Transcript

MIC CAVAZZINI: Welcome to Pomegranate Rewind where I rifle through the dusty crate of previous podcasts and pull out some timeless classics. I’m Mic Cavazzini, and today’s episode was first produced in September 2017. I won’t adding much in the way of updates except to give Professor Ric Day a warm farewell as he retires after sixty years of service at St Vincent’s Hospital, Sydney. Through his academic appoint as Professor of Clinical Pharmacology at the University of New South Wales he has published several hundred peer-reviewed papers that have been cited more than forty thousand times. Oh, and he was feted as a Member of the Order of Australia for his internationally regarded work on quality use of medicines. The other voice you’ll hear is Professor Sarah Hilmer, also a prolific and impactful scholar in her role leading the Ageing and Pharmacology research group at the Kolling Institute. Now begins the podcast.

RICHARD DAY: We say that adverse drug events, we can avoid about half. It’s a big ticket item that causes a lot of morbidity. It costs a lot of money so we really should be doing better.  

MIC CAVAZZINI: Adverse drug events cause about 5 per cent of admissions to hospital, although some studies put the number much higher. That makes at least half a million patients in Australia and 55,000 in New Zealand every year. Rheumatologist and clinical pharmacologist Richard Day of St Vincent’s Hospital, Sydney.

RICHARD DAY: Yeah, it’s a big number and you’re not safe when you get into hospital. So the rate sort of adverse drug reaction there again, 5 to 15 percent. But about a fifth of those are drug interactions.   

MIC CAVAZZINI: Drug interactions become more likely in ageing patients who develop multiple conditions. Polypharmacy is the term given when five or more medications have been prescribed to a single patient, and it’s far from uncommon.

SARAH HILMER: Yeah, look in Australia, polypharmacy is the norm for older people. Amongst older Australians in the community aged 75 years or older, about half are exposed to five or more medicines and about a fifth are exposed to 10 or more, which we call hyper-polypharmacy.  

MIC CAVAZZINI: That’s geriatrician Sarah Hilmer of the Royal North Shore Hospital, and Professor at the University of Sydney.  

SARAH HILMER: For older people in hospital, the numbers are much higher. Nearly all older people in hospital are on five or more medicines. And five or more is the number of drugs that’s been associated with adverse outcomes in older people. It seems like once you get to your fifth drug, that’s when you’re more likely to fall over, that’s when you’re more likely to get confused, become frail, die.

And that may be partly because of the reason why you’re prescribed those five drugs, it may also just be a red flag and the more drugs you’re on, the more likely you are to be on a high-risk drug. But it seems in controlled studies, where you’ve controlled as well as you can in observational data, that it’s more than that. So older people are more susceptible to adverse drug reactions and drug interactions—firstly because they take more drugs, secondly because they don’t clear them as well in the first place and thirdly because their end organs and their overall body is less resilient.

MIC CAVAZZINI: This is modern day Mic now, just to flag that the 2017 podcast was inspired by Professor Day’s review in the Internal Medicine Journal  titled “Life-threatening drug interactions: what the physician needs to know”. And another by Professor Hilmer in Medicine today with the title “Polypharmacy in older people: when should you deprescribe?” We began our interview speculating on why there’s been a three to fivefold rise in rates of polypharmacy over the last few decades.

RICHARD DAY: Yeah, I think it’s really interesting because for each of the particular conditions, as the evidence accrues about what should be done, it often includes more than one drug. So, for example, if you have a heart attack there are three or four drugs that are part of the guidelines that one should prescribe. Now, if you happen to have diabetes, well there’s a few more. And if you happen to have hypotension a few more again, and it’s unusual for any guideline to present what to do when you’ve got comorbidities.  

MIC CAVAZZINI: And Sarah, how does the prescribing cascade fit into this?

