| 1. |
- Gellerstedt, Martin, 1966-, et al.
(författare)
-
Partitioning reference values for several subpopulations using cluster analysis
- 2007
-
Ingår i: Clinical chemistry and laboratory medicine. - : Walter de Gruyter. - 1434-6621 .- 1437-4331. ; 45:8, s. 1026-1032
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: A crucial question when developing reference intervals is whether different subpopulations need their own reference interval or if a single joint reference interval can be used. It is reasonable to use partitioned reference intervals in situations where a single interval results in considerable variation in sensitivity between subpopulations. The aim of partitioning is to harmonize the sensitivity of the reference intervals, i.e., to make the sensitivity similar for all patients, regardless of patient characteristics. Statistical criteria to identify when partitioning is adequate have been developed over the last two decades. These criteria are applicable when considering two subpopulations, but recently a procedure for considering several subpopulations has been developed. When several subpopulations are considered, there is a possibility that some subpopulations could form a group or cluster that could share a common reference interval. However, there is no formal systematic approach to indicate how to divide these subpopulations into clusters. The aim of this study was to suggest such a systematic approach for clustering. METHODS: A clustering technique was applied to data including several subpopulations. The technique is based on measuring the distance between separated reference limits and successively pooling subpopulations divided by short distances. A cluster is defined by a group of subpopulations that are close to each other and that differ from subpopulations in another cluster. A cluster recruits new subpopulations as long as the subpopulations can be pooled without violating a partitioning criterion. CONCLUSIONS: We have suggested a procedure for partitioning a number of Gaussian (or Gaussian-transformable) subpopulations into clusters. This is the only formalized procedure indicating how to analyze several subpopulations and identify a suitable number of groups and reference intervals. Using a computer program developed for partitioning issues, the approach was easy to adopt.
|
|
| 2. |
- Gellerstedt, Martin, 1966-
(författare)
-
Partitioning reference values of several Gaussian subpopulations with unequal prevalence--a procedure with computer program support
- 2006
-
Ingår i: Clinical Chemistry and Laboratory Medicine. - Berlin : Walter de Gruyter. - 1434-6621 .- 1437-4331. ; 44:10, s. 1258-63
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: To be able to interpret laboratory values, it is essential to develop population-based reference intervals. A crucial consideration is whether a reference interval should be divided into subpopulations or not, so-called partitioning. There are established methods for deciding whether partitioning should be done or not. However, these methods are only applicable when partitioning into two subpopulations is considered. The primary aim of this study was to suggest a procedure that was also valid for several subpopulations. The method assumes that these subpopulations are Gaussian. Furthermore, a secondary aim was to provide a tailor-made computer program to support calculations. METHODS: The fundamental idea is to partition reference intervals if the proportions of the distributions of the subpopulations outside the combined reference limit deviate from the nominal value of 0.025. This is made possible by finding the combined reference interval using an equation solver algorithm. RESULTS: It was found that an equation solver algorithm could easily identify the combined reference interval when combining two or more subpopulations, even if these subpopulations had unequal prevalences. It was also found that this could be done even if the ratio between samples does not reflect the ratio between prevalences. Using this algorithm, it was possible to study whether the proportion outside the combined reference limits in any of several subpopulations deviated from the nominal 0.025 by such a magnitude that partitioning was recommended. When similar figures to those found in earlier studies with other methods were tested, the procedure showed consistent results with these methods. The procedure was also found to be applicable when several subpopulations were considered. As a practical result of the study, a tailor-made computer program was developed and is now provided over the Internet. CONCLUSIONS: The suggested procedure could serve as an alternative or complement to existing methods. The procedure provides calculations of the combined reference interval, even if sample fractions do not reflect prevalence fractions. The important advantage with the suggested procedure is the generalisation to the situation when several Gaussian subpopulations, possibly with unequal prevalences, are considered. Finally, since a tailor-made computer program is provided, the procedure is simple to use.
|
|
