An ongoing dialogue on HIV/AIDS, infectious diseases,
February 19th, 2018
Can We Solve the Morass of Outpatient Intravenous Antibiotic Therapy?
If you want to get an ID doctor riled up, here are a few reliable strategies:
- Get an ID consult on a complex patient just to summarize the chart for your discharge summary.
- Endorse the view that procedural doctors deserve their vastly higher salaries than MDs in cognitive specialties.
- Prescribe azithromycin for patients with bad colds.
- Discharge a patient from the hospital on intravenous antibiotics when an oral antibiotic would work just as well.
I’ve not hidden my distaste for unnecessary “outpatient parenteral antimicrobial therapy,” or OPAT on this blog, once devoting a whole post to oral antibiotics with excellent absorption.
It’s not just the inconvenience and potential dangers (blood clots, secondary infections from the IV catheters) of OPAT that I dislike. It’s also that there is essentially no support (read: money) for the clinicians charged with monitoring these complex patients, or for coordinating the many people who may be involved in their care.
Payers certainly don’t pay doctors or nurses for the substantial phone time, emails, and faxes that each OPAT patient generates. This may motivate some doctors to schedule unnecessary outpatient visits for these cases — otherwise they’d get nothing.
And since nobody is paying anyone to oversee the care of OPAT patients, there’s often a “Who me?” approach to their follow-up that is neither good for patient care or the morale of their providers.
Or just as bad, there are too many cooks in the kitchen, and no one is quite clear who’s responsible for what.
A recent post on the IDSA website by ID doctor Parker Hudson detailed this thorny issue perfectly:
Our OPAT program spends a disproportionate amount of time trying to track down labs/levels from patients on IV antibiotics who were discharged to SNFs [skilled nursing facilities]. Despite our orders and requests, most of these values are interpreted and managed by medical directors and not sent back to us — the ID docs writing the orders and following up the patients.
What followed were literally dozens of comments from other ID doctors with similar problems, problems we certainly experience on a daily basis with our OPAT program as well.
To illustrate the complexity, here’s a case — and then a very simple poll.
A 57-year-old man with diabetes and chronic renal disease is referred for admission by his primary care provider (PCP) with 2 weeks of progressive back pain, and found to have spinal osteomyelitis secondary to MRSA. A peripherally inserted central catheter (PICC) line is placed, and because of severe ongoing pain, he is discharged on hospital day 5 to a skilled nursing facility to complete 6 weeks of parenteral vancomycin and to receive physical therapy. At the time of discharge, his ID consultant (who does not do outpatient ID care) recommends that safety labs be done twice weekly, and vancomycin levels once weekly. An outpatient appointment with a different ID doctor is scheduled for 4 weeks later, as well as with his PCP.
Now, take the poll — and if you have a moment, provide the rationale for your answer in the comments section. Bonus points for any practical solutions to the OPAT morass — they would be most welcome!
Whoever wrote the orders for the nursing home, or more ideally agreed to follow the patient and lab PRIOR to the discharge.
As the ID doc recommending the treatment, we are also assuming some liability; I wouldn’t want to leave the interpretation of the results to others (even though the PCP or medical director at the SNF would probably call to discuss the results if they were abnormal, I wouldn’t want to take that chance). For optimal patient care, we should get those results, and make certain that the ID doc doing followup has those results prior to seeing the patient.
I completely agree. But a remaining problem is the routing of the results back to the responsible ID doctor(s). Once patients are discharged, there does not appear to be a foolproof way of ensuring this happens, and it takes a tremendous amount of time tracking these lab tests down.
Yes, very difficult to track down the lab work and clinical updates, but not impossible. In the final analysis, we are the ones who can best interpret and make decisions, and it really is best for the patient. I suppose our natural compulsive tendencies and compassion are why we became ID docs. I now give my cell # to the patient and/or family member (and they provide theirs to me) and insist on weekly calls. Onerous, no compensation, but gives me the comfort of knowing that there will be no ‘falling through the cracks’
All of the above – because, as described, it takes a everyone involved since there is no systematic way to ensure that any one individual can do it
This patient does not require 6 weeks of IV treatment if his isolate is susceptible to rifampicin and a 2nd agent and he shows a favourable clinical course (and inflammatory markers going down significantly), and does not have a large epidural abscess.