SARAH HILMER: Yeah, the prescribing cascade is something that we always look for in geriatric medicine when we’re doing a medication review. The prescribing cascade is when you use one medication to treat the side effects of another without realising that you’re doing it. And a good example of that is if someone is taking thiazide then they’re more likely to be taking a gout medicine. If someone is taking metoclopramide, they’d more than likely be taking an anti-Parkinson’s medicine. If someone takes a cholinesterase inhibitor, then they’re more likely to be prescribed an anticholinergic for urinary incontinence.  

MIC CAVAZZINI: And can you paint a bit of a picture of the number of different prescribers that a geriatric patient might have?

SARAH HILMER: Depends very much on the patient, but if you think that on average, every decade over 60 you pick up one chronic disease, then a person in their 80’s usually has about three and for each of those they will have a sub-specialist, they will also have their GP they may well have another GP who covers when their GP is not there. And they’ll have the geriatrician who may or may not see them in the community. And so it’s really quite the norm to have four or five different prescribers and it’s quite hard to stay on top of knowing what those four or five different prescribers are doing.  

MIC CAVAZZINI: Now there are different types of drug-drug interactions. Can you start by describing the difference between pharmacodynamic from pharmacokinetic interactions?

RICHARD DAY: Probably the commonest are what we call pharmacodynamic interactions. So that’s when two drugs do roughly the same thing. So, for example they both cause CNS depression.

More challenging in a lot of ways is the pharmacokinetic interactions. So that’s removal of the drug from the body and its absorption. Other drugs might slow down the process of elimination and the reverse can happen, where you can actually speed up the process of elimination. So if you increase the removal, that means loss of effect and if you block the removal that means too much effect. Now, if it’s a critical drug, the difference between too much and too little isn’t much. Probably the biggest area is in metabolism in the liver.

MIC CAVAZZINI: Yeah, we’ll get into the complex metabolic pathways later, but there’s also some very simple interactions in the gut, aren’t there? The example of iron supplements and absorption.

RICHARD DAY: Yeah, it’s very important that people understand that the absorption can be affected by binding agents like cholestyramine or iron preparations, and a good general rule is to separate critical drugs from those type of agents. One of the important things is that people that take the iron preparations you mightn’t know about it, because the patient’s actually purchase them themselves or some other practitioners prescribe them. So that’s one thing.

The other thing you can do is you can change the pH with some very effective agents like proton pump inhibitors. Now that can affect the absorption of a number of drugs, for example, some of the new kinase inhibitors used to treat cancer won’t be absorbed if the pH is headed towards neutral. You can also effect through the same mechanism, using antacids. 

MIC CAVAZZINI: Sarah, can you explain the physiological changes older patients that put them at greater risk of adverse drug reactions? 

SARAH HILMER: Older people are at highest risk. Firstly as we get older, we get changes in our body composition, what we call sarcopenic obesity. So we lose muscle, that’s sarcopenia and we get relatively more fat. And that has implications particularly for the volume of distribution of drugs, so a water soluble drug like gentamicin or a low molecular weight heparin would actually have a smaller distribution for the patient’s overall weight, so you’d need a smaller loading dose. And it’s partly also because you get age-related decline in your renal function, so your kidneys aren’t working as well in the first place.

The good thing about renal clearance is you can measure it, so you can actually adjust your maintenance dose of renally-cleared drugs using creatinine clearance. You do have to always be careful not to just use the serum creatinine as an estimate of a person’s renal clearance because obviously as you have less muscle, so you have less creatinine to clear. And therefore a person can have a relatively normal serum creatinine, but a very impaired creatinine clearance when they’re old.  And when it comes to hepatic clearance, There’s about a 50per cent reduction in hepatic blood flow as you get older and there’s a reduction in enzyme clearance. Your liver does not work as well as you get old.

MIC CAVAZZINI: Most of the clearance of medications in the liver and gut occurs by oxidative metabolism. The cytochrome P450 enzymes convert drugs into products that can be eliminated more easily. Unfortunately, many drugs also disturb the function of these enzymes. One example is clarithromyocin, a commonly used macrolide antibiotic. Its inhibitory effect on cytochromes can result in dangerously high concentrations some potent subject drugs. Richard Day explains.  

