An assessment of the informative value of data sharing statements in clinical trial registries.

BACKGROUND: The provision of data sharing statements (DSS) for clinical trials has been made mandatory by different stakeholders. DSS are a device to clarify whether there is intention to share individual participant data (IPD). What is missing is a detailed assessment of whether DSS are providing clear and understandable information about the conditions for data sharing of IPD for secondary use. METHODS: A random sample of 200 COVID-19 clinical trials with explicit DSS was drawn from the ECRIN clinical research metadata repository. The DSS were assessed and classified, by two experienced experts and one assessor with less experience in data sharing (DS), into different categories (unclear, no sharing, no plans, yes but vague, yes on request, yes with specified storage location, yes but with complex conditions). RESULTS: Between the two experts the agreement was moderate to substantial (kappa=0.62, 95% CI [0.55, 0.70]). Agreement considerably decreased when these experts were compared with a third person who was less experienced and trained in data sharing ("assessor") (kappa=0.33, 95% CI [0.25, 0.41]; 0.35, 95% CI [0.27, 0.43]). Between the two experts and under supervision of an independent moderator, a consensus was achieved for those cases, where both experts had disagreed, and the result was used as "gold standard" for further analysis. At least some degree of willingness of DS (data sharing) was expressed in 63.5% (127/200) cases. Of these cases, around one quarter (31/127) were vague statements of support for data sharing but without useful detail. In around half of the cases (60/127) it was stated that IPD could be obtained by request. Only in in slightly more than 10% of the cases (15/127) it was stated that the IPD would be transferred to a specific data repository. In the remaining cases (21/127), a more complex regime was described or referenced, which could not be allocated to one of the three previous groups. As a result of the consensus meetings, the classification system was updated. CONCLUSION: The study showed that the current DSS that imply possible data sharing are often not easy to interpret, even by relatively experienced staff. Machine based interpretation, which would be necessary for any practical application, is currently not possible. Machine learning and / or natural language processing techniques might improve machine actionability, but would represent a very substantial investment of research effort. The cheaper and easier option would be for data providers, data requestors, funders and platforms to adopt a clearer, more structured and more standardised approach to specifying, providing and collecting DSS. TRIAL REGISTRATION: The protocol for the study was pre-registered on ZENODO ( https://zenodo.org/record/7064624#.Y4DIAHbMJD8 ).

Assessment of a demonstrator repository for individual clinical trial data built upon DSpace.

Background: Given the increasing number and heterogeneity of data repositories, an improvement and harmonisation of practice within repositories for clinical trial data is urgently needed. The objective of the study was to develop and evaluate a demonstrator repository, using a widely used repository system (DSpace), and then explore its suitability for providing access to individual participant data (IPD) from clinical research. Methods: After a study of the available options, DSpace (version 6.3) was selected as the software for developing a demonstrator implementation of a repository for clinical trial data. In total, 19 quality criteria were defined, using previous work assessing clinical data repositories as a guide, and the demonstrator implementation was then assessed with respect to those criteria. Results: Generally, the performance of the DSpace demonstrator repository in supporting sensitive personal data such as that from clinical trials was strong, with 14 requirements demonstrated (74%), including the necessary support for metadata and identifiers. Two requirements could not be demonstrated (the ability to include de-identification tools and the availabiltiy of a self-attestation system) and three requirements were only partially demonstrated (ability to provide links to de-identification tools and requirements, incorporation of a data transfer agreement in system workflow, and capability to offer managed access through application on a case by case basis). Conclusions: Technically, the system was able to support most of the pre-defined requirements, though there are areas where support could be improved. Of course, in a productive repository, appropriate policies and procedures would be needed to direct the use of the available technical features. A technical evaluation should therefore be seen as indicating a system's potential, rather than being a definite assessment of its suitability. DSpace clearly has considerable potential in this context and appears a suitable base for further exploration of the issues around storing sensitive data.

