Clinical scoring for alcohol abuse

The previous article on tests for antibodies to Helicobacter pylori, and previous examples on smoking and urine testing demonstrate the power of clinical history in generating high pre-test probability to exploit the additional power of a test. Of course, it's been said before and better. David Sackett and colleagues [1] put it like this (for clinical history of angina):

"Look at the relative size of the likelihood ratios for a brief, immediate, relatively cheap history and a much longer, delayed, and relatively expensive exercise electrocardiogram. There is no contest. Likelihood ratios for key points in the history and physical examination, both for this and for most other target disorders, are mammoth and dwarf those derived from most excursions through high technology."

A chastening thought for those of us who have been on a lifetime excursion through high technology. Yet frustratingly there are few examples with real data comparing clinical symptoms and eventual diagnosis by some gold standard.

Example from alcohol abuse

Patients can be screened systematically for drinking problems with a simple questionnaire. The CAGE questions are just four, scoring one point for each positive answer:

Have you ever felt you should Cut down on your drinking?
Have people Annoyed you by criticising your drinking?
Have you ever felt bad or Guilty about your drinking?
Have you ever had a drink first thing in the morning to steady your nerves or to get rid of a hangover ( Eye-opener )?

Likelihood ratios

Researchers in Virginia applied the questions to the outpatient medical practice aged over 17 of an urban teaching hospital [2]. Eight hundred and thirty six patients who met the inclusion criteria were asked to participate, and 98% agreed. Of these, 36% met criteria for a lifetime history of alcohol abuse or dependence using a gold standard instrument.

The results are shown in the table, with likelihood ratios calculated for patients who scored 0, 1, 2, 3, or 4 questions answered with yes. These can be used on the nomogram to help determine the post-CAGE probability of alcohol problems.

The pre-test probability is likely to have to come from prevalence figures. This might be known locally - but figures from the USA using the same gold standard instrument as in the paper suggests a prevalence of 25% for men and 4.5% for women [3]. These might be useful starting points for men and women.

Clinical diagnosis of cirrhosis

Continuing the alcohol theme and the triumph of clinical observations over the laboratory, some Danish researchers [4] have shown that diagnosis of cirrhosis can be better made by clinical observation than by using biochemical tests.

Using data on over 300 alcohol-abusing men collected nearly 30 years ago, and with liver biopsy as the gold standard, they showed that facial telangiectasia, vascular spiders, white nails, abdominal veins, fatness and peripheral oedema could be used with high diagnostic accuracy. This was clearly superior to using biochemical variables of ESR, bilirubin, albumin, gamma globulin, liver enzymes and clotting factors.

The only problem is that the multiple logistic regression method they used is impenetrable to Bandolier . Perhaps one of our mathematically inclined readers can help.

References:

1 Clinical Epidemiology (Eds DL Sackett, RB Haynes, GH Guyatt, P Tugwell), 2nd Edition, 1991, 132.
2 DG Buchsbaum, RG Buchanan, RM Centor, SH Schnoll, MJ Lawton. Screening for alcohol abuse using CAGE scores and likelihood ratios. Annals of Internal Medicine 1991 115: 774-7.
3 G Edwards. Drug problems as everyday doctor's business. The Oxford Textbook of Medicine, 3rd Edition, 1996, 4263-5.
4 KJ Hamberg, B Carstensen, T Sørensen, K Eghøje. Accuracy of clinical diagnosis of cirrhosis among alcohol-abusing men. Journal of Clinical Epidemiology 1996 49: 1295-1301.

previous or next story in this issue