RICHARD DAY: So that’s a very good example because it does also inhibit cytochrome P450s, 3A4 and 3A5. Now, they metabolise some very important drugs—drugs used in transplantation and also cancer drugs and also some of the newer anticoagulants. So you can see the situation where someone’s maybe had a transplant, in the community, he goes to the GP, he has an infection, he’s started on antibiotic—and the drug that’s actually suppressing the rejection, its removal is inhibited. Now that can be very serious because people are over immune-suppressed with a risk of infections and so on.

MIC CAVAZZINI: In the paper you write that macrolides also inhibit warfarin metabolism by the cytochromeP450 pathway. So that would elevate concentrations, and increase the risk of bleeding?

 RICHARD DAY: Yeah, warfarin would be a good example. It’s really—the difference between too much and too little isn’t much, so if you increase the removal that means loss of effect, and if you block the removal, that means too much effect, so either bleeding or clotting, which is catastrophic.

Amiodarone, which is a widely used antiarrhythmic drug, again, a very important interaction where you can increase the effect of warfarin significantly. Now, the opposite can happen with other antibiotics that actually induce the metabolism, so rifampicin you might be using for staphylococcal infection will induce metabolism. And any of the drugs used to treat seizures, at least the old ones phenytoin and carbamazepine.

I think the thing with warfarin is it’s one of the biggest drugs for serious adverse drug reactions. If you’re starting and stopping anything, just expect an interaction, number one. Number two, you need to be monitoring more often, otherwise it’s too risky.  

MIC CAVAZZINI: And amiodarone itself has a narrow therapeutic window and is also sensitive to changes in P450 function. What are the consequences there when you’re trying to control arrhythmia?

RICHARD DAY: Amiodarone is one type 3 anti-arrhythmic, metabolised by this cytochrome P450 system. And either concentrations that are too low or too high are bad—the arrhythmia gets out of control or, in fact, you know, there’s very significant toxicity with amiodarone, the lungs, the skin and the heart. So inducers and inhibitors of cytochrome P450 are big issues.

I think there’s also in the HIV group of drugs interacting with amiodarone and you’ve got to just look that up, I’ll have to look it up because I’ve forgotten the one which does the inhibiting. But my HIV therapeutic colleagues shake in their boots when amiodarone is mentioned.

And HIV I think is a great example because now we’ve got big cohorts of people who are maintained very well on combination therapy who are now developing coronary artery disease, diabetes, depression, intermittent infections, being treated by lots more people, different doctors, polypharmacy, drugs that are through the cytochrome P450 system—so potential for mayhem unless you’re very careful.

SARAH HILMER: Yeah, certainly in geriatrics the kinase inhibitors are proving to be quite a challenge. Things like ibrutinib which people take for chronic lymphocytic leukaemia, and can take for quite a long time, have a lot of drug interactions and then they have side effects like an increased risk of atrial fibrillation, and then the drugs that you want to prescribe for the atrial fibrillation will interact with the ibrutinib. And it’s a really complex cycle and I think as a geriatrician or general physician or a GP who’s looking at the whole picture, it’s really hard to stay on top of these new drugs. But we have to remember that just because it’s a sub-specialist drug, it doesn’t mean that it doesn’t interact with all the other drugs that we’re using and someone’s got to look at all of them and think about it.

RICHARD DAY: Exactly, Exactly.

MIC CAVAZZINI: And, of course, always in the press are statins. I think it might have been in your paper, Richard. You say that around 10 per cent of older patients prescribed statins and admitted to a teaching hospital, were discovered to have potentially clinically significant drug interactions with these ubiquitous medicines.

SARAH HILMER: That was my paper. You might have cited it.

RICHARD DAY: I hope so!

MIC CAVAZZINI: Maybe I’ll let you answer that one.

SARAH HILMER: No, Rick can talk about it.

RICHARD DAY: Yeah so, look, firstly adverse reactions with statins are common, I suppose the biggest one is muscle pain, also liver function test abnormalities. But there also is metabolism by this cytochrome P450 system. So, they’re subject to inhibition and higher concentrations, so greater risk of these adverse effects. Now, in the group of statins, some are more prone to these interactions than others, so simvastatin for example is prone, whereas rosuvastatin would be much less likely to be associated with the problem.

