Healthcare delivery systems in the US continue to face an uncertain future. This is the result of rapidly evolving technology used in the practice of medicine, continued exploration of new organizational structures to provide care and manage health, as well as the seemingly endless debate regarding the future of the Affordable Care Act. How do you plan for needed facility construction or renovation in the face of all this uncertainty?
1 – Identify as many future scenarios as you can.
In the abstract, identify as many future scenarios as you can in which the new facility investment must succeed. For example, are there conceivable variations in future patient demographics, local employment, competitors, partners, government policies, reimbursement, population health, or anything else that would affect how your organization delivers healthcare to the community you serve? The answer is almost always yes, so Step 1 is to identify these potential future scenarios and list them.
2 – Design for flexibility and evaluate against multiple scenarios.
Even though you have now identified a number of possible futures, chances are there is one that you feel is most likely to occur. It’s important to resist the temptation to design your new or renovated facilities only to meet the needs of this particular scenario. If you believe a particular scenario is 80% likely to occur, then it stands to reason you are taking a 20% chance that the new facility will need to function in a different environment. In that case, it may not be successful and that is simply unnecessary risk. The best way to increase the odds of success is to place your design solutions into multiple scenarios and evaluate how well they will serve the required purpose in each of those cases.
3 – Select the most robust design solution.
After evaluating your design solutions against multiple future scenarios, select the one that will be successful in as many of those futures as possible. In this way, you can plan for what you believe will happen, while guarding against failure that could result from different circumstances. Using this mindset is perhaps the best way to mitigate risk in a time of uncertainty while continuing to move forward. After all, standing still in uncertain times is more likely to be a self-defeating strategy than a strategy based on thorough scenario planning.
What types of tools are you using to “evaluate” your potential solutions? Particularly in situations where you are experimenting with really innovative or risky potential future scenarios, what kind of analysis are you doing in order to be sure (to the extent that you can) that a design will actually operate the way you expect?
Thanks for the comment Brittany. It’s not the easiest question to answer, since the evaluation techniques will vary greatly depending on the type of scenario and solution. For example, if you are evaluating what will happen in various scenarios related to competition, you would identify the different competing organizations that might affect your services, in what way their offerings would affect your volume and patient types, and what those changes would mean to the economic viability of the facility. Once you have those impacts in mind, the client might decide that the risks are worth it and move forward, albeit with a design that is flexible enough to change and adapt as might be required. The decision to move forward may even dissuade competitors from entering the same market space. Or the client might decide that this scenario is a serious risk and that a collaborative joint venture with a competing organization would be preferable.
This may not be what you’re really asking, because the last part of your comment refers to how a design will actually operate. You may be talking about a more specific, finer scale of design decisions, like whether to have centralized or decentralized nursing stations, and how does one determine later that the decisions played out as desired. Perhaps you can elaborate a bit on your question.
Hi Randy, thanks for the quick response. You are correct that I am thinking more from the operational/nursing unit level. I am wondering about how you decide on questions like nursing stations, where to place equipment rooms, things of that sort. These would be process, flow, and congestion issues.
I am also wondering about proximity questions such as the organization of the departments on a floor relative to each other. For example, I recently heard of a hospital that is going to have a shared bed unit that sits between the ED and the OR so that the beds will flex throughout the day – the majority being used by the OR in the morning and then swinging to the ED as the day progresses and demand patterns shift. This is a really innovative concept, but my concern is how do you know how many beds to build, and what is your decision making about priority when the inevitable conflict arises? It seems to me like a major risk for the architect who needs to say “yes, this will work” – and if it doesn’t then their reputation is on the line.
My question is how you or others in the field are looking at answering these questions?
Probably more to this discussion than will work in a blog dialogue, but let me try and respond to the two main questions you’re asking.
Regarding process, flow and congestion issues, this all goes back to programming where a careful analysis of those issues needs to occur, or the design will likely miss the mark. This can involve everything from observation of current operations to simulations and/or mock-ups of future operations. The former may be limited by the current layout in which they work, and the latter may be limited by our imaginations with regard to variables that may have an impact later. It’s critical to ask lots of questions!
The specific example you cite of a hospital placing shared beds between an ED and OR suite is an interesting one. Presumably this is a hospital that has good statistics on their ED arrivals and LOS, as well as on their OR case types and recovery times (and of course how they are scheduled). It seems logical that they can determine the likely census by type for those shared beds by day of week and hour of day, but of course that would just be a projection and can vary due to the normal ebbs and flows. I presume that if the shared beds were all assigned, they would just go back to whatever they were doing before, but of course I don’t know what that was so it’s hard to evaluate.
Feel free to add further thoughts to this, or if you’d rather take this offline, don’t hesitate to email me at rcole@strattonbrook.com. It’s an interesting topic to be sure.