Home / Clinical Services / IVD Performance Studies
ISO 20916 · IVDR Articles 57–77 · CLSI · India MDR 2017

IVD Performance Studies for IVDR, FDA and CDSCO

REGHELPS runs analytical and clinical performance studies for in vitro diagnostic devices as a full-service CRO: study design and sample size, sample sourcing, ethics and regulatory approvals, testing against reference methods, data management, statistics and the final study report that feeds your Performance Evaluation Report.

Get a performance study proposal See how we run studies
2study types: analytical and clinical
Art. 58sets when authorisation is needed
CLSI EPguidelines for analytical studies
PERwhere every study result ends up
The basics

What are IVD performance studies?

IVD performance studies are planned studies that generate the analytical and clinical performance evidence an in vitro diagnostic needs for market approval. Under IVDR, they feed the Performance Evaluation Report alongside scientific validity data.

Analytical performance studies show the test correctly detects or measures the analyte. Clinical performance studies show its results correlate with the clinical condition in the intended-use population, usually by comparing results against a reference method or clinical diagnosis.

Need the regulatory documents too? Our IVDR performance evaluation page covers the PEP, PER and SSP that these studies support.
A
Analytical evidencePrecision, trueness, sensitivity, specificity, interference and stability.
C
Clinical evidenceDiagnostic sensitivity, specificity and predictive values in real patients.
✓
ISO 20916 compliantPlans, ethics, monitoring and reports to good study practice.
↻
Built for every marketOne study design supporting IVDR, FDA and CDSCO where possible.
Study types

Types of IVD performance studies

The right study type depends on your device, claims and the samples you can obtain.

Study typeWhat it showsSamplesTypical approval route (EU)
Analytical performance studyPrecision, trueness, LoD, interference, linearity, stabilityContrived, spiked or leftover samplesUsually no IVDR authorisation; laboratory study
Clinical performance study (leftover samples)Diagnostic sensitivity and specificityLeftover or biobank samplesEthics and national rules; notification if companion diagnostic
Prospective observational studyPerformance in routine useFreshly collected samples, non-invasiveEthics and national rules
Interventional clinical performance studyPerformance when results guide patient managementProspective samplesAuthorisation under Article 58(1)
Companion diagnostic studySelection of patients for a specific medicineProspective or leftover samplesAuthorisation or notification, plus medicines authority link
Self-test lay-user studyLay users can perform and read the test correctlyLay participantsEthics and national rules
Decision flow

Which approval route does your IVD study need?

A simplified view of IVDR Article 58. National rules add requirements, so we confirm the route for every country in your study.

Question 1Can leftover or archived samples answer your clinical questions?
YesLeftover sample study: ethics and national rules. Notify under Article 58(2) if it is a companion diagnostic.
NoGo to question 2.
Question 2Will the study results influence how patients are managed?
YesInterventional clinical performance study: authorisation under Article 58(1).
NoGo to question 3.
Question 3Are samples taken by surgically invasive procedures only for the study, or are there other added risks for subjects?
YesAuthorisation under Article 58(1) before the study starts.
NoProspective observational study under ethics and national rules.
Analytical performance

Analytical performance studies and guidelines

IVDR Annex I sets the characteristics to address. CLSI EP guidelines are the most widely accepted way to design the studies.

CharacteristicWhat it showsCommon guideline
PrecisionRepeatability and reproducibility across runs, days, lots and sitesCLSI EP05
Trueness and method comparisonAgreement with a reference methodCLSI EP09
Limits of blank, detection and quantitationLowest amount reliably detected or measuredCLSI EP17
Analytical specificityInterference and cross-reactivityCLSI EP07
Linearity and measuring rangeRange over which results are reliableCLSI EP06
Qualitative test performanceAgreement for positive and negative resultsCLSI EP12
StabilityReagent shelf life, in-use, transport and specimen stabilityCLSI EP25
Clinical performance

Clinical performance measures we report

01
Diagnostic sensitivityProportion of true positives the test correctly identifies.
02
Diagnostic specificityProportion of true negatives the test correctly identifies.
03
Predictive valuesPositive and negative predictive values at the expected prevalence.
04
Likelihood ratiosHow much a result changes the probability of the condition.
05
Expected valuesResults in normal and affected populations.
06
Confidence intervalsPrecision of every estimate, justified by sample size.
By market

IVD performance study requirements in the EU, USA and India

Planning for all target markets from the start avoids repeating studies later.

TopicEuropean UnionUnited StatesIndia
Legal basisIVDR Articles 57–77, Annexes XIII and XIV21 CFR 812, 21 CFR 50 and 56Medical Devices Rules 2017
Good practice standardISO 20916:2019ISO 20916 and FDA guidance; CLSI widely usedISO 20916 and Indian GCP
ApprovalAuthorisation or notification under Article 58 where required; ethics reviewIRB; IDE unless exempt under 812.2(c)(3)Form MD-24 application, Form MD-25 permission; ethics review
OutputStudy report feeding the PER and SSPData for 510(k), De Novo or PMAData for the CDSCO import or manufacturing licence
Process flow

From study design to Performance Evaluation Report

How a clinical performance study moves from idea to regulatory evidence.

Design and sample sizeEndpoints, reference method, population and statistical plan.
Study plan and approvalsClinical performance study plan, ethics and regulatory approvals.
Sample sourcingSites, biobanks or prospective collection in the intended-use population.
Testing and dataBlinded testing against the reference method, with data management.
Report and PERStatistical analysis, study report and update of the PER and SSP.
↻Post-market performance follow-up (PMPF) keeps the evidence current after certification.
How we run studies

Our IVD performance study process

REGHELPS manages your performance studies as a full-service CRO, with one team from first design to final report.

  1. 1

    Evidence gap analysis

    Strategy

    We review your claims, existing data and target markets to define the studies you need.

  2. 2

    Study design and statistics

    Design

    We set endpoints, reference methods, acceptance criteria and sample size.

  3. 3

    Study plan and documents

    ISO 20916

    We write the study plan, consent forms, case report forms and lab procedures.

  4. 4

    Approvals

    Ethics · regulators

    We manage ethics submissions and Article 58, FDA or CDSCO MD-24 applications.

  5. 5

    Sites, labs and samples

    Sourcing

    We qualify sites, laboratories and biobanks and secure the right samples.

  6. 6

    Study conduct and monitoring

    Execution

    We train sites, monitor testing and protect blinding and sample traceability.

  7. 7

    Data management

    Quality

    We clean and lock the data with full audit trails.

  8. 8

    Statistics and study report

    Report

    Our biostatisticians analyse results and we write the final study report.

  9. 9

    PER and submission support

    Close the loop

    We integrate results into your PER and SSP and support Notified Body, FDA or CDSCO questions.

Avoid findings

Common IVD performance study pitfalls

01
Wrong populationSamples that do not reflect the intended-use population.
02
Weak reference methodComparator not accepted as the state of the art.
03
Unjustified sample sizeConfidence intervals too wide to support the claims.
04
Broken blindingTesters aware of reference results.
05
Missed approvalsArticle 58 authorisation or notification overlooked.
06
Legacy data reused blindlyIVDD studies used without a gap assessment.
Our services

Our IVD performance study services

Analytical performance studiesDesign, protocols and reports to CLSI guidelines.
Clinical performance studiesLeftover, prospective and interventional studies to ISO 20916.
Companion diagnostic studiesStudy design aligned with the medicine’s development.
Self-test and lay-user studiesUsability and reading studies with lay participants.
Sample sourcing and site managementSites, laboratories and biobanks for the right population.
Statistics and study reportsAnalysis and reports ready for your PER and submissions.
FAQ

IVD performance study FAQs

What is the difference between IVD performance evaluation and a performance study?

Performance evaluation is the overall process of assessing scientific validity, analytical performance and clinical performance, documented in the Performance Evaluation Report (PER). A performance study is one of the ways to generate that evidence: a planned study, analytical or clinical, run on specimens or subjects.

Which standard applies to IVD clinical performance studies?

ISO 20916:2019 sets good study practice for clinical performance studies using specimens from human subjects. It covers planning, ethics, documentation, monitoring and reporting, and is the reference standard Notified Bodies expect under IVDR.

When does an IVDR performance study need authorisation?

Under IVDR Article 58(1), authorisation is needed when samples are taken by surgically invasive procedures only for the study, when the study is interventional (results may influence patient management), or when it involves additional invasive procedures or other risks for subjects. Companion diagnostic studies using only leftover samples must be notified under Article 58(2).

Can leftover or archived samples be used?

Yes. Many clinical performance studies use leftover or biobank samples, which avoids extra burden on patients. They still need ethics approval and data protection controls under national rules, and the samples must represent the intended-use population.

How many samples does a clinical performance study need?

It depends on the target sensitivity and specificity, the expected prevalence and the confidence intervals your claims need. We calculate the sample size statistically and justify it in the study plan, following CLSI and MDCG guidance.

Which analytical performance studies are typically required?

Precision, trueness, limits of blank, detection and quantitation, analytical specificity (interference and cross-reactivity), linearity and measuring range, cut-off, method comparison and stability. The exact set depends on the assay type and claims, usually designed to CLSI EP guidelines.

Do I need permission to run an IVD performance study in India?

Yes, for a new in vitro diagnostic device. Under the Medical Devices Rules 2017, the manufacturer applies to the Central Licensing Authority on Form MD-24 and receives permission on Form MD-25, with ethics committee approval.

Does the FDA require an IDE for IVD studies?

Many IVD studies are exempt from the full IDE requirements under 21 CFR 812.2(c)(3), for example when sample collection is non-invasive and results are confirmed by an established method. IRB review and informed consent rules still apply. Studies that do not meet the exemption need an IDE.

Planning an IVD performance study?

Share your assay, claims and target markets using the form below and we'll propose a study design and budget.

Request a study proposal

Get a free quote

Read our Privacy Policy.