Author: Elie Matar
Reviewer: Katherine George
It’s your first day of internship and you step onto the ward. You’ve seen Grey’s Anatomy, House MD, Scrubs so you have some idea what you’re in for…or at least you thought you did. A few hours in it doesn’t take long before you realise that you are underprepared for one of the most important responsibilities of your job. No, I am not talking about emergency tracheostomies, end of life discussions, inotropes … I’m talking about discharge summaries.
As part of our local redesign group at our hospital, we were tasked with regularly surveying junior medical officers to identify key issues within the hospital system that can be improved. A recurrent issue raised year after year was a resounding dissatisfaction of interns with discharge summaries. It wasn’t just that they were cumbersome (though this was undoubtedly a contributor), but rather that there was a lack of formal guidelines or education about what was required for discharge summaries. This was made worse as they moved through different rotations under different consultants and registrars with different expectations. Inevitably, this resulted in significant variability in the quality and length of discharge summaries. I personally have seen some discharge letters ranging from a single sentence to dissertations of Iliadic proportions (I recall my own magnum opus was 10 pages long!). The former situation can indicate loss of vital information, whilst the latter likely implies time that could have been better spent.
In response to this, and backed by the executive of our hospital who saw this as a priority, we set up a subcommittee of over 12 interns, residents and registrars to address the issue of what makes a good discharge summary in the hope of trying to establish some formal guidelines. In this blog, I’ll share some of the key learning points from our efforts.
But firstly, it’s worth revisiting the purpose of discharge summaries. Traditionally, we think of discharge summaries as a letter to the general practitioner (GP) detailing the events of a patient’s admission to hospital. But discharge summaries also serve other functions and there are several components of the healthcare system that rely on them. Emergency physicians are increasingly relying on previous discharge letters as reliable documentation of a patient’s medical history where this is otherwise not available. Consultants and inpatient teams from various hospitals likewise use it as a record of investigations and treatment a patient may have received at their own or other institutions. Even hospital administrators rely on discharge summaries to keep track of hospital activity and codes to ensure appropriate allocation of funding.
With this in mind, we decided that we needed to approach all the above stakeholders for our own project. So together with an enthusiastic group of interns and residents, we surveyed general practitioners, medical and surgical consultants, coding officers and emergency physicians on what they thought were essential components of discharge letters, what’s not important and what can be improved. These are summarized below.
TMI (Too Much Information)
The overwhelming consensus amongst all the groups was that the more concise a discharge summary – the better. After all, most GPs working in a typically busy practice will not have time to rifle through all ten pages of your discharge document. Indeed, as most senior consultants would tell you, discharge summaries used to be a hand-written single-page form filled in just as the patient was about to leave. Overall, the feedback was that a discharge summary should be treated as a handover document – a synthesized narrative with only the relevant facts rather than several disparate details.
The Sine qua non (What not to miss)
When asked what were the most relevant aspects of a discharge summary that should be included there was some slight variation between the stakeholders, but invariably the most agreed upon was a focus on the discharge plan – in particular for relevant pathology, appointments and actions that GPs and other specialists are required to follow through. This is a handover summary distilled to its essence, so the take home message was that if you only had a few precious minutes to write a discharge letter – focus on the plan. But there were other essential items that recurred in the surveys that many thought were absolute requirements and they include:
- Changes in medications and rationale
- Main and secondary diagnoses and issues during the admission (being as specific and concise as possible, e.g. Right middle cerebral artery infarction due to AF vs just ‘CVA’)
- Results of relevant investigations (imaging, blood test, histopathology etc)
- Any post-operative or procedural complications
- Identification of any new allergies
Room for improvement (what’s lacking)
Amongst the responses were several other very useful suggestions on what they thought was lacking from many discharges and should be included to improve their quality. Clearly omission of any of the above ‘essential items’ was immediately considered as room for improvement, and this was sadly encountered quite frequently by our stakeholders. Here, I’ve included the general comments on what stakeholders and our resident team felt was often lacking and could be improved in discharge letters:
- A comprehensive but brief opening summary was thought to be essential to contextualize the rest of the admission. This should include the main diagnosis and a brief synopsis of the major complications. Think of it as the opening title for an essay – but keep it brief.