Legal assessment tool (LAT): an interactive tool to address privacy and data protection issues for data sharing.

BACKGROUND: In an unprecedented rate data in the life sciences is generated and stored in many different databases. An ever increasing part of this data is human health data and therefore falls under data protected by legal regulations. As part of the BioMedBridges project, which created infrastructures that connect more than 10 ESFRI research infrastructures (RI), the legal and ethical prerequisites of data sharing were examined employing a novel and pragmatic approach. METHODS: We employed concepts from computer science to create legal requirement clusters that enable legal interoperability between databases for the areas of data protection, data security, Intellectual Property (IP) and security of biosample data. We analysed and extracted access rules and constraints from all data providers (databases) involved in the building of data bridges covering many of Europe's most important databases. These requirement clusters were applied to five usage scenarios representing the data flow in different data bridges: Image bridge, Phenotype data bridge, Personalised medicine data bridge, Structural data bridge, and Biosample data bridge. A matrix was built to relate the important concepts from data protection regulations (e.g. pseudonymisation, identifyability, access control, consent management) with the results of the requirement clusters. An interactive user interface for querying the matrix for requirements necessary for compliant data sharing was created. RESULTS: To guide researchers without the need for legal expert knowledge through legal requirements, an interactive tool, the Legal Assessment Tool (LAT), was developed. LAT provides researchers interactively with a selection process to characterise the involved types of data and databases and provides suitable requirements and recommendations for concrete data access and sharing situations. The results provided by LAT are based on an analysis of the data access and sharing conditions for different kinds of data of major databases in Europe. CONCLUSIONS: Data sharing for research purposes must be opened for human health data and LAT is one of the means to achieve this aim. In summary, LAT provides requirements in an interactive way for compliant data access and sharing with appropriate safeguards, restrictions and responsibilities by introducing a culture of responsibility and data governance when dealing with human data.

A standardised graphic method for describing data privacy frameworks in primary care research using a flexible zone model.

PURPOSE: To develop a model describing core concepts and principles of data flow, data privacy and confidentiality, in a simple and flexible way, using concise process descriptions and a diagrammatic notation applied to research workflow processes. The model should help to generate robust data privacy frameworks for research done with patient data. METHODS: Based on an exploration of EU legal requirements for data protection and privacy, data access policies, and existing privacy frameworks of research projects, basic concepts and common processes were extracted, described and incorporated into a model with a formal graphical representation and a standardised notation. The Unified Modelling Language (UML) notation was enriched by workflow and own symbols to enable the representation of extended data flow requirements, data privacy and data security requirements, privacy enhancing techniques (PET) and to allow privacy threat analysis for research scenarios. RESULTS: Our model is built upon the concept of three privacy zones (Care Zone, Non-care Zone and Research Zone) containing databases, data transformation operators, such as data linkers and privacy filters. Using these model components, a risk gradient for moving data from a zone of high risk for patient identification to a zone of low risk can be described. The model was applied to the analysis of data flows in several general clinical research use cases and two research scenarios from the TRANSFoRm project (e.g., finding patients for clinical research and linkage of databases). The model was validated by representing research done with the NIVEL Primary Care Database in the Netherlands. CONCLUSIONS: The model allows analysis of data privacy and confidentiality issues for research with patient data in a structured way and provides a framework to specify a privacy compliant data flow, to communicate privacy requirements and to identify weak points for an adequate implementation of data privacy.

Fostering EMA's transparency policy.

The European Medicines Agency has opened a window to access clinical trial data. This is an important step forward which deserves attention, support, and advice from all the stakeholders. Regulatory agencies are the most comprehensive repositories of clinical trial data on drugs and can also promote and develop standard practices for data sharing. The release of the EMA draft policy on publication and access to clinical trial data in 2013 has fueled a lively debate among academia, industry, and the public in general that is still ongoing. As clinical researchers and producers and users of clinical trial data, we endorse the European Medicines Agency's opening and offer a few suggestions for complete, safe, and effective data sharing.

