Abstract
From replies to a postal questionnaire used in a radioactive iodine follow-up scheme patients had to be classified as either "suspected hypothyroid" or "euthyroid." A comparisons has been made of the effectiveness of published statistical techniques in making this classification. Two main conclusions emerged. Firstly, all except one of the methods identified an acceptable proportion of the hypothyroid patients, and, secondly, the results given by these methods were remarkably similar. Thus the simplest, which required only a count of the number of symptoms present, was selected for use.
Full text
PDF
Selected References
These references are in PubMed. This may not be the complete list of references from this article.
- Anderson J. A., Whaley K., Williamson J., Buchanan W. W. A statistical aid to the diagnosis of keratoconjunctivitis sicca. Q J Med. 1972 Apr;41(162):175–189. [PubMed] [Google Scholar]
- Barker D. J., Bishop J. M. Computer analysis of symptom patterns as a method of screening patients at special risk of hypothyroidism. Br J Prev Soc Med. 1970 Nov;24(4):193–196. doi: 10.1136/jech.24.4.193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Barker D. J., Bishop J. M. Computer-based screening system for patients at risk of hypothyroidism. Lancet. 1969 Oct 18;2(7625):835–838. doi: 10.1016/s0140-6736(69)92288-0. [DOI] [PubMed] [Google Scholar]
- Billewicz W. Z., Chapman R. S., Crooks J., Day M. E., Gossage J., Wayne E., Young J. A. Statistical methods applied to the diagnosis of hypothyroidism. Q J Med. 1969 Apr;38(150):255–266. [PubMed] [Google Scholar]