Reference Range Harmonization: 4 Checks Before Labs Share an Interval

Reference range harmonization aligns clinical interpretation across different measurement procedures when SI-traceable standardization is not available. The governing references are CLSI EP28, IFCC/ICHCLR guidance and the CDC Clinical Standardization Programs, including HoSt. Our recommendation: assess analytical comparability and commutability first, then choose a derivation or verification path that fits your analytes, your sites and your population.

TL;DR:

• Proceed only after analytical performance is stable, calibrator traceability is documented, sites align units and names, and external quality assessment tracks bias.
• A single site using one platform, or sites unable to demonstrate analytical comparability, should usually verify an existing interval instead.
• Direct studies provide stronger evidence but take more recruitment and time; indirect methods scale faster, yet routine data need cleaning and prospective verification.
• Set bias limits and acceptance criteria before testing; CDC HoSt uses a panel of 40 patient samples, while thresholds vary by protocol.
• Standardize collection conditions, test names, units, rounding, and flags across sites, then monitor reagent and platform changes for drift before shared intervals are affected.

What reference range harmonization is and why it matters at the bedside

Standardization ties results to a higher-order reference, such as an SI-traceable reference measurement procedure. Harmonization is the fallback when no such procedure or reference material exists: you make results, or at least their clinical interpretation, equivalent across methods by other means. The aim is equivalent clinical interpretation rather than identical numeric ranges, and harmonization often relies on mathematically derived adjustments that translate platform-specific results into one clinical recommendation.

A hormone assay with no reference measurement procedure is a typical case. Two platforms can both be well run and still report different numbers for the same specimen.

Why this matters for patients:

• A result near a cutoff can be flagged as abnormal on one platform and normal on another, which is misclassification.
• Patients who move between health systems may face unnecessary repeat testing to confirm what is really the same finding.
• Clinicians applying one guideline threshold to different methods may start or withhold treatment inconsistently.

Experts describe harmonization as a patient-safety necessity, because consistent intervals support reliable decisions across settings and platforms.

When should you pursue harmonization, and what must you check first?

Harmonization is not the right project for every analyte. Before you commit resources, confirm these prerequisites in order:

1. Stable analytical performance, shown by in-control internal quality control data over a meaningful period.
2. Traceability evidence for your calibrators and, where available, reference materials.
3. Alignment of units and nomenclature across the sites that will share intervals.
4. Active participation in an external quality assessment scheme, so you can see your bias against peers.

The situations that call for harmonization are multi-site reporting under one clinical service, planned shared decision thresholds, and measurands that lack a reference measurement procedure or reference material.

Local verification or method-specific intervals often remain the better choice when one laboratory serves a distinct population, when a single platform is used everywhere, or when analytical comparability across methods cannot be demonstrated. In those cases, a well-run verification of a manufacturer or published interval is faster and easier to defend.

Direct, indirect and algorithmic methods: which path fits your project?

Three routes cover most projects. They differ in cost, speed and the evidence they produce.

Direct reference-interval studies. CLSI EP28-A3c is the standard protocol for establishing, verifying and transferring reference intervals. It centers on well-defined healthy reference individuals and offers transfer and verification options for laboratories that cannot afford a full study. The 2022 implementation guide (EP28IG) is the key companion for laboratories verifying existing intervals.

Indirect, big-data methods. These mine routine results and use statistical models to separate the non-pathological distribution from diseased subpopulations. A CSCC-led big-data effort derived harmonized intervals for several biochemistry analytes this way and then verified them across laboratories. The main pitfalls are diseased subpopulations hiding in the data, method or reagent changes during the collection window, and unclear specimen provenance.

Harmonization algorithms. The IFCC and ISO guidance supports protocols built on commutable harmonization materials, derived algorithms that translate results between methods, and validation runs. The ICHCLR toolbox, including the Integrated Harmonization Protocol and Step-up Design, helps teams assess feasibility for analytes without international standards.

