Welcome back for part 2 of simplified systems change. If you haven’t yet, be sure to check out part 1. I've previously discussed the basics of systems change, the rationale behind it, and finding the critical system constraints.

This post will describe how to improve constraints with systems thinking, and the importance of infrastructure and behavior to sustain improvements. It sounds difficult to focus simultaneously on the synergy of processes on a macro-level while looking for just a few of the things preventing the system from reaching its goal. The idea here is not only to agree to implement a solution, but to make the solution the top priority because of negative effects on the outcomes. While the technical nature of using statistical methods to optimize efficiency and variation of throughput in a system can be overwhelming, in applying systems thinking to health and human services, there is an opportunity to focus on the role of people in systems such as:

• Opportunities for training and improving capability

• Steps and tools to make better decisions as a team

• Understanding the impacts of individual work on both strategic goals and individual patients

• Conflict resolution

• Finding problems in processes, and recognizing solutions in people (instead of vice-versa)

• Peer learning and mediation

• Improving relationships and teamwork

Improving the system - what changes do we make?

Once you know what the constraint is, it is important to resolve it. The problem with bandaging it up and moving on, is that a bandage can fall off, and serious problems can fester underneath. You risk overburdening your team by asking them to do more improvement work without resolving the cause.

Constraints aren't anyone's fault. Our systems naturally have variation, and some areas of service delivery move faster than others, are more complex, or require more from patients to move forward. Finding true causes of constraints isn't about blame. It is looking for:

• Wasteful processes - Think about the classic lean wastes here (overproduction, transportation, excess services, service defects, over-processing, excess motion, waiting and unutilized talent).

• Unevenness - Is activity equitably distributed through the system, are certain areas, roles, or steps jam packed, while others are sitting underutilized?

• Overburden - Are you burning people out? Are impacts diminishing despite added resources?

These things are intertwined. Don't worry about checking boxes, as much as looking for cross-cutting themes. Most systems only have a few true constraints, but addressing the cause can have a big impact.

Once the team uncovers what is causing the problem, the team can shift to breaking free of these limitations by managing flow of the system.

Take advantage of:

• Time: Sustainable processes rely on having a buffer to absorb fluctuations in demands. However, the constraint is the one exception. This is the step that should always be at full capacity. This doesn't mean overburdening people, it means staffing it appropriately to meet demand.

• Capability: The constraint should be focused on only doing its primary function, with the right level of credentialing and experience. Don't have your nurses and social workers doing paraprofessional work while trying to manage a system bottleneck.

• Flow: Strive for an even, continuous stream of work going throughout the process instead of stops and starts. This can be hard to grasp. You're looking for problems in really small areas, but evaluating how service is delivered by the system as a whole.

Determining Priorities

To address the constraint, it needs to be top priority. Think about using strategies such as swarming, and rapid cycle improvement to work through it. An all hands on deck approach that gives folks a time to reflect on what needs to change will send the message that this is a priority. The goal here is exactly what I meant when I led off the post: not making the constraint the most efficient step in the process, just doing enough to not be eaten by the shark.

• Four things to keep an eye on:

  • Backlog - Throwing excess work in the constraint burns out the whole system

  • Unplanned Work - Hot jobs and firefighting cause things to go awry

  • Invisible Work - Invisible work is the cost associated with "multitasking"

    • Interrupted task switching: Having to switch from one task to another, before the first task is complete; this is a huge time suck. The human mind takes approximately 20 minutes to resume a task after an interruption. If you're doing five projects at once, you're only giving them each about 5% of your mental bandwidth, not 20%...the rest is lost to task switching. If you're interested in learning more about detriments to team and individual performance be sure to check out the work of Gerald Weinberg. It's tempting to tout our ability to multitask, and believe that a high number of projects is a badge of productivity, but in actuality it is probably hindering goals.

  • Variance in Work - We tend to underestimate how long it takes us to do things, and are poor at accounting for unplanned interruptions and challenges. Forecasting, estimating, and scheduling are important, but they are naturally prone to errors and fallacies. Doing a very similar intervention with two similar teams can yield totally different results.

• How do we effectively manage these challenges?

  • Use process and aligned subprocess (step-level) metrics to measure and address our changes related to backlog, unplanned work, coordination, and variance.

  • Critical Chain Project Management - Give yourself a buffer. This accounts for the unplanned, the invisible, and the variable. Working slower with great flow is better than "hurrying up and waiting". I find people's eyes glazing over when I try to explain Drum, Buffer, Rope, and I'm aware of its limitations (I prefer a modified version for service delivery), so my basic advice is this: The constraint should be the only step that is at full-capacity.

  • Take advantage of gains because the constraint is always ready for more work. By getting the process owners excited, and supported this can be a transformative time.

  • Take advantage of buffer time for staff in other parts of the process. If you're on an overall journey to be more quality focused, using a little excess capacity to be reflective, and doing things like brainstorming, planning, and self-directed enrichment ,instead of firefighting, can go a very long way.

Overcoming Barriers

To truly alleviate the constraint, it's critical to keep in mind some important factors:

• How will we continuously measure the change? Think about discontinuing the subprocess measures that were unique to improvement cycles, and adding the process measures to your dashboard.

• Resources to sustain the change. Ask the people doing the work, and receiving the services what is needed. This is another great way to get buy-in from staff, and implement some value adds for patients.

• Openly communicate about the change. Ask stakeholders, including staff and patients, how it's going. Once the constraint is broken, your system will be more robust and resilient. This is an opportunity to let frontline staff own the improvement and provide feedback about what they need to be motivated, capable, and feel that they are empowered to add value to the system -- not just a cog in the machine. This can mean taking a few more risks with new ideas. If you ever want to truly change things you need to challenge the status quo!

Repeat - Look for the new Constraint

Once you break the biggest constraint, unfortunately, you're not done. There is now a new constraint. While moving on to the next constraint should be an easier lift than the first one, it's not simply going back to the beginning with the next constraint. There is also an element of sustaining the broken constraint in addition to working to improve the new one. Don't become complacent!

• This requires new process maps, protocols, pathways, training, and modeling behavior.

• I find it really helpful to track each iteration of the system visually. It's a great way to recognize accomplishments, and work towards an ideal state vision of the system. The goal isn't perfection of reaching the ideal state, but it's a good conversation to have in improvement huddles

• Keep momentum by incorporating what you've heard from patients and staff both informally, and through surveys, interviews etc., as well as analyzing data on changes in service delivery aligning with patient needs.

That wraps it up. I hope you've gained some new insights about getting started with systems thinking. This has been a quick and simple introduction, meant to show how it can be applicable to service-oriented organizations. While traditionally it has been an organizational change method that is primarily focused on efficiency and improving the bottom line, I hope I have laid out how it can include employees, patients, and key stakeholders as well. If they aren't empowered in the healthcare service delivery experience, the benefits won't be fully realized. Our goal is better systems and processes as a means to a better life for staff and patients.