Making sense of taxonomies in health informatics

Simon de Lusignan

Editor in Chief, Informatics in Primary Care

University of Surrey, Guildford, UK

Copyright © 2014 The Author(s). Published by BCS, The Chartered Institute for IT under Creative Commons license


This issue of the journal starts with a leading article about the tangled taxonomies within our discipline.1 In this issue explores this further, looking at how a clear taxonomy for our discipline might improve our understanding of what is and, perhaps more importantly, is not part of health informatics.

Taxonomies, the classification or grouping of things, are well developed for living things. Many of the groupings of plants and animals take a phylogenetic perspective, namely they make the assumption that there was evolution from a common ancestry. Darwin was one of the first known to have sketched out a ‘Tree of life’ to illustrate this common ancestry of many life forms.2 We should, as an informatics community, better define the components of what makes up our discipline, as this would help us define and explain what we do. Creating such a taxonomy should not necessarily constrain us. Darwin recognised that some branches of the tree of life might die and fall away:

From the first growth of the tree, many a limb and branch has decayed and dropped off; and these fallen branches of various sizes may represent those whole orders, families, and genera, which now have no living representatives, and which are known to us only in a fossil state.

I will not speculate which areas of health informatics are likely to become extinct, and only be known to future informaticians through their fossilised remains. However, some of the failed major implementations in informatics, many of which are described in the pages of this journal,3 might be usefully conceptualised in this way.


The leading article describing the tangled taxonomies in health informatics lists some of the taxonomies that have been applied to health informatics. 1 Several of these look to define the scope of the discipline using existing or new classifications; medical subject headings (MeSHs) is one of these.46 By way of contrast, Staggers and Thompson suggested there might be (1) technology, (2) role and (3) conceptorientated definitions of Health Informatics.7 It is possible to draw these together in a two-dimensional or as a biaxial taxonomy (Figure 1). However, I have extended Staggers and Thompson’s classification by making it a general taxonomy and by extending concept to include theory.

More dimensions could be added to this definition: adding layers for multidisciplinary and interdisciplinary working, the granularity of the data being studied, its degree of patient focus and academic underpinning (Figure 2).

Figure 1 Biaxial taxonomy for health informatics.

The MeSH describes the scope, and Staggers and Thompson’s classification invites each category to also be described as predominantly a technology, role or concept.

Figure 2 Schema of a multidimensional taxonomy for health informatics.

Adding conceptual layers identified among the tangled taxonomies


The paper by Pietrzak et al. in this issue of Informatics in Primary Care, looks at whether smart home technology prevents falls in older adults?8 This paper could be firmly placed in the intersection of technology and the MeSH ‘analytical, diagnostic and therapeutic techniques and equipment’ (Figure 1). However, this paper has further dimensions, particularly patient-centred ones drawing out the incompatibility of some technologies with the seniors who might use them.

The next paper reports how telehealth and Skype are useful for supporting young people with continence problems. 9 There is a big contrast between the age groups of the subjects in these two papers and also in the acceptability of the technology. This paper, by Levy et al. is much more tightly focussed on the role of the nurse and the impact of using technology to enable nurses to use their time better.

Morrison et al. describe variation in data quality in UK hospitals and practices. The focus of their paper is on ethnicity recording.10 So far as the biaxial taxonomy is concerned (Figure 1), the paper sits in the anthropological and social phenomenon categories.

Bush et al. next report what characterised non-attenders.11 Unsurprisingly, the longer the time interval to the next appointment, the less likely people are to attend, but mid- level providers also have lower attendance rates. These are important facets of non-attendance to document.

Pandhi et al. describe an approach to optimising the use of electronic health records. The key in achieving this was the development of a mediating managerial layer between managers and clinical teams.12 This approach resulted in the tailoring of programmes to fit teams; and perhaps pruning the branches of the informatics tree of life, rather than let them overextend and fall. This paper has a clear location on the biaxial taxonomy (Figure 4). In addition, this paper drew out interdisciplinary and multidisciplinary dimensions.

In the last paper in this issue, age, this time of family physicians, comes to the fore. Prazeres observes how GP trainees are more adaptable to technology than their older GP trainers.13

Figure 3 Mapping the keywords listed by the author to the biaxial taxonomy.

The key words listed by authors are mapped to the nearest MeSH category, where keywords belong to more than one category the author


The keywords listed by the authors, or their nearest equivalent MeSHs are listed in Figure 3. They are placed in the row of the MeSH category they belong to. Some MeSH terms belong to more than one category, where this happens the dominant choice was selected. Most, perhaps unsurprisingly belonged to one of the following MeSH categories:

• Information science

• Disciplines and occupations

• Analytical, diagnostic and therapeutic techniques and equipment.