An OPAT service is great, but we should not forget, that more and more infectious diseases may be treated with oral antibiotics (maybe even endocarditis in the future).
We eagerly await publication of the OVIVA study!
The few times I have tried to run the OPAT myself have left me with great respect for the ID docs doing this work. Not unlike the dentist, few pts are happy to see you, good outcomes are forgotten, and bad ones are memorialized indefinitely. While I treat TB, I am dreading the day I need to use parenteral meds for 12 months.
Monitoring and management of vancomycin could be handled by the consulting pharmacist at the SNF.
In an ideal world you have an ID colleague or group with an outpatient clinic devoted to followup of OPAT, and insist that the patient have weekly appointments in the clinic for lab draws (and face-to-face visits with an MD as needed) for the duration of therapy. Alternately, OPAT should be managed by the MD who wrote the orders or a colleague within the practice group who is responsible for following OPAT initiated by the group’s members.
Both approaches are burdensome but manageable so long as OPAT use is rare – and the key point, I think, is that OPAT ought to be rare.
Regarding this vignette, the superiority of IV over highly bioavailable oral antibiotic therapy has not been demonstrated for bone and joint infections; on the contrary, some limited data suggest their equivalency. Our pediatric ID colleagues have been successfully treating acute hematogenous osteomyelitis (admittedly a somewhat different disease) with oral antibiotics for years. Hopefully the results of OVIVA will encourage a shift toward more rational use of oral antibiotic therapy among adult physicians.
Whoever writes the orders is responsible. Labs, sed rates, xrays, drug levels are the responsibility of whoever wrote the orders. If this is not the i.d. doc, then this person assumes the responsibility of getting this info to the i.d. doc. It goes w/o saying the i.d. doc should try to the bet of his her bility to get this. If pt is in a snf, the ultimate responsibility is w/ the medical director. I once was following a pt I was consulted to see in the snf and not in the hospital, had my orders changed and instead of a chart war, signed off the case and never went back to that snf. In my private office, I used to get hospital discharges w/ orders for home care on pts I never saw in the hospital and were never signed off to me. I told the home care co. that whoever ordered is responsible to sign and they eventually stopped calling me.
This is an area of preventable harm. At Hopkins, Sara Keller is generating scholarship around this area. One of our pharmacists (Elle Danan) has generated a process map to show where things could go wrong. There are places that have already figured it out. Cleveland Clinic is a good example. What that place has done with EPIC is remarkable. The best and most adaptable programs have some nursing and physician staff FTE’s dedicated to this.
The ideal situation would be for the ID physician who prescribed the treatment to monitor the patient. However a possible solution is for the patient to have a 2week follow up appointment with the primary care physician who should have a copy of ID physician’s consult, discharge and post discharge notes. The ID Physician should be accessible for contact from the PCP if needed.
At Pitt, we have a dedicated OPAT team, which includes two PharmDs (exclusively for OPAT), nurses, and us physicians.
I saw it evolve from a nebulous mess as a fellow to a seamless system a mere 3 years later.
Here’s our workflow:
-ID physican decides an inpatient needs to go out on IV abx
-ID physician writes something called an “antibiotic discharge note” in Cerner and emails it to the aforementioned PharmDs. (This extra step of emailing is cumbersome—we have outpatient Epic and in patient Cerner, and the “forwarding” across both systems is even more convoluted).
-The antibiotic discharge note lists all monitoring labs and follow-up plans
-The PharmDs follow the labs and route the results to the physician in Epic. For most situations, including vancomycin adjustment, they take care of it themselves and only ask us for our input when things get complicated
-With certain exceptions, all patients are seen in ID clinic before IV abx are stopped. Again, this is arranged for by the OPAT team
-The ID person seeing them as an outpatient need not be the same person who saw them in-house. I only round on the transplant ID service, and there’s only 6 of us who usually know everyone. On the general ID side, they have faculty and fellows rotate on dedicated OPAT weeks, when this is all they do.
-the nurses and PharmDs also have weekly meetings in which they discuss patients who haven’t been seen yet
It works really well. Our PharmDs will hound SNFs for labs like they’re superheroes. Some pitfalls:
-patients who live far away and can’t come. I’ve often given my blessing to stop abx
-if the SNF simply won’t send the labs
-not everyone needs IV abx
-when’s the right time to see them? Our system is pushing for everyone to be seen 1 week after discharge, but I view this as a waste of everyone’s time, as I can promise you the plan will not change. Home health nurses are seeing them weekly and call us if there are any issues. I favor prn and end of therapy visits.
