Measuring transition success from paediatric to adult epilepsy services and its association with seizure outcomes: a UK retrospective cohort study Authors: Shivana Neeta Modi, Emmanouil Bagkeris, Catalina Suarez-Rivera, Maria-Christina Petropoulos, Dougal Hargreaves Published: 2026 ABSTRACT AbstractObjectivesTo develop a working definition of successful transition from paediatric to adult epilepsy services, assess transition success, identify predictors and determine whether success is associated with improved seizure outcomes in early adulthood.DesignRetrospective cohort study.SettingPaediatric and adult epilepsy services in London, UK.ParticipantsPatients aged 20–24 with epilepsy in the paediatric service database (2009–2023) who transitioned to adult care.DemographicsDate of birth, sex, ethnicity.ExposuresAge at first documented paediatric epilepsy clinical appointment (this is not age of epilepsy onset) and number of epilepsy-related comorbidities at age 17.Transition measureSuccess was defined by four criteria: developmentally appropriate initiation, comprehensive preparation using National Institute for Health and Care Excellence guidance, attendance at joint paediatric/adult transition clinic and adult follow-up within 6 months of transition clinic.Main outcome measureChange in seizure frequency between ages 17 and 20.ResultsOf 59 eligible patients, 72.9% (43/59) achieved successful transition. Criteria were well-met: initiation (91.5%), preparation (91.5%), joint clinic (79.7%) and follow-up (88.1%). Successful transition increased the odds of seizure improvement/remaining seizure-free (adjusted OR 4.28, 95% CI 1.07 to 17.19) after adjusting for: age at first paediatric epilepsy clinical appointment, sex, ethnicity and comorbidities. Males had higher odds of seizure improvement/remaining seizure free than females (adjusted OR 4.29, 95% CI 1.26 to 14.67).No significant association was seen with ethnicity or age at first appointment.ConclusionsA novel transition success tool was developed. Larger, multicentre studies are needed to validate this tool and confirm associations with long-term outcomes. WHAT IS ALREADY KNOWN ON THIS TOPIC Despite National Institute for Health and Care Excellence guidance, epilepsy transition services remain inconsistent and their impact on clinical outcomes has not been investigated. It is important to explore factors contributing to a successful transition, as well as the positive impacts of a smooth transition to encourage practitioners to make service improvements. Despite National Institute for Health and Care Excellence guidance, epilepsy transition services remain inconsistent and their impact on clinical outcomes has not been investigated. It is important to explore factors contributing to a successful transition, as well as the positive impacts of a smooth transition to encourage practitioners to make service improvements. WHAT THIS STUDY ADDS This study is the first to propose a novel definition of transition success incorporating four measurable criteria and we use it to evaluate the University College London Hospitals-National Hospital for Neurology and Neurosurgery pathway. Successful transition was achieved in 72.9% of the cohort and was associated with greater likelihood of improved seizure control/remaining seizure-free 3 years post-transfer. However, this association should be interpreted cautiously in light of the study’s retrospective design and limitations. This study is the first to propose a novel definition of transition success incorporating four measurable criteria and we use it to evaluate the University College London Hospitals-National Hospital for Neurology and Neurosurgery pathway. Successful transition was achieved in 72.9% of the cohort and was associated with greater likelihood of improved seizure control/remaining seizure-free 3 years post-transfer. However, this association should be interpreted cautiously in light of the study’s retrospective design and limitations. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY This study proposes a structured tool to define successful transition and support national standardisation of epilepsy transition services. Seizure frequency was included as an objective clinical outcome to explore potential patient benefit. This study proposes a structured tool to define successful transition and support national standardisation of epilepsy transition services. Seizure frequency was included as an objective clinical outcome to explore potential patient benefit. Introduction Kindly note: Purple is recognised as the international colour for epilepsy, hence the deliberate use of purple schematics throughout my paper.1 The Standards for Quality Improvement Reporting Excellence (SQUIRE) checklist was used for this study.2 Epilepsy is the most prevalent significant childhood neurological disorder,3 and can have a major impact on the lives of young people with epilepsy (YPE) and their families. Effective transition from paediatric to adult care is critical to ensuring continuity of care, adherence to treatment and long-term well-being. Despite the existence of National Institute for Health and Care Excellence (NICE) guidelines, gaps remain in defining and implementing an optimal transition process, leading to varied practices and suboptimal outcomes. NICE’s epilepsy-specific transition guidelines outline key transition considerations, including driving, contraception, pregnancy planning and adherence to treatment,4 acknowledging the increased risks associated with epilepsy. However, these guidelines lack a clear distinction between ‘transfer’ (a single event when the handover of care occurs from paediatric to adult services) and ‘transition’ (a multifaceted, gradual process of planning, preparing and moving from children to adult services)5 and fail to address essential topics such as mental health, bone health and peer relationships.68 The absence of standardised transition models and long-term patient outcome evaluations post-transition leaves room for inconsistent care. Consequently, many YPE experience poorer health, non-adherence to treatment, issues with employment and reduced independence post-transition compared with those without epilepsy. Parents and carers, often central to paediatric epilepsy care, also require tailored guidance to navigate their evolving roles, ensuring positive outcomes for both YPE and their families.914 There is currently no consensus definition for measuring a successful transition from paediatric to adult epilepsy care. For example, a recent Japanese study15 defined success solely as the ‘initiation of adult care’ without requiring any process-related metrics during transition. The wider literature, however, highlights several key principles of effective transition for YPE: Transitions should be initiated at a developmentally appropriate age—not limited to those with complex epilepsies or comorbidities.12 16 Transitions should be initiated at a developmentally appropriate age—not limited to those with complex epilepsies or comorbidities.12 16 Key epilepsy-specific transition topics should be covered during the process, including contraception, driving and mental health.17 Key epilepsy-specific transition topics should be covered during the process, including contraception, driving and mental health.17 Collaboration between paediatric and adult teams (for example, through multidisciplinary team (MDT) clinics) plays a pivotal role in improving transition outcomes.4 18 Collaboration between paediatric and adult teams (for example, through multidisciplinary team (MDT) clinics) plays a pivotal role in improving transition outcomes.4 18 Long-term follow-up within adult services is essential to ensure successful transition. Success should be evaluated against both process metrics and long-term health outcomes,10 hence the need to implement longitudinal patient follow-up post-transfer. Long-term follow-up within adult services is essential to ensure successful transition. Success should be evaluated against both process metrics and long-term health outcomes,10 hence the need to implement longitudinal patient follow-up post-transfer. Developing clear metrics for success (based on the principles above) would enable healthcare providers to monitor and enhance their services and implement more targeted improvements, in line with local and national policy goals. For example, NHS England have highlighted the importance of ‘identifying services that are transitioning well’ and translating this learning to make improvements to other services across England.19 Institutions, such as University College London Hospitals (UCLH), have introduced in-house guidance (figure 1) for transition to the National Hospital for Neurology and Neurosurgery (NHNN), but these have not been formally evaluated. Service evaluation and feedback from patients or parents are both underutilised yet integral components for refining epilepsy transition services.20 Additionally, understanding how transition success impacts seizure frequency and long-term health outcomes remains an area requiring further exploration. Summary of the University College London Hospital proprietary epilepsy transition pathway to the National Hospital of Neurology and Neurosurgery, introduced in 2015. MDT, multidisciplinary team; NICE, National Institute for Health and Care Excellence. Aims In this study, we aimed to develop a working definition of a successful transition