Published
I am on my hospital's nurse practice counsel and recently the discussion turned to nurses failing to recognize declining patients before it is too late. Before I was an ER nurse, I was a floor nurse. When you have 5-7+ patient to juggle at a given time you can barely focus your attention on one patient for minutes at a time in the midst of endless med passes and documentation. What is your hospital doing to improve this issue? I am aware that the EPIC charting system has an Early Warning Score (EWS) built in based off the vital signs. However, even a slight change in mental status (including anxiety) and decreased urine output are two possible signs of deterioration other than changes in vital signs. I am trying to create a generic electronic tool that nurses, particularly med-surg and telemetry nurses with high patient loads, would fill out 6 hours into the shift which will help them determine when to be alarmed and notify the physician to intervene. What do you think about this issue and my idea?
Great thread with lots of thoughtful responses.A couple of thoughts for OP
Its unwise to "jump to solutions" until you actually know what the problem is. In my consulting years, a lot of work involved reversing the superficial "improvements" that were implemented to solve problems - AKA, it seemed like a good idea at the time. My mantra became "Today's problems were yesterday's solutions". Don't fall into that trap.
Look at the research being done on "missed nursing care", "failure to rescue" and "nursing surveillance".... there's a ton of evidence out there to inform your decisions. Better yet, start analyzing those near misses or nursing failures using a system like HFACS or a rigorous root cause analysis process to determine the systemic problems that may actually be the causative factors.
I very much agree with you here.
It is imperative to do some investigation as to why the problem occurs as well as search in the literature. This has led me to recommend solutions or action that was not obvious to me from just looking at a problem.
This is the algorithm I came up with. It is not indended for use in the critical care setting. It is indended more for adult med-surg and telemetry units. It does not replace a nurse's knowledge, it encourages it. Critical thinking is meant to be used along with it. I'm not saying that anyone needs to be told to do simple things like "treat fever" just because it is included as an option in the algorithm. However, there are some things we can do before notifying a provider - either the attending, surgeon, PA, NP, resident, or rapid response team. Note: in some situations - depending on the facility- the rapid reponse team is supposed to be activated. The idea is that someone should be notified right away and if you can't get ahold of Dr. Hoity Toity - then a rapid response should be called. The nurse should use their clinical judgement to decide if the rapid reponse team should be called NOW or if the patient can afford to wait till a provider is reached. Please let me know your thoughts.
[*]If NO, move on to the next question.
[*]Is the systolic BP less than 90 mmHg
[*]If NO, move on to next question
[*]Is the systolic BP greater than 160 mmHg or diastolic BP greater than 100 mmHg
[*]Is the patient due for scheduled BP medication or is a PRN blood pressure medication ordered?
[*]If NO, move on to the next question.
[*]Is the heart rate between 100 and 125 beats per minute (BPM)?
[*]If NO, move on to the next question.
[*]Is the heart rate greater than 125 BPM?
[*]Is the heart rate less than 50 BPM?
[*]If NO, move on to the next question.
[*]Does the patient have stable baseline bradycardia which is above 40 BPM (view trend)?
[*]If NO, move on to the next question.
[*]Is there an increase or decrease in heart rate of at least 40 BPM which cannot be explained solely by expected effects of medication administration or evidence of patient discomfort?
[*]Is the respiratory rate less than 12 breaths per minute especially after opiate or sedative administration or suddenly greater than 24 breaths per minute?
[*]Has there been a temperature greater than 38C (100.4F) or less than 36C (96.8F) in the last 4 hours?
[*]Has the oxygen saturation dropped below 93% in the last 4 hours?
[*]Has there been decreased or absent urine output in the last 6 hours?
[*]Has the patient had decreased mental status, increased confusion, or increased anxiety or restlessness, even mild, in the last 4 hours?
[*]If NO, continue to monitor patient.
Aside from mental status, those questions all relate to changes in vital signs. That is a tremendous waste of time for an RN to fill that out for each patient. However, if you could create a computer program that looked at those trends and then "popped up" with a box that alerted the nurse to a change, then that *might* be helpful. The vitals are already in the computer so it seems like it might be pretty simple to just design a program that could look at those trends instead of creating additional workload for an already overloaded nurse. You basically want the RN to re-enter info that is ALREADY in the computer. Ummm....not helpful.
What would be very helpful would be getting behind legislation that set mandatory RN-to-patient ratios. It's been done in CA. And there are other states where even though there isn't a set ratio across the board, there is a statute in place that requires that a panel of RNs set a realistic ratio in each facility--I believe WA state participates in ratios in that manner.
What about experience; how many veteran nurses and new grads are there?
This seems to be the biggest factor that our facility has identified in regards to patients deteriorating without being noted. We had a relatively high proportion of new grads (about a year ago we had somebody that had been a nurse for 5 months doing charge on a med-surg floor). There has been significant improvement as our facility wide level of experience has increased.
We've recently instituted a family initiated rapid response to help reduce the number of code that occur outside of ED & ICU. It has not been implemented long enough to garner any data.
Aside from mental status, those questions all relate to changes in vital signs. That is a tremendous waste of time for an RN to fill that out for each patient. However, if you could create a computer program that looked at those trends and then "popped up" with a box that alerted the nurse to a change, then that *might* be helpful. The vitals are already in the computer so it seems like it might be pretty simple to just design a program that could look at those trends instead of creating additional workload for an already overloaded nurse. You basically want the RN to re-enter info that is ALREADY in the computer. Ummm....not helpful.What would be very helpful would be getting behind legislation that set mandatory RN-to-patient ratios. It's been done in CA. And there are other states where even though there isn't a set ratio across the board, there is a statute in place that requires that a panel of RNs set a realistic ratio in each facility--I believe WA state participates in ratios in that manner.
