A Note on the Use and Efficiency of Graded Tuberculin Test in Case Finding and Estimation of Prevalence Rate of Pulmonary Tuberculosis

In a recent paper on the ' Diagnosis of Pulmonary Tuberculosis,' Sen and Chakrabortty^1) (1952) have proposed an inexpensive and simple method of diagnosing pulmonary tuberculosis. According to this, it is possible to select a level of dilution of tuberculin such that at that level most of the tuberculous persons will react and a very negligible proportion of non-tuberculous persons also will show reaction. The error of this test is twofold; a small proportion of tuberculous patients whose ' threshold '* level falls below the selected level will be classified as non-tuberculous and another proportion of non-tuberculous people whose ' threshold '* level falls above this selected level will be classified