Work contribution Part 5 - Improving the reliability of the history

Blog - Work contribution Part 5 - Improving the reliability of the history

Dr Mary Wyatt | Published: August 25, 2015
Patient histories are by nature unreliable as people’s recollection skills are never perfect. We are human after all. My estimate, based on research and everyday experience, is that the history of long term health problems is only about 50% accurate.
I’m a doctor but have also been a patient - not very often, but often enough to know that my recollection is just as poor as the patients I see.  
I’ve been in situations where I’ve been asked about my past history and forgotten it. Only a day or two later have I realised that I should have mentioned x or y, relevant information which I’d left out. If someone who works in the field is unable to provide an accurate history, how can we expect patients to be reliable? 
Studies suggest that the more questions you ask about prior problems, the more likely you are to get an accurate response. For example, if you ask, “Have you had any previous back pain?” a certain percentage of people will answer yes but most will answer no. 
If this is followed up with a second question, such as “Have you ever been to a chiropractor?” some of the people who answered “No” will then recall going to a chiropractor for maintenance treatment for back stiffness. 
There are also other ways you might ask the question, such as “Have you ever had any x-rays or scans of your back?”
The more questions you ask, the more likely you are to get an accurate history.  
While these approaches increase the accuracy of the patient history, limitations in recollection remain an issue. 
Patients and doctors can also have different interpretations of questions.  
As part of a standard history I ask people if they drive. Those who say they are not driving because of their condition are then asked a second question, “When did you last drive a car?” 60% to 70% then reply they are in fact driving, but only for short distances, infrequently or only if no one else is available. I still scratch my head over this one. At my end ‘are you driving’ is a Yes / No question. Clearly not for the patient. 
In taking a history, it’s quite common to follow up with further questions to check the doctor and the patient are on ‘the same page’. 
Q “Any previous back pain?”
A “Not really.”
Q “What do you mean by not really?”
A “Well, hmm, I mean no, no previous pain.”
In some situations it’s important to explain the reason a question is important. For example, “It’s important that I have a full understanding of any previous problems as that helps me to decide on what is the best treatment for you, and will help me tell how your condition is likely to progress.” Armed with an understanding of the importance of a question, some patients will be more expansive. 
Getting a good history isn’t always simple, but through engaging the patient and asking probing questions, it can produce a more thorough picture even though it takes more time. It will be time well spent.