Utilization of five data mining algorithms combined with simplified preprocessing to establish reference intervals of thyroid-related hormones for non-elderly adults

Feasibility evaluation of big data algorithms for establishing serum protein electrophoresis reference intervals using Hoffmann and refineR methods

BACKGROUND

Serum protein electrophoresis (SPE) is essential for diagnosing monoclonal gammopathies and a variety of other diseases. Despite its importance, there is a scarcity of SPE parameter reference intervals (RIs) derived from large datasets. This study seeks to fill this gap by establishing sex-specific RIs using Hoffmann and refineR algorithms and assessing the feasibility of these methods.

METHOD

We utilized two health check-up population databases to create a reference and a validation set. The reference set included 52,293 individuals with outlier removal via the Tukey method. Variance component analysis was used to evaluate the impact of sex and age on SPE parameters. The Hoffmann and refineR algorithms were applied to establish RIs, with the bias ratio method comparing RI differences. Validation data from differing timeframes helped assess the RIs' reliability and utility. Moreover, we juxtaposed our RIs with previous studies to identify potential disparities.

RESULTS

Sex-specific RIs were established using the Hoffmann and refineR algorithms, with high consistency between the two algorithms, except for a slight difference in upper limits (ULs) of α1-globulin and β1-globulin. Additionally, there are sex differences in RIs for α2-globulin, β2-globulin, γ-globulin, and Albumin. In the validation analysis, all established RIs passed verification, except for RIs determined by refineR for female α1-globulin and male β2-globulin.

CONCLUSION

This research highlights the critical impact of sex on SPE RIs and the need for tailored RIs in distinct clinical environments. The utility and feasibility of the Hoffmann and refineR algorithms for creating these customized intervals are effectively demonstrated.

Multidisciplinary approach to redefining thyroid hormone reference intervals with big data analysis

OBJECTIVES

This study aimed to employ big data analysis to harmonize reference intervals (RI) for thyroid function tests, with refinement to the TSH upper reference limit, and to optimize the TSH reflex algorithm to improve clinical management and test utilization.

DESIGN & METHODS

TSH, free T4, and free T3 results tested in Alberta, Canada, on Roche Cobas and Siemens Atellica were extracted from the laboratory information system (N = 1,144,155 for TSH, N = 183,354 for free T4 and N = 92,632 for free T3). Results from specialists, inpatients, or repeat testing, as well as from positive thyroid disease, autoimmune disease, and pregnancy biomarkers were excluded. RIs were derived using statistical models (Bhattacharya, refineR, and simple non-parametric) followed by endocrinology and laboratory review.

RESULTS

The TSH RIs for 0 to 7 days, 8 days to 1 year, and ≥1 year were 1.23 to 25.0 mIU/L, 1.00 to 6.80 mIU/L and 0.20 to 6.50 mIU/L, respectively. The free T4 RIs for 0 to 14 days, 15 to 29 days, and ≥30 days were 13.5 to 50.0 pmol/L, 8.7 to 32.5 pmol/L, and 10.0 to 25.0 pmol/L, respectively. An updated TSH reflex algorithm was developed based on the optimized TSH and free T4 RIs, with free T4 reflexed only at a TSH of <0.1 mIU/L.

CONCLUSIONS

The collaboration of a multidisciplinary team and the utilization of big data analysis led to the enhancement of thyroid function RIs, specifically resulting in the widening of the upper TSH reference limit to 6.50. Application of these optimized RIs with the TSH reflex algorithm will serve as a guide for improvement in interpretation of thyroid function tests.

Age-Specific Reference Intervals for Thyroid-Stimulating Hormones and Free Thyroxine to Optimize Diagnosis of Thyroid Disease

