Can computer software programmes be used to help clinicians with their diagnoses?

23 Jun

by Rahul Alam, Research Associate in General Practice theme

Isabel screenshot

Figure 1 Screenshot of a differential diagnosis list

General Practitioners (GPs) and nurse practitioners in general practice are required to recall a large number of illnesses and diseases as well as a large number of tests and drugs that can be ordered and prescribed. Given such breadth, arriving at the correct diagnoses can be a difficult task, particularly when patients present with unusual symptoms. As a result, diagnostic errors can, and do, occur. One potential resource for helping clinicians to make the correct diagnoses are specialised computer software programmes, otherwise known as differential diagnosis (DDx) generators. They are aimed at helping clinicians with the diagnostic process to reduce the possibility of clinicians missing, delaying or making incorrect diagnoses. In principle, clinicians can enter the patient’s clinical symptoms and the DDx generator produces a list of potentially relevant diagnoses the clinician might want to consider (please see figure 1). This information may be utilised by clinicians as part of the differential diagnostic process to whittle the potential diagnoses down to those most likely. Isabel is one such commercially available web-based DDx generator.

However, little is known about how DDx generators will be viewed and whether they can be used by clinicians in routine practice.  There is also limited evidence on how accurate programmes like Isabel are. In order to ascertain the potential feasibility and utility (benefits) of using this program, we are conducting a small one-practice study divided into two separate components.

The first component of the study will investigate the diagnostic ‘accuracy’ of Isabel in relation to cases of diagnostic uncertainty. Two situations that are likely to involve elements of diagnostic uncertainty relate to referrals to secondary care for follow-up investigations and people who frequently attend the practice (those whose attendance rate is in the top 3% of the practice). Two GPs will review and select all cases of diagnostic uncertainty and the patient details will be entered into Isabel and compared to the final discharge diagnosis from the hospital or the final recorded diagnosis to determine Isabel’s accuracy rate.

In the second component we aim to ascertain the feasibility of clinician’s using Isabel in routine practice when diagnosing patients. We have conducted 11 interviews with 9 GPs and 2 nurse practitioners to ascertain their views and experiences of using a differential diagnosis tool such as Isabel.  We then trained all clinical staff at the practice to use Isabel and they have now had a 6-month opportunity to use Isabel. A remote monitoring system captured wider data such as usage rates, links accessed and time taken to use. We are now conducting post-use interviews with the clinicians to obtain their feedback and opinions on Isabel as a diagnostic aide in routine general practice.

We hope that the findings of the two components of this study can help assess the diagnostic accuracy of Isabel in UK general practice as well provide useful insights in to the feasibility of utilising Isabel in routine clinical practice.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: