The Technical Guideline For The Application Of Real-world .

2y ago
32 Views
2 Downloads
236.74 KB
15 Pages
Last View : 22d ago
Last Download : 2m ago
Upload by : Warren Adams
Transcription

The Technical Guideline for the Application of Real-world Datain Clinical Evaluation of Medical Devices(Draft for Comments)1 / 15

Contents1. Overview . 41.1 Real-world Data & Evidences . 41.2 Advantages & Disadvantages of Real-world Data . 52. Common Resources and Classification of Real-world Data . 53. Evaluation of Real-world Data . 64. Design Principles, Common Types & Statistical Analysis Techniques of Real-worldData Research . 74.1 Design Principle of Real-world Data Research. 74.2 Common Types of Real-world Data Research Design . 74.2.1 Effectiveness Test . 74.2.2 Observational Research . 84.2.3 Other Design Types. 94.3 Statistical Analysis Techniques . 95. Common Situations in which Real-world Data Can be Considered for ClinicalEvaluation of Medical Devices . 105.1 Provide Clinical Evidences in Congeneric Clinical Evaluation Process . 105.2 Support Medical Device Registration as Supplementary Evidences . 105.3 Real-world Data from Chartered Medical Devices Can be Used asSupplementary Evidences in Registration. 105.4 External Reference Materials in Single Unit Test . 115.5 Provide Clinical Data for Setting Target Value of Single Unit . 115.6 Provide Basis for Revise of Application Area and Indications . 115.7 Provide Basis for Revise of Clinical Statement in IFU . 115.8 Support Post-marketing Research of Registered Products with Conditions . 115.9 Long-term Safety & Effectiveness Evaluation for Medical Devices with High-riskImplants and Others . 125.10 Life-cycle clinical evaluation of medical devices used to treat rare diseases . 12-speed up their market launch, and meet patient needs . 125.11 Post-market Surveillance (PMS) . 12Appendix 1: Common Statistical Analysis Methods . 121. Statistical Methods for Pragmatic Randomized Controlled Trials(pRCT). 122 / 15

2. Statistical analysis methods commonly used in observational study . 132.1 Stratified analysis . 132. 2 Multivariate regression models . 142.3 Propensity Score Analysis Method . 142.4 Instrumental Variable Analysis Method . 153 / 15

The Technical Guideline for the Application of Real-world Data inClinical Evaluation of Medical Devices(Draft for Comments)In order to solve the problems affecting innovation, quality and efficiency of medical devices,accelerate the modernization of the medical device management system and improvegovernance capacity, NMPA launched “Scientific Method of Medicine Supervision of China”,which is based on the reality of China and focused on the reform and innovation of thereview and approval system. This guideline is based on the findings of the first project ofthe action plan-" Research on How to Apply Real-world Data in Clinical Evaluation ofMedical Device Scientifically".This guideline aims at standardizing and guiding the application of real-world data in clinicalevaluation of medical devices, providing technical guidance for the application of real-worlddata in registrations and evaluation of medical devices.This guideline is a technical guidance documents for applicants and reviewers, soadministrative process like registration approval is not involved. This guideline is notcompulsory. Any other methods that can meet the requirements of laws and regulationscan be adopted too. The guideline is based on the existing scientific power and cognitivelevel and it will be improved and revised as the development of real-world data researchand changes of relevant laws, policies and standards.1. Overview1.1 Real-world Data & EvidencesThe real-world data in this guideline refers to data from real medical cases and itsresources are not traditional clinical trials. These data can reflect the health status ofpatients and the medical service process in the actual diagnosis and treatment.Real-world data research involves real-world data, many different disciplines, multidisciplinary methods and techniques in Epidemiology, Biostatistics and Evidence-BasedMedicine. The research process includes collecting real-world data in clinical conventionsand using rational epidemiological design and statistical analysis under pre-establishedresearch assumptions. It can be future-oriented or conclusion-oriented and it completestraditional clinical trials. The vast amount of clinical data generated by clinical practiceswhich is supported by information-technology laid a solid foundation for real-world dataresearch.Real-world evidences refer to clinical evidences related to product use and potentialrisks/benefits, which are based on the analysis of real-world data. However, due to thedifferent sources and types of real-world data, data quality and information they covered4 / 15

vary greatly, not all real-world data are applicable to clinical evaluation of medical devices.In the context of related requirements, real-world evidence can show the risk and benefitfeatures of medical devices in the whole service life, which may help in decision-making insupervision.1.2 Advantages & Disadvantages of Real-world DataWith fewer restrictions on patients’ conditions, real-world data studies have a widercoverage of the population than traditional clinical trials. Real-world data studies are carriedout in a realistic health care environment and its research conclusions are easy to use inspeculation. Real-world research emphasizes the comprehensive use of data fromdifferent resources, for instance, hospital electronic medical records, enrollment data,regional health care data, medical insurance data, etc. This means that we can get clinicalconclusion data on a long-term basis. Real-world studies can also be used in observingrare serious adverse events, answering questions about rare diseases, and evaluatingdifferences in clinical outcomes among different people.There are two disadvantages of real-world research. Firstly, there are many data sourcesin the real-world. In collecting and storing data, measurement/classification errors or datamissing are common occurrences. The data is not structured and the data quality needsto be evaluated. Connection between different data sources is lame. Secondly, studiesbased on real-world data often have many biases and confusions, and the results is difficultto use in stipulation.2. Common Resources and Classification of Real-world DataCommon sources of real-world data include registry databases, hospital electronic medicalrecords, regional health care data, medical insurance data, health records, routine publicsurveillance data, self-reported patient data (including home environments), and data fromother health tests (e.g. mobile devices). Real-world data suitable for medical devices alsoinclude data from whole service life of a medical device (manufacturing, marketing,transportation, storage, installation, use, maintenance, retirement and disposal), e.g.acceptance report, maintenance record, user feedback, operation environment, calibrationrecords, performance log, original image, etc.).In terms of clinical evaluation of certain devices, the above data sources can be dividedinto two categories according to the relationship between time and research launch time.The first type is the existing data resources, it means the data resources already existedwhen the research carried out. Based on the differences in the data generating process,this type of data resource can be divided into two cases as bellow:a)First type of data came from the process of providing health care services andpayment, such as electronic medical record, medical insurance records, healthrecords, etc.5 / 15

b)Second type of data refers to database which is established in an orderly manner andit is based on certain research purpose, data standard and data collection method.For instance, registration data, database based on the effectiveness clinical trials.The second type is for the purpose of clinical evaluation on certain devices, it usually hasclear data standards and data collection models. In short, these data are established in anorderly manner. Representative data includes devices registration data and effectivenessclinical trials data.3. Evaluation of Real-world DataThe quality of real-world data directly affects the reliability of evidence from real-worldresearch. So, evaluation of real-world data is the foundation of carrying out real-worldresearch. In addition, corresponding reliability assurance system and evaluation measuresneed to be established and implemented. To evaluate the data, researchers are requiredto consider reliability of data source and control data quality throughout the whole researchprocess.Firstly, in using the data from existing data sources for clinical evaluation, it is necessaryfor researchers to evaluate whether the existing data includes the all targeted population,key variables and follow-up duration. Researchers should also consider the accuracy andintegrity of medical device identification information, IFU, etc. Secondly, in specificresearch design, researchers should filter and extract existing data. In this way, we can geta real-world database.During the data collection, firstly, we need to consider the rationality and feasibility of theresearch design. There are many inclusion and exclusion criteria in real-world data, weneed to ensure quality of follow-up investigation and its time. Secondly, it is necessary toensure the authenticity, accuracy and traceability of data, establish systematic follow-upstandards, then, train and supervise researchers. It is important to identify the possibleresearch biases and confounding factors before taking action. At the same time, we shouldmeasure and record relevant confounding factors in the preparation process. To controland minimize the effects of confounding factors, we can use hierarchical analysis, multifactor analysis and tendentiousness score in data analysis.For standardize evaluation of data management, we can consider factors like managementprocesses, personnel, information systems, and data standardization. In term of datastandardization, we should establish a standardized document format and data structure,in addition, we need a standardized variable dictionary.In evaluation, we should consider the relevance and reliability of the data. In terms ofrelevance between data and the research, we should evaluate sufficiency of individualvariable and consider whether it can adequately answer the clinical questions base on theresearch purpose; in terms of adequacy, we should put our focus on accuracy of data6 / 15

collection which includes collection scope, variables, data dictionary, and collectionmethods (such as data extraction tables). In this way, we can minimize errors, and ensurethe authenticity and integrity of data.4. Design Principles, Common Types & Statistical Analysis Techniques ofReal-world Data Research4.1 Design Principle of Real-world Data ResearchIn the clinical evaluation of medical devices, we should use the real-world data in a practicalmanner and design the research based on the specific research purpose. The overallresearch plan includes clarifying research questions, confirming data sources, determininghow data is generated, filtering data and forming research teams. The overall researchdesign includes determining the design type, clarifying the research objects and researchvariables, identifying the source of confounding factors, conducting reasonable control,and formulating a statistical analysis plan. In planning, designing and implementing realworld research, we should also pay full attention to ethical factors and data security.We should pay attention to biases in using real-world data, no matter what type it is. Thismay limit the reasoning and interpretations of research results. To minimize potentialbiases, researchers need to identify them during the planning, design, implementation, andanalysis stages of the research, and formulate corresponding measures in advance to forma research plan and analysis plan carefully before acquiring and analyzing data. Applicantscan choose different research designs according to their needs. They can also choosedifferent research designs at the same time if it is necessary.4.2 Common Types of Real-world Data Research DesignThe types of real-world data research can be mainly divided into 3 types: Effectivenesstest, observational research and other design types.4.2.1 Effectiveness TestpRCT (Pragmatic Randomized Controlled Trials) is typical in interventional research and itis also an important part of effectiveness test. pRCT means using random or contrastingmethods in real or less-real medical environment, and compare results of differentintervention methods. Being different from traditional contrast tests, pRCT are usuallyconducted in real clinical environment--the research object may have comorbidities, theintervention processes is highly similar to real clinical case, however, the processes areinfluenced by skills or the experiences of intervenors. Therefore, we should consider theresearch planning and designing comprehensively.Because of different choices of groups, pRCT usually includes the following test types:a) Individual pRCT: It means that individual is intervened and monitored as one randomgrouping unit. In clinical evaluation, the unit means the patient.7 / 15

