Challenges with Building Compliance: Connecting Cause with Effect

I am up, again, at 0400, because one of my dogs had an urgent need.  As I am walking her pre-dawn, I am reflecting on why on some days, she sleeps in and some days she needs a nocturnal constitutional.  I consider all variations in process to explain this different outcome.  My dogs seem happiest when there is an established routine, so in my statistics-mind, a deviation in effect should have a precipitating deviation in cause.  So, by considering the timing of their last meal or their last walk or the weather or the amount of exercise they got the previous day, or the amount of sleep they got the previous day, I should be able to identify why some days she sleeps in and some days she does not.  Cracking this code, then, would both allow her to be more comfortable and allow me to get an extra hour or so of sleep. 

This connection between cause and effect is also critical in building compliance with patient experience behavior.  The challenge in both settings is that the number of potential causal factors is lengthy, and this does not even cover the interaction effect that multiple variables may create.  If you want to manage the outcome, then you need to manage the cause.  If you want to manage the cause, you need to identify the real or most important cause that there is.  Once you have done this, you need to link it back to an important effect that can be targeted and tracked.  Success in building compliance needs to include a clear understanding of cause and effect in your staff’s mind.  In this essay, I will discuss three important concepts in navigating the path of connecting cause and effect. 

Picking the right effect

The reason why we create new processes and procedures in patient experience is to improve the patient’s experience.  Duh.  But what does it mean to improve the patient’s experience?  There are numerous ways we might articulate this.  Some are more direct; some are more concrete.  Some are qualitative and some are quantitative.  Better PX could mean:

• Reduction in patient anxiety
• Fewer call-lights
• Higher patient satisfaction scores
• Better attention and retention of instructions
• More engagement in course of care
• Improved optimism
• Enthusiastic approach to rehab and therapy

These are just to name a few things off the top of my head.  I will get to the relative value of these measures, but first I will float the rather outrageous notion that even starting down this path is the wrong approach.  All of these measures focus on the patient’s interest.  This may seem obvious since we are talking about patient experience.  But, if you are in the process of instituting a new process and especially if this is the FIRST new process you are instituting, your primary focus on targeted effects should be not on the patient’s interest, but on the staff’s interest.

This may be offensive to some.  Let me be clear; I am not suggesting that staff don’t care about the patients.  I am simply saying that behavior modification is often rooted in self-interest.  I remember back in the early days of curb-side recycling.  Teaching people to separate their cans and bottles from their cantaloupe rinds and coffee grounds was the very definition of instituting a new process.  While many may care in the abstract about saving the planet, connecting that to training someone to rinse a bottle and put it in a different bin was not easy.  Where I grew up, there were two things implemented to assist in behavior modification.

1. The garbage truck would not pick up garbage that was not sorted.  If they saw recyclables in with the regular trash, they would leave that bag at the curb. 
2. Instituting cash deposits on cans and creating direct-to-end-user payment for aluminum recycling.  There were kiosks across the city where you could insert your cans and get money immediately for your aluminum.

Both approaches connected behavior modification to self-interest.  The first was the desire to get rid of your garbage and the second the desire to recoup some cash.  It is not that people forty or fifty years ago didn’t care about the planet.  It was that this abstract concern was less meaningful when you were in the kitchen trying to make dinner with the TV blaring and the kids screaming.  More concrete effects needed to be established to break through the easy habit of simply dumping the soup can in the trash.¹ 

 Likewise, while staff care about patients, their work-lives are filled with competing demands for time and attention.  Making a broad case for easing patient anxiety sounds great in the abstract, but when compared to charting, call lights, and providing routine care to a full complement of patients, all while trying to find five minutes to wolf down some lunch or go to the bathroom, reduction of patient anxiety will crash upon the rocks of reality. 

My first step in building a process change is creating the self-interest use-case.  Yes, hourly rounding is great for patients, but if I cannot convince nurses that it is great for them first, it won’t happen.  In my experience, one of the biggest headaches for nurses is managing call-lights.  Again, this is not because they hate patients, but because call-lights are random and ad-hoc.  Every instance of a call-light is a dice roll on whether it will be urgent, important, or inane.  So, bedside shift reports or hourly rounding are great for patients, but if I can show how they also reduce the number of call-lights, I can capture the attention of the staff.  Luckily, in the hyper-connected healthcare world, the number of and response time to call-lights is captured.  My suggestion is that before starting a new push, pull the call-light number and response time for a unit as a baseline.  That way, people can see how much of their day is taken up with call-light management.  It also becomes a point of comparison when the new process is implemented and operationalized.  If necessary, from here you can implement any sticks or carrots on individual performance as you want as well.

So, while most enter healthcare because they want to make a positive difference in their patients and their communities, you still need to leverage their basic wants and needs in motivating the change you want. 

Mind the gap

Another problem in identifying causes and effects is the fact that often we focus on outcomes that are too far removed from the behavior we want to modify.  This is easy to see in the behavior of others, but not often in ourselves.  Health maintenance is often built upon the concept that changes made today are not designed to address tomorrow but designed to address the rest of someone’s life.  Ask any smoker if they felt better the first day after quitting.  Much like a diabetic after one day of self-management, or someone on the first day on an SSRI, the smoker will probably laugh and say that they actually feel worse. 