from paediatric to adult epilepsy services, grounded in existing literature. We then assessed the rate of successful transition among YPE aged 20–24 within the UCLH pathway. Finally, we examined associations between baseline predictors of transition success and seizure frequency in early adulthood. Methods Defining a successful transition Following an extensive review of the literature and discussions with the UCLH epilepsy team, core criteria for defining a ‘successful’ transition were established. These criteria were subsequently evaluated by paediatric and adult clinicians from the UCLH-NHNN pathway to ensure their clinical validity and applicability. The agreed criteria were: Transition initiated at a developmentally appropriate age (yes/no), decided on a case-by-case basis by the patient’s lead consultant paediatrician and epilepsy specialist nurse at UCLH. The decision was individualised based on the preferences of the patient and their family as well as the clinical team’s assessment of the patient’s neurodevelopmental status, epilepsy severity and self-management abilities. For example, the adolescents with intellectual disabilities or neurodevelopmental disorders in our study required a personalised, delayed transition initiation. All patients had transferred to adult services by age 18; delayed transition initiation referred to timing of preparatory discussions rather than transfer of clinical responsibility. Transition initiated at a developmentally appropriate age (yes/no), decided on a case-by-case basis by the patient’s lead consultant paediatrician and epilepsy specialist nurse at UCLH. The decision was individualised based on the preferences of the patient and their family as well as the clinical team’s assessment of the patient’s neurodevelopmental status, epilepsy severity and self-management abilities. For example, the adolescents with intellectual disabilities or neurodevelopmental disorders in our study required a personalised, delayed transition initiation. All patients had transferred to adult services by age 18; delayed transition initiation referred to timing of preparatory discussions rather than transfer of clinical responsibility. Comprehensive coverage of all relevant topics from existing NICE epilepsy transition guidance (yes/no). See online supplemental Appendix A. Comprehensive coverage of all relevant topics from existing NICE epilepsy transition guidance (yes/no). See online supplemental Appendix A. Attendance at a joint paediatric-adult transition clinic prior to transfer (yes/no): a doctor-led clinic held in-person at UCLH (or virtually during the COVID-19 pandemic) involving the lead consultant paediatrician with a special interest in epilepsy, the paediatric epilepsy specialist nurse, a consultant adult neurologist from NHNN and the patient (and often parent/carer). Adult NHNN epilepsy specialist nurse may also be present. Attendance at a joint paediatric-adult transition clinic prior to transfer (yes/no): a doctor-led clinic held in-person at UCLH (or virtually during the COVID-19 pandemic) involving the lead consultant paediatrician with a special interest in epilepsy, the paediatric epilepsy specialist nurse, a consultant adult neurologist from NHNN and the patient (and often parent/carer). Adult NHNN epilepsy specialist nurse may also be present. Follow-up in the adult clinic within 6 months post-transfer (yes/no). This doctor-led clinic at NHNN usually involves an adult neurologist and epilepsy specialist nurse, with no paediatric team involvement. Follow-up in the adult clinic within 6 months post-transfer (yes/no). This doctor-led clinic at NHNN usually involves an adult neurologist and epilepsy specialist nurse, with no paediatric team involvement. A transition was deemed ‘successful’ when all four criteria were met and this was considered the key predictor of the outcome measure described below. These measures were derived from patient records and clinical letters, and aligned with both in-house UCLH-specific and national NICE guidance. UCLH provides a secondary care paediatric epilepsy service led by general paediatricians with epilepsy expertise rather than paediatric neurologists. NHNN functions as a local adult neurology service and receives tertiary and quaternary referrals for more complex cases. Outcome measure and confounders Our key outcome (change in seizure frequency) was defined by comparing the seizure frequency at two time points: first appointments after the 17th and 20th birthdays of the study participant to measure change from time of handover of care (transfer) to 3 years post-transfer in adulthood. Age 17 represented the final phase of paediatric care while age 20 was chosen to allow sufficient follow-up