The other thing is, there’s things called transporters that are linked very closely to the cytochrome P450 system and p-glycoprotein is the transporter which transports across membranes but particularly from the circulation into the gut. So blocking that can lead to very high concentrations of drug and it’s very much like increasing the dose, as if you double the dose or triple the dosing.

SARAH HILMER: Yeah, I’ve seen some terrible cases of digoxin toxicity when people have started on a macrolide, a chest infection in winter and suddenly they block their p-glycoprotein and wound up with digoxin toxicity and also dabigatran and bleeding.

RICHARD DAY:   Just today before coming via my registrar. An elderly patient, think, who was 68, I think. Middle aged patient. Atrial fibrillation and had a carcinoma of the lung which was removed.

SARAH HILMER:                Not elderly!

RICHARD DAY:   A young patients. A middle-aged patient. Risk of thrombosis on two fronts. So, the drug selectors rivaroxaban, one of the new oral anticoagulant- inhibiting factor 10, which we don't monitor, we just give the drug to. Now the patient’s GFR was said to be about 60, that's the kidney function, which is important, so that's reasonable. The patient's quite light, you know, 50 kilos.

So, the starting dose was 15, and the question was, well, our target's 20. But we're also starting the person on, or is on, itraconazole because they've got aspergillosis in the lung. What do you think the dose is? So in our discussion, my sense is we're inhibiting the metabolism of rivaroxaban. We've got an older person who's not very heavy, who has atrial fibrillation, who we don't want to have a stroke. I mean, I was feeling nervous, I said, “Let's stick with 15 and then try to figure out whether or not that's anticoagulation or not”. But this is a good example in this drug which has got a therapeutic index which is pretty narrow. Too little- stroke, too much- bleed to death.

SARAH HILMER:                Were you tempted to say just use Warfarin we can monitor it?

RICHARD DAY:   I hadn't thought of that but it's a very good suggestion. So that's a good one, will take that back.

MIC CAVAZZINI: A slightly updated chapter heading in 2026, though not much has changes in 9 years. Of the most prescribed drugs in Australia, statins take the first two places. On the doses per 1000 people chart, the ACE inhibitor perindopril comes next, and another three drugs targeting the angiotensin system make it to the top ten, being candesartan, telmisartan and ramipril. When prescribing these it’s important to consider pharmacodynamic effects as well as homeostasis of salts in the blood serum. Also among the top ten most prescribed drugs are a couple of antidepressants, but as we’ll hear, they’re not the only drugs which block serotonin reuptake.

SARAH HILMER: So, look, ACE inhibitors are commonly used with other drugs that also lower blood pressure which would then predispose the person to hypotension or postural hypotension and falls. If you think about a person who’s got heart failure or is just after having a myocardial infarction, they’ll most likely be prescribed an ACE inhibitor as well as a beta-blocker as part of their guideline-recommended therapy, and there you’ve got two drugs that are going to block your blood pressure.

The other interaction that we worry about with ACE inhibitors is interactions with other drugs that can impair your renal function. And we talk about the triple whammy which is when you’re on an ACE inhibitor and a diuretic and a non-steroidal anti-inflammatory drug. And the combination of those three has a very good chance of causing renal impairment and those three are very commonly used together. ACE inhibitors are commonly used for heart failure, so are diuretics and non-steroidal anti-inflammatory drugs are readily available over the counter and people will take them not necessarily discussing it with their doctor, and they’ll go from having two drugs that put their renal function on the edge, to having the triple whammy.

Another one that I often see with ACE inhibitors is spironolactone, again as part of the heart failure guidelines. And you’ve got to be very careful to monitor potassium in that situation because both drugs will increase potassium and again, in older people with borderline renal function in the first place, you can wind up with dangerous hypokalaemia very easily. And the other drug that’s commonly prescribed with ACE inhibitors is trimethoprim. Trimethoprim is an antibiotic commonly used for urinary tract infection and it can cause a type 4 renal tubular acidosis and high potassium.

