From Ms S. Bloom
Until recently I was programme manager for the North West Hospital pharmacy automation programme, a collaborative programme to implement pharmacy automation in hospitals. Between 2005 and 2008 I was responsible for co-ordinating the collaborative procurement and installation of 23 pharmacy robots across the North West.
I was disappointed to read the article on hospital pharmacy (PJ, 17 May 2008, pp599–602). I agree that too little has been written about pharmacy automation and its benefits and was therefore, at first, heartened to see an article on the subject in your publication. However, I think that the article is out of date and factually incorrect.
I draw attention to what it said about systems available in the UK. The information appears to have been taken from a 2004 article and therefore is four years out of date. That is a long time, considering the speed of development of these systems. Functionality has increased since then. For example, the Mach4 system now has automatic loading and refrigerated modules.
The article mentioned two systems that are no longer marketed in the UK and made no mention of several systems that are being actively marketed.
RoboPharma has a channel-fed dispensing system that is installed at Whiston Hospital and P@P Picking Systems has a high-volume picking system suitable for large-volume orders, such as ward supplies. It is currently installed at Kent and Canterbury Hospital. Siemens Logistics and Assembly Systems changed its name to Dematic in 2006, and currently has six installations in the UK.
The NHS Purchasing and Supply Agency (PASA) keeps a fairly up-to-date list of UK installations as well as other useful automation information on its website that can be accessed from their robotics and automated dispensing page.
Perhaps the most worrying comment in the article was the assertion that “the recognition rate of barcodes is 70 per cent and therefore less useful as a source of information for storing and selecting products”. Barcodes are fundamental to pharmacy automation, but if the recognition rate is only 70 per cent how can installation be justified?
I acknowledge that the reading of some barcodes is problematic due to print quality, colour or product packaging and is a cause for far too many products being unsuitable for automation.
However, barcodes have high reliability. Accuracy rates of one error in 1.7 million have been cited, far more reliable then manual human intervention and are the reason why installing automation has been shown to reduce or even eliminate product selection errors.
I was commissioned by the PASA (www.pasa.nhs.uk) to write a toolkit on implementing hospital pharmacy automation to share the learning that has been gathered from the NW Pharmacy Automation Programme.
Should any of your readers wish to obtain more up-to-date advice on implementing automation they can download the toolkit from the PASA website’s robotics and automated dispensing page.
Shoshana Bloom
Independent Pharmacy Automation Consultant
Stockport, Cheshire
Stephen Goundrey-Smith, author of the article, responds
The published wording of the section on automated dispensing systems available in the UK suggests that the detail of this section is entirely from Swanson’s 2004 paper. This is not the case; I am grateful to various pharmacy automation suppliers for information on recent robot developments.
Current developments with the ARX ROWA systems and the Mach4 systems are mentioned. I approached representatives of ARX, Mach4, Swisslog and Baxter while researching the article earlier this year. I was unable to make contact with a representative of Robopharma within the timescale of preparing this article, although every effort was made to do so.
I am grateful to Ms Bloom for providing information to PJ readers on the current UK activities of Dematic, Robopharma and P@P Picking Systems. Unfortunately, there does not seem to be any information on UK hospital pharmacy applications on their respective websites, and centres where these systems have been installed have not, to my knowledge, published their experiences in the professional literature. This lack of disseminated information on developments in automated dispensing is one of the points made in my article.
Ms Bloom indicates that barcodes may not be readable because of print quality, colour or product packaging. I am aware of anecdotal evidence from work in medicines management software analysis that a proportion of medicinal product barcodes are not machine readable, due to data anomalies in the barcodes. I concede that the rate of barcode recognition of pharmaceutical products is probably now higher than 70 per cent, but it is by no means 100 per cent.
I am not suggesting the process of product identification by barcode is not accurate when the barcodes are correctly machine readable. However, I would be interested in any information that readers have on this issue in a UK context, either from research they are aware of, or from official sources.
From Professor B. Dean Franklin, MRPharmS
I would therefore like to draw readers’ attention to a study published in the International Journal of Pharmacy Practice earlier this year,1 which was not mentioned in the article. This is probably the most comprehensive evaluation of dispensing robots in a UK hospital to be published to date; it explored the impact of two different dispensing robots in a controlled study at Imperial College Healthcare NHS Trust.
The study showed that installation of a dispensary robot has modest benefits in terms of reduced dispensing errors, reduced picking times, increased staff satisfaction and increased storage capacity; there was no conclusive impact on prescription turnaround times. These findings seem to be independent of the type of robot installed.
I fully support Mr Goundrey-Smith’s assertions that quantitative evaluations of such developments are needed; it is all too easy to assume that anticipated benefits will be borne out in practice without finding out if this is indeed the case.
Bryony Dean Franklin
Director, Centre for Medication Safety and Service Quality
Pharmacy Department, Imperial College Healthcare NHS Trust/The School of Pharmacy, University of London
References
1. Franklin BD, O’Grady K, Voncina L, Popoola J, Jacklin A. An evaluation of two automated dispensing machines in a UK hospital pharmacy. International Journal of Pharmacy Practice 2008;16:47–53.