What to weigh across the three:

• Direct studies give the strongest evidence but demand the most recruitment and time.
• Indirect methods scale quickly across sites but depend on data quality and statistical choices.
• Algorithms suitcases where methods differ by a predictable bias that commutable samples can reveal.
• When no reference measurement procedure exists, alternatives include multiples of upper reference limits or procedure-specific percentiles.

Pro Tip: Pair an indirect derivation with a small prospective direct verification. The big-data step gives you a candidate interval cheaply, and the prospective check proves it holds in your own patients.

If you are mapping early operational steps, our primer on lab data harmonization covers where teams usually start.

Traceability, commutability and verification: the technical checks behind a common interval

Three concepts decide whether a shared interval is defensible.

• Traceability: documented linkage of your calibrators to the highest available reference. Operationally, collect calibrator lot documentation and the manufacturer’s traceability statement for each platform in scope.
• Commutability: whether a material behaves like a fresh patient specimen across methods. Without it, bias estimated on processed material may not apply to real samples. Test candidate materials against patient samples on every platform.
• Allowable bias and imprecision: the error budget you will tolerate before two methods can no longer share an interval. Set limits from clinical need before you look at the data.

For verification, a common approach tests a small panel of local samples, such as the 40 patient samples used in the CDC HoSt program, and accepts the proposed interval if an agreed proportion falls within its limits. Acceptance thresholds vary by protocol, so document the criterion you choose before the run.

Support all of this with outside reference points:

• EQA panels to track bias against peers over time.
• CDC HoSt panels of patient samples, which let participants assess bias and recalibrate.
• Periodic surveillance, so drift after reagent lot or instrument changes is caught early.

A stepwise checklist for implementing harmonized reference intervals

1. Scope. List the analytes and stakeholders, and write down the clinical decisions each interval drives.
2. Inventory. Record methods, units, calibrators and recent EQA results for every site. Gather commutable sample panels.
3. Choose a path. Select a direct, indirect or harmonization-algorithm route, and fix the statistical method before analysis begins.
4. Verify. Run prospective or reserve samples and record pass or fail criteria in advance, with documented outcomes.
5. Roll out and monitor. Update the LIS, brief clinicians on what changes and why, and set a surveillance schedule for lot and platform changes.

Pro Tip: Brief clinicians before the LIS change goes live. A shifted flag that surprises a physician can look like an error even when it is a correction.

Our post on analysis-ready lab data is a useful companion for step 2, especially for teams preparing routine data for indirect derivation.

What multicenter studies tell us about feasibility

Consortium work gives decision-makers an evidence base rather than a leap of faith. Across multicenter projects, harmonized intervals appear feasible for some analytes, but results depend on the analyte and methods used:

• Multicenter CSCC and IFCC-linked efforts show that harmonized intervals are feasible for many common biochemistry analytes when methods are comparable.
• Some analytes fail verification or yield overly wide intervals, so feasibility must be judged analyte by analyte.
• Prospective validation is what turns a statistically derived interval into one a laboratory can adopt with confidence.

The practical reading is straightforward: treat big-data results as strong candidates, not finished intervals, and verify before you release them to clinicians.

What verification and validation criteria should a proposed harmonized interval meet?

Verification asks whether an interval fits your local population and method. Validation asks whether it is fit for the clinical decision it supports. A proposed harmonized interval should clear both.

Start with the analytical gate. Confirm that bias between methods sits inside the allowable limit you set from clinical need, and that imprecision is stable. If a method cannot meet that gate, no statistical adjustment will rescue a shared interval.

Then apply the sample gate. Run a defined panel of local samples against the proposed limits and judge the result against a pre-declared acceptance rule. The CLSI EP28IG guidance describes verification for existing intervals, and many laboratories follow the familiar pattern of a small panel with a high proportion required inside the limits.

Add a subgroup gate. Check whether age, sex or other partitions that mattered in the source data also hold in your samples. An interval verified on a pooled adult group can still fail in children or older adults.

Finally, apply a clinical gate. Ask whether the interval would change flags on results near key decision points, and review those cases with clinical stakeholders.

