More thoughts on the ASHP national survey results for informatics and pharmacy practice

By | July 23, 2015

On Monday I spoke briefly about two articles in AJHP that summarize two recent ASHP surveys. The first covers Pharmaccy Informatics in U.S. Hospitals(1), while the second focuses on pharmacy practice in acute care hospitals(2).

Both surveys contain a wealth of information, and provide a snapshot of what pharmacies in the U.S. are doing. While conducted at different times by different groups, I think it’s more interesting to look at the two surveys together. As I mentioned in my podcast, the adoption of automation and technology goes hand in hand with pharmacy operations. You can no longer have one without the other.

One of the most confusing things about the pharmacy practice survey is their use of a “design-based analysis” with weighted results. It created a bit of confusion for me, which makes me wonder how many others are in the same boat, or worse, didn’t pay close enough attention to notice.

Let me give you an example of my confusion. Below is an image showing hospital characteristics from each paper – informatics survey on the left, practice survey on the right. I’m not really a statistics kind of guy, so I’m having a lot of trouble wrapping my brain around this. The percentage of respondents from the informatics survey (left) add up to 98.1%. Taking into account rounding, etc, that makes sense to me. The percentages of survey respondents for the practice survey (right) add up to 209.4%. You have to dig into the text to understand what’s going on. “To account for the sampling method, weights were assigned to respondents to adjust their contribution to the population estimate.” Ok, what impact does that have on the results? I still don’t fully understand exactly what happened there.

ASHP Survey Tables

Items that I thought were worth highlighting can be found in the table below. Some items are mature, such as BCMA and ADCs, while others are still relatively new, such as I.V. Workflow Management Software. The maturity of the technology is reflected in the adoption rates.  Something that is not seen in the table is the disparity between large and small hospitals. Both surveys show that hospital size plays a role in the adoption of technology, with larger hospitals adopting faster than their smaller siblings. It’s unclear why there’s such a disparity, but it may have something to do with budget size and/or available resources to manage projects. I’ve written about this in the past.

Two items that are worth noting are the adoption rates for ADCs and BCMA. Whether you agree with their use or not, both have become universally accepted as best practice.

ASHP Survey Comparison

Items of note:

  1. ADCs are still the most common medication distribution method for hospital pharmacies. Practice survey showing 63.5 percent of hospitals decentralized with ADCs, while the informatics survey showed 79.2 percent. I find that even though a hospital says they’re “decentralized”, what they’re really saying is that they decentralize as much as they can. Not everything can be put in an ADC. The push for decentralization of medication distribution waxes and wanes over time. I’ve come across more than a few facilities in the past couple of years that are decreasing ADC use in favor of more direct control of the process.
  2. Compounding robots still aren’t ready for primetime. The pharmacy practice survey shows a 2.9% adoption rate (0.3% for hazardous drugs), up from 2.5% (0.1%) in 2011. However, it’s unclear whether or not adoption increased or surveys simply landed in the hands of a few more pharmacies with compounding robots. I personally don’t think this area of pharmacy technology is expanding. In my opinion robotic technology hasn’t matured enough to become mainstream in pharmacy practice.
  3. The use of remote video supervision of pharmacy technicians’ i.v. preparation activities was used by 10.2% of hospitals.”(2). What the heck does that mean? Do pharmacists actually watch technicians prepare iv’s in real time? I would hate to think that that is a real thing.
  4. According to the Informatics Survey(1), “66.9% did not have plans to acquire i.v. drug workflow systems”. I’m surprised and disappointed by that number. The iv room is the most dangerous place in the pharmacy. It’s also the most ignored when it comes to innovation for safety. It’s not likely to change until pharmacies are required to have some form of technology for CSP production. Unfortunately, the terrible truth is that regulatory requirements drive change in pharmacy.
  5. I didn’t see anything on whether or not pharmacies were still using the pull-back method for CSP preparation. Would like to have that information. I know people still do it, and it’s one of the most dangerous practices in pharmacy.
  6. Are there really acute care pharmacies in the U.S. that don’t use a computerized pharmacy information system (PhIS)? According to the informatics survey there are. The article reports that 5.1 percent of respondents didn’t use a computerized PhIS.(1) That’s insane. What do they use?
  7. According to the Pharmacy Practice survey 65% of hospital pharmacies have a USP <797> compliant cleanroom.(2) That number is completely skewed by the larger hospitals, with more than 93% of the hospitals with more than 600 beds reporting compliance. I’m not surprised by this. It goes hand in hand with the adoption of technology, i.e. budgets and resources. I wonder what the smaller hospitals are using for CSP production? Probably using CAIs or CACIs.
  8. It would be interesting to know which hospitals responded to both surveys. From my experience, people that respond to surveys will respond to multiple when given the opportunity, and those that don’t, won’t.

As previously stated, both surveys provide a wealth of information, and I would encourage everyone to go through in detail. If you can’t get access to the articles due to the paywall, just drop me a line and I’ll snail mail you my hard copies of the journals. That’s still allowed, right?


Leave a Reply

Your email address will not be published. Required fields are marked *