I’ve learned that communication is key.
I feel bad for the PCP who has to follow up on every lab result.
An example where he should not receive OPAT unless the MRSA is XDR. Rifampicin + second agent with speedy oral switch should be most ID / microbiologists advice. If XDR then OPAT VA with oral linezolid follow-on at the very worst!
I am not in the USA, but where I have done most of my work we always have a “co-ordinating doctor” (by various names that might be confusing) who is responsible for the patient and who consults/calls upon experts in various disciplines as required. They should be able to work out whether it is necessary or desirable for results to be seen by whomever. Of course, this isn’t always perfect and mistakes have occurred but in most cases all the people involved take care to see that there is proper co-ordination and co-operation. I shouldn’t get too political, but perhaps it does help that we all work for the same organisation!
the tightest control would be for the original ID doc to sign out carefully with the medical director, and for that ID doc to be available for questions from the medical director. that seems to contain the least chance of chaos.
The primary care doctor is ALWAYS ultimately responsible for the care of the patient, especially when it means making sure consultants are doing their job. Good primary care eliminates this problem.
I think that once the recommendation is made by the inpatient ID doc, it is up to the facility that accepts the patient to take responsibility, not only for monitoring, but also the ordering the meds in the new facility. That should fall to the doctor at the SNF who is caring for the patient and who orders the antibiotics and any associated drug levels or labs (who is not necessarily the medical director) to monitor them and follow up with ID as necessary. Having a consulting pharmacist for the facility with experience with drug levels and dosing would be ideal. If the ID doc being used for follow up or the PCP who is not caring for the patient at the SNF are expected to be responsible, a four week follow up would not be appropriate.
whoever orders the test results is responsible to follow-up on them. We should take responsibility on following up the results of the tests we have ordered.
The ID doc is generally responsible for monitoring “safety labs” and vancomycin levels. If the inpatient ID doc sees outpatients it is relatively straightforward. In the above case we would have the inpatient ID doc check labs until the patients sees the (new) outpatient doc. Although it’s not fair for several reasons it is not practical nor in the best interests if the patient to have the primary care doc or the snf doc responsible for the monitoring as they do not have the expertise and likely won’t adjust the vancomycin levels accordingly etc. We do spend a fair amount of time chasing down these labs etc. and clearly there should be some compensation for this important responsibility.
In only a slightly sarcastic comment, I’d say that the patient’s daughter would be the one who would be in the position to pay the most attention to the details!
Here is OVIVA abstract link.
Thanks! Looking forward to reading the paper when it’s published.
The Emergency Medicine Doctor will ultimately be responsible for everything, when the SNF inevitably transfers the patient to the Emergency Room, for some inexplicable reason.
At Kaiser Orange County, the patients we ID docs see in the inpatient setting, we continue to see in the outpatient setting if they have OPAT needs. We have a dedicated team of pharmacists, home health, case managers who help arrange the transition between inpatient and outpatient. All of the labs are part of the same EMR used for both inpatient and outpatient, which the dedicated ID doc follows.
It works because it is an enclosed, integrated system. I also know it’s not representative of the “real” world.
Here in Winnipeg, we have it is referred to as the Community IV Program (CIVP), which seems to work pretty well. Because I work solely in an inpatient setting, the process I usually see is:
If a patient is ready for discharge, other than requiring IV antibiotics the admitting service will consult the hospital ID service. The ID team assesses patient suitability for CIVP program as well as if they are on an appropriate antibiotic regimen (and that they actually need IV therapy), and then refer on to the CIVP team. A CIVP nurse then sees patient in-hospital to make sure that patient is suitable and, if so, the patient is discharged to CIVP, where they will be managed and monitored by the CIVP interdisciplinary team (ID physician/pharmacist/nurse).
There’s a short summary (from a pharmacist’s perspective) available here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2826947/ (some of the details have changed, and the program has evolved a bit, but the broad strokes are the same)
The safety tests done twice weekly, if there are no obvious deviations in the test results, the medical director is suitable to review. If he has difficulty in interpretation or if he feels a change of treatment plans, he can always contact the ID specialist who recommended it because he knows the patient very well.
