Safety and quality in primary care: whither goest?

“The fundamental problem … in the overall approach to primary care interventions is that they have focused on individual entities – doctors and their prescribing, pharmacists and their dispensing, private hospitals and their medication management. However, nobody has yet bitten off the enormous challenge of the highest risk part of all of it: the interfaces between all of those elements, and the things that can go wrong when patients move between independent providers who are not working from common datasets.”

A FEW weeks ago, I received a discharge notification letter and medication list for a patient discharged from a private hospital. The medication list was issued by the hospital pharmacy and indicated two changes to medications that had occurred during her admission. The discharge notification letter was prepared by the nursing staff of the hospital.

The pharmacy documented that the patient’s telmisartan 40 mg daily had been ceased, and they should take candesartan 8 mg daily. I pulled up their medical record and made the appropriate changes, thinking little more of it, noting that they had an appointment a couple of weeks out.

About two weeks later, a letter from the specialist arrived written after a post-discharge review. It did not specifically document changes to medication, so I did not undertake a medication reconciliation. I failed to remember that changes had been advised by the pharmacy.

Fast forward another week and I received a fax from the pharmacy requesting a renewed prescription for the ceased telmisartan. I opened the file, found the amended medication list, and telephoned them and advised that it had been ceased in hospital. They confirmed that they had also received the pharmacy discharge summary, but the patient had presented requesting telmisartan, and it had been dispensed.

By this point, I was utterly confused. And worried.

I reverted to the specialist letter, which stated that the patient was stable on 80 mg telmisartan, and had improved markedly, so should continue with this dose. I telephoned the specialist to try and work out what was supposed to be happening, explaining that the hospital pharmacy had advised that telmisartan had been ceased and candesartan prescribed by him. He in turn telephoned the patient, who confirmed that she was indeed taking telmisartan, and the specialist telephoned me back, confirming telmisartan 80 mg and no candesartan.

I then telephoned the community pharmacy, who were absolutely sure that the patient could not be taking telmisartan 80 mg as they had never dispensed this, and the consumption rate of the 40 mg tabs indicated that the patient was taking 40 mg per day. They had not dispensed any candesartan.

With my brain quietly exploding, I telephoned the patient, and we walked through her medications. She was not taking any candesartan and she was taking telmisartan 40 mg – exactly the same medications she had been admitted to hospital on, with no changes.

Now, my own failings in this sorry saga are myriad, and I am not seeking to evade that. I can see several points where I had an opportunity to intercept this. My purpose in writing about it is that, as a long time manager of systems for safety, there was no system for me, as a GP, to turn to. There was no feasible way I could engage with my pharmacy and specialist medical colleagues to undertake an incident analysis, to learn and prevent this sort of thing from happening again. The best I could do was ask the specialist to communicate with the hospital pharmacy and ask the private hospital to undertake an analysis of how this had happened.

In the public hospital system, errors like this are reported in an incident management system. They are then categorised according to a severity assessment and managed with whatever level of gravity the severity assessment warrants.

There is simply no equivalent system available to us in the primary care non-system. We are very loosely connected private businesses with no common IT systems, no interoperability. Eventually, My Health Record may help with this gap, but its coverage is very patchy, with only 9.16% of specialists and 46.93% of private hospitals participating. This means that an absence of something in the record does not mean it has not happened. It is also vulnerable to simply replicating the errors that were in the summary I received. If this had been a public hospital patient, I could have logged into the excellent Health Provider Portal and obtained direct information about the admission. But it was a constellation of private hospital, private hospital pharmacy, community pharmacy, specialist physician and me.

Not one health professional in that constellation had an accurate picture of what was happening with this patient. Fortunately, she was well, and improved markedly, despite there being no actual change to her medications.

My own analysis of my telmisartan/candesartan saga is that there was a cascade of errors, fortunately, none of them resulting in patient harm. Here are some that I can identify on a cursory review. None of these suggestions is intended to point a finger – the entire system just didn’t work, and understanding what went wrong and why is fundamental to preventing it again.  Blaming individuals achieves nothing.

1. The hospital pharmacy provided inaccurate information in their discharge summary (one possible explanation for this is below at point number 8).
2. The hospital discharge notification did not provide medication information, rather just relied on the separate communication by the pharmacy.
3. The hospital appears not to have had a discharge discussion/briefing with the patient about her medications.
4. The specialist letter was prepared apparently without reference to the pharmacy summary (and vice versa, but the temporal lag between explains the pharmacy not seeing the specialist letter).
5. The community pharmacy received the hospital pharmacy summary but didn’t make any notes in the patient’s records. They therefore did not question the patient’s request for ongoing telmisartan dispensing – which by a stroke of luck turned out to be a good thing – it is how we discovered this situation.
6. I received the hospital pharmacy summary and noted it, but failed to recognise the difference between the specialist’s and the pharmacy’s medications.
7. I didn’t recall the patient for a consultation to discuss the changes, but rather accepted the pharmacy’s summary as truth.
8. The patient did not leave hospital with a discharge prescription of 80 mg telmisartan, so she kept taking her 40 mg dose.

This episode came on the back of several fax/email webster pack prescription renewal requests from pharmacies, where I was asked to provide prescriptions for medications that I had ceased, that I had advised the patient were to be ceased, and even for medications that I had specifically written to the pharmacy notifying that they were ceased.

I have now reached the point where I return a full medication summary with every webster-pack renewal, and ask the pharmacist to notify urgently if there is any variance between what they are dispensing and what I have in records.