Record every criterion, the data and the sign-off. If the interval fails, narrow the claim: adopt it for the analytes and populations that passed, keep method-specific intervals for the rest, and schedule a repeat assessment after any reagent or platform change.

How do pre-analytical and post-analytical factors affect harmonization?

Analytical agreement is only part of the picture. Harmonization that focuses solely on the analytical phase often falls short, because reporting, nomenclature and units also shape how clinicians read a result.

Pre-analytical factors change the numbers before the analyzer sees the sample. Fasting status, posture, time of day, tube type, additive, transport delay and storage temperature can all move results for sensitive analytes. If one site draws on ice and another does not, a shared interval may look wrong for reasons that have nothing to do with the assay. Reference samples should be collected under conditions that match routine practice, and those conditions should be written down.

Post-analytical factors change what the clinician sees. Mixed units for the same measurand, different names for the same test, and inconsistent decimal precision all create avoidable confusion. A glucose reported in two unit systems across sites, or a test with three local names, undermines a harmonized interval even when the values agree.

Practical controls include:

• A single catalog of test names and units for all participating sites.
• Standard collection and handling instructions in every site’s manual.
• LIS rules that apply the same rounding and flagging logic.
• Report layouts that show the interval, its source and the unit consistently.

Treat these items as part of the harmonization scope from step 1, not as cleanup after the statistics are done.

What challenges limit harmonization, and how can you work around them?

The common obstacles are predictable, which means they can be planned for.

Missing reference systems. Without a reference measurement procedure or reference material, you cannot anchor results. The workaround is a protocol such as the ICHCLR approach, supported by commutable materials and an agreed allowable bias.

Biased source data. Routine data can hide diseased subpopulations, and method changes can create artificial shifts. Clean the data, document the collection window, and verify prospectively.

Analytes that will not harmonize. Some produce wide or unstable intervals. Accept that and keep method-specific intervals for them, rather than forcing a shared range.

Population differences. Age, sex, ancestry and local physiology can justify separate partitions. Test for them explicitly instead of assuming a pooled interval fits.

Operational drag. Coordinating sites, LIS changes and clinician communication takes sustained effort. A named project owner, a written timeline and agreed pass or fail rules keep the work moving.

Drift over time. Reagent lots, calibrators and platforms change. Build surveillance into the plan from the start, since a harmonized interval is only as good as its latest verification.

A multi-site standardization program that treats platform choices, reporting and integration as shared levers tends to run more smoothly than one that tackles each in isolation.

What regulatory and compliance requirements apply to harmonizing reference ranges?

Requirements for establishing, verifying and documenting reported intervals may depend on your accreditation framework and jurisdiction. Confirm the applicable requirements with the body that inspects your laboratory.

Several practical expectations recur:

• Documented basis. Keep the source of each interval, whether a direct study, a transfer, a verification or a harmonization protocol, along with the data behind it.
• Change control. Treat an interval change as a controlled change, with approval, an effective date and a record of who was told.
• Quality system fit. Tie harmonized intervals into procedures, competency records and EQA review, so they are not a one-off project outside your quality system.
• Reference services. The CDC’s Clinical Reference Laboratory is accredited to ISO/IEC 17025 and ISO 15195, and CDC’s programs support using established clinical decision points when assays are certified.
• Assay claims. If you adjust a manufacturer’s interval, record the rationale and the verification, since the change sits within your own laboratory’s responsibility.

Before rollout, involve your quality manager and medical director. Their sign-off, and a clear audit trail, matter as much as the statistics.

How does harmonization change clinical decisions and patient outcomes?

The value of harmonization shows up when a result leaves the laboratory. A clinician reading a flag, comparing a value with last year’s, or applying a guideline threshold is relying on the interval being consistent.

Consider a patient followed across two health systems that use different platforms. If the intervals differ, a stable condition can look like progression or recovery simply because the flag changed. That prompts repeat testing, extra visits and sometimes unwarranted treatment changes. Shared intervals remove one source of that noise.

