Website performance
We intend to monitor the performance of this website and will periodically publish some statistics here.
The site opened for public access on 3rd April 2011 and had had 10,269 distinct visitors by 11th December 2011 from 106 countries, 1289 of whom had registered with the site and 1020 people had successfully completed the online questionnaire.
Of patients with scores of 40% or greater, 7/59 had normal test results - ie 87% of these subjects had CTS, for questionnaire scores above 70 the CTS confirmation rate is 90%. It is likely that some of these cases with high symptom scores and normal NCS are examples of false negative neurophysiology.
We are particularly interested in how good the questionnaire is at excluding CTS as a diagnosis as we would not want to miss patients who would benefit from treatment because the questionnaire gets it wrong.
By 15th December 2011 we had seen 252 people whose questionnaire scores were less than 20%. (23% of all patients referred for testing have scores this low). 24 of these patients (10%) had previously documented CTS, often previously operated. The questionnaire is not calibrated for use in patients whose diagnosis is already established and who may have failed some previous treatment. Of course such patients are not likely to be misled by a low score here. Of the remaining 228, 7 patients, on clinical review, clearly had problems other than CTS which were the dominant cause of their symptoms. The neurophysiological grade of CTS detected in these patients ranged from grade 1 to grade 3 and none were in immediate need of treatment for the CTS. 13 had grade 1 or 2 cts - these cases do not need urgent surgery and a significant number will remit spontaneously. It is therefore probably not unreasonable to adopt a waiting strategy with such patients and review the situation after a few weeks if the symptoms persist.11 patients (5%) had CTS of grade 3 or above which probably was the main cause of their symptoms and these represent the patients whose treatment might be delayed if the questionnaire answer was accepted as significant evidence that the problem was not CTS. 9 of these patients (3%) were subjects who we would normally refer directly for surgery (grade 4 or above) after testing. Two of these patients had scores of exactly 20%. All of these patients, when interviewed gave somewhat atypical clinical histories of CTS as judged by myself.
We have now gone on to perform a formal prospective assessment of the accuracy of the diagnostic predictions made by the website questionnaire. This has been published in an open access journal so that anyone who is interested can read the full report HERE. Overall the performance of the website in predicting the likelihood that evidence of CTS will be found on nerve conduction studies is very good but it should be remembered that the website returns a probability that the patient will turn out to have evidence of CTS - not a guaranteed answer to the question "does this patient have CTS?" In statistical terms it is also of interest that patients with higher diagnostic scores on the website have a marked tendency to have more severe CTS.
We will continue to monitor these figures and investigate apparent exceptions and in due course we will also publish some of the clinic activity data and clinical outcomes of treatment.
Revision date - 27th September 2014
