“The young physician starts life with 20 drugs for each disease, and the old physician ends life with one drug for 20 diseases.”
Sir William Osler (1849 – 1919)
The world has recently been devastated by a strange and rather specific apocalyptic event that has left us with the following:
1. Clean water
2. A proper sewer system
3. Adequate shelter
4. An abundance of good food
5. and sunshine
6. Adults only
7. All available vaccines
8. No chronic infectious diseases such as HIV and TB
9. No surgical or obstetrical conditions
10. No politicians or guidelines (hey, this is OUR fantasy…allow us to dream!)
A Thought Experiment
What drugs, currently available today, do you think are really the most useful? Imagine a world where you could only have 20 drugs to provide the care for your patients. Assume cost is NOT an issue and one can use all available dosage forms.
Which ones would be in your list?
Here is what we came up with. Amazingly, the mean date of release/patenting of these 20 medications was the mid 1950s (before any of us were born).
Here is the basis for our decisions
MORPHINE
It was felt an opioid pain medication should be included because of their established value in treating moderate to severe chronic and acute pain. Morphine was chosen because we are unaware of any convincing evidence that other newer opioids provide an important clinical advantage over morphine and it comes in a number of useful dosage forms.
ACETYLSALICYLIC ACID/ASA/ASPIRIN
Non-steroidal anti-inflammatory drugs (NSAIDs) have a broad range of therapeutic uses including pain/fever control and in some cases cardiovascular protection. There was considerable debate as to which NSAID should be included but ASA was agreed upon because we are unaware of any solid evidence suggesting other NSAIDs such as ibuprofen or naproxen are more effective for pain/inflammatory/fever conditions or safer overall than ASA. In addition, ASA has some other useful indications with evidence it reduces the risk of cardiovascular events in both acute and chronic cardiovascular situations.
EPINEPHRINE/ADRENALIN
Allergic reactions to foods, medications, and insect bites can, for some individuals, have immediate, severe and even life-threatening complications. Given the severity of this condition it was felt epinephrine/adrenalin was an important addition to the list. Its usefulness in cardiac arrest is an added bonus.
INSULIN
Insulin is basically essential for the survival of patients with Type 1 diabetes. In addition, it can be used to control blood sugars in patients with Type 2 diabetes, in particular those who are symptomatic from their elevated glucose. Insulin is usually given as a subcutaneous injection and is available in many different forms, which allow for different onsets of action and duration.
LEVOTHYROXINE
Levothyroxine, is routinely in the top 3 most-prescribed medications worldwide and hypothyroidism and subclinical hypothyroidism occurs in roughly 0.5-1% and 5-10% of people respectively. In true clinical hypothyroidism, access to levothyroxine is a simple, life-saving product.
DEXAMETHASONE
Corticosteroids have a very broad use ranging from acute life threatening allergic/pulmonary conditions to more chronic uses in pulmonary, rheumatological, gastrointestinal, neurological (Bell’s palsy and multiple sclerosis flares), and numerous dermatological and ophthalmological conditions. We chose dexamethasone over agents like prednisone, prednisolone and betamethasone solely because it is available in all the essential dosage forms (oral, topical, inhaler, nebulizer, and parenteral). In addition, we are unaware of any convincing evidence that supports the superiority of one corticosteroid over another. Steroid development in the 20th century represented one of the true “miracle drugs” – the other ones being insulin and thyroid.
ORAL CONTRACEPTIVE
Oral contraceptives are used by tens of millions of women worldwide. In addition to birth control, they are effective for controlling dysmenorrhea, premenstrual syndrome, acne and could even potentially be used to help control menopausal symptoms.
SALBUTAMOL/ALBUTEROL
Salbutamol was included on the list because of its important use in acute and chronic pulmonary conditions. No other beta-agonist has been shown to be more effective or safer than salbutamol.
QUETIAPINE
Including this medication in the list may be somewhat of a surprise. It was felt having access to medications for depression, schizophrenia, and other psychological disorders was important. While quetiapine is not the treatment of first choice for depression, anxiety or even bipolar disorder, it does have evidence for those conditions as well as schizophrenia. We felt this medication is sort of the “jack of all trades; master of none” in this class.
LORAZEPAM
While benzodiazepines can lead to dependence, they do play a useful clinical role when used judicially for anxiety, insomnia, acute seizures, as well as for acute sedation for medical procedures and aggressive patient’s issues. Lorazepam was selected because it is relatively short acting but can also be used chronically and it comes in useful dosage forms.