within NHNN adult services to assess early adult outcomes while ensuring adequate data completeness. Seizure frequency at each time point was categorised by clinicians as: seizure-free, infrequent (≤4 seizures per year), frequent (5–10 seizures per year) or very frequent (>10 seizures per year). The change in seizure frequency between the two time points was coded in a new variable with two levels as either: (1) improved (whose frequency of seizures decreased) or remained seizure-free, or (2) worsened (frequency of seizures increased) or remained bad (retained an infrequent, frequent or very frequent seizure frequency) from time point 1 (age 17) to time point 2 (age 20). Seizure-free individuals under ongoing paediatric follow-up were included and transitioned in the same way as those with active seizures. All individuals in the cohort were under active follow-up until at least the age of 20. Factors associated with transition success included age at first UCLH paediatric epilepsy clinical appointment (continuous variable), as earlier presentation allows longer follow-up within the service and may predispose individuals to a more structured transition. Notably, this variable reflects age at first paediatric epilepsy clinical appointment post referral from emergency services/primary care/other healthcare services, rather than age of epilepsy onset. The number of epilepsy-related comorbidities at baseline (aged 17), ethnicity (white vs non-white) and sex (female vs male) were also included. The primary exposure was successful transition. To assess its association with seizure reduction or remaining seizure-free (our key outcome), we adjusted for a-priori confounders identified through literature and clinical relevance as a minimum adjustment to determine true effect: Age at first paediatric epilepsy clinical appointment. Age at first paediatric epilepsy clinical appointment. Ethnicity. Ethnicity. Sex assigned at birth. Sex assigned at birth. Change in epilepsy-related comorbidities between age 17 (transfer) and 20 (adulthood). This was categorised as ‘improved’, ‘worsened’ or ‘remained the same’ between transfer (aged 17) and adulthood (aged 20). Epilepsy-related comorbidities were recorded as a numerical, continuous variable at both time points. online supplemental Appendix B was used to assess the number of epilepsy-related comorbidities.8 Based on this guidance, the comorbidities included within our study were outlined in online supplemental Appendix C Change in epilepsy-related comorbidities between age 17 (transfer) and 20 (adulthood). This was categorised as ‘improved’, ‘worsened’ or ‘remained the same’ between transfer (aged 17) and adulthood (aged 20). Epilepsy-related comorbidities were recorded as a numerical, continuous variable at both time points. online supplemental Appendix B was used to assess the number of epilepsy-related comorbidities.8 Based on this guidance, the comorbidities included within our study were outlined in online supplemental Appendix C Cohort selection criteria Of the 146 patients initially identified, sequential exclusions for age criteria, incomplete transition and missing follow-up data reduced the cohort to 59 patients with complete data for analysis (figure 2). Retrospective case-note review inclusion procedure based on selection criteria and available patient data. NHNN, National Hospital for Neurology and Neurosurgery; UCLH, University College London Hospitals. Statistical analysis Descriptive statistics were used to characterise the cohort and explore factors associated with successful transition from paediatric to adult care. Median and IQR were reported for non-normally distributed, skewed data (‘age first seen at UCLH’, ‘age at which transition was initiated’, ‘number of epilepsy-related comorbidities at baseline’ and ‘age at transfer’). For categorical data, frequencies and percentages were reported (‘sex’ and ‘ethnicity’). Descriptive statistics were used to characterise the cohort and explore factors associated with successful transition from paediatric to adult care. Median and IQR were reported for non-normally distributed, skewed data (‘age first seen at UCLH’, ‘age at which transition was initiated’, ‘number of epilepsy-related comorbidities at baseline’ and ‘age at transfer’). For categorical data, frequencies and percentages were reported (‘sex’ and ‘ethnicity’). Associations between each predictor and the change in seizure frequency (a binary outcome) were examined in univariable logistic regression models. A multivariable logistic regression model was subsequently developed to control for confounders. Data were analysed using SPSS software V.29 and Stata and p values below 0.05 were deemed statistically significant. Associations