A unidirectional mapping of ICD-8 to ICD-10 codes, for harmonized longitudinal analysis of diseases

Abstract

Periodic revisions of the international classification of diseases (ICD) ensure that the classification reflects new practices and knowledge; however, this complicates retrospective research as diagnoses are coded in different versions. For longitudinal disease trajectory studies, a crosswalk is an essential tool and a comprehensive mapping between ICD-8 and ICD-10 has until now been lacking. In this study, we map all ICD-8 morbidity codes to ICD-10 in the expanded Danish ICD version. We mapped ICD-8 codes to ICD-10, using a many-to-one system inspired by general equivalence mappings such that each ICD-8 code maps to a single ICD-10 code. Each ICD-8 code was manually and unidirectionally mapped to a single ICD-10 code based on medical setting and context. Each match was assigned a score (1 of 4 levels) reflecting the quality of the match and, if applicable, a "flag" signalling choices made in the mapping. We provide the first complete mapping of the 8596 ICD-8 morbidity codes to ICD-10 codes. All Danish ICD-8 codes representing diseases were mapped and 5106 (59.4%) achieved the highest consistency score. Only 334 (3.9%) of the ICD-8 codes received the lowest mapping consistency score. The mapping provides a scaffold for translation of ICD-8 to ICD-10, which enable longitudinal disease studies back to and 1969 in Denmark and to 1965 internationally with further adaption.

Keywords: Conversion table; Crosswalk; Data harmonization; Denmark; Diagnosis; Disease codes; ICD-10; ICD-8; International classification of diseases; Mapping.

Fig. 1

The hierarchical structure of ICD-8 and ICD-10. The structure is illustrated by expanding the endocrine main chapters down to the most detailed diabetes related subcategories. ICD-8 (top) has only a single diabetes category, in contrast, ICD-10 (bottom) categorises different types of diabetes, thus enlarging the diabetes related classification options. The chapter headings are followed by the number of block categories within that chapter. Chapter names are shortened from the official names to increase illustration readability

Fig. 2

Illustration of the mapping process. The mapping process is based on a careful examination of the ICD-8 and ICD-10 structure (the structure of main chapters, blocks of categories, categories and subcategories). The context of a specific ICD-8 code is taken into consideration when identifying a list of possible ICD-10 matches. If the granularity of the ICD-8 context is similar to the granularity of the ICD-10 options, the mapping is relatively straightforward. If not, mapping to a common higher level is considered. The frequencies of the codes in each ICD revision were used to reflect the actual use as the aim is to combine medical history that span the two terminologies

Fig. 3

Mapping scores by chapter. The chapters with the best mapping scores are Infectious, Neoplasms, and Congenital malformations. However, the Mental chapter, Pregnancy chapter, the chapter for Conditions originating in the perinatal period, and the chapter for Factors influencing the health status are the chapters with the lowest level of best matches. Chapter names are shortened from the official names to increase illustration readability

Fig. 4

ICD-8 to ICD-10 mapping permits expansion of the timespan for which data sets can be analysed. Annual distribution of diagnoses in the Danish National Patient Registry by ICD-10 chapter shows consistency over the ICD-8 and ICD-10 time periods. The 40-year data span covers both ICD-8 diagnoses (1977–1993, left) and ICD-10 diagnoses (1994–2017, right); ICD-8 diagnoses are collapsed to ICD-10 main chapter categories

Fig. 5

Illustration of the mapping between Danish ICD-8 and ICD-10 codes. A line links each of the 8,596 ICD-8 codes to its match in ICD-10