MIC CAVAZZINI: And Rick, you also mention in the paper that ACE inhibitors can lead to lithium toxicity in patients being treated for bipolar disorder?

RICHARD DAY: Yeah, that’s very common. It’s very dramatic, people turn up confused, they’re twitching, people are red in the face, their blood pressure might be labile, tachycardic. It looks in a lot of ways perhaps like an infection or some sort of systemic illness, very unpleasant. And one of the common interactions is with a diuretic that started with a thiazide. Anything that interrupts the sodium economy in the body because sodium and lithium exchange, so that changes the potential difference across nerve and muscle membranes, making them more excitable. And the remedy is to get rid of the lithium by flushing it out with sodium essentially.  

MIC CAVAZZINI: And another condition that’s often comorbid with aging and heart disease is depression and SSRIs are, of course, a useful drug in its treatment, but there are many other drugs that non-specifically block serotonin reuptake. What are some of these drugs?

RICHARD DAY: Yeah, so this is a big issue, so the serotonin syndrome from minor manifestations to really very dramatic and troublesome ones. So, the serotonin reuptake inhibitors like sertraline or citalopram or escitalopram can cause this in their own right. But if you actually combine drugs that have that same effect—as examples, there’s a very commonly used opioid analgesic called tramadol which has a component that inhibits reuptake of serotonin.

There’s also, of course, St John’s wort. We haven’t talked about complimentary medicines much, but it’s a serotonin reuptake inhibitor and it’s commonly taken. So, it’s not unusual to find out later when someone turns up twitching in the ED department with their blood pressure going up and down or tachycardic with clonus, that in fact there’s been more than one drug. And it’s important to be aware of what the serotonin syndrome is actually like. I’ve had people who’ve had clonus for weeks after taking, you know, one or two of these things, occasionally with drugs of abuse like ecstasy on top of an antidepressant like sertraline.

MIC CAVAZZINI: Well, drugs of abuse or recreational drugs or even complimentary medicines are not something that patients will often talk about with the prescriber. Have you got an approach for bringing that up, Sarah?

SARAH HILMER: I always explicitly ask. I try to ask in a non-judgmental way, I explain why it’s important that I understand what they’re taking and that it might interact with their other drugs, or partly explain how their health is at the moment. And the gold standard really is to get a pharmacist to go out to their house and do a home medicines review and actually have a look around and get them to show you exactly what they’re taking. You know at clinics we say, “Come in and bring all your medicines in a brown bag, anything you take,” and we usually get a range of complimentary medicines.

Older people, on average, 50 percent of them take at least one complementary medicine. Of those people who take a complimentary medicine, they take an average of three and they’re usually the things that are related to aging which modern medicine hasn’t really made big gains in yet. Things like ginkgo biloba in the hope that it might help your memory or various antioxidant vitamins in the hope that it might help your aging. And so I always ask and I explain why I need to know and generally I find out, I think.  

MIC CAVAZZINI:                So, some immunosuppressants need to be used very carefully. For example, the enzyme that detoxifies azathioprine is inhibited by allopurinol, used to treat gout. So, what are the outcomes of this interaction?

RICHARD DAY:   Yeah, this is really considered to be one of the most important drug interactions not to miss because if it is missed, the azathioprine being an immunosuppressant and cytotoxic drug can have catastrophic effects by knocking out the bone marrow. Allopurinol blocks one of the elimination pathways for azathioprine so the concentrations can become excessive. So, the advice is they're contraindicated. But the other advice is well, you can use them together if you reduce the dose of allopurinol very substantially. So in fact, gastroenterologists have found that actually using the combination is helpful for inflammatory bowel disease. We are able now to measure the concentrations of this metabolite that's active and potentially toxic, the thioguanines, so that you can actually get the doses right. But this is not for general use really.

MIC CAVAZZINI: The last important interaction that should be mentioned. The non-steroidal immunosuppressant methotrexate. What are the concerns you have about prescribing this?