between each predictor and the change in seizure frequency (a binary outcome) were examined in univariable logistic regression models. A multivariable logistic regression model was subsequently developed to control for confounders. Data were analysed using SPSS software V.29 and Stata and p values below 0.05 were deemed statistically significant. Data confidentiality All data collection and storage adhered to confidentiality regulations, with no identifiable patient information recorded (see online supplemental Appendix D). Data extraction from Epic was conducted in a secure environment to ensure patient privacy (see online supplemental Appendix E). All data collection and storage adhered to confidentiality regulations, with no identifiable patient information recorded (see online supplemental Appendix D). Data extraction from Epic was conducted in a secure environment to ensure patient privacy (see online supplemental Appendix E). This study was carried out with no funding, and all authors had no competing interests to declare. This study was carried out with no funding, and all authors had no competing interests to declare. Results Characteristics of the study population The total number of people included in the statistical analyses was 59 after the elimination process (figure 2). Of the 59 patients, 59.3% (N=35/59) were white and 55.9% (N=33/59) were female. The median age at which transition (the gradual, planned MDT process) was initiated was 16.0 years (IQR 14.0–17.0, range 12.0–18.0) while the median age at transfer (ie, single handover of care event) was 17.0 years (IQR 17.0–18.0, range 15.0–18.0). Successful transition Of the 59 patients, 72.9% (N=43/59) had achieved a successful transition from childcare to adult care meeting all four criteria: (1) 91.5% (N=54/59) commenced at a developmentally appropriate age, (2) 91.5% (N=54/59) covered all recommended NICE transition topics, (3) 79.7% (N=47/59) attended a joint paediatric-adult clinic and (4) 88.1% (N=52/59) had an adult follow-up within 6 months post-transfer. Patient characteristics and successful transitions No significant differences in successful transitioning were identified among participants of different sex, ethnicity, epilepsy-related comorbidities at transfer or ages at first UCLH paediatric epilepsy clinical appointment (table 1). Table 1Characteristics of the study participants and their association with successful transitionSuccessful transitionTotalN=59P valueUnsuccessfulN=16SuccessfulN=43Age first seen in UCLH paediatric epilepsy clinic, median (IQR)14 (12–15)14 (13–16)14 (13– 16)0.34*Epilepsy-related comorbidities at baseline (recorded at transfer), median (IQR)1 (0–3)0 (0–2)0 (0–2)0.08*Ethnicity, n (%) 0.77† White9 (25.7)26 (74.3)35 (100.0) Non-white7 (29.2)17 (70.8)24 (100.0) Sex, n (%) 0.58† Female8 (24.2)25 (75.8)33 (100.0) Male8 (30.8)18 (69.2)26 (100.0) The brackets contain the IQR for the medians or the % of total for the means.*P value obtained from Mann-Whitney U test (continuous variable+non-parametric, could not do independent t sample test as not normally distributed).†P value obtained from Pearson’s χ2 test (parametric assumptions satisfied).UCLH, University College London Hospitals. Characteristics of the study participants and their association with successful transition The brackets contain the IQR for the medians or the % of total for the means. P value obtained from Mann-Whitney U test (continuous variable+non-parametric, could not do independent t sample test as not normally distributed). P value obtained from Pearson’s χ2 test (parametric assumptions satisfied). UCLH, University College London Hospitals. The median age at first presentation to the UCLH paediatric epilepsy clinic was 14 years in both groups. Patients who transitioned unsuccessfully had a higher median number of epilepsy-related comorbidities at transfer (1 vs 0). A successful transition was achieved among 74.3% (26/35) of the white population compared with 70.8% (17/24) of the non-white population. The majority of both the female and the male study participants achieved a successful transition (75.8% vs 69.2%, respectively). Seizure outcomes in the cohort Between ages 17 and 20, 19/59 patients showed improved seizure control, 8/59 remained seizure-free and 32/59 had unchanged or worsened seizure control (table 2). Overall, 45.8% (27/59) experienced improvement or maintained seizure freedom, while 54.2% (32/59) had persistent or worsening seizures. Table 2The spread of change in seizure frequency between transfer and adulthood in patient cohort (n=59)Adulthood (aged 20) At transfer (aged 17) Seizure-freeInfrequentFrequentVery frequent Total Seizure-free820010Infrequent1093224Frequent13026Very