# Interpreting Clinical Pathology Reference Intervals in Laboratory Animals


## Key Takeaways

- Reference intervals, typically defined as the central 95% of values from a healthy reference population, are crucial for interpreting clinical pathology data in laboratory animals. These intervals are derived using statistical methods, ideally nonparametric, with a minimum of 120 individuals for robust determination, and are specific to species, strain, age, sex, husbandry, and analytical method.
- Partitioning by sex and age is frequently necessary, as significant differences in hematologic and biochemical analytes exist between males and females (e.g., higher hemoglobin in male Sprague-Dawley rats) and change predictably with maturation (e.g., increasing RBC indices and decreasing WBC count with age in beagles).
- Breed and strain genetic backgrounds can lead to clinically meaningful variations in analyte values (e.g., microcytosis in Abyssinian cats, differing leukocyte counts in inbred mouse strains), necessitating strain- or breed-specific intervals to avoid misinterpretation.
- Reference intervals are analyzer-specific and subject to temporal drift; values generated on one instrument or from historical populations may not be valid for current use, requiring validation against local populations and contemporary analytical platforms.
- Distinguishing treatment effects from physiologic variation or analytical artifacts requires understanding the reference population's characteristics, the statistical methodology, and the expected biologic variation for the analyte, with serial monitoring and reference change values offering more sensitive detection of meaningful individual changes.

---

Clinical pathology data from laboratory animals are interpreted against reference intervals that describe the dispersion of values in healthy individuals. A reference interval is typically defined as the central 95% of values from a healthy reference population, calculated according to international recommendations such as those published by the Clinical and Laboratory Standards Institute and the American Society for Veterinary Clinical Pathology [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/)(https://pubmed.ncbi.nlm.nih.gov/19737162/). For the veterinary researcher, the practical question is not whether a value falls inside a printed range, but whether the comparator population resembles the animal under evaluation in species, strain, age, sex, husbandry, and analytical method.

This article provides a framework for selecting, validating, and applying reference intervals in common laboratory species including rats, mice, rabbits, and beagles. It addresses how strain, age, sex, and breed modify hematologic and biochemical values, how to detect reference intervals that do not fit a study population, and how to use biologic variation to distinguish meaningful changes from analytical noise. The intended reader is a veterinary researcher or clinical pathologist who must interpret individual animal results or group data in toxicology, safety assessment, and investigative studies.

The central diagnostic reasoning task is to determine whether an observed value represents a treatment effect, a physiologic variant, or an artifact of the reference interval itself. This distinction requires knowledge of the reference population from which the interval was derived, the statistical method used, and the expected biologic variation for the analyte in question.

## At a Glance

| Parameter | Decision or Fact |
|---|---|
| Reference interval definition | Central 95% of healthy reference population values, per Clinical and Laboratory Standards Institute C28-A3 and ASVCP guidelines [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/) |
| Minimum reference sample size | 120 individuals for a priori nonparametric determination, smaller sets require transference or validation [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/) |
| Sex partitioning | Required when male and female distributions differ clinically, demonstrated in Sprague-Dawley rats for most hematologic and biochemical analytes [Sex-specific reference intervals of hematologic and biochemical analytes in](https://pubmed.ncbi.nlm.nih.gov/29261747/) |
| Age effects | RBC, hemoglobin, hematocrit, MCV, and platelet count increase with age in laboratory beagles, WBC count decreases [Reference intervals, intraindividual and interindividual variability, and reference change](https://pubmed.ncbi.nlm.nih.gov/25651086/) |
| Breed effects | Breed-specific intervals matter when differences cause misinterpretation, such as microcytosis in Abyssinian cats or high creatinine in Holy Birmans [Hematological and biochemical reference intervals of four feline breeds](https://pubmed.ncbi.nlm.nih.gov/23906706/) |
| Analyzer specificity | Reference intervals are instrument-specific, values from one analyzer cannot be assumed valid for another [Canine reference intervals for the Sysmex XT-2000iV hematology analyzer](https://pubmed.ncbi.nlm.nih.gov/21884184/) |
| Temporal drift | Hematology values in healthy cattle changed between 1957 and 2001, including increased neutrophil counts and N:L ratio [Comparison of bovine hematology reference intervals from 1957 to](https://pubmed.ncbi.nlm.nih.gov/20070646/) |
| Biologic variation | Intraindividual and interindividual variation determine whether a serial change is clinically significant, expressed as reference change values [Reference intervals, intraindividual and interindividual variability, and reference change](https://pubmed.ncbi.nlm.nih.gov/25651086/) |