DIPHENHYDRAMINE
An antihistamine was considered to be an important addition to the list for its use in both acute and chronic allergic reactions and also as an antinauseant and potentially for insomnia. We chose diphenhydramine because it is available orally, parenterally and topically. There is no convincing evidence that any antihistamine is superior when it comes to efficacy. While diphenhydramine is typically thought of as a somewhat more sedating antihistamine, the sedation can be minimized by starting with much lower doses and if it does occur, tolerance develops within a couple of days.
OMEPRAZOLE
A proton pump inhibitor (PPI) was chosen given its usefulness for many upper gastrointestinal conditions including heartburn and H.pylori infections when used in conjunction with antibiotics. Histamine 2 receptor antagonists like ranitidine have been shown to be less effective than PPIs for heartburn. Omeprazole was chosen because it was the first in the class and it is available both orally and parenterally. However, given there is no evidence that any PPI is more effective than another; the choice is somewhat arbitrary and should likely be based on cost.
FUROSEMIDE
Furosemide is on the list given its ability to provide rapid relief when given parenterally for acute pulmonary edema and it is very much the mainstay of the symptomatic treatment for heart failure. It also has some use for edema and fluid overload associated with liver and renal disease.
METOPROLOL
Beta-blockers have a very broad range of important clinical uses including hypertension, heart failure, angina, post MI, migraine prevention, atrial fibrillation, essential tremor, and anxiety disorders. There is likely very little meaningful difference among beta-blockers and metoprolol has some of the best evidence in heart failure and has been used clinically for decades.
LOSARTAN
Blockers of the renin-angiotensin system have numerous clinically important uses for common conditions including hypertension and heart failure. Despite angiotensin converting enzyme (ACE) inhibitors having more evidence and experience when it comes to treatment of hypertension, heart failure and post-myocardial infarction patients, an angiotensin receptor blocker was chosen because of the issue of cough limiting their use in 10-25% of patients. Losartan was chosen because it was the first in the class but an angiotensin receptor blocker choice would be based on cost as we felt there is little difference between them.
AMOXICILLIN/CLAVULANATE
Deciding which antibiotics to include created considerable discussion and debate. Ultimately three different agents were included in the list. It was felt this agent provided broad coverage against many different organisms and could be used for many different types of infections. Amoxicillin/clavulanate gram-positive coverage includes S. pneumonia, pyogenes, and viridans, enterococci, and S. aureus methicillin sensitive. From a gram-negative perspective, it covers non-betalactamase and betalactamase producing E. coli, K. pneumonia and H influenza. It also covers S. marcescens, P. mirabilis, Acinetobacter sp., Citrobacter sp., Enterobacter sp. along with anaerobes above the diaphragm. Amoxicillin/clavulanate can be used for a variety of infections including pulmonary, genitourinary, skin and soft tissue infections, and dental infections.
DOXYCYCLINE
Doxycycline covers S. pneumonia, Staph aureus, including community acquired methicillin resistant species, H. influenza, and atypical organisms like mycoplasma, legionella, and chlamydia. It has uses in pulmonary and skin and soft tissue infections and it can be used for acne, Rocky Mountain spotted fever, cholera, syphilis, Q fever (yes we know you don’t see this every day), and Lyme disease. In addition, it can be used in penicillin allergic patients.
METRONIDAZOLE
Metronidazole covers a number of organisms missed by the previous two antibiotics. Metronidazole covers anaerobes above and below the diaphragm, and can be used for bacterial vaginosis, trichomoniasis, rosacea, and C. difficile.
FLUCONAZOLE
Fungal infections in the vagina, throat, mouth and skin create a considerable source of morbidity and can also cause systemic infections. For these reasons an antifungal was included in the list. Fluconazole is as effective as any of the other available antifungals.
POLYETHYLENE GLYCOL (PEG)
Constipation causes considerable morbidity to a large percentage of the population especially in the elderly and people being treated with narcotics. Because of this, it was felt an effective and well-tolerated laxative should be in the top 20. PEG was selected because the evidence suggests it is as good or better than any other laxative.
Drugs that didn’t make the top 20
Clearly with only 20 available spots available a number of very useful medications were not included. Our rationales for exclusion are as follows:
Chlorthalidone – while chlorthalidone should likely be considered the drug of first choice for most people with elevated blood pressure because it has few other applications, it was felt betablockers and ARBs would provide more overall societal value. In addition, because we included furosemide, chlorthalidone’s diuretic effect really didn’t add any value
Paracetamol/acetaminophen – while this drug is useful for mild to moderate pain and fever control, NSAIDs are as good as or better for a number of pain conditions and fever. In addition, eradicating H. pylori – which could be done with drugs on the list – very much reduces the risk of gastrointestinal complications secondary to chronic use of NSAIDs.