In Seattle, at the UW and Fred Hutch, we monitor OPAT with a coordinated team of dedicated ID PAs, pharmacists, and a ID physician. It is tedious, not reimbursed, and very relentless. However, I do feel it is a place wherein a mid-level provider could be used and help extend the services of the ID physician.
I often wonder whether we cannot find an IT solution. An app for OPAT patients, where labs get automatically fed into from the lab together with patient clinical information. Doctors get alerts when they are due, have arrived and possibly even have an alert system if they are out of a pre-determined range. The app could also alert the doctor about scheduled follow up. If the system were set up properly one could put an extra layer into this that the doctor or nursing group administering the antibiotics get alerts to remind them about labs to be drawn. If still no lab is done within 24 hours then an administrative assistant from the ordering ID docs groups could get an alert who then calls SNIF, home nursing when labs are not reported/done to avoid the docs spending time on tracking down labs. Not foolproof but maybe a bit better than the usual paper trail.
May I contribute to the discussion from the barren lands of socialised medicine?!
Our OPAT service is a partnership between an NHS hospital group and a – administratively, managerially, and financially separate – NHS community provider. Referrals are made electronically via EPR and the ID/micro team (comprised of doctors and specialist nurses) screen and approve all referrals to the service. Complex cases are usually already known to us via our consult service – and the OPAT plan made by us. Our specialised nurses ensure the correct form of IV access is in place.
The referral is sent simultaneously to our community partner – and they liaise with our nurses to ensure the discharge is smooth, and all clinical details correct.
Any problems at home (drug reactions, failure to progress) are alerted to our specialist nurse team by the community nurse – usually whilst they are still with the patients. Our nurses are prescribers and can review in clinic, discuss with ID doctors and make drug changes including IV/oral switch.
Complex cases have bloods taken at home weekly – and in our semi-integrated system the majority of these are analysed in labs accessible via our IT systems. Once a week those on long term treatment (e.g. cerebral abscesses etc) are discussed at a weekly virtual round involving the community nurse team caring for the patient at home. Patients are brought up to clinic if considered necessary and we liaise with other specialist teams as needed.
The service is part funded by the hospital – anticipating admissions avoided and early discharges – and the community provider anticipating reduced antibiotic use due to ID oversight and clinical governance provision.
Ah, we could only dream! Thanks for your comments.
We do not have ID docs in town so PCPs usually end up responsible for labs and follow ups. If a patient is in SNF, whoever is taking care of the patient on behalf of PCP. ID docs can be consulted remotely but not responsible for patient care. There is an infusion clinic but there is no dedicated team of pharmacist, home health, or nurses etc to help PCPs. So it can be a mess. It would be nice once more oral options are available but I feel most practitioners in town feel uncomfortable keeping those patients on oral vs IV even if there is a good evidence of harm from prolonged IV therapy.
I would say the medical director, until the patient’s first ID follow up. BUT I would recommend a follow up within 1-2 weeks, not 4 weeks later. We as ID don’t go to SNF, can’t control what the medical director does over there. I have had cases when the antibiotics are changed/stopped by them.
For patients that go home on IV, if they are arranged through the ID docs infusion center obviously the follow up is easy. When it is arranged by another home infusion company I only write for 1-2 weeks of Abx which will be renewed/extended upon patient’s follow up in office, that way we make sure they follow up
A decision support system that reports results back to all care providers listed above. Why are we wasting time performing tasks prone to error when technology can allow us to time to develop assessment and treatment plans, interpret findings, and make appropriate changes when indicated to improve patient outcomes
A fever is not an indication of ceftriaxone deficiency.
I spent several years as the OPAT pharmacist in the VA. Any patient that needed to go home on IV therapy had to have n ID PharmD review. Do you know how many interns have laughed at me for calling them and asking if the patient had running water/electricity!? (we learned early on not to assume- simple things that will lead to failure of OPAT!). . I would then track down weekly labs and document progress notes with the ID doc co-signed and ensure they got into clinic with an ID provider as outpatient. It is probably the most fool-proof system to have a designated person- but it was tough! (and only a small part of my day to day). A closed system like the VA makes things much easier though- I guess it’s the closest thing we have to socialized medicine in the US. I have a hard time seeing so many patients go out our door on IV antibiotics at my new job in a community hospital with no ID/PharmD review at all when it could be avoided so many times.