Socioeconomic factors and effect of evidence-based patient information about primary prevention of type 2 diabetes mellitus--are there interactions?

BACKGROUND: Having shown in a recent randomized controlled trial that evidence-based patient information (EBPI) significantly increased knowledge on primary prevention of diabetes compared to standard patient information, we now investigated interaction between socioeconomic status (SES) and the effect of an EBPI. FINDINGS: 1,120 visitors (aged 40-70 years, without known diabetes) to the "Techniker Krankenkasse" and the "German Diabetes Center" websites were randomized. The intervention group received a newly developed on-line EBPI, the control group standard on-line information. The primary outcome measure was knowledge, classified as "good/average/poor". We analyzed associations of knowledge with socioeconomic variables (education, vocational training, employment, subjective social status) combined with intervention effect including interactions, adjusted for possible confounding by knowledge before intervention, self-reported blood glucose measurements, blood pressure, blood lipid levels, age and gender. Logistic regression models were fitted to the subpopulation (n = 647) with complete values in these variables.Education (high vs. low) was significantly associated with knowledge (good vs. average/poor); however, there was no significant interaction between education and intervention. After adjustment, the other socioeconomic variables were not significantly associated with knowledge. CONCLUSIONS: Socioeconomic variables did not significantly change the effect of the intervention. There was a tendency towards a lower effect where lower educated individuals were concerned. Possibly the power was too low to detect interaction effects. Larger studies using SES-specific designs are needed to clarify the effect of SES. We suggest considering the socioeconomic status when evaluating a decision aid, e.g. an EBPI, to ensure its effectiveness not only in higher socioeconomic groups. TRIAL REGISTRATION: Current Controlled Trials ISRCTN22060616 (Date assigned: 12 September 2008).

Within-trial economic evaluation of diabetes-specific cognitive behaviour therapy in patients with type 2 diabetes and subthreshold depression.

BACKGROUND: Despite the high prevalence of subthreshold depression in patients with type 2 diabetes, evidence on cost-effectiveness of different therapy options for these patients is currently lacking. METHODS/DESIGN: Within-trial economic evaluation of the diabetes-specific cognitive behaviour therapy for subthreshold depression. Patients with diabetes and subthreshold depression are randomly assigned to either 2 weeks of diabetes-specific cognitive behaviour group therapy (n = 104) or to standard diabetes education programme only (n = 104). Patients are followed for 12 months. During this period data on total health sector costs, patient costs and societal productivity costs are collected in addition to clinical data. Health related quality of life (the SF-36 and the EQ-5D) is measured at baseline, immediately after the intervention, at 6 and at 12 months after the intervention. Quality adjusted life years (QALYs), and cumulative costs will be estimated for each arm of the trial. Cost-effectiveness of the diabetes-specific cognitive behaviour group therapy will be analysed from the perspective of the German statutory health insurance and from the societal perspective. To this end, incremental cost-effectiveness ratio (ICER) in terms of cost per QALY gained will be calculated. DISCUSSION: Some methodological issues of the described economic evaluation are discussed. TRIAL REGISTRATION: The trial has been registered at the Clinical Trials Register (NCT01009138).

[Assessment and importance of clinical trial endpoints in various areas of medicine]. / Bewertung und Bedeutung von Studienendpunkten in unterschiedlichen Bereichen der Medizin.

Numerous generic and disease-specific endpoints for clinical trials are available and in use. Standardised quality criteria for the assessment of endpoints have been developed which form the basis of many current reviews. In many clinical fields these endpoints were shown to pose major problems, though. These include a lack of standardisation and poor methodological quality, problems related to the applicability of endpoints and their limited appropriateness for specific research questions. In the meantime standardisation and harmonisation ecommendations have been developed (e.g., EMEA/FDA, RECIST, IMMPACT and OMERACT). In future, the selection of endpoints for clinical trials will have to be based on the specific research question (disease, hypothesis, intervention), taking account of recommendations, standardisations as well as harmonisations.