## The Statistical Basis of Reference Intervals

International recommendations specify a nonparametric rank percentile method for reference interval calculation when at least 120 reference individuals are available [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/)(https://pubmed.ncbi.nlm.nih.gov/19737162/). The nonparametric approach makes no assumption about the underlying distribution of data, which matters because many hematologic and biochemical analytes are not normally distributed. The 2.5th and 97.5th percentiles define the interval, and 90% confidence intervals around each limit express the precision of the estimate.

When fewer than 120 animals are available, the reference limits become imprecise and alternative strategies are required. Transference validates an existing interval by measuring a smaller number of healthy individuals and confirming that a defined proportion, typically 90%, falls within the established range. This approach is acceptable under international guidelines but requires that the analytical platform and preanalytical conditions match those used to generate the original interval [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/).

The selection of reference individuals is the most critical step. Inclusion and exclusion criteria must be documented extensively, and the animals must represent the population to which the interval will be applied. A reference interval derived from young, specific-pathogen-free rats cannot be applied to aged animals with subclinical disease or to a different outbred stock without validation.

## Sex as a Partitioning Variable

Sex-specific reference intervals are required when male and female distributions differ to a clinically relevant degree. In a study of 500 healthy Sprague-Dawley rats, 250 of each sex, most hematologic and biochemical analytes were significantly influenced by sex [Sex-specific reference intervals of hematologic and biochemical analytes in](https://pubmed.ncbi.nlm.nih.gov/29261747/)(https://pubmed.ncbi.nlm.nih.gov/29261747/). Males had higher hemoglobin, hematocrit, red blood cell count, and red cell distribution width than females. These differences are large enough that a single combined interval would misclassify normal females as anemic or normal males as polycythemic.

The decision to partition by sex should be based on statistical testing followed by clinical judgment. A statistically significant difference does not always justify separate intervals if the overlap is substantial and the clinical interpretation would not change. Conversely, analytes with modest statistical differences may still warrant partitioning when the reference limits shift enough to alter diagnostic decisions.

In laboratory beagles, the only clinically relevant sex effect among hematologic variables was lower platelet counts in females [Reference intervals, intraindividual and interindividual variability, and reference change](https://pubmed.ncbi.nlm.nih.gov/25651086/)(https://pubmed.ncbi.nlm.nih.gov/25651086/). This finding illustrates that sex partitioning is analyte-specific and species-specific. The researcher should not assume that a sex effect documented in one species or strain applies to another.

## Age and Maturation Effects

Age-related changes in hematologic parameters follow predictable patterns that must be accounted for when selecting reference intervals. In laboratory beagles aged 9 to 36 months, red blood cell count, hemoglobin, hematocrit, MCV, MCH, red cell distribution width, and platelet count increased with age, while white blood cell count decreased [Reference intervals, intraindividual and interindividual variability, and reference change](https://pubmed.ncbi.nlm.nih.gov/25651086/). These changes reflect ongoing bone marrow maturation and immune system development during the first years of life.

The practical consequence is that a reference interval established in young adult animals may misclassify older animals as polycythemic or thrombocytotic. Conversely, applying an adult interval to juvenile animals can produce spurious leukocytosis or anemia. Age-stratified intervals are preferable when study designs span developmental periods, and the researcher should verify that the reference population matches the age range of the study animals.

Weight had no effect on any hematologic analyte in the beagle study, which simplifies reference interval selection for dogs within a normal body condition range [Reference intervals, intraindividual and interindividual variability, and reference change](https://pubmed.ncbi.nlm.nih.gov/25651086/). This finding does not extend to other species without verification, and strain-specific body weight differences in rodents may still influence blood volume and dilutional effects during collection.

## Breed, Strain, and Genetic Background

Breed-specific reference intervals matter when genetic background produces clinically meaningful differences in analyte values. In a comparison of four feline breeds, Abyssinian cats showed microcytosis and high alpha-2-globulin concentrations, while Holy Birman cats had high serum creatinine, alpha-2-globulin, and glucose concentrations [Hematological and biochemical reference intervals of four feline breeds](https://pubmed.ncbi.nlm.nih.gov/23906706/)(https://pubmed.ncbi.nlm.nih.gov/23906706/). Norwegian Forest and Siberian cats showed high alkaline phosphatase activity and elevated calcium and phosphate concentrations.

