Prevalence, predictors, and moderators of transfer in primary care

Abstract

Objective

Health care transition from pediatric to adult primary care is understudied, and factors associated with successful transfer to adult care are not well understood.

Methods

The current study surveyed a sample of patients (N = 110) of transition age (18 – 22, M_age = 19.42, SD_age = 1.16) who attended a pediatric primary care practice. Mixed methods were used to assess the prevalence of successful transfer to adult primary care, factors associated with successful transfer, as well as barriers, facilitators, and recommendations for improving transition support.

Results

45 % of the sample (n = 50) reported attending an adult primary care appointment. Contextual factors assessed (e.g., socio-economic status, disease complexity) and transition constructs (e.g., transition supports, transition behaviors, and transition attitudes) were not associated with transfer. No statistically significant predictors were identified. Exploratory patterns suggested that transition behaviors (e.g., managing medications, making appointments independently) may merit further investigation as a moderator. We report common barriers (e.g., logistics), facilitators (e.g., provider and family support), and recommendations (e.g., earlier transition support).

Discussion

Our preliminary findings suggest that given the association between behaviors and transition readiness, pediatric primary care settings might benefit from using screening to prioritize which patients might benefit from enhanced transition support.

1. Introduction

Health care transition refers to the process of adolescents and young adults (AYAs) moving from the pediatric to the adult health care system. Health care transition includes: 1) preparation of adolescents in the pediatric health care setting as early as 12 years of age; 2) transfer out of the pediatric and into the adult health care setting; and 3) full integration into the adult care setting by 26 years of age.¹ Transition-aged youth (i.e., AYAs between the ages of 12 and 26 years) make up 20 % of the United States population,² yet only 14–17 % of transition-aged youth receive adequate health care transition preparation and support.³

Health care transition is important for all AYAs; however, the majority of health care transition research thus far has focused on AYAs with special health care needs (i.e., chronic health conditions) in specialty care settings. Limited research exists on health care transition for AYAs without special health care needs and/or AYAs in the primary care setting.⁴ It is important for research to begin addressing this gap in the literature as preliminary research5, 6 suggests that AYAs without special health care needs may be at greater risk for inadequate transition readiness compared to AYAs with special health care needs because they see fewer providers less frequently and transition supports are offered at a lower rate in primary care compare to specialty care.

1.1. Health care transition in primary care

Pediatric primary care provides patient- and family-centered continuous, comprehensive, and coordinated preventative, acute, and chronic health care services for youth from infancy to young adulthood.⁷ Thus, health care providers in pediatric primary care settings are in a unique position to build a foundation for successful health care transitions. Practice recommendations for successful health care transitions for AYAs with and without special health care needs exist; however, many transition guidelines are limited in their acceptability and sustainability.⁸ Got Transition recommends six core elements of successful health care transition including: (1) Transition and Care Policy/Guide, (2) Tracking and Monitoring, (3) Transition Readiness, (4) Transition Planning, (5) Transfer of Care, and (6) Transfer Completion that allow for the health care transition process to be individually tailored to AYAs’ specific health needs.¹ Despite these and other existing practice recommendations, there is limited consensus on how to sufficiently support the health care transition in the pediatric primary care setting for AYAs with1, 9 and without special health care needs.¹⁰ Further, research suggests that existing health care transition practice recommendations are followed less than 25 % of the time in the pediatric primary care setting.¹¹

1.2. Barriers and facilitators of successful health care transitions

The Social-ecological Model of AYA Readiness for Transition (SMART) describes the complex relationships among AYAs, caregivers, and health care providers involved in optimizing AYAs’ ability to manage their health independently, while providing supports and interventions as appropriate.¹² Both pediatric and adult health care providers play a critical role in supporting AYAs and their families upon the initial navigation of the adult health care system transition process. Unsuccessful transitions may be a product of unrealistic self-management expectations¹³ and variability in the complexity of the health condition,¹ while other transition difficulties may be more reflective of barriers within the health care system (e.g., limited time to discuss the transition during visits, tracking who has received information, support to ensure successful transfer of records to the adult team) and broader systemic barriers (e.g., regional location, general accessibility to care, health insurance coverage, and racism).¹⁴ Specific challenges within adult care facilities include ensuring that the transferred patient understands their rights in the medical decision-making process, promoting independent self-management, reviewing confidentiality policies, and outlining procedures on how the newly transitioned patient can access resources outside of their routine visits. Other studies suggest problems with disjointed communication about AYA health care transition services, limited collaboration between pediatric and adult health care systems, and even a lack of providers available to ensure a successful transition.1, 13 This is problematic, as providers themselves have identified that the maintenance of care continuity is an imperative component of health care transition success.¹⁵

