What is the frequency of symptomatic mild hypoglycemia in type 1 diabetes in the young?: assessment by novel mobile phone technology and computer-based interviewing

Background: The prevalence of mild hypoglycemia is difficult to document, particularly, in young people with diabetes. The usual method is to ask for subject recall using written 'diaries'. Objective: In 2004, we investigated if new technology could be used to ascertain an accurate prevalence of mild hypoglycemia, particularly self-treated. We compared the use of 'text messaging' and computer-based interviewing with the standard diary method. Participants: Thirty-seven participants, aged 7-18 yr, with type 1 diabetes (T1D) for > 1-yr duration. Method: Open comparison of three systems to collect the data on frequency of hypos (all severity): diary, mobile phone and computer-based interview (CBI), with qualitative analysis of patient feedback. Results: One hundred thirty-two hypos were found over 705 recorded days. All were graded mild or moderate and none severe. Calculated frequency was 5.2 hypos per month: 13.6% subjects had no recorded episode, 36.4% had 1-4, 31.8% 5-9 and 18.2% > 10. Mean blood glucose level at the onset of hypoglycemia was 3.0 mmol/L (1.0-5.2). Response rate of occurrence of hypoglycemic episode recorded by three systems is as follows - diary: 24 (65%) of the 37 subjects reported episodes, mobile: 18 (95%) of 19 subjects and CBI: 16 (89%) of 18 subjects. Sixty-five percent of subjects preferred the mobile and 54% of subjects preferred CBI compared with the diary. Fifty-five percent and 30.8% of subjects found the mobile and the CBI, respectively, easiest to fit into their everyday life. Conclusions: Mobile phone text messaging and CBI are alternatives to written diaries as methods of data collection. Each has its own strengths and weaknesses, but both have the advantage of daily reminders, rapid response and quick data analysis. Using this technology, it was found that the frequency of hypoglycemia was higher (> 3 times) than that previously recognized.