The same logic applies to laboratory species. Inbred mouse strains differ in baseline leukocyte counts, serum chemistry values, and coagulation parameters. A reference interval established in C57BL/6 mice may not apply to BALB/c or outbred CD-1 mice. The researcher must match the reference population to the study strain or validate the interval using a small number of healthy animals from the specific strain.

Breed differences do not always require separate intervals. In the feline study, most new breed-specific intervals overlapped with published intervals, and the use of breed-specific data would minimally affect clinical interpretation for most analytes [Hematological and biochemical reference intervals of four feline breeds](https://pubmed.ncbi.nlm.nih.gov/23906706/). The decision to generate breed-specific intervals should focus on analytes where misinterpretation would lead to incorrect diagnostic conclusions, such as interpreting breed-related microcytosis as iron deficiency or breed-related high creatinine as renal disease.

## Analytical and Temporal Variability

Reference intervals are analyzer-specific. A study establishing canine hematologic reference intervals for the Sysmex XT-2000iV analyzer found that impedance and optical methods for red blood cell and platelet counts correlated well but differed significantly [Canine reference intervals for the Sysmex XT-2000iV hematology analyzer](https://pubmed.ncbi.nlm.nih.gov/21884184/)(https://pubmed.ncbi.nlm.nih.gov/21884184/). Platelet aggregation was also scored, highlighting that preanalytical variables affect results even when the analyzer is standardized.

Temporal drift adds another layer of complexity. Comparison of bovine hematology reference intervals from 1957 and 2001 in healthy Holstein cows showed that mean neutrophil counts increased significantly while lymphocyte, monocyte, and eosinophil counts decreased [Comparison of bovine hematology reference intervals from 1957 to](https://pubmed.ncbi.nlm.nih.gov/20070646/)(https://pubmed.ncbi.nlm.nih.gov/20070646/). The neutrophil to lymphocyte ratio increased to 1.17. Changes in genetics, management, subclinical disease prevalence, and analytical methods all contribute to such drift.

The researcher should therefore treat published reference intervals as starting points instead of fixed truths. Validation against a local healthy population, using the same analyzer and collection protocol, is the standard approach. When local validation is not feasible, the interval should be transferred from a source that matches the study population in species, strain, age, sex, and analytical platform [Reference values: a review](https://pubmed.ncbi.nlm.nih.gov/19737162/).

## Building a Species-Specific Interpretation Framework

The practical application of reference intervals in laboratory animal medicine begins with a simple question: does the interval being used actually describe the population from which the patient was drawn? For a laboratory beagle, intervals derived from pet dogs may mislead, particularly for platelet counts where sex-specific differences have been documented in laboratory colonies. For a Sprague-Dawley rat, intervals that ignore sex will obscure meaningful differences in red cell mass and biochemical analytes. The interpreter must therefore verify three features before applying any interval: the source population, the analytical platform, and the statistical method used to generate the limits.

### Verification Sequence for Interval Applicability

A structured verification sequence reduces the risk of misinterpretation. First, confirm that the reference population matches the patient for species, strain or breed, age class, sex, and husbandry conditions. Second, confirm that the analytical platform and methodology match those used to generate the interval. Hematology analyzers vary in their measurement principles, and intervals generated on one instrument may not transfer to another. Third, confirm that the interval was generated using current statistical recommendations, ideally the nonparametric method with at least 120 reference individuals. When an interval fails any of these checks, the interpreter should either locate a more appropriate interval or use the interval with explicit caveats documented in the record.

### Decision Points That Change Interpretation

Several clinical scenarios require the interpreter to shift from population-based intervals to alternative comparators. In toxicology studies, concurrent control animals provide the primary comparator, and reference intervals serve a secondary role. When a study uses a small number of animals, the statistical imprecision of reference limits becomes substantial, and the interpreter should rely more heavily on control group data and on the magnitude of change relative to baseline values.

Age-related changes demand particular attention. In laboratory beagles, red blood cell count, hemoglobin, hematocrit, MCV, MCH, red cell distribution width, and platelet count increase with age, while white blood cell count decreases. Applying a single interval across a wide age range will misclassify young and old animals. The interpreter should use age-stratified intervals when available, or at minimum track serial values within an individual to detect trends that cross the interval boundary.