Although there are significant barriers, facilitators of successful health care transitions have also been identified. Specifically, assistance from AYAs’ pediatric providers to develop a plan to address all relevant health care needs beyond physiological needs (e.g., emotional, social), as well as teaching skills for independent disease management.¹⁵ Successful adult health care transitions also involve providers who delegate self-management responsibilities that are consistent with the AYA’s chronological and developmental age, prepare to discuss common challenges for the given health condition, promote the AYA’s confidence in personal responsibility and self-efficacy, individualize the transfer of care, and connect families with a care transfer coordinator to ensure an effective transfer of care for the AYA.¹⁶ Within the modern world, primary care providers are well-equipped to bridge the gaps in the pediatric to adult health care transition by using technology.¹⁴ Electronic medical records, for example, may enhance the communication and transfer of health care information between providers, as well as provide a record of the patient’s transition readiness, as they achieve greater degrees of self-management prior to their transfer to an adult provider. Additionally, clinical decision support systems may help providers in making choices for care based on patients’ transition readiness, and using telehealth visits may help break down barriers to care, such as transportation.

1.3. The current study

There are a multitude of barriers and facilitators that impact the likelihood of successful transitions from pediatric to adult health care, and it is important to understand whether these barriers and facilitators previously identified in the literature are applicable across primary care facilities. Therefore, the primary aim of the present study was to identify variables associated with AYAs’ health care transition status. To do this, we: (1) identified the prevalence of AYAs who successfully transferred to adult care, (2) assessed how transition support recommendations are associated with successful transition, (3) evaluated how medical complexity, socio-economic status (SES), and transition readiness impact this association, and (4) identified barriers and facilitators to successful transfer and report patient-identified recommendations for transition support.

2. Methods

2.1. Participants

Participants were recruited from a dataset generated from a large, Midwestern, urban pediatric hospital’s electronic health record (EHR). Patients were included in the dataset if they were ages 18–22 years old and attended an appointment at any of three academic pediatric primary care offices between January 2021–December 2023. Details of this sample can be found in Murphy et al. (under review). Largely, patients of these practices are 73.82 % Black/African American, 17.57 % White, and 8.61 % Other in terms of race, and 90.47 % Non-Hispanic, 8.40 % Hispanic in terms of ethnicity, and 80 % publicly insured. Participants were excluded if they were currently a patient in the pediatric hospital’s Adolescent Medicine clinic (transfer age is 22 in this clinic versus 18 in pediatric primary care clinics) or if they had a disability (e.g., intellectual disability, blindness, receptive speech disorders) or circumstance (e.g., incarceration, unable to complete a survey in English) that prevented them from being able to complete an online survey independently. Low health literacy was not an exclusion criterion unless general literacy precluded completion of the survey.

2.2. Procedures

Study team members contacted participants identified in the EHR dataset by phone, text, or email to determine interest and direct them to the e-consent. The survey was administered via RedCap,¹⁷ a secure online survey platform. Individual links for the survey were sent to all potential participants who were contacted unless the patient or their parent specifically declined. The survey took 10–15 min to complete, and participants were compensated with a gift card by mail. All procedures were approved by the hospital's Institutional Review Board.

2.3. Measures

2.3.1. Demographics

Participants completed a demographic survey, which included age, gender, race, ethnicity, zip code, and self-report of chronic disease(s). These data were used descriptively and assessed as potential covariates. Report of diseases were recoded into three groups: no chronic disease, one chronic disease, and two or more chronic diseases.¹⁸

Participants also reported their zip code, which the authors then converted to a Social Deprivation Index (SDI)¹⁹ quintile as a proxy for socio-economic status (SES). The SDI is a measure of neighborhood disadvantage based on seven categories associated with socio-economic variation in health outcomes, including housing, employment, education, and vehicle ownership. Zip codes are assigned an SDI score from 0 to 100, and these scores were converted to quintiles (a score of 1–5) for the purpose of analysis. Scores correspond with the following levels of development: 1 = high level of development, 2 = middle-high, 3 = middle, 4 = low-middle, and 5 = low levels of development.

2.3.2. Transfer questionnaire

The de novo created “Transfer Questionnaire” was designed for the current study by authors BM, AL, and LC based on tools available as a part of Got TransitionTM.²⁰ The survey asked participants if they have attended a visit with an adult provider, the time in months between their last pediatric appointment and first adult appointment, follow-up questions about their visit (e.g., who scheduled it, who attended), and barriers and facilitators to transfer. Items included both open-ended and multiple-choice responses.