Webster packs are great – they allow continuation of medication semi-independence for patients on complex and bewildering arrays of drugs. Generic medications are also great, they help keep the cost of Pharmaceutical Benefits Scheme medications down. The combination of webster packs and generic medications with current supply chain instability means that patients often actually have little idea what they are taking, unless they have the street-smarts and vigilance to read the list on the webster pack (not all webster packs have one) and match it up.

This week’s little pink tablet might be yellow next week, and this makes it so hard for patients to be equal partners in their medication management once it is taken out of the manufacturer’s packaging.

Medication errors are the biggest causes of iatrogenic morbidity in the world. In 2022, the theme of the World Health Organization International Patient safety Day was “Medication without Harm”. State-funded public health systems have enormous resources to invest in safety and quality, largely because of the political implications of high profile safety breaches, and despite the fact that more services to more people every day are delivered in general practice, there are no dedicated safety and quality resources that we can call on to help. There is not even anywhere I can report; hence my decision to publish this story, so that others can learn from it.

National mechanisms have been dominated by the acute care sector, in particular the public hospital acute care sector, with some emphasis on private hospitals and residential aged care, but very little focus on general practice.

It’s demoralising to read the National Strategy for Quality Use of Medicines and discover that there is no mention at all of general practice or primary care. The National Medicines Policy would be eligible to vote by now, if it were a person, having been written in 2000.

Yes, you read that correctly, the National Medicines Policy has not been refreshed in 22 years. The previous Health Minister forecast a review, but it is quite shameful that it languished for 19 years before a Health Minister turned their mind to this issue.

Medication safety is part of the overall patient safety environment. The Australian Commission on Safety and Quality in Healthcare (The Commission) has historically had an unfortunate bias in focus on the acute care sector. There is a Primary Care Committee, and the current workplan for the Commission actually acknowledges that there has not been sufficient focus on primary care but it remains to be seen if this will translate into real safety improvements.

There has been steady, but glacially slow progress with a small but ground breaking study in 2006, a literature review (June 2009), a discussion paper (2010), a consultation report (2011), an “evidence check” (2015), and a second consultation phase (2017 to 2018) around safety and quality in primary care, with one final output being the National Safety and Quality Primary and Community Care Healthcare Standards (2021). The literature reviews and consultation papers are excellent documents that capture the problems of the system eloquently. However, they don’t fix it. An incident-reporting system might happen one day, but its currently only at the “scoping” stage.

Another potential engine room for primary care medication safety could be the Primary Health Networks (PHNs). The PHN Program Performance and Quality Framework makes no mention of medication safety. A review of PHN websites nationally (here, here and here) shows that some PHNs have initiated medication safety programs, but the majority have no visible presence in this space. The problem with PHN activity is that it will likely leave out the non-primary care players in this space, namely hospitals, hospital pharmacies and private practice specialists.

Among much angst, the National Prescribing Service has recently been defunded, and the functions for quality use of medicines transferred to the Commission. Even though it is encouraging to see that they have already advertised for a senior position with extensive knowledge of general practice, the fundamental problem I see in the overall approach to primary care interventions is that they have focused on individual entities – doctors and their prescribing, pharmacists and their dispensing, private hospitals and their medication management. However, nobody has yet bitten off the enormous challenge of the highest risk part of all of it: the interfaces between all of those elements, and the things that can go wrong when patients move between independent providers who are not working from common datasets.

Safety and quality in the acute sector always focus on system issues first – how can we create safe systems. Safety and quality in the primary and community care system, likely due to the fragmentation, and in fact the absence of any “system”, focuses on individual practices and practitioners, with an over-reliance on individual vigilance, which is known to be a highly unreliable safety modality. The first step will be a national primary care clinical incident reporting system, which will allow us to quantify the problem and make the business case for investing in systems to improve patient safety. Several countries already have this, for example, the UK and Switzerland.

Until there is such an incident reporting system, the magnitude of the problem will remain invisible, and pressure to fix it won’t exist. Individual practitioner reports to the Australian Health Practitioner Regulation Agency (AHPRA) are not the answer – using a practitioner-focused regulator is no substitute for a systems-thinking approach.

In line with that systems thinking approach, I am prompted to wonder, what pressures are our pharmacy colleagues under, and how pressed are they with the increasing burden of documentation?  What do they need changed to be better able to support patient safety, and how can we advocate for it?  We need to be mindful that all parts of the system are under pressure, not just GPs, and avoid the temptation to blame, but rather advocate for all players in the system, in the interests of patient safety.

General practice can walk towards this problem, acknowledging our slice of the error pie, so that the system can learn and improve. We can demand that investment in this critical gap is addressed. Implementing it will be painful. If experience in the acute system is anything to go by, it will start out perceived as a “dob-in” system, where grievances against individuals are inappropriately logged. A well designed system will prevent this, and ensure that reports about what happened and not about who “stuffed up” generate true system reforms.

Until we start to tackle this, the system is doomed to continue to repeat errors, sometimes with devastating outcomes. Not all patients are as tolerant of dose variation as my very gracious patient at the centre of this story.

Dr Jillann Farmer is a Brisbane-based GP, and former Medical Director of the United Nations.

The statements or opinions expressed in this article reflect the views of the authors and do not necessarily represent the official policy of the AMA, the MJA or InSight+ unless so stated.

Subscribe to the free InSight+ weekly newsletter here. It is available to all readers, not just registered medical practitioners.

If you would like to submit an article for consideration, send a Word version to mjainsight-editor@ampco.com.au.