Should patients with abnormal liver function tests in primary care be tested for chronic viral hepatitis: cost minimisation analysis based on a comprehensively tested cohort

Table 5 Yield, sensitivity and Positive Predictive Values (PPV) of different detection strategies

Strategy for viral testing	No. of patients*	Hepatitis cases*	Viral tests	Cases detected	Sensitivity (%)	PPV (%) 95% Confidence Limits
A. If repeat LFT panel is abnormal	1124	11	955	11	100	1.15 (0.64-2.05)
B. If ALT abnormal on primary test	1064	12	418	8	67	1.91 (0.97-3.73)
C. If ALT > 2 upper limit of normal on primary test	1064	12	77	6	50	7.79 (3.62-15.98)
D. If patient born in a country of intermediate to high viral hepatitis prevalence.	1208	13	170	11	85	6.47 (3.65-11.21)
E. If patient born in a country of intermediate to high viral hepatitis prevalence and ALT > 2 upper limit of normal on primary test.	1041	12	16	5	42	31.25 (14.16-55.60)
F. If patient born in a country of intermediate to high viral hepatitis prevalence, or ALT > 2 upper limit of normal on primary test.	1041	12	215	11	92	5.12 (2.88-8.93)
G. Test all cases	1236	13	1236	13	100	1.05 (0.62-1.79)

Testing patients for viral infection on the basis of country of origin is more sensitive and has much higher positive predictive value.
* The sample of patients available to evaluate each strategy varies because of patterns of missing data, as follows:
A requires a complete panel of follow-up LFTs, the missing data in the two cases that were not abnormal might have led to an exagerated estimate of sensitivity;
B and C both require an initial ALT test;
D requires information on country of birth;
E and F require an initial ALT together with country of birth.
All evaluations require results of viral tests for both Hepatitis B and Hepatitis C.
ALT: Alanine aminotransferase;
LFT: Liver function test.

ISSN: 2731-4553