Attending an adult appointment was coded as yes (1) or no (0), and those who selected yes reflect a successful transfer. Time to first adult appointment was measured in months. Participants provided text responses for six open-ended questions regarding barriers and facilitators. Text responses were coding using partial thematic analysis (see details below). The Transfer Questionnaire is included as a supplemental material.

2.3.3. Transition healthcare satisfaction

Elements of the Got TransitionTM protocol²⁰ were measured using the Transition Healthcare Satisfaction questionnaire. The current study used six items that corresponded with the transition protocol at the institution conducting the study. Participants responded yes or no to whether their pediatric provider completed recommendations (e.g., “Explain the transition process in a way you could understand”). After confirming internal consistency was adequate (alpha = .813), “Yes” responses were totaled, and the sum was used in analyses. Higher scores indicated higher exposure to transition supports in the pediatric setting.

2.3.4. Transition readiness

The Self-management in Adulthood with Rx²¹ (STARx) is an 18-item measure used to assess readiness behaviors and attitudes in five areas: medication management, provider communication, engagement during appointments, adult health responsibilities, and resource utilization. Because our sample was from the general population, we excluded three disease-specific questions. The remaining items comprise two validated subscales from the measure: Transition Attitudes and Transition Behaviors. These subscales are described below and analyzed separately. Following the reverse scoring of two items, all items were summed to create a total score. Internal consistency was good for the current sample (alpha = .833).

Nine items measured behaviors in the last three months (e.g., “How often did you take your medicines on your own?”), and participants responded on a 5-point Likert scale of Never (scored as 1) to Always (scored as 5). They could also rate “Not needed for my care” (scored as 0 per STARx scoring instructions). Two items were reverse scored, then responses were summed. Higher scores indicated more readiness to complete healthcare transition behaviors. Internal consistency for these 9 items was acceptable (alpha = .791).

Six items measured attitudes about healthcare transition readiness. Participants responded on a 5-point Likert scale (1 = Very Hard to 5 = Very Easy) about how hard it is for them to engage in healthcare transition behaviors (e.g., “How easy is it for you to talk to your doctor?”) in the past three months. Responses were summed, and higher scores denoted more favorable attitudes towards transition. Internal consistency for this subscale was also adequate (alpha = .805).

2.4. Coding text responses

Five coders analyzed the free-text responses to each open-ended question to identify and quantify codes. Coders included a licensed pediatric psychologist (n = 1), clinical psychology doctoral students (n = 3), and a medical student (n = 1). A participant’s text response to each open-ended question was analyzed in isolation from the text responses to the other open-ended questions since many respondents did not provide a text response to one or more of the open-ended questions and text responses were abbreviated (i.e., typically ranged from 1 word to 2 sentences in length). All coders received training in the coding process, which combined principles of Delphi coding²² and thematic analysis.²³ First, coders independently reviewed text responses to identify and operationalize codes for the data. Codes and operational definitions were discussed and finalized as a group. Then, coders independently coded the responses according to the agreed upon operational definitions. Finally, independent coding sheets were aggregated to reach consensus. Any responses that did not meet 100 % agreement were discussed as a group to reach consensus (i.e., at least 80 % agreement). Each individual response could be coded into 1 or more codes, thus, prevalence of codes per open-ended question do not always sum to 100 %.

2.5. Data analyses

We conducted analyses in Stata (StataCorp). Missing data were handled using multiple imputation by chained equations (MICE) with 20 imputations. Continuous variables (i.e., Healthcare Transition Satisfaction total, STARx subscales, SDI quintile, days since last pediatric visit) were imputed via linear regression; comorbidity was imputed using multinomial logistic regression and specialist status using logistic regression. The binary outcome (successful transfer to adult care; 0/1) was included as a predictor in the imputation models but was not imputed given there was no missing data on this variable. Estimates were combined using Rubin’s rules, and robust (Huber–White) standard errors are reported.

Primary models were hierarchical logistic regressions predicting transfer: (1) Satisfaction-only; (2) Satisfaction × transition readiness (tested separately with STARx behaviors and STARx attitudes); and (3) Covariate-adjusted versions adding SDI quintile, comorbidity, specialist status, and days since last pediatric visit. For moderation, we derived marginal predictions to illustrate the Satisfaction slope at representative STARx values (10, 20, 30, 40) on an odds-ratio scale (reference line at OR=1). Model fit was summarized with McFadden’s pseudo-R² and change in pseudo-R² (ΔR²) across steps.

