Pulse oximeters are widely used to monitor for low oxygen levels among #Covid19 patients at home.

But does pulse oximetry improve outcomes over monitoring for symptoms of shortness of breath?

Our team provides some answers today in @NEJM 🧵👇
nejm.org/doi/full/10.10… #SGIM22
..First some context, early in the pandemic there were widespread accounts of #Covid19 patients presenting to the ER with relatively minor symptoms, but with markedly low oxygen levels, a phenomenon known as “silent hypoxia"...
nytimes.com/2020/04/20/opi…
...It was thought that use of a pulse oximeter could detect declining oxygen levels before symptoms develop, and provide an opportunity for the patients to get to the ER faster where they could be stabilized faster...nytimes.com/2020/04/24/wel…
...In the meantime, within two weeks of the first local #Covid19 diagnosis in March 2020, our health system, @PennMedicine, launched an automated text-messaging program called COVID Watch to check-in on newly diagnosed #Covid19 patients at home...pennmedicine.org/news/news-blog…
...Patients who text-back that they experiencing symptoms of shortness of breath would receive phone calls within 1 hour from an on-call nurse to either guide them to the ER, arrange an urgent tele med appointment, or continue to self-monitor...
...Other health systems (e.g. Kaiser, VA), states (e.g. Vermont), and whole countries implemented programs to monitor #Covid19 patients at home with pulse oximeters...england.nhs.uk/nhs-at-home/co…
...So after a scramble to get grant funding from
@PCORI our team set out to determine:

1) How do patients who receive symptom based monitoring (COVID Watch) do compared with no monitoring;

2) Can COVID Watch be improved with the addition of home pulse oximetry?....
...We found that enrollment in the symptom-based monitoring program (COVID Watch) was associated with a significant reduction in mortality with faster presentation to the ER compared with no monitoring

See thread from our paper in @AnnalsofIM below...
...We then set up a pragmatic randomized trial from Dec 2020 – Feb 2021 in which patients enrolled in COVID Watch as part of usual care were randomized to receive an overnight mailed pulse oximeter or just the standard COVID Watch program...
..Patients randomized to pulse oximetry were monitored for symptoms of shortness of breath. Then in addition to be asked about shortness of breath, they were asked to provide their pulse ox levels twice daily. An alert was triggered for declining oxygen levels or a level =<90%...
We hypothesized that the addition of pulse oximeters could help in two ways:

1) Enable patients who feel “dyspnea” (aka shortness of breath) but have normal oxygen level to continue to self-monitor at home

2) And help patients who develop silent hypoxia to the hospital faster
..To capture both of these potential outcomes, we made our primary outcome days alive and out of hospital at 30 days after enrollment because patients would prefer to stay home if safe, but if needed, get treatment faster reducing time in the hospital and the risk of mortality...
...Under the outstanding, tactical leadership of co-lead authors @katleemd and @AnnaUma we enrolled 2,097 during the winter 2021 #Covid19 wave. The primary analysis was specified to be among the 1,217 with confirmed positive Covid19 tests within our health system....
...Among patients in the pulse oximetry group, 78% submitted at least one pulse oximetry reading; these patients submitted a mean of 10 readings, corresponding to a response rate of 69% to pulse oximetry check-ins...
...The main result: There was no difference in days alive and out of the hospital between the standard program and program + pulse oximetry. In sensitivity analyses this held up for all randomized patients and when just looking at those who actually used their pulse oximeter...
...We also set out to ensure we would have a sufficiently large study sample to assess whether there were any differences within racial and ethnic subgroups to assess for equity in effects. We didn't find any differences...
...This is important because recent research has raised concerns that fingertip pulse oximeters may be less likely to detect low oxygen levels in Black patients (this paper came out in the middle of our trial)...nejm.org/doi/full/10.10…
...Across a large number of exploratory outcomes, we did not find any differences. This included patient reported anxiety, helpfulness, and confidence in managing #Covid19, time to return to the ER, rates of ER visits, hospitalizations, and death...
...However, there was one difference. The group who received pulse oximeters did generate more alerts to the nurses and more phone calls indicating more clinical resources were required to manage the population...
...So what can we conclude? The addition of pulse oximetry to an established symptom based remote monitoring program did not improve clinical outcomes despite increased cost and resource use....
...This implies that low tech, symptom based monitoring with access to an on-call nurse is a scalable approach to continue to monitor large populations of #Covid19 patients...covidwatch.waytohealth.org...
..Important caveat. We did not compare handing out pulse oximeters to no monitoring in this trial. So for patients that don’t have access to on-call clinicians or a remote monitoring program, using a pulse oximeter may be reasonable. But so is monitoring for shortness of breath..
This was a pandemic labor of love involving a huge team of researchers, clinicians, innovators. Special shout outs to @katleemd (implementation lead), @AnnaUma (Covid Watch medical director), @khchaiyachati (co-PI), and @76Dalton @PM_Innovation leader and everyone on the team...
...And special thanks to @PCORI for funding this effort.

With the limited word count in the paper, figure some extra context would be helpful.

