When to Use Antibiotics: Persistent Cough Case Study for Nurse Practitioners
A patient comes in with a cough that's been hanging on for three or four weeks and says, “I need an antibiotic.” How do you figure out whether they actually need one—and when does a persistent cough need more evaluation?
In this case study, I walk through a patient with a 3–4 week cough, including the differential diagnosis, how I think through bronchitis vs. pneumonia and sinusitis, when antibiotics may or may not be appropriate, and when to consider a chest X-ray.
2026 Update: Originally published in 2020 and reviewed for clinical accuracy in 2026. The overall clinical approach remains useful, but recommendations around bacterial sinusitis, antibiotics for bronchitis, and when to obtain chest imaging have changed or need clarification. See the Clinical Updates as of 2026 below the video.
For current guidance, see the updated Persistent Cough: Subacute + Chronic Cough Evaluation Cheat Sheet.
Listen
Watch
What I Covered in This Episode:
How to approach a patient with a cough lasting about 3–4 weeks
History questions that help narrow the differential, including what they've tried, whether this has happened before, and whether symptoms are improving or worsening
The differential diagnosis for a subacute cough, including postinfectious cough, sinusitis, bronchitis, pneumonia, postnasal drip, asthma, GERD, and medication-related cough
Clinical findings that can raise concern for pneumonia or another more serious respiratory illness
Why purulent or colored sputum alone doesn't tell you whether an infection is bacterial
How to think through bronchitis vs. pneumonia
When a patient with persistent respiratory symptoms may need a chest X-ray
When antibiotics may or may not be appropriate
Supportive care and patient education for persistent respiratory symptoms
Clinical Updates as of 2026
A few recommendations have changed or need clarification since this episode was recorded:
A cough lasting 3–8 weeks is considered subacute, and postinfectious cough is a common cause. If the cough began with a respiratory infection, the other symptoms have resolved, the patient is otherwise well, and the cough is gradually improving, further testing or antibiotics may not be necessary.
Acute bacterial rhinosinusitis isn't diagnosed by duration alone. Rather than using a simple “7 days = viral, longer = bacterial” cutoff, look for persistent symptoms ≥10 days without improvement, severe symptoms at onset, or worsening after initial improvement (“double worsening”).
Don't treat a persistent cough as “bacterial bronchitis” simply because it has lasted several weeks. Uncomplicated acute bronchitis and postinfectious cough generally do not require antibiotics. Consider specific treatable infections such as pertussis when the history or exposure suggests them.
A chest X-ray isn't automatically needed because a cough has lasted 3–4 weeks. Consider imaging when the clinical picture raises concern for pneumonia or another pulmonary process—for example, abnormal vital signs or oxygenation, focal lung findings, dyspnea, pleuritic chest pain, systemic illness, an atypical course, or a cough that isn't improving.
If pneumonia is diagnosed, severity scores support rather than replace clinical judgment. The Pneumonia Severity Index (PSI) is generally preferred over CURB-65 to help inform outpatient vs. inpatient management in adults with community-acquired pneumonia.
COVID-19 wasn't included in the original differential because this episode was recorded before the pandemic. Consider SARS-CoV-2 along with other respiratory infections when clinically appropriate. For a subacute cough, ask whether it began with a recent COVID-19 or other respiratory infection and whether the other symptoms have resolved.
For a current step-by-step approach, see the Persistent Cough: Subacute + Chronic Cough Evaluation Cheat Sheet in the Digital NP Binder.
Key Takeaways - 2026 Updates
Duration matters: acute <3 weeks, subacute 3–8 weeks, chronic >8 weeks.
Postinfectious cough is common in the 3–8 week window.
Colored sputum alone doesn't mean bacterial infection or justify antibiotics.
Reassess the whole picture: trajectory, vitals, oxygenation, lung exam, and red flags.
Chest X-ray is selective, not automatic.
Think beyond infection when the cough persists.
At 8 weeks, shift into a chronic cough evaluation rather than continuing empiric treatment for a respiratory infection.
Need a Step-by-Step Approach to Persistent Cough?