3. Results

3.1. Participants

Out of 2203 AYAs in the EHR dataset, the study team contacted 1133 potential participants, who were determined to be representative of the entire dataset in terms of race, gender, age, and zip code. 206 (18.2 %) completed the eligibility screening form, and 110 (9.7 % of eligible participants that were contacted) provided data that were included in the analyses. See Fig. 1 for the study flow chart.

Fig. 1.

Study flow chart.

Participants in the final sample (N = 110) were predominantly female (n = 62, 56.4 %), Black or African American (n = 79, 71.8 %), and Non-Hispanic/Latine (n = 107, 97.3 %). Ages ranged from 18 to 22 years old (M = 19.42, SD = 1.16), and the mean SDI was 3.69 (SD = 1.32) which corresponds to middle to middle-high neighborhood distress. See Table 1 for full demographic information.

Table 1.

Demographics.

Demographic	M (SD) /n(%)
Age	19.42 (1.16)
Gender
Male	49 (44.5 %)
Female	60 (54.5 %)
Nonbinary	1 (1 %)
Race
White	25 (22.7 %)
Black/African American	78 (70.9 %)
Middle Eastern	1 (0.9 %)
Self-describe / Prefer not to Say	6 (5.4 %)
Ethnicity
Hispanic / Latine	3 (2.7 %)
Not Hispanic / Latine	90 (81.8)
Self-describe / Prefer not to Say	7 (6.3 %)
Social Deprivation Index Quintile (n = 103)	3.69 (1.32)

Note. N = 110 unless otherwise noted.

3.2. Prevalence of successful transfer

Of the 110 survey participants, 50 (45 %) reported they have attended a visit with an adult provider. The mean time to first appointment was 12.56 months (range 1–60 months, SD = 12.52).

Twenty-eight participants (25 %) provided a response to indicate why they were not connected with an adult provider. The most commonly reported reasons for not being connected with an adult provider included: they were still trying to find an adult provider to schedule with (n = 8), they were still under the care of a pediatric provider (n = 2), they experienced barriers to transition (e.g., insurance barriers, transportation difficulties; n = 8), and they did not perceive a need to schedule with an adult provider (e.g., no immediate need or being generally healthy; n = 5).

3.3. Contextual predictors

In the hierarchical logistic regression model assessing potential predictors of transition, the following variables were not significantly associated with transfer: Healthcare Transition Satisfaction, SDI quintile, comorbidity, having a specialty physician, transition behaviors, and transition attitudes (Satisfaction OR = 1.08, 95 % CI [0.86, 1.35], p = .50; STARx behaviors OR = 1.05, 95 % CI [0.96, 1.14], p = .28; STARx attitudes OR = 0.95, 95 % CI [0.86, 1.05], p = .33). Days between the last pediatric appointment and survey completion was included as a covariate but was not significant. See Table 2.

Table 2.

Logistic Regression Analyses.

Predictor	OR	SE	p	95 % CI
Satisfaction	1.08	0.12	0.50	0.86 – 1.35
Transition Attitudes	0.95	0.05	0.33	0.86 – 1.05
Transition Behaviors	1.05	0.04	0.28	0.96 – 1.14
SDI Quintile	0.99	0.16	0.93	0.72 – 1.35
Comorbidity	1.19	0.34	0.55	0.68 – 2.07
Specialist	1.29	0.62	0.60	0.50 – 3.28
Days since last pediatric visit	1.00	0.00	0.12	1.00 – 1.00

Note. Dependent variable = successful transfer to adult care. OR = odds ratio. CI = confidence interval. SDI = social deprivation index, a proxy for socio-economic status. SE = standard error.

3.4. Transition as a moderator

3.4.1. Satisfaction-only model

Satisfaction was not significantly associated with transfer, OR = 1.10, 95 % CI = 0.90, 1.35, p = 0.36.

3.4.2. Moderation by STARx behaviors

In the unadjusted interaction model, higher Satisfaction (OR = 2.79, 95 % CI = 0.79, 9.88, p = 0.11) and higher STARx behaviors (OR = 1.17, 95 % CI = 0.99, 1.37, p = 0.07) were associated with higher odds of transfer in magnitude, though not statistically significant (see Table 3). The Satisfaction × Behaviors interaction was not significant (OR = 0.97, 95 % CI = 0.93, 1.01, p = 0.14), with a small improvement over the Satisfaction-only model (Δ pseudo-R² = 0.04). In the covariate-adjusted model, estimates were similar. Additional fit gain was small (Δ pseudo-R²=0.02). See Fig. 2.

