6

Medicine is not my field, nor do I have mesothelioma. I'm also not seeking any clinical trials or experimental treatment. However, sometimes I find myself trying my read medical papers, either out of sheer interest or if I have a serious complication. I want to stress that I don't think self-diagnosis is a good idea, I just can't help myself sometimes.

I find that quite a few papers I get from PubMed appear to be a "study of one". Consider this paper https://academic.oup.com/ejcts/article/41/6/1393/420831

Or this: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5728971/

I'm not saying this is a bad paper, I have no idea. However, I am curious as to how common and accepted this type of paper is within the field of medicine.

On one hand, I can see that for pushing the limits of surgery, or any technique, it is interesting to see what people have had success with, even if it is just one patient. On the other hand, it is just one patient. How can anything be generalized from this?

Is there a difference between treatment with medications and surgical technique? I can imagine that surgical techniques can be replicated with far more accuracy than results from medication, which are plagued by placebo and unclear method of operation (I'm making this stuff up, so please correct me).

Again, I have no grounds of criticizing the work of either of the papers, they are just used as examples, and as mentioned, there are quite a few such papers. Indeed, if I do a search for "we present a case of", I get a ton of results for actual papers that lead with this. I'm sure there are many other, similar phrases that are also used for these single study cases.

My question is simply, what does the academic medical field think of this type of paper? Can such a paper be a good paper? Is it usually considered a "bad paper", when compared to a paper that has, for example, many patients that received the same treatments, where variations and complications are discussed? Is this sort of paper simply part of the cutting edge of medical science, that all results are useful?

7

To build off of what JohnP said, scientific evidence has a hierarchy of reliability. Some types of papers, by their very nature, are more academically rigorous and likely to lead you to the truth.

Randomized controlled double-blind trials and meta-analyses are at the top of the hierarchy. These studies often have thousands of participants and are set up to produce statistically significant results. These studies form the backbone of evidence-based medicine.

A case study, which looks at a finding or condition seen in one patient, is at the bottom of the hierarchy. It's impossible to draw true evidence-based conclusions off of this type of paper, since it essentially says "here's what we saw, here's what we did, here's what happened".

However, case studies should not be written off as useless. There are two primary ways that these types of studies can be helpful.

  • If you are treating a patient with an extremely rare condition or constellation of symptoms, a case study is the best guide you're going to get. If someone else did something which worked in a similar patient, that treatment may work in your patient. Many of those rare diseases will simply never have enough patients to fuel a randomized controlled trial.
  • If there is an emerging phenomenon that has not yet been described in the literature. 2 notable examples are HIV/AIDS and the much more recent vaping pneumonitis. Both of these conditions were first identified by case reports (or case series, which is the same thing but with a handful of patients). This brought the issue to wider attention, and once this happened many other similar reports started pouring in. This can direct more academically rigorous research in that direction.

References:

First case series of patients with complications of AIDS

First case series of vaping pneumonitis

  • So randomized, controlled double-blind trials is at the top of the hierarchy, case studies is at the bottom. What is in the middle of the hierarchy? – AlphaCentauri Oct 24 at 19:44
  • 3
    There are lots of intermediate types of studies. Observational trials are used when the question can't be practically or ethically answered using a randomized controlled trial (for example, "Are abused children more likely to develop depression as adults?"). Here is a brief overview (keep in mind these aren't strict, a poorly done RCT is less trustworthy than a well-done prospective trial): libguides.winona.edu/c.php?g=11614&p=61584 – Nate Oct 24 at 20:04
5

"Studies of one" are also known as case studies, N-of-1 studies or case reports. I will refer to them below as case reports, for simplicity. They vary in quality just like any other type of studies. Case reports have a place in biomedical research and can often be very valuable, widely cited and/or otherwise influential. Double-blind randomized controlled clinical trials (RCTs), or, better, meta-analysis of multiple such trials, are the current gold standard. But they are obviously not always feasible for a variety of reasons.