There was minority representation in the following categories:

• Persons

• Diseases

• Psychiatry and psychology.

There was no representation of:

• Anatomy

• Anthropology, education, sociology and social phenomena

• Check tags

• Chemicals and drugs.

Perhaps this tells us something about where our discipline is situated.

Where in the taxonomy does the Editor place these papers

Finally, the Editor would have mapped the themes of the papers to different places in the biaxial taxonomy that the key words chosen by the authors. The core themes of the papers in this issue are mapped onto the biaxial taxonomy (Figure 4). The papers sit differently than they did when classified by their key words. However, with the exception of papers also classified in the ‘anthropology, education, sociology and social phenomena’ and ‘pharmacological actions’ categories, the rest of the categories are the same as those identified through the keyword classification. Maybe this provides insight into what are the prevailing themes that might make up a taxonomy for health informatics.

Figure 4 Mapping the predominant areas of the papers in this issue to the biaxial taxonomy for health informatics.

Papers listed by first author, mapping based on Editors appraisal of key themes


In this issue of Informatics in Primary Care, we publish three papers that report how younger age appears to be associated with better uptake or use of IT. Using IT was harder for older people and established practitioners and easier for younger people with chronic disease and trainees.

• We report problems for older people, who struggled to use a technology to reduce falls,8 and also how there were greater challenges in using IT for experienced GPs compared with their trainees.13

• By way of contrast, there were fewer problems among younger people with continence problems who readily used Skype to help in their management;9 and similarly, GP trainees (who are generally younger than their trainers) did better with IT uptake.13

In this issue reminds us of the broad scope of our discipline but that within it are certain key themes. It is time to untangle the tangled taxonomies that limit how we communicate what is and isn’t within the discipline of health informatics.


1. Barrett D. et al., Unravelling the tangled taxonomies of health informatics. Informatics in Primary Care 2014;21(3):152–155.

2. Darwin C. The tree of life. Natural History Museum. URL:

3. Cresswell K, Morrison Z, Crowe S, Robertson A and Sheikh A. Anything but engaged: user involvement in the context of a national electronic health record implementation. Informatics in Primary Care 2011;19(4):191–206.

4. Dixon BE, Zafar A and McGowan JJ. Development of a taxonomy for health information technology. Studies in Health Technology and Informatics 2007;129(Pt 1):616–20.

5. Sperzel WD, Broverman CA, Kapusnik-Uner JE and Schlesinger JM. The need for a concept-based medication vocabulary as an enabling infrastructure in health informatics. Proceedings of the AMIA Symposium 1998:865–9.

6. Boonstra A and Broekhuis M. Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions. BMC Health Services Research 2010 Aug 6;10:231. doi: 10.1186/1472-6963-10-231.

7. Staggers N and Thompson CR. The evolution of definitions for nursing informatics: a critical analysis and revised definition. Journal of the American Medical Informatics Association 2002;9:255–61. PMCid:PMC344585.

8. Pietrzak E, Cotea C and Pullman S. Does smart home technology prevent falls in community-dwelling older adults: a literature review. Informatics in Primary Care 2014;21(3):105–12.

9. Levy S, Henderson L and McAlpine C. Growing up with confidence: using telehealth to support continence self-care deficits amongst young people with complex needs. Informatics in Primary Care 2014;21(3):113–17.

10. Morrison Z, Fernando B, Kalra D, Cresswell K, Robertson A and Sheikh A. The collection and utilisation of patient ethnicity data in general practices and hospitals in the UK: a qualitative case study. Informatics in Primary Care 2014;21(3):118–31.

11. Bush RA, Vemulakonda VM, Corbett ST and Chiang GJ. Can we predict a national profile of non-attendance paediatric urology patients: a multi-institutional electronic health record study. Informatics in Primary Care 2014;21(3):132–38.

12. Pandhi et al. Approaches and challenges to optimising primary care teams’ electronic health record usage. Informatics in Primary Care 2014;21(3):142–151.

13. Prazeres F. How do GPs versus GP trainees adapt to electronic health records? A Portuguese pilot study. Informatics in Primary Care 2014;21(3):139–41.


  • There are currently no refbacks.

This is an open access journal, which means that all content is freely available without charge to the user or their institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles in this journal starting from Volume 21 without asking prior permission from the publisher or the author. This is in accordance with the BOAI definition of open accessFor permission regarding papers published in previous volumes, please contact us.

Privacy statement: The names and email addresses entered in this journal site will be used exclusively for the stated purposes of this journal and will not be made available for any other purpose or to any other party.

Online ISSN 2058-4563 - Print ISSN 2058-4555. Published by BCS, The Chartered Institute for IT