frequent2211419 Total 211641859Green=seizure frequency improved/remained seizure-free.White= seizure frequency stayed bad/worsened. The spread of change in seizure frequency between transfer and adulthood in patient cohort (n=59) Green=seizure frequency improved/remained seizure-free. White= seizure frequency stayed bad/worsened. Successful transition and seizure frequency after adjustment for confounders The multivariable regression model examined the associations between successful transition and seizure frequency improvement after adjusting for confounders including age first seen in the UCLH paediatric epilepsy clinic, ethnicity, sex assigned at birth and the difference in number of comorbidities between transfer and adulthood (table 3). Successful transition was found to be significantly associated with seizure frequency. The odds of improved seizure frequency/remaining seizure free in patients with a successful transition were more than four times higher compared with patients with an unsuccessful transition, adjusted OR=4.28 (95% CI 1.07 to 17.19, p=0.04). Ethnicity, age at first paediatric clinical appointment and changes in comorbidities were not associated with seizure frequency, while sex was. In particular, males compared with females had more than four times higher odds of seizure improvement/remaining seizure free, adjusted OR=4.29 (95% CI 1.26 to 14.67, p=0.02), indicating a robust independent association. Table 3Univariable and multivariable regression models with predictors of the outcome: having improved seizure frequency/staying seizure free between transfer (aged 17) and adulthood (aged 20)OR (95% CI)P valueAdjusted OR (95% CI)P valueSuccessful transition UnsuccessfulRef. Ref. Successful3.45 (0.96 to 12.41) 0.06 4.28 (1.07 to 17.19) 0.04 Age first seen in UCLH paediatric epilepsy clinic1.01 (0.85 to 1.20)0.951.05 (0.85 to 1.30)0.66Ethnicity WhiteRef. Ref. Non-white0.76 (0.27 to 2.16)0.600.68 (0.20 to 2.29)0.53Sex FemaleRef. Ref. Male3.20 (1.10 to 9.34) 0.03 4.29 (1.26 to 14.67) 0.02 Difference in number of comorbidities between transfer (17) and adulthood (20) Stayed the sameRef. Ref. Worsened1.06 (0.23 to 4.87)0.950.85 (0.15 to 4.75)0.85 Improved0.59 (0.16 to 2.09)0.410.81 (0.19 to 3.35)0.77Statistically significant/near significant results are highlighted in red.UCLH, University College London Hospitals. Univariable and multivariable regression models with predictors of the outcome: having improved seizure frequency/staying seizure free between transfer (aged 17) and adulthood (aged 20) Statistically significant/near significant results are highlighted in red. UCLH, University College London Hospitals. Discussion Main findings In this cohort, 72.9% of YPE achieved a successful transition, defined by meeting all four of the defined success metrics; undergoing a joint clinic was the limiting factor to achieving a higher success rate. Successful transition was associated with improved seizure control or maintenance of seizure freedom in adulthood, underscoring the importance of structured, comprehensive support. Proportions and characteristics in context Our rate of successful transition is high compared with published estimates, although direct comparison is challenging given the absence of a standardised transition success metric. A 2024 meta-synthesis by Abildgaard Hanson et al20 reported that fewer than two-thirds of YPE experience a smooth transition, with marked variability across services. Among the cohort in this study, the most frequent point of failure was undergoing the joint clinic (achieved by 79.7%). The 2022 systematic review by Goselink et al11 highlighted joint clinic attendance as both the most impactful and the most inconsistently implemented transition element—consistent with our local findings. This finding is also congruent with a recent single-centred evaluation of a cohort of 70 patients with complex epilepsy syndromes which similarly identified joint multidisciplinary encounters as instrumental to a ‘smooth landing’ in adult services.11 In the UK, joint clinics have also been shown to prompt diagnostic reassessment with 16.3% of patients undergoing further investigations following MDT review—an approach that can optimise long-term medical and psychological outcomes.11 21 In our cohort, younger age at transfer, absence of intellectual or psychological comorbidities and stable seizure control were more common among those who transitioned successfully, reflecting patterns seen internationally. Interestingly, age at first paediatric epilepsy clinical appointment did not influence outcomes, challenging the assumption that earlier engagement improves transition success and health outcomes. Males exhibited better post-transition