- Details of appointments should be included with details such as contact numbers and addresses
- A summary of conclusions from any consults obtained from other non-admitting teams, e.g. If the renal team was consulted on a cardiology patient. Make sure to include the name of the consultant as well for future reference.
- Allied health review, conclusions and interventions were felt to be lacking from discharge letters. It is crucial to remember that these are sometimes more important than the medical interventions. Examples include occupational therapy (e.g. driving restrictions), speech therapy (swallowing precautions), physiotherapy (mobility aids etc).
Bonus (Specialty) content
Every specialty has some disease-specific information that most consultants agree should be included in discharge letters not only because they were relevant to the current admission but also to serve as a reference for future management. I have termed these ‘bonus specialty headings’ and each intern and resident should be mindful of these as they start a new rotation. An effective way to find out what these are would be to ask previous residents who have rotated through the term, or ask registrars and consultants on that rotation. You may not think it – but they are just as invested on the quality of discharge letters as you are, and most good consultants would be more than happy to answer these questions (just don’t wait until the end of the term!). To give a flavour of this I have provided some examples but these are by no means exhaustive.
- Geriatrics – social history, allied health reviews, designated next of kin, advanced care directives
- Cardiology – troponins, ECG and echocardiography and angiography results, dry weight (heart failure), pacemaker checks, models of implanted devices
- Respiratory – spirometry (on admission and discharge), organisms identified, arterial gas results, oxygen requirements
- Renal dialysis – mode of dialysis, pre- and post-dialysis weight, electrolytes
- Surgical – use of DVT prophylaxis, peri-operative anticoagulation or antiplatelet use, histopathology, peri-procedural or anaesthetic complications, analgesia
- Oncology – side effects of chemotherapy, palliative care reviews, histopathology
- Obstetrics – gender of the baby, method of delivery, complications, newborn check
What to leave out
Having established that most stakeholders agree that too much information can make a discharge worse, we asked them which areas in discharge letters could be downsized. The verdict – the presentation. All agreed that the presentation on admission should be limited to one or two sentences. Only salient aspects of the presenting history should be included. Likewise, only relevant examination findings and investigations should make it into the discharge summary. Sure, a normal neurological and PR exam in a pneumonia patient may be a sign of a thorough ED resident, but if the patient did not develop a stroke or PR bleeding during the admission, this can be safely left out from the discharge summary.
Similarly, most stakeholders felt that investigation results (pathology, imaging etc) should only be included if they were significant and/or needed to be followed up.
Surgical consultants specifically felt that routine inclusion of social history is not necessary and should only be included if relevant (i.e. risk factors for illness or if there were barriers to discharge planning).
The importance of discharge summaries should not be understated. An admission to hospital can be life-changing and an understanding of a patient’s health can often be wholly reliant on the adequate documentation of this event. It is amazing to think that this information rests with that single lowly-regarded discharge letter and even more baffling that the great responsibility of creating this letter falls to the most junior member of the medical team. It may not be as thrilling as an emergency airway, but realising that many components of the health system from consultants to administrators are reliant on this crucial document may go some ways in validating all those hours doctors spend on discharge summaries in their formative years. Our small project also highlights a need for more formal education in medical school and internship on how discharge summaries should be composed.
I’d like to thank all the members of the Royal Prince Alfred Hospital discharge summary group who were part of this project. A. Bagust, A. Clarke, A. Snir, B. Chen, C. Yang, E. Black, G. Boots, J. Paterson, J. Soares, K. Hulme, K. Hill, L. Wells, M. Naidoo, S. Mason, S. Murray, V. Korczak, W. Hoang, Y. Kamaladasa and Y. Kong.