## Reference Interval Decision Table for Common Laboratory Species

The following table summarizes key interpretive considerations for the most common laboratory species. It is not a substitute for species-specific published intervals, but rather a decision aid for selecting and applying appropriate comparators.

| Species | Primary Partitioning Variables | Common Pitfall | Recommended Approach |
|---------|-------------------------------|----------------|----------------------|
| Mouse (inbred strains) | Strain, sex, age | Strain differences in leukocyte counts and serum proteins are common, pooled intervals obscure them | Use strain-specific intervals when available, document strain in all records |
| Rat (outbred stocks) | Sex, age, stock | Sex differences in red cell mass and biochemistry are substantial | Use sex-stratified intervals, verify stock (e.g., Sprague-Dawley vs Wistar) |
| Rabbit (laboratory breeds) | Age, sex, breed | Young rabbits have higher alkaline phosphatase, breed differences exist | Use age-matched intervals, confirm breed origin of interval |
| Non-human primate (macaques) | Species, age, sex, origin | Captive vs wild-caught differences, seasonal variation | Use colony-specific intervals generated from the same facility |
| Beagle dog | Age, sex, colony | Platelet counts differ by sex, age affects red cell indices | Use colony-derived intervals, verify age range |

### Applying the Table in Practice

The table guides the interpreter toward the correct partitioning strategy before a result is judged abnormal. For a 6-month-old female Sprague-Dawley rat with a hemoglobin value at the lower end of a combined-sex interval, the interpreter should re-evaluate against a female-specific interval before concluding that anemia is present. For a 24-month-old laboratory beagle with a platelet count below the published interval, the interpreter should check whether the interval was derived from younger animals, since platelet count increases with age in this population.

## Breed and Strain Effects on Interpretation

Breed-specific intervals matter in species where genetic bottlenecks have produced distinctive phenotypes. In cats, breed-related differences have been documented for red cell indices, serum proteins, creatinine, glucose, and alkaline phosphatase activity. Some of these differences, such as microcytosis in Abyssinian cats, could lead to misinterpretation if a general feline interval is applied. The same principle extends to laboratory species, where inbred mouse strains and purpose-bred dog colonies show heritable differences in hematologic and biochemical parameters.

For laboratory beagles, the practical question is whether intervals derived from one colony transfer to another. The available evidence indicates that weight has no effect on hematologic variables, but age and sex effects are clinically relevant. A colony-derived interval that accounts for these variables provides the most reliable comparator for studies using the same population.

## Temporal and Methodological Variability

Reference intervals are not static. Changes in analytical platforms, reagent lots, and laboratory procedures can shift results without any change in the animal population. The interpreter should periodically verify that intervals remain valid for the current analytical system. This is particularly important when a laboratory changes its hematology analyzer, since instrument-specific intervals are required for reliable interpretation.

Historical comparisons also reveal that animal populations themselves change over time. In Holstein cattle, mean neutrophil counts increased and lymphocyte counts decreased between 1957 and 2001, producing a significant shift in the neutrophil to lymphocyte ratio. This finding demonstrates that intervals published decades ago may not describe current populations, even when the breed and production system appear unchanged. The interpreter should prefer intervals generated from contemporary populations and should be cautious when relying on historical published values.

## Documentation and Reporting Standards

The veterinary record should document which reference interval was used, its source population, and the statistical method by which it was generated. When an interval is transferred from another laboratory or another instrument, the verification process should be recorded. This documentation supports defensible interpretation and allows subsequent reviewers to assess whether the chosen comparator was appropriate.

For research studies, the [Guide for the Care and Use of Laboratory Animals](https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf) emphasizes the importance of veterinary oversight and accurate health monitoring. Clinical pathology data form part of this monitoring, and the reference intervals used for interpretation should be documented in study records. When intervals are generated de novo, the selection of reference individuals must follow documented inclusion and exclusion criteria, and the analytical procedures must be quality controlled.

The [MSD Veterinary Manual](https://www.msdvetmanual.com/) provides species-specific clinical context that supports interpretation of laboratory results in context. The interpreter should use such resources to understand how a given analyte change relates to clinical disease, instead of relying on the reference interval alone to determine clinical significance.