And check out the supplement for the all the details and release below.

(nejm.org/doi/suppl/10.1…)

Thanks!/
pennmedicine.org/news/news-rele…
…PS if you are at #SGIM22, COVID Watch Medical Director and co-lead author @AnnaUma will be presenting this paper on Friday. Check out her presentation!

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with M. Kit Delgado, MD, MS

M. Kit Delgado, MD, MS Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @kit_delgadoMD

Jan 3
Some thoughts from seeing a ton of Covid in the ER this week

Good news: Aside from a few patients with compromised immune systems, didn't have to hospitalize anyone who was vaxxed/boosted

Bad news: Health systems are stressed to the max

We need to adapt...quickly

Thread👇
Below I’m going to cover

1) Clinical presentations based on vax status
2) The 3 types of hospitalization: For Covid, chronic disease exacerbated by Covid, incidental Covid
3) Differences in severity/hospital needs
4) Disruptions in care Covid+ patients
5) Health system adaption
This version of Covid is crazy contagious. Our ER hit record highs in patient volumes. Hospitals everywhere over capacity during this surge of people coming in with Covid.

There is a strikingly consistent pattern based on vax status…
Read 23 tweets
Dec 8, 2020
>310,000 people died from gun injuries in the U.S. from 2009-17. But unlike #Covid19 stats, the total # of cases is not well established due to data gaps.

Proud of @ElinoreJKaufman who led our new paper (jamanetwork.com/journals/jamai…) that dives deep into this epidemic

THREAD👇
2/ We can get exact counts of how many deaths there are per year. Those data are available on death certificates (wonder.cdc.gov/ucd-icd10.html) collected by @CDCgov. Here’s a nice overview by @jasongoldstick @cartpatr @StrohCunningham

jamanetwork.com/journals/jamap…
3/ However, estimates of nonfatal cases have been shaky because these stats are traditionally sampled from ~100 of nation's 5,000 ERs. That sample changes year to year and there’s wide variability in ER gunshot injury volume. See @fivethirtyeight @trace:

fivethirtyeight.com/features/the-c…
Read 24 tweets
Aug 18, 2020
1/ Had a patient present to the ER with severe nausea and vomiting. As has become my practice, I asked "Have you struggled with pain killers, heroin, or fentanyl?" and to my surprise he answered yes. This was his first episode of severe opioid withdrawal....

cont
2/ I gave him buprenorphine and within minutes, his symptoms were relieved. Avoided labs, CT scan. Wrote him a prescription to get him through next week. Couple of weeks go by, got a message from a primary care doc colleague that he has been doing great in clinic. Made my day
3/ I've learned from asking this question regularly that there's a lot of occult opioid withdrawal out there and patients are ashamed to bring it up, but once you ask and offer a path out, they are very willing to engage. Working on making universal screening routine. stay tuned
Read 5 tweets
Jun 4, 2020
A friend of mine of who runs an essential, office-based business with ~50 employees and 1 employee just tested + for #COVID19 after finding out about an exposure over Memorial Day weekend family gathering

A thread on running a business in the time of #coronavirus👇
2/ The business has had a zero-tolerance policy on coming to work with any symptoms. This employee who tested positive apparently felt a little tired last week, but no fever, cough, or obvious symptoms. The employee was notified via contact tracing of the exposure on Memorial day
3/ Upon being notified of the exposure, the employee was able to get a same day #COVID19 test which turned out positive. This triggered same-day tests of everyone exposed at work and shutting down the office for deep cleaning
Read 10 tweets
May 31, 2020
1/ Many lawmakers are citing preventing preventing "deaths of despair" (drug use, overdose, alcoholism, suicide) as the rationale to reopen prematurely. @meganranney & @drjessigold have written a must-read piece in @statnews addressing this. A thread 👇

statnews.com/2020/05/31/dea…
2/ First, while we have ample evidence of the impact of #COVID19 (>100,000 deaths, ~4x as many hospitalizations lasting 1-2 weeks, a lasting debility among younger adults that has yet to be quantified), we don't have any data yet on deaths of despair

3/ Second, the economic conditions that lead to deaths of despair tend not to be temporary. For example,
@atheendar found that it took 5 years after factory closures to see full impact on opioid overdoses
jamanetwork.com/journals/jamai…
Read 10 tweets
May 28, 2020
Appreciated the opportunity to comment on 2 excellent cohort studies led by @jLewnard (Kaiser) and @leorahorwitzmd (NYU) in @bmj_latest that give us a much better understanding of who gets sick from #COVID19. A thread putting this work into perspective👇

bmj.com/content/369/bm…
2/ Take away #1: It's not just the elderly who get sick.

Most who were hospitalized for #COVID19 were *under* the age of 65:
- 53% (n=1,452) in NYU cohort
- 60% (n=1,108) in Kaiser
3/ Take away #2: Those who were hospitalized spent 1-2 weeks in hospital. A trip to the hospital is no joke.

Median length of stay (75th percentile)
- NYU cohort: 7 days (13 days)
- Kaiser cohort: 10 days (17 days)
Read 11 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(