Background: Thyroid-stimulating hormone (TSH) and subsequent free thyroxine (FT4) concentrations outside the reference interval (RI) are used to diagnose thyroid diseases. Most laboratories do not provide age-specific RIs for TSH and FT4 beyond childhood, although TSH concentrations vary with age. Therefore, we aimed to establish TSH and FT4 age-specific RIs throughout life and aimed to determine whether using these RIs would result in reclassification of thyroid disease diagnoses in adults. Methods: This multicenter retrospective cross-sectional study used big data to determine indirect RIs for TSH and FT4. These RIs were determined by TMC and refineR-analysis, respectively, using four different immunoassay platforms (Roche, Abbott, Siemens, and Beckman Coulter). Retrospective data (2008-2022) from 13 Dutch laboratories for general practitioners and local hospitals were used. RIs were evaluated per manufacturer. Age groups were established from 2 to 20 years by 2-year categories and decade categories between 20 and 100 years. Results: We included totally 7.6 million TSH and 2.2 million FT4 requests. TSH upper reference limits (URLs) and FT4 lower reference limits were higher in early childhood and decreased toward adulthood. In adulthood, TSH URLs increased from 60 years in men, and from 50 years in women, while FT4 URLs increased from 70 years onward. Using adult age-specific RIs resulted in a decrease in diagnoses of subclinical and overt hypothyroidism in women above 50 and men above 60 years in our Roche dataset. Conclusion: This study stressed the known importance of using age-specific RIs for TSH and FT4 in children. This study also showed the clinical relevance of using age-specific RIs for TSH in adulthood to reduce diagnoses of subclinical hypothyroidism in older persons. Therefore, implementation of adult TSH age-specific RIs should be strongly considered. Data are less uniform regarding FT4 age-specific RIs and more research should be performed before implementing these in clinical practice.

Pre-analytical stability and physiological fluctuations affect plasma steroid hormone outcomes: A real-world study

Since steroids are crucial for diagnosing endocrine disorders, the lack of research on factors that affect hormone levels makes interpreting the results difficult. Our study aims to assess the stability of the pre-analytical procedure and the impact of hormonal physiological fluctuations using real-world data. The datasets were created using 12,418 records from individuals whose steroid hormone measurements were taken in our laboratory between September 2019 and March 2024. 22 steroid hormones in plasma by a well-validated liquid chromatography tandem mass spectrometry method were measured. After normalization transformation, outlier removal, and z-score normalization, generalized additive models were constructed to evaluate preanalytic stability and age, sex, and sample time-dependent hormonal fluctuations. Most hormones exhibit significant variability with age, particularly steroid hormone precursors, sex hormones, and certain corticosteroids such as aldosterone. 18-hydroxycortisol, 18-oxocortisol. Sex hormones varied between males and females. Levels of certain hormones, including cortisol, cortisone, 11-deoxycortisol, 18-hydroxycortisol, 18-oxocortisol, corticosterone, aldosterone, estrone, testosterone, dihydrotestosterone, dehydroepiandrosterone sulfate, 11-ketotestosterone, and 11-hydroxytestosterone, fluctuated with sampling time. Moreover, levels of pregnenolone and progesterone decreased within 1 hour of sampling, with pregnenolone becoming unstable with storage time at 4 degrees after centrifugation, while other hormone levels remained relatively stable for a short period of time without or after centrifugation of the sample. This is the first instance real-world data has been used to assess the pre-analytic stability of plasma hormones and to evaluate the impact of physiological factors on steroid hormones.

Insulin reference intervals in Brazilian adolescents by direct and indirect approaches: validation of a data mining method from laboratory data

OBJECTIVE

To determine reference intervals (RI) for fasting blood insulin (FBI) in Brazilian adolescents, 12 to 17 years old, by direct and indirect approaches, and to validate indirectly determined RI.

METHODS

Two databases were used for RI determination. Database 1 (DB1), used to obtain RI through a posteriori direct method, consisted of prospectively selected healthy individuals. Database 2 (DB2) was retrospectively mined from an outpatient laboratory information system (LIS) used for the indirect method (Bhattacharya method).

RESULTS

From DB1, 29345 individuals were enrolled (57.65 % female) and seven age ranges and sex partitions were statistically determined according to mean FBI values: females: 12 and 13 years-old, 14 years-old, 15 years-old, 16 and 17 years-old; and males: 12, 13 and 14 years-old, 15 years-old, 16 and 17 years-old. From DB2, 5465 adolescents (67.5 % female) were selected and grouped according to DB1 partitions. The mean FBI level was significantly higher in DB2, on all groups. The RI upper limit (URL) determined by Bhattacharya method was slightly lower than the 90 % CI URL directly obtained on DB1, except for group female 12 and 13 years old. High agreement rates for diagnosing elevated FBI in all groups on DB1 validated indirect RI presented.