Two short-term outcomes after instituting a national regulation regarding minimum procedural volumes for total knee replacement.

BACKGROUND: Several studies have demonstrated positive relationships between high hospital volume and improved outcome following total knee replacement. To our knowledge, it has not been demonstrated whether improved outcomes are causally determined by selective referral to high-volume hospitals. We therefore evaluated the effect of a national regulation regarding minimum hospital volume for total knee replacement on two short-term outcome parameters. METHODS: We performed a comparison of the years before (2004, 2005) and after (2006) the implementation of a national regulation on minimum hospital volume for total knee replacement through a secondary analysis of a national database on the quality of inpatient care in Germany as reflected by the number of cases per hospital and the postoperative rates of wound infection and wound hematoma or secondary hemorrhage. RESULTS: We analyzed 110,349 cases from 2004, 118,922 cases from 2005, and 125,322 cases from 2006. Implementation of the regulation had a significant effect on the number of cases per hospital. Of the hospitals that had performed one to forty-nine cases in 2005, 35.6% moved to higher-volume categories and 21.2% dropped out in 2006. Multiple logistic regression analysis adjusting for patient characteristics demonstrated risk reductions of 22.5% (odds ratio, 0.775; 95% confidence interval, 0.700 to 0.857) for postoperative wound infection and of 44% (odds ratio, 0.562; 95% confidence interval, 0.531 to 0.596) for wound hematoma or secondary hemorrhage from 2005 to 2006. For wound infection, approximately half of the improvement was attributable to the effects of the minimum-volume regulation. For wound hematoma and secondary hemorrhage, the improvement could not be explained by the minimum-volume regulation. CONCLUSIONS: Implementation of the minimum-volume regulation for total knee replacement resulted in more patients being managed at higher-volume hospitals than expected. Following the implementation of a minimum-volume regulation, effects on two short-term outcome parameters were observed, but definite conclusions could only be made regarding wound infection, with the minimum-volume regulation resulting in a decreased rate of infection.

[Utilities: a solution of a decision problem?]. / Utilities--eine Lösung für ein Bewertungsproblem?

Utility is a concept that originates from utilitarianism, a highly influential philosophical school in the Anglo-American world. The cornerstone of utilitarianism is the principle of maximum happiness or utility. In the medical sciences, this utility approach has been adopted and developed within the field of medical decision making. On an operational level, utility is the evaluation of a health state or an outcome on a one-dimensional scale ranging from 0 (death) to 1 (perfect health). By adding the concept of expectancy, the graphic representation of both concepts in a decision tree results in the specification of expected utilities and helps to resolve complex medical decision problems. Criticism of the utility approach relates to the rational perspective on humans (which is rejected by a considerable fraction of research in psychology) and to the artificial methods used in the evaluation of utility, such as Standard Gamble or Time Trade Off. These may well be the reason why the utility approach has never been accepted in Germany. Nevertheless, innovative concepts for defining goals in health care are urgently required, as the current debate in Germany on "Nutzen" (interestingly translated as 'benefit' instead of as 'utility') and integrated outcome models indicates. It remains to be seen whether this discussion will lead to a re-evaluation of the utility approach.

Evaluating compulsory minimum volume standards in Germany: how many hospitals were compliant in 2004?

