Our multicentre study compared CH patient care documented in HypoDok with evidence-based recommendations. The majority of the surrogate parameters which assessed the adherence to the core statements were comparable in both groups, irrespective of guideline publication. However, few significant differences between the two cohorts before and after guideline introduction were apparent. This continuously high level of care can be partly explained by the fact that the data of the registry is derived from paediatric endocrinologists, who were directly or indirectly involved in its development as being members of the guideline developing professional society.
The recommendations regarding diagnostics (structural quality) as well as therapy control (process quality) were implemented, but not fully documented.
The low percentage of patients with documented TT4/fT4 serum concentration at initial diagnosis is one example of suggested documentation gaps, as low TT4/fT4 levels are precondition for diagnosis confirmation. This may arise due to a structural fault of the database, that the date of diagnosis can be entered without underlying TT4/fT4 levels.
Another example is the low number of documented hearing tests at the start of treatment, although newborn hearing screening has been a standard service of public health insurance since 01/2009 in Germany and has been regularly reimbursed since 10/2010 . This problem of incomplete documentation in the HypoDok, especially at the beginning of treatment, is well known and has already been described by Ellerbroek et al. . The reason for this may be that many patients were only referred to paediatric endocrinologists over the course of the first months, as the guideline only recommends referral within 3 months. Subsequent documentation seems incomplete.
Furthermore, improvement of psychomotor development assessment documentation is required, especially in group B (24.8 %; group A 49.3 %). It can be assumed that these assessments were only carried out during the course of therapy if there was a clear clinical indication for developmental retardation. Experienced examiners seem to preselect patients who need a formal assessment, which is however in contrast to current guideline recommendations. The rate of IQ/EQ testing may be underestimated in group B, as some patients are still very young and therefore not all data have been collected so far. Further reasons why psychomotor development has not been regularly checked may be additionally a lack of compliance in the patient family. This may be due to fear of receiving an unfavourable result, a lack of understanding of the importance of the test and the time required for the extensive testing procedure . Here, the implementation of shorter screening tests could increase patient compliance.
In general, a very variable level of adherence to existing guidelines has already been observed by the treating physicians for a wide range of paediatric diseases. Guideline adherence regarding the use of antibiotics in infectious disease is a highly investigated subject [15-17]. The studies showed no change of management in urinary tract infections after introduction of the guideline, whereas the use of antibiotics in CAP was evidence-based. Overall, it could be shown that disregarding evidence-based recommendations carries risks.
Vezzani et al. examined guideline adherence in adult patients with primary hypothyroidism . Here, a satisfactory adherence to the European recommendations was found, with the exception of the L-T4 starting dose.
The time period, in which guidelines are strictly adhered to has been analysed by Ament et al. The authors could show that already one year after the publication of guidelines the degree of their implementation is starting to decrease . Since the importance of guidelines can also be demonstrated by a systematic review , existing considerations for improving the implementation of guidelines should also be considered for CH. The successful integration of guideline recommendations can lead to a high, uniform standard of care, save financial resources and improve patient outcome . The aim to improve guideline implementation should also include the elevation of patient compliance. Provision of practice aids like checklists or reminders on the side of clinicians and implementation of diaries on the side of patients may improve the problem next to a clear documentation of patient history by healthcare providers .
Matlock et al. followed this two-sided approach and presented methods to equally promote guideline adherence by improvement of patient care through the responsible physicians and also in a CH patient-based manner: : these included, e.g., compiling and continuation of patient records in a database, and education programs of treating physicians and patient families on guideline recommendations. Furthermore, patient registries should also be strengthened as a suitable instrument for quality assurance .
The HypoDok registry documents a satisfactory adherence to the core statements of the guideline regarding treatment. Thus, the basic preconditions for achieving the best possible patient outcome with adequate therapy were given in both cohorts [25-27]. The adherence concerning diagnostics and treatment monitoring by professionals however can be improved in both groups. Alternatively, guideline recommendations on IQ/EQ-testing in routine care might require re-evaluation.
Overall, the results cannot easily be transferred to the overall quality of care of patients with CH in Germany and Austria, as only 17.5% of expected patients with CH in Germany and even 6.7% in Austria are currently documented in HypoDok at the time interval analysed here. This low reporting frequency is due to the fact that the database is known and used among experts (paediatric endocrinologists), but not widely used by paediatricians in private practice. This could indicate that a significant number of patients with CH are not seen by paediatric endocrinologists. The use of the HypoDok registry should be seen as an opportunity to document and ensure the quality of care for patients with CH and to adapt it to current clinical standards. This way, the treating physician can also compare and adapt patient care with the existing data on the basis of regular database reports. Compiling of genetic sequencing data combined with clinical information and phenotypic data including blood tests could further help to subtype patients suffering from CH into clinically relevant subgroups. Overall, management of patients in databases has not only been identified by the professional society as one of the appropriate instruments for quality assurance. Administrative bodies such as the European Commission or the US-American National Institute of Health also started programs for rare diseases. One example is the European initiative of European Reference Networks (ERN) for different specialities like Endo-ERN for rare endocrine conditions in order to share knowledge to assure and maintain high level of patient care across Europe. As part of this initiative, patient cohort documentation in databases in the care of rare diseases such as CH is important to ensure structural quality. In case of CH in Germany and Austria, the centre-specific obligation for documentation will have to be strengthened, for example by means of financial incentives. The authors acknowledge that the current setup of the registry (local documentation with periodic extraction of pseudonymized data) may not be optimal for paediatricians taking care of very few patients with CH. Long-lasting adherence to guidelines and respective documentation remains difficult, since in principle, it is easier to motivate participating professionals and patients over a limited period of time .
Furthermore, a legal requirement to keep these patients in registries should be discussed. As the low number of documented CH patients shows, the implementation of the guidelines recommendations of the professional society can only be checked and improved by either obligation or standardized acquisition of data that facilitates documentation in order to guarantee adequate care.
The fact that legal requirements and financial support by the state are helpful in this respect is not only reflected in the reporting frequency of newborn screening, but also in other fields of paediatrics. For example, in oncology approximately 95 % of childhood cancers are documented in the German Childhood Cancer Registry by appropriate legal requirements .