The Persistent Cough: Subacute + Chronic Cough Evaluation Cheat Sheet helps you work through a cough that isn't going away—including red flags, common causes by duration, when further workup is needed, initial chronic cough testing, and when to consider referral. Get the Persistent Cough Cheat Sheet in the Digital NP Binder.
Resources that correspond with this episode:
If you liked this post, also check out:
-
Liz Rohr:
Hey there. It's Liz Rohr, from Real World NP, and you're watching NP practice made simple though, weekly videos to help save you time, frustration and helped you learn faster so you can take the best care of your patients.
So it is the season for runny noses and coughs and patients asking for antibiotics, whether or not they need them. And the time period between three and four weeks of having this kind of persistent cough I find to be the most confounding for new and experienced clinicians alike, honestly. And so in this case study, in this week's case study, we'll be talking all about that and going over the differentials, the workup, the treatment, who needs antibiotics and who doesn't, and then the role of chest x-ray. So hopefully it's really pertinent to what you're seeing every day in clinical practice for primary care. So without further ado, I'm going to share my screen with you.
So let's jump in. This is Deborah. She's 57. She has a persistent cough for three to four weeks. Again, this is not her real name or her photo. Lovely photo though. So chief complaint is, "I need an antibiotic." So I'm sure you hear that quite a bit. The story is, again, the persistent productive cough for three to four weeks. It's "a cold that won't go away" or "nasty cough." She has some sick contacts. She works in a daycare so she's around sick kids and sick adults kind of all the time. She's "tried everything." She's a nonsmoker. She drinks alcohol occasionally. Her past medical history, past surgical history are hypertension, cholecystectomy, and she has some seasonal allergies from time to time. No family history that she knows of.
She has one male sexual partner and she's menopausal so doesn't need any contraceptives. She takes losartan, 50 milligrams daily, although not taking any other meds. Blood pressure of 128/72 today, heart rate of 72. She's afebrile and her oxygen is 98%.
So for her review of systems, she has rhinorrhea, so runny nose, sore throat, coughing, throat clearing, and some facial pressure. Negative though for shortness of breath and wheezing, chest pain, ear pain, fever, chills, nausea, vomiting, or diarrhea. On physical exam ... I kind of mushed this all together so it was on one slide, but effectively it was negative except for a mild facial tenderness over maxillary and frontal sinuses, enlarged turbinates in her nose, purulent mucus kind of present in her nostrils, cobblestoning of her pharynx with some postnasal discharge. Actually, I didn't write that down. Her eyes were clear. Her TMs bilaterally were dull, but not red or bulging. Her lung sounds were a coarse sound, but otherwise clear.
So what other questions do you want to ask? I left a couple of things out. So what medications specifically is she taking? Not just, "I've taken everything." That's really kind of important to get that full history. And has she had this happen before? I find that that question, when I ask that, is really, really helpful because it helps me from asking a bajillion things that could easily be elicited if I just kind of simply started with that question. And then I always ask, are they better, worse, or the same? That really gives you a great clinical picture of where they're at in their illness. Are they almost completely better or do they feel like they're getting a lot worse? Things like that.
And then respiratory rate. I did not include that in the vital signs. That's something you really want to think about when you have a respiratory illness, of not just, what is the standard that everyone seems to get, right? What is the actual respiratory rate? Because something that's 20, 22, 24 and up are a little bit more concerning and more significant in terms of your physical assessment for a respiratory complaint.
So what are some differential diagnoses here? So sinusitis, bronchitis. There's a note that there's both viral and bacterial causes. So pertussis, chlamydia, and mycoplasma pneumonia. I may or may not be saying that correctly, but those are the pathogens we're talking about there. Pneumonia, and this can be from viral influenza or bacterial. And actually it's indistinguishable on a chest x-ray if pneumonia from one to the other, fun fact. And then you can get people with a persistent cough after they've had an infection. So they have had a bronchitis and this kind of dry, annoying cough lasts for several weeks. But typically those people don't feel unwell. They just have this annoying cough that hangs on for a while. And that you can kind of uncover with the history.