Most of the basic questions do relate to vital signs - however, there are many sub questions which would pop up to determine if there is a quick treatable reason to the changes in vitals signs and to determine if these are truly "danger signs" or just the patient's baseline. It uses more nursing judgement than just a computer program spouting out abnormal vital signs. It also incorporates urine output. A computer program would depend on a CNA documenting I&O in a timely fashion which as we know does not happen as frequently as it should in real life. Is is easier for the nurse to ask the patient if they have used the bathroom in the past 6 hours.
I am all for better nurse to patient ratios, however it took California 20 years to get those ratios and that was in just one state!
It is designed to take about 10 minutes or less to fill out once the nurse gets the hang of it. Like I said, some other form of "petty" documentation should be taken away before implementing this.
I currently work in a PCU, so I guess this algorithm technically doesn't apply to me. In my hospital system, we have an "Early Warning Score" that automatically populates when vital signs and GCS are charted. A score of 0-2 is considered stable. If the score is 3-4, we need to chart some kind of intervention or acknowledgement of the vitals. Per policy, a score of 5 or above warrants calling a rapid response. However, our system allows for some degree of nursing judgement...obviously there are patients whose baseline vital signs or mental status keep them at a 5 or above, and there are patients with scores of 0 who have a sudden change or just don't look right, and in that case we call a rapid response anyway. Either way we need to document that we are aware of the score and then have the option to chart, "patient at vitals baseline," "patient at neurological baseline," "provider at bedside," etc. This system exists on our med surg and tele floors as well, so I think your algorithm would be a little bit redundant because it addresses a lot of the same things. Also, I feel like it would be very difficult to take the time to fill this out for each patient 6 hours into each shift, and if there is a problem at that point it may already be too late to intervene.
BUT...having worked as a med surg nurse for 3 years, I definitely agree that there is an issue with failing to recognize early signs of deterioration on non-monitored units and I appreciate what you're trying to do! But from my experience, the main issue is definitely staffing. When I had 7 patients and a constant stream of discharges and admissions, there were times when a patient would not be assessed or even have eyes on them for hours. We rarely had CNAs on our med surg floors and if we did they were responsible for 25-30 patients on their own. It's literally impossible to be on top of every tiny change in status when ratios are that high. Another issue I think is experience...I know the hospital where I work has a very young staff, especially on 3-11 and 11-7. On my unit I'm the second most experienced RN, and I've only been working for five years. When I was first starting out, I would often ask a more seasoned nurse to take a second look at my patient if I had a bad feeling but couldn't pinpoint exactly what the problem was. Without experienced nurses to consult with or ask questions, it's hard for new grads to pick up on subtle changes. It's honestly a recipe for disaster in my opinion, and the only way to fix this problem is to push for better staffing, which will in turn lead to improved staff retention.
I feel like the Modified early Warning Score is a very helpful tool. We have to get our own shift vitals where I work and use it as part of our assessments. It is calculated using MEWs. This intervention has assisted me and others with catching that a patient's condition has changed. You can have the patient that may look okay (and you think you got it under control), but the MEws calls for a rapid response team to be called. They then determines to transfer the patient off the floor. When the score is a 3 or above, we do vitals every 2 hours, have to call the doctor, and inform the charge nurse. When the score is a 5 or above, we have to call the rapid response. It is very useful.
Honest opinion, if my hospital acknowledged that nurse patient ratios were not safe but refused to add more staff and instead gave me some form like that to fill out I would be looking for another job. I don't need a computer telling me how to treat my patient. I need frequent face to face interaction and assessment and time to review the chart and my assessment findings so I know what is happening with my patient.
I get that you are trying to help but I feel like this will only contribute to the problem and piss people off.
I personally think it is more of a retention and patient ratio numbers game that a checklist just won't solve. The EWS only truly works with frequent VS and I&Os in itself and decreased urine output is a later sign, as you know, but easy to miss without a foley or up to date documentation.The last time I did CPR (not but a few months ago mind you) I was on a floor at shift change and happened to be the most experienced nurse there between both shifts with everyone else having a year or less experience. I work nightshift so that means all of day shift, including their charge, was that experienced.
Dude straight up just flat lined, no pulse, but by the grace of powers beyond me, I had a pulse back and a fairly stable patient by the time the code team got there because the code was called later than it should because no one except me really new how to call it. Instead of just calling security to overhead page they had been pushing a button on the wall that apparently wasn't working yet as it was a new build floor!
I hope you do find more helpful comments that don't take quite as long to implement for safety!
Excellent points. If the nurses are too inexperienced, then another tool won't help as much as knowledgeable nurses and a better skill mix.
nutella, MSN, RN
1 Article; 1,509 Posts
Yes - it is a major teaching hospital with the resources.
I now work at a different place that does not have the same resources and there is definitely a lot of more "waiting" going on based on the fact that nurses do not "like to bother" the MD or are unsure if they should call a rapid response.
The basic triggers are the same for everybody on med/surg floors and of course once evaluated and found to be ok the MD can change an order for example for tele to be adjusted.