Nicholas Summerton

The medical history is a powerful diagnostic technology. However, in seeking to establish an appropriate balance between the history and the other diagnostic modalities more explicit consideration must be given to the performance characteristics of the medical history. Building on recent work undertaken in the UK and elsewhere in Europe it is now feasible to develop a library of setting-specific likelihood ratios and kappa statistics for key elements of the medical history. Of particular importance to those working in primary care, statistically adjusted combinations of information from the medical history can be generated; furnishing clinicians with likelihood ratios of significant magnitudes. It is suggested that developing a more rational approach to the use of the medical history could lead to improvements in diagnostic efficiency and effectiveness, with benefits for individual patient care in addition to the overall NHS budget. When diagnosis is viewed as a processing pathway founded on a robust medical history, it becomes clear than in some situations investigations may become unnecessary and, in other circumstances, their impact will be enhanced.

The disease Can the disease be clearly defined? Is the condition an important problem in terms of prevalence and incidence or morbidity and mortality? Is it a policy priority? Does the condition present a diagnostic problem (inaccuracy or inefficiency), and would it be useful to have better diagnostic tools? Is there evidence of current variation in diagnostic practice (or inappropriate variations in treatment, morbidity, or mortality resulting from diagnostic variability)? Could the diagnostic processing pathway for the disease be improved by obtaining information in a less risky fashion or in a manner more acceptable to patients?

Background: Many medical specialities have reviewed the statistical content of their journals. To our knowledge this has not been done in general practice. Given the main role of a general practitioner as a diagnostician we thought it would be of interest to see whether the statistical methods reported reflect the diagnostic process. Methods: Hand search of three UK journals of general practice namely the British Medical Journal (general practice section), British Journal of General Practice and Family Practice over a one-year period (1 January to 31 December 2000). Results: A wide variety of statistical techniques were used. The most common methods included t-tests and Chi-squared tests. There were few articles reporting likelihood ratios and other useful diagnostic methods. There was evidence that the journals with the more thorough statistical review process reported a more complex and wider variety of statistical techniques. Conclusions: The BMJ had a wider range and greater diversity of statistical methods than the other two journals. However, in all three journals there was a dearth of papers reflecting the diagnostic process. Across all three journals there were relatively few papers describing randomised controlled trials thus recognising the difficulty of implementing this design in general practice.

A case-control study design was used to examine consultation patterns during the three years leading up to the diagnosis of an internal malignancy, within the context of the registered patient list of a single large general practice at Winterton, North Lincolnshire. Using a combination of matching, consultation subclassification, and conditional logistic regression, account was taken of the major confounders affecting consultation rates. Generally, the odds of cancer rose in tandem with increases in the average time between new consultations. This trend was significant for all breast cancers (at the 5% level) and persisted after adjustment for occupation, smoking, and marital status (P = 0.03), as well as after the exclusion of patients identified by routine screening (P = 0.05).

There is little information available to assist general practitioners (GPs) in deciding which patients with haematuria are likely to have a malignancy. To derive discriminant functions for specific items or clusters of clinical history information in relation to the categorisation of patients presenting to the 'open access' haematuria clinic in Hull. Recruitment of patients via an 'open-access' haematuria clinic. A consecutive series of 363 patients aged between 18 and 80 years who attended the clinic. Between February 1999 and October 1999 clinical history information derived from the participating patients was compared with the patients' diagnoses. Diagnoses were established by a combination of cystoscopy and radiological assessments and rechecked against the patient records and the hospital patient administration system two to three months later. A number of individual variables seemed to be particularly helpful in discriminating malignancies. However, when indicants were combined using regression shrinkage techniques, only the following variables were preserved: age, sex, type of haematuria, number of episodes of haematuria, hesitancy, poor urinary stream, smoking history, and history of urinary tract infections. It is possible to generate helpful discriminant information to assist GPs in making more appropriate decisions in a difficult area of clinical practice. However, it remains a matter of judgement as to how representative the study population is likely to be compared with all haematuria patients encountered in primary care. We have reasonable confidence in the general applicability of the rules for macroscopic haematuria: however, it seems likely that the prediction rules outlined for microscopic haematuria have their greatest relevance once a patient has been referred by a GP. In developing the work further and testing out the discriminators identified in this study, we propose that a primary care-based project now needs to be undertaken focusing on microscopic haematuria with a particular emphasis on addressing selection biases. In addition, there is a more general need to assess the reliability of all the suggested items of clinical discriminant information.