Table 3.

Logistic Regressions for Moderators.

Predictor	OR	SE	p	95 % CI
Transition Behaviors as a Moderator:
Step 1
Satisfaction	2.79	1.79	0.11	0.785 – 9.88
Transition Behaviors	1.17	0.10	0.07	0.99 – 1.37
Step 2
Satisfaction x TB	0.97	0.02	0.14	0.93 – 1.01
Transition Attitudes as a Moderator
Step 1
Satisfaction	1.01	0.54	0.98	0.36 – 2.88
Transition Attitudes	0.96	0.09	0.68	0.80 – 1.16
Step 2
Satisfaction x TA	1.00	0.02	0.86	0.96 – 1.05

Note. Dependent variable = successful transfer to adult care. TB = transition behaviors. TA = transition attitudes. SE = standard error.

Fig. 2.

Note. The red line is odds ratio = 1.00. Below the line is lower odds of transfer; above the line is higher odds of transfer.

3.4.3. Moderation by STARx attitudes

In the unadjusted interaction model, results were largely null: Satisfaction OR = 1.01, 95 % CI = 0.36, 2.88, p = 0.98), STARx attitudes OR = 0.96, 95 % CI = 0.80, 1.16, p = 0.68), interaction OR = 1.00, 95 % CI = 0.96, 1.05, p = 0.86); Δ pseudo-R² < 0.01. Covariate-adjusted models showed a similar pattern with a Δ pseudo-R² = 0.03.

3.5. Understanding of transition, facilitators, barriers, and recommendations

3.5.1 Understanding of Transition. Eighty-five participants (77 %) provided responses regarding their understanding of the key differences between pediatric and adult health care. Participants identified “age” (36.5 %, n = 31), meaning they indicated that pediatric health care primarily works with children and adult health care works with adults. Participants also identified differences regarding the scope of practice (12.9 %, n = 11), quality of communication (12.9 %, n = 11), and the degree of independence (11.8 %, n = 10). A few individuals mentioned differences in the cost of services provided (3.5 %, n = 3). Although many participants displayed some understanding of the differences between pediatric health care and adult health care, many indicated that they don’t know the difference between the two (17.6 %, n = 15) and some indicated that there were no differences between pediatric health care and adult health care (5.9 %, n = 5). AYAs’ responses also commonly fell into other (18.8 %, n = 16), which reflects responses that did not fall into the categories previously described, such as those that broadly acknowledged that there was a difference, but did not specify what the difference was and those that commented on their own developmental place in health care (e.g., “I’m an adult now”). Refer to Table 4 for a summary of responses.

Table 4.

Understanding the differences between pediatric and adult health care.

Code	Operational Definition	Exemplar Quote(s)	n	%
None	No differences between pediatric and adult health care	“Nothing”; “None”	5	5.9 %
Don’t Know	Not sure of the differences between pediatric and adult health care	“I do not know of any difference”; “I’m not sure”	15	17.6 %
Age	The age of the patients is different	“that one is for kids and one is for adults”; “my understanding[ing] is 1 clinic is 18 and under and the adult is fo[r] 18 + ”	31	36.5 %
Scope	The scope of care, services provided, and/or content discussed are different	“Pediatric Healthcare: Focuses on growth, development, and family-centered care…Adult Healthcare: …The focus is more on maintaining health, managing conditions, and preventing complications”	11	12.9 %
Communication	The way in which services are delivered	“Pediatric healthcare is more personal and they take their time where an adult doctor can rush through the exam”	11	12.9 %
Cost	Differences in expenses / cost of services between the two settings	“It’s more expensive….”	3	3.5 %
Independence	Greater independence is expected in adult health care compared to pediatric health care	“I have to advocate for myself rather than my mom doing it”; “becoming independent”	10	11.8 %
Other	Did not fit into above categories or did not answer the question	“Everything”; “both follow you through that stage of life”	16	18.8 %

Note. N = 85 participants provided responses to this free response item.

3.5.1. Barriers

Eighty-seven participants (79 %) provided a response regarding barriers to health care transition. The most commonly reported barrier to healthcare transition was logistics, such as finding a provider, scheduling an appointment, etc. (34 %, n = 30). Access (17 %, n = 15) and lack of information (15 %, n = 13) were also identified. Other barriers identified included lack of support from providers, family members, or others (8 %; n = 7), discomfort (8 %; n = 7), and fear of transitioning (3 %; n = 3). Notably, 3 % (n = 3) had not transitioned yet and 24 % (n = 21) of respondents reported no barriers to transition. See Table 5.

Table 5.

Barriers to transitioning from pediatric to adult health care.

Code	Operational Definition	Exemplar Quote(s)	n	%
None	No barriers identified	“there weren’t any”; “no barriers”	21	24 %
Not Yet Transitioned	Have not yet transitioned to adult health care	“haven’t yet”; “No, have not transferred”	3	3 %
Lack of Info	Did not know where to start to make the transition and lacked information on how to go about transitioning	“lack of knowledge”; “Asking the right questions”	13	15 %
Logistics	Difficulties navigating the logistics of health care transition (e.g., finding a doctor who accepts their insurance and/or has availability)	“Trying to find the right doctor and filling out paperwork”; “PCPs not accepting new patients”	30	34 %
Access	Geographical location, cost, and/or insurance posed barriers	“Transportation”; “Trying to [stay] health[y] cause I don’t have insurance”	15	17 %
Lack of Support	Difficulties independently managing their health and navigating the transition to adult health care without the support of other s(e.g., parents, providers)	“Lack of communication and coordination”	7	8 %
Discomfort	Discomfort with leaving their pediatric provider who they had a positive relationship with and establishing a positive relationship with a new / adult provider	“Finding a doctor that I can trust and will be comfortable with”; “They weren’t as easy to talk to”	7	8 %
Fear	Expressed fear or being scared of the transition to adult health care	“being scared”	3	3 %
Other	Did not fit into the above categories or did not answer the question	“being a adult”; “Got older”; “don’t know”	12	14 %

Note. N = 87 participants provided responses to this free response item.

3.5.2. Facilitators

Eight-six participants (78 %) provided a response regarding facilitators to health care transition. Participants identified support from the healthcare team (18.6 %, n = 16) and support from family (11.6 %, n = 10) as two primary facilitators for healthcare transition. Positive interactions with providers (7 %, n = 6) and accessibility of services (7 %, n = 6) were also identified, with some (7 %, n = 6) noting that the transition was easy, and therefore facilitators were not needed. A few (n = 7; 19 %) indicated that they had not yet transitioned yet and therefore did not identify any facilitators. The most common coded response was “other,” which included a variety of responses including “I don’t know,” references to the transition being difficult, and reports of no change. See Table 6.

Table 6.

Facilitators in making the transition to adult healthcare.

Code	Operational Definitions	Exemplar Quotes	n	%
None	Nothing was helpful	“N/A”; “I had not help really. I’ve had to figure it out on my own”	16	18.6 %
Not Yet Transitioned	They have not yet made the transition to adult healthcare	“I have not made the transition to an adult provider”	7	8.1 %
Ease of Transition	Nothing was most helpful because the transition was incredibly easy	“It was easy”; “It was easy to find a doctor”	6	7.0 %
Accessibility	Ease of access to providers and/or clinics	“Having a doctors office near my area and being able to get attended fast”; “How easy it is to make an appointment”	6	7.0 %
Independence	It was helpful that they took charge of their health and had the knowledge and skills necessary to navigate the transition independently	“being able to go to the doctor alone”; “My parents preparing me for it as I got older. I’d make my own appointments, talk to the dr myself, pick up my own prescriptions, etc.”	4	4.7 %
Positive Interactions	They felt heard and validated by providers	“That I feel more heard about my concerns”; “having a nice doctor”	6	7.0 %
Family Support	They received support from family members	“Family connections”; “My mom explaining it to me”	10	11.6 %
Healthcare Support	They received informational and/or physical supports from providers and/or the healthcare system	“Referrals”; “My insurance card contains a website that providers comprehensive information about various doctors, enabling me to conveniently locate and select the most suitable healthcare professional for my needs.”	16	18.6 %
Other	Did not fit into above categories or did not answer the question	“Not much of a change to be honest”; “get all information I need”	23	26.7 %

Note. N = 86 participants provided responses to this free response item.

3.5.3. Recommendations

Seventy-five participants (68 %) provided a response regarding recommendations for their health care providers to support the transition to adult health care. The majority of AYAs (66.7 %, n = 50) did not have any suggestions for improving the transition from pediatric health care to adult health care. Of those who provided recommendations, participants suggested that if would be helpful if providers provided them with more information about the health care transition process (13.3 %, n = 10), if providers began preparing patients earlier in the transition process (9.3 %, n = 7), and if providers implemented more patient-centered care practices (9.3 %, n = 7). Other comments (13.3 %, n = 10) that did not fit within the previously discussed categories included ensuring that providers are available. Some participants did not have any suggestions, noting they were fearful about the transition. See Table 7.

