Jill E. Maddison, BVSc, DVCS, PhD, FACVSc, MRCVS
In problem-based clinical reasoning each significant clinico-pathological problem is considered separately before being related to the other problems that the patient may present with.
An alternative approach is to try to remember all diseases that fit the 'pattern' of clinical signs/pathological abnormalities that the animal is showing. This may be relatively simple for common disorders, if a disorder has a unique pattern of clinical signs or if you have a huge amount of experience and knowledge, as well as an excellent memory. However, using pattern recognition can also lead to errors of omission.
Pattern recognition is unsatisfactory when the clinician is faced with a patient exhibiting multiple clinical signs that are not immediately recognisable as a specific disease or, more commonly, if the pattern of clinical signs is suggestive of certain disorders but not specific for them. This can lead to dangerous tunnel vision where the vet pursues his/her initial diagnostic hunch to the exclusion of other diagnostic possibilities and interprets all subsequent data as favourable to their initial diagnosis including ignoring data that don't 'fit' their preferred diagnosis.
For experienced veterinarians, pattern recognition combined with 'fishing expeditions' (i.e., 'I have no idea what's going on so I'll just do bloods and hopefully something will come up!') can result in a successful diagnostic or therapeutic outcome in perhaps 70-80% of medical cases in general practice.
But the other 20-30% of cases do not yield their secrets so readily using these approaches and it is these cases that frustrate veterinarians and make practice less pleasant than it should be.
What is Problem Based Clinical Reasoning?
In problem-based clinical reasoning, the pathophysiological basis and differential diagnosis for the most specific clinical signs the patient is exhibiting are considered before a pattern is sought. This approach ensures that one's mind remains more open to other diagnostic possibilities than what might seem to be initially the most obvious and helps prevent pattern-based tunnel vision. Pattern-based tunnel vision can result at best in a speedy, correct diagnosis but, at worst, in wasted time, money, and it sometimes endangers the life of the patient.
If there are multiple clinical signs e.g., vomiting, polydipsia, and a pulse deficit, each problem is considered separately first and then in relation to the other problems to determine if there is a disorder (or disorders) that could explain all the clinical signs present. In this way, the clinician, if she/he has developed a logical approach to each clinical problem, will be able to easily assess the potential differentials for each problem and then relate them rather than trying to remember every disease process that could cause that pattern of particular signs.
Thus, one does look for patterns but not until we have put in place an intellectual framework that helps prevent tunnel vision too early in the diagnostic process.
The cornerstone of a problem-based approach is to always approach each problem logically and (usually) to rigorously pursue, as appropriate, some or all of the following questions:
What is the problem?
What system is involved?
Where within the system is the problem located?
What is the lesion?
Identify the Problem
Example: The owner reports that the dog is vomiting. Is the animal really vomiting or regurgitating?
Until the problem is appropriately defined, diagnosis is not possible. It is probably at this first step that many clinicians most commonly make mistakes, e.g., assuming the animal is vomiting when in fact it is regurgitating. Failure to appropriately identify the problem can lead to wasted time and money.
Identify the Body System Affected and How it is Involved
The question can be posed as 'Do I have a primary (i.e., structural) problem of a body system or a secondary problem (i.e., the system is affected by other factors)?'
Example: Is vomiting due to primary or extra-gastrointestinal disease?
Failure to appropriately identify the system involved can also lead to wasted time and money as well as client and veterinarian frustration.
In fact, if you do nothing else in assessing a case before seeking the diagnostic 'pattern', ask yourself for each of the specific problems--'what system could be involved and how?'. This simple question will immediately open up your mind to diagnostic possibilities you may never have contemplated if you were just focusing on the 'pattern'.
An alternative, though closely related, question for some problems is 'Is the problem local or systemic?'
Epistaxis--due to local nasal disease or systemic disease e.g., coagulopathy, hyperviscosity?
Melaena--GI bleeding due to local disease (ulceration--which in turn may be due to primary or secondary GI disease) or systemic disease e.g., coagulopathy
Identify the Anatomical Location of the Lesion
Example: Having determined that vomiting is due to primary gastrointestinal disease, where in the gastrointestinal tract is the lesion located?
In this example, by asking this question you will select the most appropriate method to either answer the question or to move on to the next step.
For example, if you believe that your history and physical examination and other ancillary data indicate a lower small intestinal lesion, endoscopy is not going to be an appropriate method of visualising the area or obtain biopsies. On the other hand, if all information you have suggests a gastric lesion, endoscopy would be appropriate
Identify the Lesion
Example: The patient has a gastric lesion--is it a tumour, foreign body, or ulcer?
This question will require visualisation and/or biopsy to answer but it would have been a waste of time asking the question until you had arrived at the right location.
The final result
Instead of thinking 'I wonder if it has a gastric foreign body or renal failure or a liver tumour?' when faced with a vomiting patient, your initial energies are directed at defining the problem, then the system etc. In this way, the diagnosis is made logically and thoughtfully, and, during the process, all diagnostic options can be considered as the need arises.
Does Pattern Recognition Have a Place?
Having said all of this, it is important to reiterate that pattern recognition for many cases is appropriate and justified--depending on your level of experience. For example, if a potbellied terrier with bilaterally symmetrical alopecia, seborrhoea, hyperpigmentation and comedones walked into your consulting room and the owner reported that the dog was drinking lots of water, was ravenously hungry and seemed to be panting excessively, obviously hyperadrenocorticism is the most obvious diagnosis and going through the motions of assessing each specific problem would be ridiculous (but not if you had never seen a dog with hyperadrenocorticism before!).
However, it is important to be aware that pattern recognition is only foolproof if the pattern is virtually unique to the disease and its value is very dependent on the clinician's experience, depth of knowledge, ability to sort data quickly and efficiently etc.
Of course, once you have considered each individual problem, you do in fact look for a pattern in the clinical signs. But the insertion of that initial step of considering each specific problem individually and then relating it to the other problems present should ensure that you don't miss the less obvious possible diagnoses that you may not have initially been aware of.
There are some combinations/patterns of clinical signs that make the diagnostic options very limited: for example, a patient with PU/PD who is also polyphagic. If the PU/PD and polyphagia have been present for the same length of time then they are almost certainly due to the same disorder and it is quite appropriate to assess them together. There are very few conditions that will cause this pattern of clinical signs so it is quite appropriate to concentrate on these first. Even if the polyphagia is not due to the same pathology there are a number of causes of PU/PD that simply aren't feasible in a dog that is also eating well (because they would decrease not increase appetite).
In addition, the process of developing a sound problem-orientated approach can enhance your ability to recognise patterns, because you have a greater understanding of the reasons why you believe a certain pattern suggests some disorders more than others.