RICHARD DAY:   So, it's really heavily used and look, it's extremely safe if the dose is right and it's monitored properly. However, I guess I've seen a couple that have been catastrophic where people have been on methotrexate legitimately for something like rheumatoid arthritis, not in the excessive dose—renally impaired or on the edge of renal impairment—all going well, trimethoprim-sulfamethoxazole is started. Now, each of those drugs actually interfere with folate metabolism. The elements of the bone marrow are suppressed. So, there have been cases, and they've ended up in courts, where people have been sued for negligence because of using the combination. But it's usually not just the antibiotic, it's usually renal impairment.

Now it's also difficult because on interaction alerting systems that will often pop up that is a cause of a drug interaction and they're not unless you're giving huge doses of methotrexate. And this illustrates that flagging a potential drug interaction seems like a very obvious thing to do. But what we know is that doctors quickly turn off reacting because they see so many of them. And so how do you select the ones to put into an alerting system so it will make sense to an individual doctor?

MIC CAVAZZINI: You can keep informed about dangerous drug interactions through numerous online resources, which we've linked to on our website. A few example are the Australian Medicines Handbook, MIMS Online and the Drug Burden Index. But it’s good practice in patients already taking more than five drugs to review them carefully. Sarah Hilmer outlines a strategy for deprescribing, and consideration of non-pharmacological solutions.

SARAH HILMER: Yeah, look, we try very hard to get patients down to their top four or five and the first step is to look at what their problems are and what their goals are and to see how you can align those. And then—people do this in different ways, some people imagine that they’re starting from scratch and say “OK, well if this patient has these five comorbidities and his main goal is to maintain their physical and cognitive function, which five drugs could I use to achieve that?” The other strategy is to start with what they’re on and look at each drug and say, “Is there still a valid indication for this, is there evidence of toxicity?” Weigh up the risk benefit for each drug and then work out whether those drugs are going to be in the top five or not.

The important thing is to make sure that you explain to the patient that you are doing this to help them achieve their goals, you’re not doing this because you’ve given up on them, because they’re not worth treating, because you don’t want to spend money on their drugs, you are doing this to help them achieve whatever their goal may be.

Sometimes that is about optimizing the treatment of one particular disease that is actually the thing that is their rate-limiting step for example, if you had stage four heart failure and bad arthritis and you also had macular degeneration, it may well be that you’ve got everything in place to manage getting around with your arthritis, and your house is set up for your impaired vision, but the thing that is actually stopping you from getting out of bed and going to sit out in the sun is your shortness of breath. And in that case, there would be good justification for doing what you could to treat the heart failure. However, if the treatment of the heart failure then made you incontinent or made you so dizzy that you feel over and you couldn’t get outside anyway, then there would really be no point.

MIC CAVAZZINI: The point about functional outcomes I could imagine in the case of incontinence for example, that you might be able to avoid pharmacological treatment using pads or something else?

SARAH HILMER: Continence is a great example. The drugs we use for incontinence are generally quite toxic to older people and are not terribly effective. For urge incontinence, which is the incontinence you get when you can’t get to the toilet in time, the main drug class that we use are the anticholinergics. And on average they reduce your episodes of going to the toilet by one or two per day, that brings down from an average of 12 to 10, it doesn’t make a huge difference to your lifestyle.

But the drugs themselves, being anticholinergic, give you dry mouth, they give you constipation, they make you confused, they give you blurred vision, so yeah, we often suggest things like timed toileting. Particularly if someone is living in residential aged care, it’s been well established that if you just take someone to the toilet every couple of hours you can keep them dry. Or continence pads and continence aids. They’re also—for the more highly functioning people with good cognition, there are various bladder training exercises and things that you can do which, again, are about as effective as the drugs and not nearly as toxic.

MIC CAVAZZINI: You say in your review that treating or preventing a specific disease in a very elderly person may alter the cause of death without substantially impacting on life expectancy. Can you elaborate on this?

