DOCTORAL thesis - dissertation


The task of ensuring Patient Safety is, more than ever, central in Health Care. The report “To Err is Human” [Kohn et al. 2000], was revealing alarming numbers of incidents, injuries and deaths caused by deficiencies in Health Care activities. The book initiated assessment and change of healthcare methods and procedures. In addition, numerous reports to Swedish HSAN (Medical Responsibility Board) have shown a high rate of information and communication deficiencies in Health Care has a direct or indirect cause of incidents, injuries and deaths. Despite numerous of new sophisticated tools for information management in recent years, e.g., tools such as Electronic Health Records (EHR) and Clinical Decision Support Systems (CDSS), the threats to patient safety have not been redeemed. Rather to the contrary. Underlying reasons for this paradox are twofold. Firstly, advancements in diagnosing techniques have given rise to increasing volumes of data at the same time as the number of patients has increased due to demographic changes and advancements in treatments. Secondly, the information processing systems are far from aligned to related workflow processes. In short, we do not at present have interoperability in our healthcare systems.

In this doctoral dissertation, we present an in-depth analysis of two different “HSAN-typical” cases, where Patient Safety was jeopardized by incomplete information flows and/or information breakdowns. The cases are mirroring the apprehension of Simplicity, that is, Occam´s Razor of Diagnostic Parsimony. A well-known protocol used in Health Care and implemented in most (knowledge based) CDSS. This rule of thumb is the foundation for the well-known adage: “when you hear hoof beats, think horses, not zebras”. Hickam´s Dictum is one well known objection to the simplifications of Occam´s  Razor stating "Patients can have as many diseases as they damn well please". Of course, this Dictum is harder to implement effectively! In the thesis we suggest a visualization tool Visual Incidence Anamneses (VIA) to provide middle out compromise between Ockham and Hickam but providing means to increase patient safety.

The findings of our Study for the thesis have resulted in a number of Aspects and Principles as well as Core-principles for future CDSS design. That is, tools and methodologies that will support designing and validating Interoperability of Health Care systems across patient-centric workflows. The VIA tool should be used as the initiating point in a patient (individual) centered workflow, quickly visualizing vital information such as symptoms, incidents and diagnoses, occurring earlier in the medical history, at different times, to ground further vital decisions on. The visualization will enable analysis of timelines and earlier diagnoses of the patient, using visually salient nodes for visualization of causalities in context. Furthermore, support for customization of the tool to the views of stakeholders, members of health care teams and empowerments of the patient, is crucial.


The printed version of the thesis will  contain a Book Chapter. Estimated delivery of the printed thesis: December 1st 2011.

New! Digital version of the printed thesis with book chapter available here. The thesis is also available in BTH Research database.

Doctoral defence January 10th 2012. Download "Spikblad".



New site:

Please visit the site after January 15th 2012!


Share Dela