seizure outcomes compared with females, consistent with the existing literature consensus.2224 Hormonal fluctuations, such as those during menstruation, pregnancy and menopause, can exacerbate seizure frequency. This highlights the need for sex-specific transition planning—integrating reproductive and psychological support. Impact of successful versus unsuccessful transition Poorly executed transition carries significant risks: service disengagement, missed opportunities for diagnosis or treatment optimisation and increased seizure recurrence.10 19 These risks are not merely clinical: seizure recurrence in adulthood can remove driving eligibility, restrict employment and educational opportunities and worsen mental health. The meta-synthesis attributes these adverse outcomes to loss of continuity, abrupt changes in care and reduced self-efficacy; gradual, well-communicated handover, on the other hand, fosters adherence and stability.20 The psychosocial burden of seizures is substantial: a UK study25 involving 70 YPE reported significantly higher rates of depression, anhedonia and social anxiety compared with the 70 matched controls without epilepsy. Among the 70 YPE in the cohort, high seizure frequency was strongly associated with reduced self-esteem and tonic–clonic seizures were specifically linked to greater depressive symptoms. In our cohort, post-transition challenges such as substance misuse, gambling, difficulty adjusting to higher education, missed appointments and safeguarding concerns likely contributed to poorer seizure control.7 8 The physical and emotional needs of subgroups—such as those with refractory epilepsy or requiring dietary therapy—warrant particular attention. For example, the limited availability of adult providers for treatments like the ketogenic diet26 27 creates vulnerability during transition and highlights the importance of addressing and accommodating the diverse needs of YPE.11 UCLH-NHNN service strengths and limitations Several features may have contributed to the high success rate. The incorporation of structured transition tools such as ReadySteadyGo, and emerging questionnaires like EpiTRAQ, supports patient-centred care and readiness assessments.28 29 UCLH benefits from epilepsy specialist nurses and in-house psychological medicine, consistent with current NICE guidance recommending nurses as lead coordinators.4 This resource-rich environment may not be generalisable to all centres, where staffing and funding constraints limit provision.18 30 31 In contrast, NHNN lacks dedicated psychological clinics. Despite this, follow-up rates within 6 months of joint clinic attendance in the adult service were high (88.1%), even during the COVID-19 pandemic. Ongoing discrepancies between paediatric and adult service provision highlight the need for better continuity. Expanding services for 16–25 years old19 could reduce adult service strain. Recent initiatives, including the appointment of a youth worker to support patients until age 25, and the exploration of weekend groups11 and transition mentoring programmes, mirror evidence from other chronic disease transition programmes showing improved engagement and navigation of adult care. Study strengths and limitations This is the first service evaluation assessing the UCLH-NHNN epilepsy transition pathway, an integrated and well-established service with shared electronic health records across paediatric and adult care. Four novel assessment variables, developed through expert consultation and a comprehensive literature review, provided a structured framework for evaluating transition success. A literature review across four major databases (MEDLINE, Embase, Web of Science, Cochrane Library), ensured a broad and rigorous evidence base. Analysis included all patients transitioning from UCLH to NHNN up to 2023, using the complete Epic database. Robust statistical analyses and engagement with epilepsy teams across both services strengthened the validity and clinical relevance of the findings. Recognition of case-by-case variability in transition timing improves applicability to real-world practice. One limitation was sample size, partly due to the exclusion of patients with substantial missing data (22/81). As a retrospective cohort study relying on clinical documentation, analyses depended on the completeness and accuracy of recorded information; some transition processes—such as nurse-led informal phone calls—were not always formally documented, particularly before full Epic (Electronic Care Record) implementation in 2019. This introduced potential information bias. Exclusions of patients with incomplete records may have also introduced selection bias, potentially under-representing patients with poorer