On the use of patient data for the definition of reference intervals in clinical chemistry

Separating disease and health for indirect reference intervals

The indirect approach to defining reference intervals operates ‘a posteriori’, on stored laboratory data. It relies on being able to separate healthy and diseased populations using one or both of clinical techniques or statistical techniques. These techniques are also fundamental in a priori, direct reference interval approaches. The clinical techniques rely on using clinical data that is stored either in the electronic health record or within the laboratory database, to exclude patients with possible disease. It depends on the investigators understanding of the data and the pathological impacts on tests. The statistical technique relies on identifying a dominant, apparently healthy, typically Gaussian distribution, which is unaffected by the overlapping populations with higher (or lower) results. It depends on having large databases to give confidence in the extrapolation of the narrow portion of overall distribution representing unaffected individuals. The statistical issues involved can be complex, and can result in unintended bias, particularly when the impacts of disease and the physiological variations in the data are under appreciated.

Statistical methods used in the calculation of geriatric reference intervals: a systematic review

Geriatric reference intervals (RIs) are not commonly available and are rarely used. It is difficult to select a reference population from a cohort with a high degree of morbidity. Also important are the statistical approaches used to determine health-associated reference values. It is the aim of this study to examine the statistical methods used in the calculation of geriatric RIs.

A search was conducted on EMBASE and Medline for articles between January 1989 and January 2014. Studies were selected if they: 1) were English primary articles; 2) performed a clinical chemistry test on a blood fraction; 3) had a population sub-group consisting of individuals ≥65 years of age; and 4) calculated a RI for the subgroup ≥65 years of age.

There were 64 articles identified, of which 78.1% described the RI calculation method used. RI calculation was performed by non-parametric (21.9%), parametric (42.2%), robust (3.1%), or other (17.2%) methods. Outlier detection (SD, Grubb’s test, Tukey’s fence, Dixon) was infrequently used and although most studies performed partitioning, only 57.8% tested the statistical significance of the partitions. Few studies (17.2%) reported confidence intervals for the RI estimates. Overall, only 14.1% of studies provided RI estimates which followed the CLSI guideline EP28-A3c.

Statistical methods for RI calculation and partitioning varied considerably between studies and many failed to provide adequate descriptions of these methods. Challenges in analyses arose from insufficient sample sizes and heterogeneity in the elderly population. Geriatric RIs, although present in the literature, may not be properly calculated and should be carefully considered before applying them for clinical care.

Verifying reference intervals for coagulation tests by using stored data

OBJECTIVE

The purpose of the study was to verify the reference intervals for prothrombin time (PT) and activated partial thromboplastin time (APTT), using stored data of ambulatory pre-op subjects with exclusion of certain clinics, according to age and sex.

MATERIALS AND METHODS

Results of test requests (13,600 PT and 14,083 APTT) of subjects aged 15?80 made from outpatient clinics of surgical departments before surgical interventions in 2008 were retrieved from the electronic medical record. Thromborel S and Actin (Dade Behring, Germany) were used on the Sysmex® CA-1500 coagulation analyzer. Extreme values were determined by using Horn's algorithm after Box-Cox transformation, and the upper and lower reference limits were determined as the 2.5th and 97.5th percentiles of the cleaned data.

RESULTS

The values outside the interval of PT data 10.5-17.0 seconds and the interval of APTT data 20.6-35.8 seconds were excluded from the analysis. There were significant differences among age subsets of PT measurements ( p < 0.0001) and of APTT measurements ( p < 0.0001). Accordingly, the data were tested for gender differences and a significant difference was found in PT ( p = 0.002). APPT results did not differ statistically between men and women.

CONCLUSION

Although we found values different from the limits stated in the kit insert, it would be better to confirm our findings with the direct method, especially in APTT for patients under the age of 40 and over the age of 59, and also for sex differences in PT.

Reference intervals: the way forward

New facts have recently enhanced interest in the topic of reference intervals. In particular, the International Organization for Standardization standard 15189, requesting that 'biological reference intervals shall be periodically reviewed', and the directive of the European Union on in vitro diagnostic medical devices asking manufacturers to provide detailed information on reference intervals, have renewed interest in the subject. This review presents an update on the topic, discussing the theoretical aspects and the most critical issues. The basic approach to the definition of reference intervals proposed in the original International Federation of Clinical Chemistry documents still remain valid. The use of data mining to obtain reference data from existing databases has severe limitations. New statistical approaches to discard outliers and to compute reference limits have been recommended. On the other hand, perspectives opened by the improvement in standardization through the implementation of the concept of traceability suggest new models to define 'common' reference intervals that can be transferred and adopted by different clinical laboratories in order to decrease the proliferation of different reference intervals not always justified by differences in population characteristics or in analytical methodology.

Estimation of individual reference intervals in small sample sizes

In occupational health studies, the study groups most often comprise healthy subjects performing their work. Sampling is often planned in the most practical way, e.g., sampling of blood in the morning at the work site just after the work starts. Optimal use of reference intervals requires that the population, on which the reference interval is based, is representative for the study group in question. The International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) recommends estimating reference interval on at least 120 subjects. It may be costly and difficult to gain group sizes of that order of magnitude for all topics in question. Therefore, new methods to estimate reference intervals for small sample sizes are needed. We present an alternative method based on variance component models. The models are based on data from 37 men and 84 women taking into account biological variation from various variables such as gender, age, BMI, alcohol, smoking, and menopause. The reference intervals were compared to reference intervals calculated using IFCC recommendations. Where comparable, the IFCC calculated reference intervals had a wider range compared to the variance component models presented in this study. The presented method enables occupational health researchers to calculate reference intervals for specific groups, i.e. smokers versus non-smokers, etc. In conclusion, the variance component models provide an appropriate tool to estimate reference intervals based on small sample sizes.