SARAH HILMER: Well, once you get to a certain age and you’ve got a lot of different chronic diseases going on at the same time and frailty, we could talk about the concept of competing morbidity. And the idea is that even if you really optimize the treatment of one of the diseases, then one of the other ones will probably catch up with you. And a number of preventative medicines will take at least two years to have any difference. And if a person’s life expectancy is less than two years, then clearly there’s not a lot of point in taking the medicine, you’re getting the side effect and there’s no evidence at all that you would ever get any benefit.

And the other really important thing is to monitor closely after you start deprescribing. You also have to remember that when you withdraw a drug, you can basically reverse an existing drug interaction and Ric touched on that a bit with the warfarin. If you stop something that was inhibiting cytochrome P450, then you’ve got to bear in mind that now cytochrome P450 will be working better and it might actually affect the clearance of some of your other remaining drugs and so that’s one of the things that you have to monitor for.

MIC CAVAZZINI: One of my editorial committee talked about this culture where clinicians are so caught up in the importance of their area of specialty and the disease that they’re treating can act quite single-mindedly to cure that particular problem. Is this something you see, Ric?

RICHARD DAY: Yes, absolutely. We we’re taught to be physicians first and -ologists second, but it’s the hardest lesson to get across. All of us have an obligation as physicians to do a simple check on what people are taking. It does require time and some expertise , and people haven’t got time and maybe not the expertise. Well, I think that’s a big area that physicians can lead on.  We’re the main prescribers or main influencers of prescribers. Because general practice of course is writing most of the prescriptions, but a lot of them start by the sub-specialists for these particular patients and people, once started—look, stopping your own prescriptions are hard enough, but someone else’s is even a tougher one.  

SARAH HILMER: As one of my really very respected colleagues put it, we don’t like to mess with other people’s drugs. We see ourselves as specialists in one area and other people are specialists in other areas and we don’t really like to mess with what they’re doing. But if we do recognize a potential drug-drug interaction, then, yeah we shouldn’t ignore it. We need to talk to the other sub-specialist about it. Talk to the GP about it, make sure the patient knows. And I think as specialists, one of the things we need to bear in mind is that when we give our advice back to GPs, we give advice for the condition that we’re treating, but give advice for them to change in view of the patient’s overall condition, particularly as that changes over time. Because a lot of the time when patients get very old and very frail, they don’t get out to see their sub-specialist anymore and the advice that was given when they were in great shape when they were 75, may no longer apply at 85 when they’ve picked up a couple of more comorbidities with drug-drug and drug-disease interactions. And we need to almost give permission to adjust the treatment as appropriate as the patient’s condition changes.  

MIC CAVAZZINI:                And clinicians are sort of obviously follow the guidelines as best they can. Sarah, in your review in Medicine Today, you say that clinicians are often more knowledgeable about the effectiveness of medications and how to start prescribing them than about how to identify the harms and when to stop them. Why do you think that is?

SARAH HILMER:                Look, I think it reflects partly our training. Partly the way that our postgraduate training is done and the way that drugs are promoted, partly the way drugs are regulated, whereby there's nothing specifically in the product information on when to stop a drug or how to stop a drug. And I think it's also partly because we're physicians and we like to help and we like to think we can do something with a drug that can help. That's the main intervention we have as a physician.

MIC CAVAZZINI:                Is there sometimes resistance from patients themselves?

SARAH HILMER: That’s where I was getting. Yeah, look the other factor really is the patient side of the equation. And whenever we ask patients, “Would you be happy to stop a drug if your doctor said that you could”, over 90% of patients in any setting will say yes, “I would be happy if my doctor said I could”. A lot of patients don't like the idea of sort of I rattle when I walk and would like to be on less medicines, but they trust us as their doctors and they presume that they're on all their medicines for a good reason and they need to be on them. S,o I think it's really up to us to talk to our patients about it. I've been involved in a few studies. And it looks like depending on your setting and your intervention, you can usually stop an average of about two drugs per patient. And survival is the same, so you don't have to worry you're gonna kill people.

MIC CAVAZZINI:                Is that a real fear that clinicians have?

