Sometimes doctors prescribe steroid tablets to suppress pain for diseases, such as arthritis. Why do they prescribe it even though they know the side effects?


3 Answers 3


I think a missing bit of information that might help you get a better sense of this practice is: steroids are miracle drugs.

OK, that was in jest - no miracles here. Truth be told, though, if there is a single class of drugs that has added more quality-adjusted life-years to human history than any other, steroids must be competing with just a few antibiotic classes for that title.

To make clear what we’re talking about, the term “steroid” as a label for drugs generally refers to glucocorticoids (GCs) - drugs that act like cortisol, an endogenous steroid hormone. Commonly used GCs include:

  • Short acting: hydrocortisone, cortisone
  • Intermediate-acting: prednisone, prednisolone, mehthylprednisolone, triamcinolone
  • Long-acting: betamethasone, dexamethasone, paramethasone

Your question poses a specific example of one indication for steroids but seems to be asking about the use of these drugs more generally. As others have brought up, any decision about drug treatment involves weighing the risk-benefit ratio.


Although going through the efficacy data for various conditions is beyond the scope of this answer, I list a sampling of the common indications for GC treatment, and in the concluding paragraph I will provide specific efficacy data for arthritis.

  • Replacement therapy: due to either primary adrenal insufficiency (Addison’s disease) or secondary/tertiary insufficiency (at the level of the hypothalamus or pituitary.

  • Inflammatory conditions (deep breath in): asthma, eczema, inflammatory bowel disease, allergic rhinitis/sinusitis, eosinophilic anything, inflammatory arthritis, anaphylaxis, septic shock (in very specific circumstances), polymyalgia rheumatica, polyarteritis nodosa, temporal arteritis, minimal change glomerulonephritis, autoimmune hemolytic anemia, urticaria, autoimmune hepatitis

  • Other things that aren’t (necessarily) obviously inflammatory: increased intracranial pressure, intractable nausea, acute leukemia, sarcoidosis, cluster headaches, dermatomyositis


First, please note that a few of the indications for GC require only a very brief burst of high-dose GC. In this time course, these drugs actually very few serious side effects (psychiatric effects and hyperglycemia are exceptions, but rarely outweigh the benefit). Over the longer term, however, supra-physiologic doses of GC have an array of adverse effects that fall into a few categories:

  • Suppression of the hypothalamus-pituitary-adrenal (HPA) axis: Exogenous GCs suppress the hypothalamic pathway that stimulates the adrenal gland, resulting in adrenal atrophy. This means that if the exogenous GC is abruptly discontinued, adrenal crisis can ensue. The rule of thumb I learned was that >3 weeks treatment with >10 mg prednisone (or equivalent) can cause HPA suppression. As a result, GC treatment that meets these criteria is generally ended with a taper, allowing the adrenal to regenerate, whereupon it can resume normal function.

    This side effect is usually a non-issue if the medication is tapered correctly. Therefore, the benefits of treatment will generally outweigh the risks.

  • Cushing’s syndrome: This term is generally applied to the constellation of symptoms occuring with chronic administration of supra-physiologic doses of GC.

    • Hyperglycemia
    • Hypertension
    • Psychiatric effects: insomnia, precipitation of mania or psychosis in susceptible individuals
    • Salt retention1
    • Decreased bone mineral density
    • Weight gain: generally in a pattern characterized by central adiposity and peripheral muscle atrophy
    • Increased intra-ocular pressure

So, after all that:

Why do doctors prescribe steroid tablets though they knew the side effect?

Because the benefit outweighs the risk.2 For the example you gave — arthritis3 — the best paper I found was a meta-analysis that used a statistical method that compares the number needed to treat (NNT) with the number needed to harm (NNH) for long-term (>1 year) treatment of rheumatoid arthritis. The ratio there NNH/NNT was 0.25, which is considered “good.” I’ll let you read their methods if you’d like the details. For further reading on this methodology and insights into how we quantitatively assess the comparison between risks and benefits, this article by Guo et al. provides some good bedtime reading.

1. This is actually a mineralocoricoid rather than a GC effect, only present for those GCs whose receptor specificity overlaps, most prominently the short-acting agents hydrocortisone and cortisone.

2. Usually. Steroids are probably over-used in some settings because most people know that steroids tend to “make everybody feel better.” They can be an easy way to make patients happy if the side effects aren’t fully appreciated. Thus (in case you needed me to tell you!), your question is a valid one.

3. I’m assuming here rheumatoid arthritis, because this is the most common type of arthritis for which steroids are appropriate. For the more common osteo-“arthritis” — a.k.a. degenerative joint disease — I’m sure steroids would make patients feel better, but the benefit there rarely outweighs the risk.