Table 7.

Recommendations.

Code	Operational Definition	Exemplar Quote(s)	n	%
None	No recommendations provided	“No”; “All the doctors and nurses were good and respectful and always answered my question”	50	66.7 %
More Info	Encouraged providers to give more information and/or resources	“make a pamphlet or paper giving details about how to change healthcare providers and what to expect”; “It would help to know who I should move to”	10	13.3 %
Prepare	Encouraged providers to begin preparing patients earlier for transition and to revisit the discussions at each visit	“Start having these discussions at least 12 months before adolescent meets the age they are suggested to begin seeing adult healthcare”; “Start talking about earlier”	7	9.3 %
Patient Centered	Encouraged providers to make attempts to understand and listen to the patient and focus on building the patient’s confidence	“just hearing my concerns about going to difference places for care vs one bldg”; “They couldn’t just been more supporting and guiding to me”	7	9.3 %
Other	Did not fit into above categories or did not answer the question	“I do not know”; “Prefer not to answer”	10	13.3 %

Note. N = 75 participants provided responses to this free response item.

4. Discussion

Transition for AYAs moving from pediatric to adult primary care is largely understudied, thus this study aimed to report the prevalence and experiences of transfer, assess factors associated with successful transfer, and report patient-identified barriers, facilitators, and recommendations. In our sample, nearly half of the participants successfully transferred to adult care. According to the SMART model,¹² contextual factors can impact transition, however SES, medical complexity, race, and gender were not significantly associated with transfer in the current sample. Qualitative findings revealed that logistics and access were primary barriers and family support, such as help from parents, and support from the healthcare team were primary facilitators.

Though not reaching statistical significance, transition behaviors emerged as a potentially clinically relevant moderator. For patients reporting fewer transition behaviors (lower transition readiness), such as managing medication, the association between the reported support received at their pediatric appointments and successful transfer appeared stronger than for patients reporting more transition behaviors. Although nonsignificant, this exploratory pattern is consistent with prior recommendations and ongoing assessment of transition readiness and using that information to inform and prioritize transition supports for patients reporting lower readiness. It also supports the six elements staged approach to transition over the adolescent period, with repeat assessments essential to building more transition behaviors.

These conclusions are echoed in the coding of free response items, as support from the healthcare team was identified as a primary facilitator for successful transfer. Additionally, patient-identified barriers such as logistics, access, and lack of information can be addressed at pediatric appointments if the healthcare team is aware of the patient’s specific needs, which may be identified with appropriate, repetitious, and timely assessment of transition readiness. Adult providers can also work with pediatric providers to share information on access, capacity, and support services available. Our finding is commensurate with established guidelines, such as Got Transition,²⁰ which recommends beginning transition support, including readiness assessment, at age 12.

Notably, the pattern with transition behaviors did not emerge with transition attitudes, or the difficulty level of transition activities (e.g., how easy it is to schedule appointments, manage medications). Therefore, these findings indicate that when assessing transition readiness knowledge and skills in this context, measuring behaviors may be more salient than measuring attitudes. Over the past few decades, the study of healthcare transition is fairly new relative to other areas of medicine.²⁴ Therefore, researchers are still working to identify the predictors and outcomes that are the most effective in promoting healthcare transition.²⁵ Pinpointing these constructs could lead to more parsimonious and better fitting models to define transition success and understand factors associated with successful transition. Our findings indicate that models may benefit from behavioral measurement.

Additional strengths of this study include qualitative comments from AYAs specifically offering recommendations to improve the transition experience. The voice of the patient is often under-reported in policy recommendations, and incorporating direct lived experiences from AYA who have successfully transitioned, as well as those still attempted to connect to adult care, is imperative. These learnings from AYAs can help inform more targeted interventions as well as illuminate the priorities and/or reasons for seeking future healthcare. For example, our participants highlighted family support as a facilitator in addition to support from the health care team, which may be useful information for knowing who to include when developing a transition protocol.

4.1. Limitations and future directions

Across all models, effects were small-to-moderate in magnitude with wide confidence intervals and no statistically significant associations. The behaviors specification showed the most suggestive (but nonsignificant) pattern and conclusions should remain cautious given the sample size and event distribution. Although these moderation patterns did not reach statistical significance, they may be useful as hypothesis-generating signals for future research and should be interpreted cautiously alongside the qualitative findings.

