Today, Michel Koeken has defended his master thesis. The thesis relates to a project conducted in collaboration with Siemens and provides valuable insights into the HIS procurement methods of Dutch academic hospitals.
Thesis title
Identifying and ranking the selection criteria in European tenders regarding the procurement of HIS/EHR systems in Dutch academic hospitals.
Management summary (excluding tables)
The thesis focused on the procurement of Hospital Information System / Electronic Health Record (HIS/EHR) systems at Dutch academic hospitals. As Dutch academic hospitals are considered to be a public institution and the value of the procurement of a HIS/EHR system exceeds the threshold value of €200,000 they are obliged by the BAO (Besluit Aanbestedingsregels voor Overheidsopdrachten) to conduct a European tender. The BAO is the Dutch implementation of the European guideline 2004/18/EG.
Each European tender contains a Research For Proposal (RFP). The RFP provides a detailed set of requirements of what the academic hospital wants to acquire and the score model that the academic hospital will use to evaluate the proposals of the vendors. The detailed set of requirements is divided into subtopics and these subtopics are considered as the selection criteria. The score model states how much points can be scored by the vendors by fulfilling a specific selection criterion. The vendor with the highest score and therefore the most economically advantageous offer (i.e. best price-quality ratio) wins the tender. In The Netherlands are three European tenders conducted by the Dutch academic hospitals in order to acquire a HIS/EHR system. All three European tenders were won by different vendors and this is interesting when one take into account that:
- a vendor had to be selected through a strict tendering process which is based on the three principles (equal treatment, transparency, and non-discrimination) and therefore ensures for total transparency since all decisions are open for involved parties, no possibility for negotiation, and an equally review of every vendor;
- the Dutch academic hospitals did not differ that much;
- and there were only five (six at the time of the first two European tenders) available vendors from which the Dutch academic hospitals could choose from.
One would expect that at least two out of the three European tenders were won by the same vendor, but this was not the case. Therefore it is interesting to analyze these European tenders to investigate how Dutch academic hospitals used the European tenders to come to their results.
To answer the raised issue the thesis started with analyzing the three score models in the RFPs in order to identify and rank the selection criteria in terms of importance. The three included RFPs were those from the European tenders conducted by Leiden University Medical Center + University Medical Center Utrecht, University Medical Center St. Radboud, and Erasmus Medical Center + University Medical Center Groningen. This analysis showed that the three European tenders could not be compared one on one as they used slightly different terminology and wording for their selection criteria. This problem was tackled by selecting one European tender as the reference tender and the selection criteria of the other two European tenders were matched with the selection criteria of the reference tender. This matching was done by two vendor experts and followed the two round Delphi method (Skulmoski, Hartman, & Krahn, 2007). Now the three European tender could be compared as they contained the same selection criteria. The comparison showed that the European tenders scored their selection criteria differently.
As academic hospitals are often an example for general hospitals this identification and ranking could be an important insight for general hospitals. It is even possible that general hospitals use this as a blueprint for determining their own selection criteria. Obviously also other academic hospitals can use this insight to their advantage and consultancy companies can use it in advising their clients.
The differences between the percentages were analyzed with the help of a vendor expert and as the tenders occurred sequentially trends were determined. It is important to state that the trends could be based on coincidence as there were only three available data sets. Nonetheless it is interesting and useful to discuss the possible trend and especially the most important ones.
The possible trends can be beneficial for every stakeholder in The Netherlands or elsewhere as they show which topics are hot or not. Vendors could use it to determine their business strategy, the consultancy companies can advise their clients about it, and the hospitals can take it into account in their procurement processes.
The full thesis text can be downloaded here. Michel also provided a Powerpoint file with an embedded audio recording of his (trial) presentation.