engagement, although the direction of this bias cannot be determined retrospectively. The database also lacked key confounders, including parental demographics and socioeconomic status. Seizure type or syndromic diagnosis was not analysed separately, limiting the specificity of seizure-burden comparisons. Residual confounding therefore cannot be excluded. Seizure trajectories in early adulthood may partly reflect the natural history of childhood-onset epilepsies, with longitudinal cohort studies demonstrating spontaneous remission in some and worsening in others—despite optimal treatment and adherence.32 Accordingly, seizure improvement at age 20 cannot be attributed solely to transition quality. For example, 8 out of 59 patients were seizure-free at age 17 and remained seizure-free at age 20; this maintenance of remission may reflect the underlying course of their epilepsies rather than transition processes. More broadly, the retrospective observational design limits causal inference between transition success and seizure outcomes. Although this was a single-centre study and findings cannot be directly generalised nationally, the UCLH–NHNN transition pathway is a valuable model for audit and research. Co-located paediatric and adult services enabled robust longitudinal follow-up; other services may not have this same advantage. Using the proposed criteria, this pathway achieved a 72.9% transition success rate. While 100% remains the aspirational standard, stricter catchment criteria have since been introduced and the UCLH–NHNN service no longer manages out-of-area patients for transition, improving appointment attendance—particularly joint clinic attendance, previously the main barrier to higher success rates—and service communication. Policy and practice recommendations There is currently no clear NICE guidance on transition outcome measurement, representing an important policy gap. National adoption of standardised success metrics based on the four core elements identified in this study could support consistent service evaluation. Targeting these elements—particularly joint clinic attendance as a high-yield intervention—may improve transition outcomes. Tailored pathways should identify vulnerable patients requiring bespoke transition planning, with sex-specific support available throughout. Services must also accommodate late entrants to paediatric services, such as those diagnosed with epilepsy aged 16 or older. Expanding developmentally appropriate services for young adults aged 16–25 could reduce strain on adult neurology and better align with the neurodevelopmental trajectory. Future work should include multicentre audit using these measures to compare transition performance across services and explore associations between patient characteristics, service factors and adult outcomes post transfer. A prospective multicentre cohort study examining the four pillars of transition success, with prospective monitoring of seizure frequency and seizure type, would strengthen the evidence base for improving epilepsy transition and long-term outcomes for YPE. doneAlthough successful transition was associated with a fourfold increase in the odds of improved seizure frequency or remaining seizure-free, the wide 95% CI (1.07 to 17.19) reflects the limitations of a single-centre sample. Larger multicentre studies are needed to provide more precise estimates. While seizure frequency is an important clinical outcome, future research should also examine broader outcomes of transition success, including patient-reported quality of life scores, cognitive function, psychological well-being and medication burden, requiring collaboration and dedicated funding across centres. Conclusion This single-centre study demonstrated a structured approach to epilepsy transition, with 72.9% of YPE meeting our criteria for successful transition from UCLH to NHNN: developmentally appropriate initiation, NICE-recommended topic coverage, joint clinic attendance and follow-up within 6 months. Successful transition was associated with improved seizure outcomes between ages 17 and 20, although this should be interpreted cautiously given the retrospective design and wide CI. Nevertheless, if effective transition can improve long-term outcomes for YPE, this should be a powerful incentive to continually improve transition services. While many factors contribute to successful transition, including psychological and family influences, a universal standard should include these core four elements. Adoption of this national framework could provide an objective benchmark to monitor services over time, compare performance across centres and drive improvements in long-term outcomes for YPE. Supplementary material