Indeed, the most significant limitation impacting the findings was the sample size, as our response rate was lower than anticipated. Additionally, the contextual constructs (e.g., SES, race, gender) that have been found to be associated with successful transfer in other studies²⁶ were not significant in our sample. This could be attributed to homogeneity in the sample, but the small sample size also contributed to wide confidence intervals and reduced certainty in the estimates. Homogeneity in the sample may limit generalizability to other pediatric practices. Alternatively, unmeasured contextual factors may be driving differences in transfer, thus our models were unable to capture which factors are most salient. Future studies may assess the current model with a larger sample size of patients that represent multiple settings (e.g., rural, suburban).

Another limitation was use of retrospective self-report using some measures (i.e., Transition Checklist) with limited psychometric information available. Participants were asked to report their memory of their appointments, which may be subject to inaccuracy or social desirability bias. Though this method is often used to reflect the parts of transition support that are internalized by patients,²⁷ it may not represent a fully accurate account of the totality of the transition support provided (e.g., patient may not remember all supports they received). Furthermore, the measure of transition behaviors asked the participants to report on the past three months. Future studies that utilize a prospective versus a retrospective approach would be able to assess factors that are predictive of transfer and transition.

Though the psychometric properties of the measures are potentially limiting, we took measures to mitigate this where possible. For example, we used established subscales for the STARx, and each of the subscales achieved adequate internal consistency. The Transition Checklist also had good internal consistency in our sample, and the measure is highly face valid. However, future studies are still needed to formally validate the transition checklist.

Our study aimed to focus on patient perspectives of transition and transfer. However, these milestones are housed within larger systems, including providers, parents, and practices. The current study did not capture these broader perspectives. Our findings (e.g., parental support being identified as a facilitator) indicate these voices may be important to understanding the systems impacting transition and transfer more broadly. Future studies focusing on or including these perspectives will be valuable to the field.

Finally, our study is intentionally focused on traditional primary care in an office-based medical home. However, there are other ways that AYAs seek medical care (e.g., on-demand care, care when you need it, telehealth), and future studies are needed to understand trends of where AYA are seeking care, especially for traditional primary care services, such as vaccines, screenings, and preventative care.

5. Conclusions

Health care transition for healthy AYAs is understudied in primary care. In this survey of AYAs, nearly half (45 %) reported a successful transfer to adult care. An exploratory pattern suggested that transition behaviors (e.g., managing medications, making appointments) may be a more clinically informative correlate of transfer status than attitudes. Qualitative responses converged on provider and family support as key facilitators and logistics/access/information gaps as common barriers. Participants recommended clearer information and earlier preparation. Taken together, our results align with existing guidance to assess transition readiness early, routinely, and repeatedly, with particular attention to patients reporting lower readiness behaviors. Larger, prospective studies in diverse settings are needed to test these signals and clarify who most benefits from enhanced supports.

CRediT authorship contribution statement

Constance A. Mara: Writing – review & editing, Writing – original draft, Methodology, Formal analysis, Data curation. Amy C. Lang: Writing – original draft, Methodology, Data curation, Conceptualization. Bridget N. Murphy: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Conceptualization. Lori E. Crosby: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Margaret N. Jones: Writing – review & editing, Conceptualization. Mary Carol Burkhardt: Writing – review & editing, Supervision, Methodology, Conceptualization. Thomas Elliott: Writing – review & editing, Conceptualization. Elysia M. Smith: Writing – review & editing, Project administration. Jessica T. Hinojosa: Writing – review & editing, Writing – original draft, Formal analysis, Data curation.

Ethics Statement

Ethical approval for this study/case/case series was obtained from Cincinnati Children’s Hospital Medical Center’s Institutional Review Board (ID: 2024–0333).

Written informed consent was obtained for anonymized patient information to be published in this article.

Funding

The project described was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health (NIH), under Award Number 1UM1TR005265 and the NIH Clinical and Translational Science Award (CTSA) program, grant UL1TR001425. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Lori Crosby reports financial support was provided by National Institutes of Health. Lori Crosby reports a relationship with American Society of Hematology that includes: funding grants. Lori Crosby reports a relationship with Health Resources and Services Administration that includes: funding grants. Lori Crosby reports a relationship with The Foundation for Sickle Cell Disease Research that includes: funding grants. Lori Crosby reports a relationship with Pfizer that includes: funding grants. Lori Crosby reports a relationship with Forma Therapeutics Inc that includes: consulting or advisory. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

Supplementary material

Data availability

Data will be made available on request.