I always rather enjoy being processed by the NHS. Instead of my usual panoramic (and perhaps highly misleading) view I’m down in the scrub. What struck me in my latest encounter was the extreme primitiveness of the records.

The medical part of the encounter was well managed. I woke with a prominent floater in my right eye (I can see it now), and I fretted that I might have a retinal detachment. I’m a high myope and have been almost waiting for my retina to detach. I set off for an eye casualty department. The “rapid assessment nurse” decided that I did need to see the ophthalmologist, the nurse practitioner dilated my pupils, and the ophthalmologist used some expensive machinery to look at every part of my retina and declared it undetached. All very satisfactory.

What was unsatisfactory was the records part of the process. For each of the two people in front of me in the queue the receptionist had to pull four sheets of paper from a rack of papers and write the name of the patient on each sheet. He then took a punch, which had the words “Do not remove” taped onto it, and punched two holes in each sheet. Next, he threaded the four sheets together with what I think is called a treasury tag. This 19th century manoeuvre was completed by him writing the name of the patient into a log book.

With one patient the process was complicated by the patient being a child. Perhaps because he’d written the name of the father where he should have written the name of the child or vice versa he had to cross out names and write in other ones—on both the paper and the log book.

All of this took some 15 minutes while I stood and waited. I didn’t mind, I was fascinated by how archaic it all was, but it’s as well that I didn’t have anything that was more of an emergency. The only information that was exchanged during all of this was the first name of the father of the boy. Otherwise, it was all writing, punching, and tagging.

When it came to my turn the same process needed to be repeated, but we hit a snag when the receptionist asked for the name of my GP. I remembered that she had retired, and I didn’t know who was now my GP. Nor was I sure of the name of the practice. I knew the road and the district, but I didn’t know the post code. This is, of course, a failure on my part but not, I guess, an unusual failure.

The receptionist did have access to a computer, so he began a search for my GP, typing with one finger. I don’t know what he was searching—some sort of NHS database presumably—but it wasn’t going well. So I searched with Google on my Blackberry and was able to tell him the postcode. He could then find the information he needed.

Inevitably I was struck by the contrast between the sophistication of the medical process, including use of the expensive equipment, and the crudeness of the record keeping. I thought too of my visits to the supermarket where because of my loyalty card the supermarket knows every last thing about me the second I swipe my card. Why are NHS records so hopeless?

I see two reasons. One is that records have a lower priority than skilled staff, state of the art medical equipment, and very expensive drugs. Records are not “frontline.” They are “back office.” They are boring. This is, of course, a wholly misplaced view. In a knowledge enterprise like healthcare records are of fundamental importance, and they become increasingly so as healthcare is concerned more and more with elderly people with multiple problems dealt with by many different healthcare workers and organisations. Records are central to the “integration” that is so fashionable and necessary.

The second reason for the hopelessness of NHS records is that we spent billions trying to fix the problem and blew it. Connecting for Health was a disaster—a top down megaproject that didn’t pay attention to how people worked and what mattered most.

This is where I must declare a conflict of interest. I’m the chair of Patients Know Best, a still emerging company started by a young doctor who has been a lifelong patient and who understands software. We can provide secure, online records plus much more—and one key feature is that the records belong to the patients, who can then share them with whomever they want avoiding excruciating problems of data protection and the compatibility of different systems.

Progress is slow because health workers are nervous of electronic records, a nervousness that has been hugely amplified by the failure of Connecting for Health. But surely if I return to the eye clinic in ten years’ time it won’t still be necessary to write my name on four pieces of paper, punch them, and connect them with a treasury tag.

Conflict of interest. RS is the chair of Patients Know Best. He is not paid but has equity in the company.

RS was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.