## Handling Uncertainty and Borderline Results

A result that falls just outside the reference interval does not necessarily indicate disease. The 95% interval by definition excludes 5% of healthy individuals, so approximately one in twenty healthy animals will have a value outside the interval for any given analyte. The interpreter should consider the magnitude of the deviation, the presence of concurrent abnormalities, and the clinical status of the animal before concluding that a result is clinically significant.

Serial monitoring provides additional interpretive power. A value that remains stable across multiple time points, even if slightly outside the interval, carries different weight than a value that has shifted markedly from an individual's baseline. Reference change values, which account for both analytical and biological variation, offer a statistically grounded method for detecting meaningful changes in serial samples. Where such values are available, they should be used in preference to a simple comparison with the population interval.

## Recognized Failure Modes and Early Detection

Reference intervals fail in predictable ways. The most common failure is silent transference, where an interval generated on one analyzer, anticoagulant, or population is applied to another without verification. This produces systematic shifts that mimic treatment effects. Detect it by running a small verification set of 20 to 40 healthy animals from the local colony and comparing the proportion of results falling outside the established interval. More than 5% outside the central 95% interval, or any result outside the extremes, triggers formal revalidation per the transference procedures described in [veterinary reference value guidelines](https://pubmed.ncbi.nlm.nih.gov/19737162/).

Preanalytical error is the second major failure mode. Hemolysis, delayed separation, prolonged tourniquet time, and sample storage temperature each alter specific analytes. Potassium, lactate dehydrogenase, and aspartate aminotransferase rise with hemolysis. Glucose falls rapidly without fluoride oxalate. Platelet counts decline with storage. Detect these by reviewing sample quality indicators, including serum color, clot retraction, and time from collection to analysis, before interpreting any result.

The third failure mode is cohort drift. A breeding colony changes over time, through genetic drift, subclinical disease, or dietary reformulation. Intervals validated two years ago may no longer describe the current population. Monitor this with quarterly quality control samples from a stable sentinel group and by tracking the proportion of healthy animals whose results fall near interval boundaries.

## Common Errors in Interpretation

Less experienced clinicians frequently compare a single abnormal value against the interval without considering the magnitude of change from that animal's own baseline. A creatinine rising from 40 to 90 micromol per liter in a rabbit remains within many published intervals but represents a doubling of renal workload. Reference change values, calculated from intraindividual and interindividual variability, provide a more sensitive threshold for meaningful change than population intervals alone, as demonstrated in [laboratory beagle hematology studies](https://pubmed.ncbi.nlm.nih.gov/25651086/).

Another common error is ignoring the direction of sex-specific partitioning. Male rats have higher hemoglobin, hematocrit, and red cell counts than females, and applying a combined interval can mask mild anemia in males or falsely flag normal females as polycythaemic. The sex-specific intervals established for Sprague-Dawley rats illustrate the magnitude of these differences and the need for partition-specific interpretation.

Students also overinterpret single borderline values. A single analyte marginally outside the interval in an otherwise healthy animal is usually a false positive, given that the central 95% interval by definition excludes 5% of healthy individuals. Repeat the measurement, assess the trend, and interpret the full panel instead of the isolated value.

| Observation | Likely cause | Discriminating check |
|---|---|---|
| Systematic shift in potassium and AST | Hemolysis | Inspect serum color, repeat on fresh sample |
| Low glucose in all samples | Delayed separation | Verify time from collection to centrifugation |
| Platelet count falling with storage | EDTA-induced clumping or time delay | Recollect, analyze within 2 hours |
| Creatinine above interval in one sex only | Sex partitioning omitted | Apply sex-specific intervals |
| 10% of healthy animals outside interval | Interval transference without validation | Run 20 to 40 healthy animals, compare proportions |

## Limitations of Current Evidence

The evidence base for laboratory animal reference intervals remains uneven. Many published intervals derive from fewer than the recommended 120 reference individuals, and the resulting limits are imprecise. The [review of reference value methodology](https://pubmed.ncbi.nlm.nih.gov/19737162/) notes that small sample sizes produce reference limits with wide confidence intervals, and this imprecision propagates into every subsequent interpretation.

Expert opinion differs on whether strain-specific intervals are necessary or whether species-level intervals with breed or strain adjustment suffice. Feline breed studies show that most breed-specific intervals overlap with published values and that only a few analytes, such as red cell indices in Abyssinians, require separate intervals. The same pattern likely applies to laboratory rodents, but the data are sparse. Until strain-specific intervals are available for common strains such as C57BL/6 and BALB/c, use species-level intervals with caution and verify against local colony data.