Some other causes, post-nasal drip. You can have a viral or allergic cause, and that's just inherently the problem that's causing that cough. Asthma is definitely one you want to think about. GERD, reflux, and people who are on ACE inhibitors can have this kind of persistent cough, and then some others. So what we're talking about here is a subacute cough. So acute cough is less than three weeks, subacute is considered to be three to eight weeks, and chronic cough tends to be greater than eight weeks. So what we're talking about is the differential diagnosis, not only for this patient, but for somebody with a subacute type of cough. So she's right around the three week mark, but she's not really sure it could maybe could be four weeks.
And I have starred here the sinusitis, bronchitis and pneumonia, because those are kind of our top three that we want to rule out. Couple of notes, I'm not really going to get into the other things, the post-nasal drip, the asthma, GERD, things like that. If you have somebody who's not really presenting like this patient ... this patient is kind of presenting as having had a cold for a very long time. But if you have kind of a subacute to chronic cough that doesn't have that kind of illness type of presentation, you want to consider that second column a little bit more. And for those, the treatment is dependent on what's going on.
And then to just make a note, you just want to make sure that you're not kind of like missing a CHF kind of gradually worsening situation, or a chronic bronchitis is something to consider with somebody who smokes, who has kind of this persistent cough with a whitish sputum. But again, those patients are a little bit less of a acutely infectious presentation like this patient is.
So I have a differential diagnosis table. I used to really, really resist. I was very stubborn in school with doing these because I found them to be very tedious. However, they're very helpful for organizing your thoughts and I'm going to be doing one, so you can use this tool as well. It is helpful sometimes, but I'm just stubborn.
So sinusitis is the first one. So inflammation of your sinus cavities, it can be viral or bacterial. They can start with cold symptoms and have this kind of a classic history of a double sickening where they get sick, they get better, and they get sick again. They may or may not have purulent sputum, and this is kind of a non-specific finding. So if it's less than seven days, it's considered to be viral. This is the general thought. If it lasts greater than seven to 14 days, it's typically more on the bacterial side. But you can have a viral cause with purulent sputum if it's under that seven days.
So things to watch out for if you come to this conclusion. So you're looking for high persistent fevers. It's pretty concerning if you have sinusitis. You want to make sure that they don't have periorbital swelling or erythema because worst case scenario if someone had sinusitis is that you're progressively getting worse and it's spreading to your ... around your eye to your face, and worst case scenario, to your brain. So abnormal extraocular movements, cranial nerve palsies, vision changes. These are all the things that I'm charting negatively if I'm diagnosing this in somebody. Severe headache, meningeal signs, altered mental status. I'm kind of making sure that I paint the picture of writing that note, that they don't have any of those things.
So bronchitis, this is, I feel like, the hardest part, I think for me, and for clinicians in general, is bronchitis versus pneumonia for people who just have this hanging on cough that are just like, "I really need antibiotics," right? So I just really want to break it down here. Bronchitis, it's characterized by a cough of about one to three weeks, and the median is about 18 to 21 days, so it can be a little bit longer. And it can start with "cold symptoms," like a sinusitis type of symptoms, like runny nose and sore throat and things like that. They, again, may or may not have purulent sputum. This is non-specific and does not help you diagnose if it's bacterial or not, really unfortunately.
So they may have wheezing, mild dyspnea, and they could have some ronchi that clear with coughing. And it can be from a viral etiology, so from a regular garden variety cold, or influenza can cause it. And bacterial is pretty rare, but it is an option, right? So less than 6%, depending on the numbers that you're looking at. Pertussis is kind of the main one that is underdiagnosed, it sounds like, in adolescents and adults, and that is that kind of whooping cough, where that cough starts with cold-like symptoms and then it continues with this persistent cough. Mycoplasma, chlamydia pneumonias can actually self-resolve. They can cause not only a bacteria ... not only bronchitis, but also a pneumonia. But fun fact, they can also self-resolve or they can progress, so something to think about.