SARAH HILMER:                Absolutely, well we're prescribing these drugs because we want to help. And particularly the preventative drugs. We are concerned that if we stop them, that something bad might happen. And a real barrier is the medico-legal concern that you stop, say, a cardiovascular preventative drug, and then someone has a stroke or a heart attack, and you're in court trying to explain why you stopped it.

There is growing data that you can use to support your actions. The first being a lack of efficacy of a number of these drugs in the very old. The second piece of evidence you can use is the fact that if a person's having side effects, those side effects may well be worse for their overall function and their quality of life than any potential benefit of preventing an event. And, the third and growing body of evidence is the evidence around the safety of deprescribing. And when people have done this in a randomized control trial way, survival is no worse. Vascular events are common in older people. But they will happen whether you're on the drugs or not.

There is increasing evidence about deprescribing, which can help practitioners feel comfortable about what to do and how to do it. There's increasing evidence about deprescribing particular drug classes and that it's relatively safe. It doesn't increase mortality and how to monitor it. And generally the patient-level feedback is very good. So, I think the evidence is probably best around stopping psychotropics in people who have behavioural and psychological symptoms of dementia. And there is very good evidence that you can stop these. And in even a placebo-controlled randomized trial, rigorous trials, people actually wind up doing better in terms of survival and also doing slightly better in terms of their neuropsychological symptoms. You don't have to worry that when you stop them people are suddenly going to have terrible behaviours. Behavioural symptoms of dementia fluctuate over time.

MIC CAVAZZINI:                Clinical trials obviously exclude older people precisely because of the many confounding factors. Is there any way to feasibly consider all those comorbidities and drug interactions during clinical testing?

RICHARD DAY:   Well, I think that there's all sorts of developments about how we gather data, which might be described as more real-world data and make sense of it, and even translate that into evidence and guidelines and even regulatory interventions. We've got a long way to go, people are wary about trials that are not pristine. So, you need to deal with that and it's challenging. Now increasingly we think we can. Big data, we're talking about it a lot.  But a lot of the data that we use to feed the big data is administrative and it's not collected in a way that we would in trials. These are things to deal with. However, there are interactions and being observed using these systems. An even spontaneous reports. So, for example, there are some huge databases at FDA, also at WHO. And you can use statistical techniques to identify outliers. That is another way to deal with this conundrum of the real world.

SARAH HILMER:                One of the challenges with some of the adverse effects we see in older people is they're very hard to recognise specifically as adverse effects because as we get older people present with non-specific presentations that could be due to a number of things. So, if someone presents with a fall or with confusion, it could be partly an adverse drug reaction, it could just be that they're getting older, it could be an environmental factor, it could be a disease. Chances are it's a combination of all of those things. And so, recognizing and reporting those as adverse drug reactions is quite difficult.

MIC CAVAZZINI: Many thanks to Professor Sarah Hilmer and Professor Ric Day for their contribution to Pomegranate Health, even if it was nine years ago now. There’s much more recently recorded lecture from Professor Hilmer on Practical Prescribing in the College Learning Series at elearning.racp.edu.au. It’s just one of several lectures on Pharmacology and Toxicology that have been developed around the Basic Training Curriculum.

At the podcast website, there’s a transcript of this interview and links to the reviews about life-threatening drug interactions and deprescribing in the elderly which we discussed. Just click on the shortcut in the show notes or go to racp.edu.au/podcast.

At the landing page you can also sign up to our email alerts so you don’t miss a new episode, or the best way to subscribe is through some pod aggregator like Apple podcasts, Spotify, Castbox, Overcast and others. And if you click on Episode Browser you’ll find a neat search tool that allows you to filter episodes by specialty or by domain of professional practice. And remember, you can record your time spent listening as a Category 1 CPD activity. Just find the MyCPD link in the shownotes and most of the details will be prefilled for you.

This podcast was recorded on the lands of the Gadigal clan of the Yura nation. The RACP is committed to equitable healthcare for all Aboriginal, Torres Strait Islander and Māori people. I’m Mic Cavazzini, thanks for listening.