There are specific guidelines for writing case reports, for example:

Riley D.S., et al (2017) J Clin Epidemiol. 89:218-235. doi: 10.1016/j.jclinepi.2017.04.026. CARE guidelines for case reports: explanation and elaboration document. https://www.sciencedirect.com/science/article/pii/S0895435617300379

This above article also lists a number of peer-reviewed journals that explicitly accept case reports (Table 1), important historical examples of case reports (section 1.1), and different types of case reports with specific examples (section 2).

Types of case reports:

  • Research where the number of patients is limited for any reason. For example, research conducted the beginning of a potential disease outbreak, research on uniquely informative patients, reports of rare drug side effects, drug-drug and food-drug interactions. Others in this thread have also listed rare condition treatment and emerging phenomena.

  • Research that carries high cost. For example, whole genome or whole exome sequencing when those studies were still very expensive.

  • Research that cannot be applied in RCT for any reason, such as limitations of input, treatment scarcity, ethical considerations, etc.

Specific examples and references:

  1. Chen H., et al (2014) Lancet. 383(9918):714-21. doi: 10.1016/S0140-6736(14)60111-2. Clinical and epidemiological characteristics of a fatal case of avian influenza A H10N8 virus infection: a descriptive study. https://www.sciencedirect.com/science/article/pii/S0140673614601112

We report the first human infection with a novel reassortant avian influenza A H10N8 virus.

  1. Byun M., et al. (2010) J Exp Med. 207(11):2307-12. doi: 10.1084/jem.20101597. Whole-exome sequencing-based discovery of STIM1 deficiency in a child with fatal classic Kaposi sarcoma. http://jem.rupress.org/content/207/11/2307.long

Whole-exome sequencing-based discovery of STIM1 deficiency in a child with fatal classic Kaposi sarcoma.

  1. Garrett-Bakelman F.E., et al. (2019) Science. 364(6436). pii: eaau8650. doi: 10.1126/science.aau8650. The NASA Twins Study: A multidimensional analysis of a year-long human spaceflight. https://science.sciencemag.org/content/364/6436/eaau8650.long

[...] significant changes in multiple data types were observed in association with the spaceflight period; the majority of these eventually returned to a preflight state within the time period of the study. These included changes in telomere length, gene regulation measured in both epigenetic and transcriptional data, gut microbiome composition, body weight, carotid artery dimensions, subfoveal choroidal thickness and peripapillary total retinal thickness, and serum metabolites.

Addressing your specific questions about "studies of one", N=1 studies, case studies, or case reports:

What does the academic medical field think of this type of paper?

It depends on the field, and the quality of the specific study, and can vary a lot between respected, influential, etc, and irrelevant, not even acceptable for publication in high quality journal.

Can such a paper be a good paper?

Yes, again, depending on the field, and the quality of the specific study.

Is it usually considered a "bad paper", when compared to a paper that has e.g. many patients that received the same treatments, where variations and complications are discussed?

All other things being equal, obviously a case study with N=1 would be inferior to a study with multiple independent patients. In cases like this, N=1 papers would have a hard time getting through peer review in high quality journals, because the reviewers are likely to be aware of the current standards in that specific field, such as RCTs, epidemiological or association studies. I doubt N=1 study on the effects of a well-known statin on cholesterol would be published, if multiple double-blind RCTs are already available on orders of magnitude more patients.

Is this sort of paper simply part of the cutting edge of medical science, that all results are useful?

In the context of this question, one can assume that a high quality case study is indeed "cutting edge", and there is a reason why N=1 was the best one could get at the time.

3

The difference is that you are looking at a case study, versus a scientific research study.

The basic difference is that a case study is an in depth look at a single instance of something which may not be repeatable by others, and a scientific study is a broader examination of a group with results and experiments that can be repeated by others.

As an example: Your second study is a specific surgical intervention in a case of a man that had a recurrent tumor in a specific spot with malignant pleural mesothelioma. This is not something that is likely to be repeatable over and over by others, but it still worth examining in case it happens again. So it's written up as a case study, and presented as such.

  • So a "case study" isn't frowned upon by medical academics? – AlphaCentauri Oct 24 at 15:49
  • 4
    @AlphaCentauri No, they're not. They're recognized for what they are. They're individual cases and only convey information to other doctors treating similar patients. They don't and can't establish standards of care. – Carey Gregory Oct 24 at 19:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.