Guo JJ, Pandey S, Doyle J, Bian B, Lis Y, Raisch DW. A review of quantitative risk-benefit methodologies for assessing drug safety and efficacy-report of the ISPOR risk-benefit management working group. Value Health. 2010 Aug;13(5):657-66.

Liu D, Ahmet A, Ward L, Krishnamoorthy P, Mandelcorn ED, Leigh R, Brown JP, Cohen A, Kim H. A practical guide to the monitoring and management of the complications of systemic corticosteroid therapy. Allergy Asthma Clin Immunol. 2013 Aug 15;9(1):30.

Mycek RJ, Harvey RA, Champe PC. Pharmacology. Lippincotts’s Illustrated Reviews, 2nd Ed. Lippincott, 1997: 272-276.

Ravindran V, Rachapalli S, Choy EH.Rheumatology (Oxford). Safety of medium- to long-term glucocorticoid therapy in rheumatoid arthritis: a meta-analysis. 2009 Jul;48(7):807-11.

  • 7
    +1 - This answer could not be any better. Great information and references. Commented Apr 4, 2015 at 6:57
  • 2
    @anongoodnurse Couldn't agree more. This is one of the best answers I've seen! Should be the accepted answer. Commented Apr 6, 2015 at 4:39

This is a very good question.

Simply put: among doctors we call them wonder drugs or life saving drugs. In a number of emergencies, such as severe drug reactions, life-threatening asthma or allergies, or inflammatory conditions (such as your example of arthritis), no other drugs act like steroids. They act quickly and effectively, reducing inflammation and decreasing allergic responses, opening swollen airways, etc. They are used because they are so effective.

Steroids do have side effects; they can cause kidney stones, decrease bone density, may cause diabetes, nervous system related diseases and many more, but only when used continuously for more than 3 to 6 months or so. Usually, however, steroids are only used until the patient is out of danger or experiences relief of their sudden severe pain.

Steroids are not alien to our bodies; they are naturally produced by the adrenal gland, which lies above kidney. Therefore they are unlike paracetamol (which we take for fever) and other drugs which are synthetic (not natural). By giving steroids to patients, we are just strengthening a natural response our bodies attempt to mount.

I suggest you not take any steroid for more than 3 months, and only under a doctor's supervision. Your doctor can advise you of any alternatives to steroids if available.

  • 2
    Good information, but hard to read. Can you consider rewriting it to make it more clear? Commented Apr 1, 2015 at 20:25
  • 2
    This post has the makings of a very good answer, but here on Health, we strongly encourage using references. They are the only way in which we can tell if information is reliable or not. If you are struggling to find good, reliable sources, check out this post. If you want to learn more about our site's stance on answers without references, check out this meta post. Thanks :)
    – michaelpri
    Commented May 28, 2015 at 1:11

The reason drugs with adverse side effects are prescribed are

  1. Generally the benefits outweigh the risk,
  2. Every side effect that was discovered during trials, no matter how rare, must be listed, and
  3. Hopefully the FDA in the US or the similar administration in the other countries did their due diligence in order to protect consumers from drug companies.

Doctors only prescribe untested drugs to terminal ill patients who sign numerous waivers in order to take them. So on average, a patient should only rarely experience the side effects and usually the milder ones will present; however, some people will present with the more serious adverse reactions.

Additionally, if any side effect does present itself, the patient should consult their physician since alternatives may be available.

  • I'm unsure what the FDA has to do with this particular question. Steroids have been used for decades and their side effects have long been recognized. Commented Apr 4, 2015 at 1:26
  • @medica it is basically saying as long as whatever agency for your country is looking at for the consumers, there shouldn't be too much to worry about but this don't take all the risk off.
    – user139
    Commented Apr 4, 2015 at 2:10
  • I disagree. As I said, the FDA has long known about the risks of steroids, yet they are still widely used. The FDA is not in the position of protecting the public from commonly known side effects. They are in the position of assuring proper labeling. That really is the main role of the agency. If drugs which have unacceptable side effects (i.e. heart attacks, sudden death, strokes) compared to the benefit of the drug, the FDA will sometimes not approve the drug; however, if it's labelled correctly, often the drug will still be approved. See Avandia for example. Commented Apr 4, 2015 at 2:18
  • 1
    @medica you misinterpreting what I meant. What I mean is if they are doing their job correctly, the risk of absolutely horrible drugs coming on the market will be low. That is, physicians can take some solace in knowing the drugs should be okay for human consumption. However, if they didn't care, you would have to be wary of every drug in the market.
    – user139
    Commented Apr 4, 2015 at 2:25
  • 1
    You do have to be wary of every drug on the market. I can't tell you how many times I've heard patients who took Tylenol overdoses say "But if it wasn't safe, why would the government allow it to be sold over the counter?" This was when Paracetamol was the leading drug for suicide in the UK. Commented Apr 4, 2015 at 2:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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