BACKGROUND: Minimum hospital procedure volumes are discussed as an instrument for quality assurance. In 2004 Germany introduced such annual minimum volumes nationwide on five surgical procedures: kidney, liver, stem cell transplantation, complex oesophageal, and pancreatic interventions. The present investigation is the first part of a study evaluating the effects of these minimum volumes on health care provision. Research questions address how many hospitals and cases were affected by minimum volume regulations in 2004, how affected hospitals were distributed according to minimum volumes, and how many hospitals within the 16 German states complied with the standards set for 2004. METHODS: The evaluation is based on the mandatory hospital quality reports for 2004. In the reports, all hospitals are statutorily obliged to state the number of procedures performed for each minimum volume. The data were analyzed descriptively. RESULTS: In 2004, 485 out of 1710 German hospitals providing acute care and approximately 0.14% of all hospital cases were affected by minimum volume regulations. Liver, kidney, and stem cell transplantation affected from 23 to hospitals; complex oesophageal and pancreatic interventions affected from 297 to 455 hospitals. The inter-state comparison of the average hospital care area demonstrates large differences between city states and large area states and the eastern and western German states ranging from a minimum 51 km2 up to a maximum 23.200 km2, varying according to each procedure. A range of 9% - 16% of the transplantation hospitals did not comply with the standards affecting 1% - 2% of the patients whereas 29% and 18% of the hospitals treating complex oesophageal and pancreatic interventions failed the standards affecting 2% - 5% of the prevailing cases. CONCLUSION: In 2004, the newly introduced minimum volume regulations affected only up to a quarter of German acute care hospitals and few cases. However, excluding the hospitals not meeting the minimum volume standards from providing the respective procedures deserves considering two aspects: the hospital health care provision concepts by the German states as being responsible and from a patient perspective the geographically equal access to hospital care.

Do Germans keep patients too long in hospital? A prospective randomized trial.

Cost-effectiveness reduces hospital stay for all patients with thyroid surgery but lacks information on medical comparability and patients' fulfilled expectations. The aim of this study was to assess if a hospital stay of 2 days after thyroid surgery had a negative influence on the medical quality or on health-related quality of life. In a controlled prospective randomized trial with 238 patients, a postoperative hospital stay of 2 days was compared to one longer than 2 days. The postoperative medical investigation included serum calcium levels, laryngeal nerve function, and suction drainage volume. Health-related quality of life was assessed on the day of admission before the operation and again 14 days after discharge. Fourteen days after discharge patients were also asked about their subjective health. Despite the study design, it was necessary, for ethical reasons, to let the patients decide when to leave the hospital. In the 2-day study group, 56.6% of the patients preferred hospitalization for more than 2 days (most choosing 3 days). Medical reasons were hyperthyroidism (p < 0.02) and postoperative hypocalcemia (p < 0.03). In the control group 28% left the hospital after 2 days. Only 35% of the patients left the hospital at the second postoperative day, but 60% of these patient supported this shorter hospitalization. Health-related quality of life and self-rated health was significantly higher in patients leaving the hospital on the second postoperative day. A 2-day hospital stay after thyroid surgery is possible and does not show medical or health-related quality of life disadvantages in patients with an uncomplicated postoperative course who consider themselves healthy.

Does size really matter--using a decision tree approach for comparison of three different databases from the medical field of acute appendicitis.

Decision trees have been successfully used for years in many medical decision making applications. Transparent representation of acquired knowledge and fast algorithms made decision trees one of the most often used symbolic machine learning approaches. This paper concentrates on the problem of separating acute appendicitis, which is a special problem of acute abdominal pain, from other diseases that cause acute abdominal pain by use of an decision tree approach. Early and accurate diagnosing of acute appendicitis is still a difficult and challenging problem in everyday clinical routine. An important factor in the error rate is poor discrimination between acute appendicitis and other diseases that cause acute abdominal pain. This error rate is still high, despite considerable improvements in history-taking and clinical examination, computer-aided decision-support, and special investigation such as ultrasound. We investigated three databases of different size with cases of acute abdominal pain to complete this task as successful as possible. The results show that the size of the database does not necessary directly influence the success of the decision tree built on it. Surprisingly we got the best results from the decision trees built on the smallest and the biggest database, where the database with medium size (relative to the other two) was not so successful. Despite this we were able to produce decision tree classifiers that were capable of producing correct decisions on test data sets with accuracy up to 84%, sensitivity to acute appendicitis up to 90%, and specificity up to 80% on the same test set.