate the use of the same non customizable readiness assessment. To compare importance and confidence scoring of AYAs with and without SHCNs, a prospective study with pre-determined levels of complexity for AYAs with SCHNs would be indicated as there is a broad range of complexity among AYAs with SHCNs. Qualitative studies are also recommended to gain an understanding of perspectives of Hispanic/Latino AYAs in different communities with respect to HCT. Qualitative research could also help to gain a better understanding of an AYA’s thought process when scoring importance and confidence.

5. Conclusion

This study offers baseline scores on two key motivational interviewing questions that can help clinicians learn how engaged the AYA is in discussing the changes needed to move to adult care and guide the clinician to start earlier than just before transfer that often occurs between ages 18 and 21. Mean AYA scores were generally higher for confidence than importance. This study encourages consideration of the associations of age, ethnicity, and type of insurance with ratings of importance and confidence when interpreting scores. Results can assist clinicians to optimally address transition readiness and to provide guidance for AYAs initiating the HCT process.

CRediT authorship contribution statement

Teresa Hickam: Writing – review & editing, Resources, Methodology, Investigation, Conceptualization. Carrie Cuomo: Writing – review & editing, Resources, Methodology, Investigation, Data curation, Conceptualization. Lisa Lestishock: Writing – review & editing, Writing – original draft, Visualization, Supervision, Resources, Project administration, Methodology, Investigation, Data curation, Conceptualization. Patience White: Writing – review & editing, Writing – original draft, Supervision, Methodology, Conceptualization. Margaret McManus: Writing – review & editing, Writing – original draft, Conceptualization. Evan Muzzall: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Methodology, Investigation, Formal analysis, Data curation. Michele Maddux: Writing – review & editing, Resources, Methodology, Investigation, Data curation, Conceptualization. Tisa Johnson-Hooper: Writing – review & editing, Resources, Methodology, Investigation, Data curation, Conceptualization.

Ethical Statement

All co-authors verify that neither this manuscript nor parts of it have been published previously. It is not under consideration for publication elsewhere. The article's publication is approved by all authors and if accepted, the article will not be published elsewhere in the same form, in English or in any other language, including electronically without the written consent of the copyright-holder.

Funding/Financial Statement

Biostatistician support was funded by Got Transition to Evan Muzzall, PhD. No additional financial support was provided for this study.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The authors do not have permission to share data.