So things to watch out for when you're concerned somebody has bronchitis. You want to watch out for a fever because usually people with bronchitis don't have a fever. That's more of a pneumonia. Tachnypnea, hypoxia, again, more consistent with pneumonia than bronchitis. Posttussive emesis, paroxysms of coughing, an inspiratory whoop. Again, those are kind of those classic signs of pertussis. So you're kind of screening for that in the back of your mind. Are they at risk for pertussis, which is treatable?
And again, the reason why so many people get Z-Packs, if you're wondering about that ... excuse me, azithromycin. Most people call Z-Packs ... is because the bacterial bronchitises that we're talking about our pertussis, mycoplasma, and chlamydia, which are all treated with azithromycin, spoiler alert. Pleuritic chest pain. Again, that's more consistent with pneumonia. And confusion, especially in older adults, is really concerning and that's more concerning for a pneumonia process. Decreased breath sounds and crackles, you want to be really mindful of because those are more consistent with the pneumonia. And those people should all get a chest x-ray because you're wanting to make sure that they're not at risk for pneumonia.
So pneumonia, last one. So cough, fever, and pleuritic chest pain are pretty classic. Tachypnea and hypoxia and tachycardia as well actually. So respirate of in the mid-20s and beyond and hypoxia below their baseline. In tachycardia, there isn't really a normal range for that, but can be found. They can have mucopurulent sputum, or they can have a scant/watery sputum, which is kind of more consistent with the atypical pathogens according to my reading. Again, people talk about rust colored for that pneumococcal pneumonia, but it's kind of rarely seen, so it's not that helpful. Nausea, vomiting, and diarrhea may also occur with pneumonia.
Unfortunately, the kicker for this is that there's no clear constellation of symptoms. It really doesn't correlate. In studies where they've done chest x-rays, verified pneumonias, and compare their symptoms, there isn't necessarily a clear constellation really unfortunately, which is such a painful part.
So things you're watching out for, they're kind of like obvious overt ones, that are kind of those vital sign instability. And it's kind of more clear, like, "Oh, you have pneumonia." That's not that common, unfortunately. I mean, fortunately for us, but unfortunately for our diagnostic process, you don't see it that much. So CURB-65 is one of those scoring systems that you can use to look at their risk factors. And off the top of my head, I apologize, I forget what the B stands for, but it has to do with confusion, the BUN, the respiratory rate, and if they're 65 and older. And you kind of plug that in, and that score will tell you whether or not they need to go to the hospital or can be safely treated outpatient.
And again ... oops. Need a chest x-ray. That is the gold standard for diagnosing pneumonia. So you can look at all those symptoms and say most likely they have pneumonia. But really you should have that, again, because the studies just don't pan out in terms of, these symptoms correlate with pneumonia. So false negatives can occur. And if it's less than 24 hours to the onset of their symptoms, if they have neutropenia, if they're dehydrated ... and there's a type of pneumonia called PCP pneumonia, and I'm not going to pronounce it correctly so I'm just going to keep abbreviating it to PCP. And it actually has been renamed to a different name. But anybody who's immunocompromised can get this type of pneumonia and it may not show up on a chest x-ray. So HIV, immunocompromised, things like that. So sputum culture, it's not indicated as an outpatient. It's really all based on your chest x-ray.
So again, going back to the table here, so sinusitis, pertinent positives and negatives. You've got the sinus tenderness, greater than 14 days post-nasal drip, nasal mucus. It doesn't have that classic presentation. Bronchitis, she's definitely had some coarse lung sounds. It's an acute onset of a persistent cough. And she doesn't have any symptoms of pneumonia. And also, I did make a note about this before, is that if somebody has underlying COPD, you want to treat those a little bit differently because you're looking more at exacerbations of COPD versus bronchitis if that makes sense, I can definitely make a video about COPD if you're interested, but kind of exclude these people from this presentation.
So sinus tenderness is more specific to sinusitis, but she could have concomitant sinusitis and bronchitis really. Pneumonia, she has an acute persistent cough and it's mucopurulent. Again, she doesn't have those kinds of other signs that we think about, chest pain, normal vital signs, no crackles on exam. But based on what I said before, the fact that you can't really differentiate one from the other just based on symptoms alone, that's really why you need to do a chest x-ray.
So management for her. So I'm going to diagnose her with a sinusitis because of her facial ... just because of the full history and it matches up a little bit better, right? So saline sinus rinse is really the kind of first-line thing that I advise people to do and do not skip this and only take antibiotics, because this is going to help you feel so much better. And we'll talk about that in a second.
Amoxicillin clavulanate for about seven days as a treatment. So you can either do ... if they have no risk factors for resistance, you can either do amoxicillin or amoxicillin clavulanate at the regular dose for seven days. And the range is about five to 10 days, but the research doesn't necessarily support the longer duration because it has more side effects and it doesn't necessarily have that great of a benefit. There's a whole bunch of risk factors for resistance, which I can include if you're interested in that. I just didn't want to make this too long.
So plus or minus nasal steroids. I feel like I've seen this go back and forth a couple of times since I graduated, but it's recommended it can be beneficial by evidence that you could give an inhaled fluticasone or something like that. I'm trying not to use brand names so I'm just going to say it as that. You know what I'm talking about though.
And antipyretics/analgesics, so acetaminophen, ibuprofen. And then other treatment really kind of lacks evidence. So decongestants, antihistamines, things like that can cause more issues than not.
So before I kind of move on to just kind of recapping, I just want to say with ... this is the trickiest part, I think about this mucopurulent cough for about three weeks. When you have a chest x-ray that's negative, you could stop. And if they don't have any of the other risk factors, and you're pretty convinced that it's a bronchitis, most of the time, it's going to be like a viral bronchitis. But if you do find that it's a ... if you're really worried about a bacterial bronchitis, that's why you kind of see those Z-Packs, I guess, if that makes sense.
But hopefully this kind of presentation overall is helpful for you in terms of determining that path, and again, just reassurance in your own self, knowing that the risk of bacterial bronchitis is actually pretty low. So as hard as it is to say no to antibiotics ... I said yes in this case, because it was more consistent with a sinusitis, but if you have somebody more consistent with a viral bronchitis, then you've kind of got that evidence behind you. But again, you do have the tools if you need to treat somebody.
So we reviewed her antibiotic treatment, just finishing the full course, right? Nasal sinus rinse irrigation instructions. So you have to make sure that it's boiled and cooled water, not just plain tap water. You can either use distilled water from the pharmacy. You need to use the included salt, buffered salt packets, and not table salt, because that's going to burn your nose really bad, which kind of freaks patients out and it makes them not want to necessarily follow those directions. But it's kind of a long conversation about this is going to help you feel better sooner than the antibiotics.
And alarm signs reviewed. Again, those kind of scary things you think about with sinusitis of like extraocular movements that are impaired, that redness and swelling in the face, things like that. And her blood pressure is controlled, which is awesome. So she can come back in about three to six months. And that really just depends on her personal comfort level of how well controlled she feels it is. I'm fine with her coming back in six months, but if she feels like, you know what? I'd really rather check in in three months, totally fine.
Did you like this video? If so, hit like and subscribe, and share with your NP friends, so together we can reach as many new grads as possible to help make their first years a little bit easier. And don't forget to head over to realworldnp.com and hop on the email list. You'll get your copy of the ultimate resource guide for the new NP. And you'll get these videos sent straight to your inbox every week with bonus content that I really just don't share anywhere else.
If you're watching this when it's going live, it's Thanksgiving week. And so I'd love to hear from you. Leave a comment below what's something that you're really grateful for? I'd just love to say that I'm really grateful for you for watching this video or watching the other videos if you've been following along. It really means so much to me to kind of be here with you. And I hope that you find them really helpful. Thank you so much again for watching. Hang in there, and I'll see you next time.
Click Here for Your Ultimate Resource Guide for the New NP
Snag our favorite, time-saving, evidence-based resources to help you take the best care of your patients (while keeping your sanity). PLUS you'll get helpful articles, podcasts, and YT video episodes sent to your inbox every week or so.
© 2026 Real World NP. For educational and informational purposes only, see realworldnp.com/disclaimer for full details.