Nitroglycerine – is an extremely effective agent for the acute and chronic control of angina acute exacerbations of heart failure and was certainly a potential top 20 option. However, because we have beta-blockers and furosemide, nitroglycerin wouldn’t add much benefit. We know leaving this product off the list will clearly have disappointed those people with tennis elbow and anal fissures and especially those who develop tennis elbow by trying to treat their anal fissures.
Warfarin/novel oral anti-coagulant – this was another very difficult choice. It reduces stroke in atrial fibrillation by two thirds and is very helpful in a variety of conditions like pulmonary embolism, deep vein thrombosis or mechanical heart valves. The biggest reason it did not make the list is we couldn’t figure out which of the 20 to remove.
Metformin – while metformin is currently the most popular agent used in T2DM, the actual evidence for benefit in T2DM is highly inconsistent. In addition, insulin is in our list and would be able to control glucose alone.
Statin - while statins represent one of the most frequently used medications in the world, we felt despite the fact they very likely do reduce the risk of cardiovascular disease they really provide no additional benefit on any symptomatic conditions. We do realize statin’s exclusion will likely expose us to the wrath of those people who believe statins should be in the water supply.
Amitriptyline, nortriptyline, bupropion, fluoxetine – while having a number of potential uses including neuropathic pain, depression, anxiety and smoking cessation these agents didn’t make the list because they offered little additional value when lorazepam and quetiapine where already in the top 20.
DISCUSSION
Interestingly, when confined to only 20 medications, all of our choices included medications whose uses include control of symptoms. None of our top 20 medications were strictly preventive agents like statins and bisphosphonates. Because we were limited to 20 medications, a number of the included agents are not considered as first line agents for specific medical conditions however they often had more than a single use and were available in a number of useful dosage forms. In fact, many of our specific within class choices were dictated by the availability of multiple routes of delivery options. While we did not explicitly use cost as a determinant, we believe most medications within a class are therapeutically interchangeable. Therefore, if this exercise was done in a cost-conscious society, we would also use the within class lowest cost product to determine our agent of choice. Finally, it is worth mentioning like Osler’s quote, we often chose medications (ex. ASA, dexamethasone, lorazepam) that have a multitude of uses for a wide range of conditions.
Overall, with the exception of quetiapine, losartan, PEG and diphenhydramine the drugs in our list are also included in the WHO Model List of Essential Medicines – 300 or so drugs “that satisfy the health care needs of the majority of the population; they should therefore be available at all times in adequate amounts and in appropriate dosage forms, at a price the community can afford.”
From an infectious disease perspective, we realise we are lacking coverage against organisms such as P. aeruginosa and hospital acquired methicillin resistant staph but it was felt these organisms were not a frequent enough cause of infections to merit inclusion in the top 20 list. However, if any of us were infectious disease specialists as opposed to generalists, all 20 drugs would likely have been anti-microbial agents.
What is somewhat sobering is the fact the top 20 medications we would want in 2022 were released onto the market on average in the mid 1950’s. Only 5 of the twenty medications were marketed after 1970, and the 1960s was the decade with the largest number (7/20) of medications. It shows while new medications come on the market all the time, very few are both unique and impactful.
We realise our top 20 list carries with it many subjective decisions. We encourage readers of this paper to suggest other potential medications to include, BUT if you decide to add another medication you must also decide which one to remove. What is most important in this exercise is not whether there are “right or wrong” answers. The importance is in critically evaluating drug therapy in light of overall value to our patients and to the diseases and conditions from which they unfortunately suffer.
James McCormack, Pharm D
In conjuction with
Tina Korowynk, MD
Mike Kolber, MD
Mark McConnell, MD
G. Michael Allan, MD
James McCormack is a Professor with the Faculty of Pharmaceutical Sciences at the University of British Columbia in Vancouver. He recently published a lay-public book called The Nutrition Proposition in which he reviews the best available evidence and puts that evidence into context so people can make evidence-based decisions around what they eat.
Disagree on PEG - very often pushed by pediatricians starting from 6 months on when the drug itself is not allowed for anyone under 17 and for longer than a week. Yet here they are pushing it. Not even getting into the aggressive behaviors caused by PEG. No thanks for me or my family.
Fascinating list. I learned a great deal, thank you.