b)c)Group pRCT: It means that the group (such as hospitals, clinics, districts and schools)is considered as one intervention unit. In analyzing the real-world data, we shouldconsider both group effect and evaluate group members case by case.Hierarchical pRCT: It is a special type of pRCT, group members accept interventionrandomly in batches at different time.Selection of outcome indicators of pRCT is based on research targets which includes safety,effectiveness, treatment adherence, healthy economics and so on. Generally, the clinicalindicators which have important clinical meaning for patients (or the user of this researchresult) will be chose as outcome indicators. Intermediate indicators like biology indicatorsand imaging science indicators are often excluded. If the intervention cannot be donerandomly, we recommend you to chose outcome indicators such as stroke and tumor sizewhich are not influenced easily by intervention. In common cases, the sample calculationshould be done according to specific research design.Generally, we choose routine treatments, standard treatments and effective treatments ascontrasting group, placebo effect treatments are not applicable. The evaluation of longterm outcome is the focus of pRCT, so we should conduct multi-time outcomemeasurement. Accordingly, the follow-up time should be longer and the follow-upfrequency should be lower than routine pRCT.The focus of pRCT is mainly intervention effects in real clinical case, however, the researchenvironment and conditions must be combined with disease features and clinical reality forthe final judgement. The research objects should be typical in patient groups.4.2.2 Observational ResearchIn observational research which is based on real-world data, the real-world data qualityvaries because of different data resources. The measurement of outcome and exposuremay different from research definition. Generally, doctors allocate different treatments topatients, so it is not arbitrary. The conclusion is, it is an important to recognize and controlconfounding biases in design and analyzing stage in observational research. Otherwise,biases may limit reasoning. In using real-world data in observational research, to clarifyrelationship between exposure and outcomes, we suggest researchers to considercarefully about key factors and procedures and then, design research plan and statisticalanalysis plan.There are 3 steps for designing the observation research in real-world,①confirming theresearch purpose ②resolving the problems of the devices occurred in clinical evaluation③clarifying the research hypotheses. While establishing research hypotheses, the keyfactors of research should be emphasized, it means that if P (Population), I (Intervention),C (Control), O (Outcome), T (Timing) can be generated based on the data from real-world,including①if the data of study population meeting the requirements will be picked up fromthe data of real-world. ② if the unified intervention plan or standard intervention plan will8 / 15

be made, ③if the comparable contrast can be set,④if the outcome indicators andmeasure results which are necessary for research will be included.Observation research including array research (prospective, retrospective, deque), Casecontrol study and derivative design (Nested case-control study), as well as self-contrastcase series and other design types. The applicator can choose property study design aswell as the other designs at the same time, according to research purpose and the featuresof data from real-world. The applicator should have a comprehensive identification forpossible biases during the process (selection bias, measurement bias, etc.) ,establisheffective measures to control measures.4.2.3 Other Design TypesUsing real-world data as external control for single unit trial is a design type, from whichwe can get clinical evidences. Among them, the historical control is usually not comparablebecause of the differences in clinical practice, the changes in follow-up time, and the lackof consistency in the standards of diagnosis and outcome measurement. It is an effectiveway to improve these limitations by selecting the contemporaneous control rather than thehistorical control and collecting the detailed and accurate information of relevant variables.4.3 Statistical Analysis TechniquesIn real-world data research, researchers need to apply reasonable statistical methodsbased on the research purpose, data type and research design type. See Appendix 2 forcommon statistical analysis met

The Technical Guideline for the Application of Real-world Data in . Design Principles, Common Types & Statistical Analysis Techniques of Real-world Data Research 4.1 Design Principle of Real-world Data Research In the clinical evaluation of medical devices, we should use the real-world data in a practical .

Related Documents:

May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)

Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .

On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.

̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions

Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have

Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original

Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được

10 tips och tricks för att lyckas med ert sap-projekt 20 SAPSANYTT 2/2015 De flesta projektledare känner säkert till Cobb’s paradox. Martin Cobb verkade som CIO för sekretariatet för Treasury Board of Canada 1995 då han ställde frågan