Temporal drift adds another layer of uncertainty. Bovine hematology intervals changed substantially between 1957 and 2001, with neutrophil counts rising and lymphocyte counts falling. Whether similar shifts occur in laboratory rodents over decades of breeding is unknown, but the possibility argues for periodic revalidation instead of indefinite use of historical intervals.

## Escalation and Consultation

Refer to a veterinary clinical pathologist when results are persistently abnormal without a corresponding clinical explanation, when the pattern suggests a preanalytical or analytical artefact that local troubleshooting cannot resolve, or when reference interval verification fails repeatedly. Clinical pathologists can advise on transference, calculate reference change values, and recommend alternative statistical approaches for small populations.

Involve the laboratory when analyzer-specific issues are suspected. Different hematology analyzers produce different intervals for the same species, as shown in [canine Sysmex XT-2000iV reference interval studies](https://pubmed.ncbi.nlm.nih.gov/21884184/), and instrument-specific intervals are required for accurate interpretation.

Regulatory reporting applies when abnormal clinical pathology findings indicate an unexpected adverse effect in a toxicology study, a potential zoonotic agent, or a welfare concern requiring protocol modification. Institutional animal care and use committees and attending veterinarians should be notified according to the [Guide for the Care and Use of Laboratory Animals](https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf) and institutional policies. Where findings suggest a notifiable disease, consult the [WOAH terrestrial animal health standards](https://www.woah.org/en/what-we-do/standards/codes-and-manuals/terrestrial-code-online-access/) for reporting obligations in your jurisdiction.

## Frequently Asked Questions

### How Should I Proceed When My Facility Cannot Recruit 120 Healthy Animals for a Valid Reference Interval?

When fewer than 120 reference individuals are available, the nonparametric method becomes unreliable and reference limits may be highly imprecise. Use the transference or validation approach instead. Select a published interval from a comparable population, then run a small verification sample of 20 to 40 healthy animals from your colony. If fewer than 10% of results fall outside the published limits, the interval is acceptable for your use. Document the verification process and the source of the original interval. This approach follows international recommendations for veterinary clinical pathology and is preferable to calculating an unstable interval from a small sample.

### What Is the Most Practical Way to Handle Analyte Values That Fall Just Outside the Reference Interval?

A single value marginally outside the interval does not establish disease. Evaluate the magnitude of the deviation relative to the width of the interval, the direction of change, and the pattern across related analytes. For example, an isolated 2% increase in alanine aminotransferase activity in a rat with normal bile acids and no histologic lesion is unlikely to be biologically meaningful. Repeat the measurement after 24 to 48 hours to assess persistence. Compare the result with the reference change value if one has been published for that analyte and species, as this accounts for intraindividual and interindividual variability better than the population interval alone.

### How Do I Explain a Breed-Specific or Strain-Specific Result to a Supervisor Who Expects a Single Universal Interval?

Present the evidence that genetic background alters clinical pathology values. In cats, breed-specific intervals have been established because breeds such as Abyssinians show microcytosis and high alpha-2 globulin concentrations, while Holy Birmans have higher creatinine, glucose, and alpha-2 globulin values. In laboratory dogs, age and sex affect several hematologic analytes, with platelet counts lower in females. A single universal interval would misclassify these healthy animals as abnormal. Frame the discussion around diagnostic accuracy: using the appropriate partition reduces false positives and false negatives, which protects study data integrity and animal welfare.

### What Records Should I Keep When Validating or Transferring a Reference Interval for My Colony?

Maintain a validation file that includes the source of the original interval, the statistical method used, the number and characteriztics of verification animals, the date of sampling, the analyzer and reagent lot numbers, and the proportion of results falling outside the interval. Record any deviations from the source population, such as differences in age range, sex distribution, or housing conditions. Update this file whenever the analyzer is serviced, reagents change, or the colony genetic background is altered. This documentation supports regulatory review and provides a defensible basis for interpretation during study audits.

### How Do I Interpret Results When My Laboratory Uses a Different Analyzer Than the One Used to Establish the Published Interval?

