Opinion

Mind – and mine – your general practice data


Dr Chris Hogan 19/06/2018 3:14:12 PM

Dr Chris Hogan calls for better use of the data routinely collected by GPs.

News teaser
How can GPs best use the data they collect every day?

Being a GP is an amazing privilege. Many of us are able to stay in the same practice, or practices, for years, allowing us to build lasting relationships with our patients and communities.
 
We have access to the unfiltered thoughts, hopes, ambitions, fears, tragedies and triumphs of our patients. People share with us with a frankness they show to almost no one else, not even their partners.
 
Of course, this is not always deliberate on our patients’ parts.
 
We are trained to relax our patients and put them at ease, and then to respectfully ask intimate questions and often conduct intimate examinations. We are trained observers and interrogators. Also, we are good clinicians, trained to analyse and detect patterns.
 
We get to know our patients, how they react to illness, their strengths and weaknesses. We also hear what other people like friends, relatives and even adversaries say about them.
 
We effectively live in small towns – the community we serve – and privacy is a matter of politeness and ethics.
 
These records that we produce year in, year out comprise a remarkable insight into our communities, our behaviours, our relationships, and illnesses and health.
 
But what do we do with all this knowledge except to collect it, condense it and record it as data in our electronic records? What use are we making of this data and accumulated experience, information and insight? For the most part, it simply sits in our records.
 
In the past age of paper records and scribbled aide memoires, the ink would fade and the paper turn yellow. These paper records would be destroyed seven years after the patient left the practice, or the doctor retired or died. The evidence of a patient’s experience and a doctor’s life work of compassion, sweat and toil would all become just so much trash.   
 
How does electronic data die? By apathy and expedience.
 
By apathy, I mean that our records are often incomplete, unfiltered and unchecked. Old scripts and such are allowed to clog up active summaries. Expedience means that records are often not checked, not upgraded and not processed. Time-poor GPs often leave sorting through old records for another day.
 
Does this matter? In a word, yes.
 
Old records may hold vital clues for the health of the patients and of blood relatives or other family members.
 
But the data we gather may well have wider uses – and not just for individual patients.
 
Right now, there are many organisations offering to collect our clinical data for their uses, well intentioned or otherwise. Some even suggest that our data, our observations, our deductions and our insight should be available at effectively no cost to insurance companies and the like because the information apparently belongs to the patient alone.
 
If an expert is someone who makes their job look easy, I fear GPs have done too good a job because no one seems to value what we do. Outsiders think it is easy, but it is not. And our records, our data, are testament to that.
 
I believe that GPs are sitting on a treasure trove of data. We should be the instigators of our own clinical studies. We should mine our own data.
 
First, of course, we must mind our data: collect it, care for it, make it coherent.
 
Then mine it for our use.
 
When we rely on memory alone to assess our practice, we rely not on what happens, but what is memorable. The devil, as always, is in the detail.
 
Analyse your data so you can allocate resources where they are needed. Identify gaps in patient care. Build on your strengths.
 
Build up a picture of each patient from family history, past history, occupational and recreational history, examination and special tests. We can use this in conjunction with the RACGP’s Guidelines for preventive activities in general practice (Red Book) to plan what tests and interventions patients need to maintain and even improve their health. We need this before we even consider opening the Pandora’s Box of DNA testing.
 
In my experience, when GPs have a research question but do not pursue it is often due to:

  • not knowing whom to ask
  • being too shy to ask
  • not knowing how to answer the question
  • feeling they do not have the time or resources
  • not having a mentor to see them through. 
This is a shame, as there is a crying need for research into primary healthcare systems, such as:
  • clinical trials support (mentor personnel and research nurses in practice)
  • resources and workforce
  • converting evidence into practice
  • quality and safety.
One major challenge is the lack of a decent data-extraction tool. These tools used to be part of our clinical software, but are now less user-friendly. Most add-on tools are expensive and rely heavily on clean data.
 
If there are affordable data extraction tools available that are compatible with our clinical software, I would love to hear about it.
 
Useful datasets we compile in our work as GPs include:
  • number of patients
  • demographics and their distribution throughout the local area
  • numbers of patients from the same address attending our clinic
  • occupations
  • age
  • medications per person and number of scripts written per patient per year
  • medical conditions
  • number of referrals written per year
These types of datasets give us a clue as to how well we are meeting the needs of our patients and whether we need to develop or extend our services
 
To be effective, data collection for, say, a clinical study must be:
  • straightforward and simple
  • part of routine data collection
  • immediately useful to the person collecting it
  • processed rapidly then returned to those who collected it at a prearranged time.
In short, we need to know what we are doing and what works. Our patients want to know if they are getting value for the money, time and inconvenience of treatment. We need to measure not only cost-effectiveness, but treatment effectiveness for the patient. We need to evaluate if what we are doing works.
 
The answers may lie in our dusty data troves, sitting right there and waiting to be used. 


THE AUTHOR:


clinical datadatadata miningelectronic records



Leave message


 Security code