Harmonization also supports guideline use. Decision thresholds in clinical guidance assume results are comparable across methods. Consistent intervals enable reliable decision-making across clinical settings, which is the practical meaning of the patient-safety argument.

Be measured about outcome claims. The clearest, best-supported benefits are fewer misclassified results and better comparability of serial values. Direct links to improved clinical outcomes are harder to demonstrate and depend on the analyte and the decision involved. If you plan to claim outcome benefits internally, track something concrete: repeat-test rates, flag discordance between sites, or clinician queries about intervals, before and after rollout.

Communication is part of the effect. Clinicians who understand why a flag moved will trust it, so a short note explaining the change is part of clinical impact, not an afterthought.

What does harmonization cost, and when does it pay off?

The costs are real, and most of them are labor. Expect time for data collection and statistical analysis, sample panels and their testing, LIS build and validation, documentation, clinician communication and ongoing surveillance. Direct studies cost the most because of recruitment and testing; indirect and transfer approaches reduce that burden but need careful data curation.

The benefits are harder to put on a ledger but concrete. Fewer discordant flags between sites means fewer repeat tests and fewer clinician queries. One interval to maintain, rather than several method-specific ones, simplifies LIS configuration and training. Results become easier to pool for research and multi-site care.

A simple way to decide is to compare scope against need:

• A single-site laboratory on one platform may get more value from verifying an existing interval than from a harmonization project.
• A multi-site network or a shared-care service has the strongest case, because discordance there has a visible cost.
• Analytes that rarely sit near a decision threshold are lower priority than those that drive treatment.

To keep spend under control, start with a small set of high-impact analytes, reuse existing EQA and routine data, and use a laboratory partner for panels and logistics rather than building that capacity in-house. Track a few measures from day one, such as discordant flags and repeat-test rates, so the return can be shown rather than assumed.

How a laboratory partner can support your harmonization project

Harmonization projects stall on logistics more often than on statistics. Samples have to be collected, shipped, tested on comparable methods and returned as clean data, usually across several sites.

We support that work with centralized testing, specimen logistics, LIS and data integration. Running testing through one laboratory gives your team a consistent dataset to analyze. It also frees your staff to focus on analyte selection, acceptance criteria and clinician communication.

This support can help your team plan its verification work. If you are scoping a project, a short planning conversation can clarify which analytes to start with, what panel design fits, and how results should be delivered.

— Kohealth Labs

Laboratory and research support for harmonization and verification projects

We support pathology, clinical trial laboratory services, specimen logistics and LIS integration for harmonization and verification projects. Whether you need a method-comparison panel, centralized testing for a multicenter study or a clean data feed into your existing systems, our team works with yours from scoping to delivery.

Explore our pathology laboratory services or our clinical trials laboratory services, then request a technical scoping call or specimen panel coordination through KoHealth Labs. We will help you plan the next step.

FAQ

What is the mean reference range?

A reference range is the interval of results expected in a defined healthy reference population. It describes where most healthy people fall, not a pass or fail line. It guides interpretation alongside the clinical picture.

What does interval mean in medical terms?

An interval is the span between a lower and an upper limit that results are compared against. In laboratory medicine it usually refers to the reference interval for a test. CLSI EP28-A3c is the standard protocol for establishing and verifying such intervals.

What are the normal reference ranges for blood tests?

There is no single universal range, because intervals depend on the method, the population and sometimes age and sex. Always interpret a result against the interval printed on your own report. Harmonization efforts aim to make clinical interpretation equivalent across methods, not to make every number identical.

What does “biological reference interval” mean?

It is the reference interval derived from the biological variation of a defined, healthy population, as opposed to a manufacturer’s claim or a clinical decision limit. Laboratories establish it directly or verify a transferred one. Pre-analytical conditions should match routine practice when the interval is set.

Sources

• Pubmed
• About Clinical Standardization Programs | CSP | CDC

For more information about our diagnostic testing, clinical research laboratory services, or partnership opportunities, visit https://kohealthlabs.com or contact KoHealth Labs at (469) 792-8439 and ask for Amy Patterson.