Instrument-specific intervals are required because analyzers differ in measurement principles. For example, impedance and optical methods for red blood cell and platelet counts correlate well but produce significantly different values. If your analyzer differs from the published source, do not assume the interval transfers. Run a method comparison using 20 to 40 fresh samples analyzed on both instruments if possible, or verify the published interval on your analyzer using healthy animals. If verification fails, contact the instrument manufacturer for analyzer-specific intervals or collaborate with a reference laboratory that uses the same platform.

### What Are the Cost-Effective Options for Establishing Intervals When Commercial Reference Laboratory Services Are Unavailable?

Prioritize analytes most relevant to your study endpoints instead of generating a full panel. For rodents, a limited hematology profile covering red blood cell count, hemoglobin, white blood cell count, and platelet count, combined with a small biochemistry set including glucose, creatinine, alanine aminotransferase, and total protein, covers most toxicology and efficacy endpoints. Use in-house analyzers with validated quality control materials and calibrators. Where analyzer purchase is not feasible, consider sharing equipment with a collaborating institution or using a referral laboratory with established intervals for your species. The [Guide for the Care and Use of Laboratory Animals](https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf) emphasizes that veterinary care decisions should be based on sound diagnostic information, and the [MSD Veterinary Manual](https://www.msdvetmanual.com/) provides species-specific interpretive guidance that can supplement limited local data.

## Related Clinical & Scientific Guides

* [Refining IACUC Protocols to Minimize Animal Pain and Distress](/knowledge/veterinary-medicine/laboratory-animal-science/refining-iacuc-protocols-minimize-animal-pain-distress)
* [Health Monitoring Programs for Laboratory Animal Facilities](/knowledge/veterinary-medicine/laboratory-animal-science/health-monitoring-programs-for-laboratory-animal-facilities)
* [Anesthetic Risk Assessment in Laboratory Animals: Preoperative Evaluation](/knowledge/veterinary-medicine/laboratory-animal-science/anesthetic-risk-assessment-in-laboratory-animals-preoperative-evaluation)


## References and Further Reading

- [Sex-specific reference intervals of hematologic and biochemical analytes in Sprague-Dawley rats using the nonparametric rank percentile method.](https://pubmed.ncbi.nlm.nih.gov/29261747/). 2017.
- [Reference intervals, intraindividual and interindividual variability, and reference change values for hematologic variables in laboratory beagles.](https://pubmed.ncbi.nlm.nih.gov/25651086/). 2015.
- [Canine reference intervals for the Sysmex XT-2000iV hematology analyzer.](https://pubmed.ncbi.nlm.nih.gov/21884184/). 2011.
- [Hematological and biochemical reference intervals of four feline breeds.](https://pubmed.ncbi.nlm.nih.gov/23906706/). 2014.
- [Reference values: a review.](https://pubmed.ncbi.nlm.nih.gov/19737162/). 2009.
- [Comparison of bovine hematology reference intervals from 1957 to 2006.](https://pubmed.ncbi.nlm.nih.gov/20070646/). 2010.
- [Guide for the Care and Use of Laboratory Animals, 8th Edition](https://grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf). National Academies Press, 2011.
- [NC3Rs Resources on Replacement, Reduction and Refinement](https://www.nc3rs.org.uk/). NC3Rs.
- [MSD Veterinary Manual, Professional Edition](https://www.msdvetmanual.com/). MSD Veterinary Manual.

## Related Articles

- [Anesthesia for Laboratory Rabbits: Protocols and Monitoring](/knowledge/veterinary-medicine/laboratory-animal-science/anesthesia-for-laboratory-rabbits-protocols-and-monitoring)
- [Anesthesia Machine Safety Checks for Laboratory Animal Use](/knowledge/veterinary-medicine/laboratory-animal-science/anesthesia-machine-safety-checks-laboratory-animal-use)
- [Anesthesia Monitoring Parameters for Laboratory Animals](/knowledge/veterinary-medicine/laboratory-animal-science/anesthesia-monitoring-parameters-for-laboratory-animals)
- [Welfare Assessment Tools for Laboratory Rodents](/knowledge/veterinary-medicine/laboratory-animal-science/welfare-assessment-tools-for-laboratory-rodents)
- [Anesthesia Equipment for Small Laboratory Animals: Setup and Maintenance](/knowledge/veterinary-medicine/laboratory-animal-science/anesthesia-equipment-for-small-laboratory-animals-setup-and-maintenance)

> This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.