But in building a long-term horizon, it becomes difficult to manage the moment.  In the winter, we are only six months away from swimsuit weather and in the summer we are just six months away from holiday parties.  We always have a reason to want to lose weight.  But it is rare if this point-in-time six months out is powerful enough to avoid that one donut today.  I mean, it is only one donut.  If I break that donut into 180 pieces, one for every day until that Christmas party where I want to wear that super-cool stylish suit, it is literally just a crumb a day.  Given all the other decisions I will make between now and then, it literally won’t make a difference.²

This is how staff look at your request for hourly rounding on Day 1 of a patient-admit.  The number of things that have to go right for a positive patient experience is legion.  Is this really going to make-or-break the experience?  This is exacerbated by the fact that the effect that the leader is managing is improvement with the survey results.  So, the logical flow generally presented to staff is

• Do this today
• The patient will be discharged four days after that
• The patient will get a survey two weeks after that
• The patient will have up to forty-two days after that to complete the survey
• We will check the data ninety days after that, so we have enough data to understand patterns

By making “improvements in the data” as our ultimate effect, we have inserted 90-100 days between the bedside shift report or the hourly-round and its impact on performance.  Telling staff to act today for benefit that will come eventually is like encouraging someone to quit smoking because in twenty years, there may be some marginal benefit to their health. 

Successful efforts here, then, shorten the time between when we can see value in the cause and effect.  While quantitative measures like call-light logs can be helpful, I often recommend something more personal, if less scientific.  Talking to staff about the process on Day 1 (and Day 2, etc.) allows you to measure two things.  First, you can evaluate compliance. 

• How often is it happening? 
• When it doesn’t happen, why is it not happening? 
• Is there something we can tweak to make it happen more often? 

Providing interest lets staff know that the process is important and that evaluation and effectiveness of the process is a two-way street.  Second, you can immediately start connecting the cause with any effect. 

• When you did it, did you notice any difference? 
• Did your patients seem less anxious? 
• Did they ask better questions? 
• Did they seem less demanding? 

Connecting the behavior immediately back to outcomes and then further tying them to outcomes that nurses would value, is the best way to shorten the distance between cause and effect.

Right-size the relationship

I have stressed before that no PX process will be 100% effective.  Patient experiences and complexity of interactions mean that doing ONE thing will never convert 100% of patients to give you a top-box score.  Everything is about improving margins, maximizing odds.  The team with the best record only wins about a quarter of the World Series or Superbowls.  This doesn’t mean you should try to be 100% compliant in a plan; it only means that you should not expect to be perfect in outcomes.  Doing the right thing consistently improves your chances of success, but it does not guarantee them.  It also does not guarantee universal success.  At this moment, there are people being discharged from Mayo or Johns Hopkins who thought the experience was mediocre or even poor. 

By making it seem that doing Process X is the magic elixir to lose weight, grow hair, and find true love, you are insulting your audience twice.  First, you are not acknowledging that success is a complex web of relationships and while we are focusing on a key piece, it is only one of a number of pieces.  When doctors criticize PX as far less important to their success than, say, saving lives, I know that they have been sold a story that undervalues other things, like competence, staff relationship, staff skill, as well as compassion, communication and a bunch of other stuff.  It is a mistake to convince them that if they do this consistently that their other problems go away.  It is, again, far easier to convince them to do these things because it makes their life marginally easier.  I remember talking with an ED doctor who was complaining about a comment that a patient left on Google.  She said, “if only they had said that to me, we could have discussed it!”  I asked, “Did you give the patient a chance to have that conversation?”  As we discussed the event, it turned out that she did not go back into the room during discharge, so there was no opportunity for the patient to share these thoughts with her.  I was then able to connect having a final check-in with a patient with a self-interested goal of not being put on blast in social media.  She still needed to do the work, but this brief conversation with the patient would have also saved her the headache of a one-sided public exposure.

Second, by laying it on thick about how the value of a process will cure all ills, you are overselling the process, especially when your overselling ignores what people see with their own eyes.  When the process does not deliver on its outsized promise, staff will become frustrated that they are over-promised and under-delivered.  And if the leader comes back and says, “No, it does work!  You just didn’t do it right!” they will lose that staff in a heartbeat.

We are pattern-seeking animals who want to be able to control outcomes.  Even those of us who are not statisticians are seeking out patterns that will help make sense of what is happening around them.  If we don’t help people find meaningful patterns to leverage, they will construct their own patterns.  What is astrology or numerology but efforts to bring structure and meaning to life that otherwise feels random?  If healthcare leaders do not help connect cause to effect, they will not succeed in creating surviving processes to change the direction of patient perception.  Staff will then lean on their usual set of stereotypes, like “new moms are too demanding” or “old patients on medical units are just never happy” or “our emergency department is so overburdened and slow, we can never give patients a great experience.”  These are rarely true, but very comforting.  But perhaps I only say this because I am a Sagittarius.

¹One could argue that the ubiquitousness of recycling bins next to trash cans was an outcome of all that behavior modification.  So, you are welcome.

²Please note that I use the word “literally” twice in its proper definition and not, as many will use it, as “figuratively.”