CONCLUSION

The present study demonstrates that Bhattacharya indirect method to determine FBI RI in adolescents can overcome some of the difficulties and challenges of the direct approach.

Calculation of reference intervals for the concentrations of α-tocopherol and retinol in serum using indirect data-mining procedures

BACKGROUND

Establishing adequate reference intervals (RIs) for vitamins A and E is essential for diagnosing and preventing deficiencies. Due to the current boom in data mining and its easy applicability, more laboratories are establishing RIs using indirect methods. Our study aims to obtain RIs using four indirect data-mining procedures (Bhattacharya, Hoffmann, Kosmic, and RefineR) for vitamins A and E.

MATERIAL AND METHODS

8943 individuals were collected to establish the RIs. After using different data cleaning steps and checking whether these data should be divided according to age and gender based on multiple linear regression and variance component analyses, indirect RIs were calculated using specific Excel spreadsheets or R-packages software.

RESULTS

A total of 2004 records were eligible. For vitamin A, the RIs obtained were (1.11 - 2.68) µmol/L, (1.13 - 2.70) µmol/L, (1.13 - 2.71) µmol/L, and (1.17 - 2.66) µmol/L using the Bhattacharya, Hoffmann, Kosmic and RefineR approaches, respectively. For vitamin E, these intervals were (17.3 - 49.9) µmol/L (Bhattacharya), (17.3 - 48.9) µmol/L (Hoffmann), (19.6 - 50.3) µmol/L (Kosmic), and (19.4 - 50.9) µmol/L (RefineR). In all cases, the RIs were comparable.

CONCLUSIONS

Suitable RIs for vitamins A and E were calculated using four indirect methods that are suitable and adapted to our population's demographic characteristics.

Development of next-generation reference interval models to establish reference intervals based on medical data: current status, algorithms and future consideration

Evidence derived from laboratory medicine plays a pivotal role in the diagnosis, treatment monitoring, and prognosis of various diseases. Reference intervals (RIs) are indispensable tools for assessing test results. The accuracy of clinical decision-making relies directly on the appropriateness of RIs. With the increase in real-world studies and advances in computational power, there has been increased interest in establishing RIs using big data. This approach has demonstrated cost-effectiveness and applicability across diverse scenarios, thereby enhancing the overall suitability of the RI to a certain extent. However, challenges persist when tests results are influenced by age and sex. Reliance on a single RI or a grouping of RIs based on age and sex can lead to erroneous interpretation of results with significant implications for clinical decision-making. To address this issue, the development of next generation of reference interval models has arisen at an historic moment. Such models establish a curve relationship to derive continuously changing reference intervals for test results across different age and sex categories. By automatically selecting appropriate RIs based on the age and sex of patients during result interpretation, this approach facilitates clinical decision-making and enhances disease diagnosis/treatment as well as health management practices. Development of next-generation reference interval models use direct or indirect sampling techniques to select reference individuals and then employed curve fitting methods such as splines, polynomial regression and others to establish continuous models. In light of these studies, several observations can be made: Firstly, to date, limited interest has been shown in developing next-generation reference interval models, with only a few models currently available. Secondly, there are a wide range of methods and algorithms for constructing such models, and their diversity may lead to confusion. Thirdly, the process of constructing next-generation reference interval models can be complex, particularly when employing indirect sampling techniques. At present, normative documents pertaining to the development of next-generation reference interval models are lacking. In summary, this review aims to provide an overview of the current state of development of next-generation reference interval models by defining them, highlighting inherent advantages, and addressing existing challenges. It also describes the process, advanced algorithms for model building, the tools required and the diagnosis and validation of models. Additionally, a discussion on the prospects of utilizing big data for developing next-generation reference interval models is presented. The ultimate objective is to equip clinical laboratories with the theoretical framework and practical tools necessary for developing and optimizing next-generation reference interval models to establish next-generation reference intervals while enhancing the use of medical data resources to facilitate precision medicine.

Comparison of results and age-related changes in establishing reference intervals for CEA, AFP, CA125, and CA199 using four indirect methods

•The reference intervals calculated using RefineR, Kosmic, TMC, and non-parametric methods are similar.•TMC algorithm is more robust, demonstrates a high pass rate among the four methods and has the ability to automatically isolate outliers.•The reference intervals of CA125 and CA199 showed significant differences between age and sex.