Standardisation of rates using logistic regression: a comparison with the direct method

Table 1 Direct standardisation calculations for subclinical hyperthyroidism

Age group	Sex	Cases in study (r_ij)	Age-sex distribution of the study population (n_ij)	Age-sex specific prevalence rate in study (per 100) $({\hat{p}}_{i j} = \frac{r_{i j}}{n_{i j}} \times 100)$	Age-sex distribution of E & W in 100's (N_ij)	Expected cases N _ij× ${\hat{p}}_{i j}$	$\frac{N_{i j}^{2} \times {\hat{p}}_{i j}}{n_{i j}}$
65–69	male	16	945	1.6931	11306	19142	22902048
65–69	female	13	981	1.3252	12149	16100	19938220
70–74	male	14	916	1.5284	9541	14582	15189849
70–74	female	14	839	1.6687	11228	18736	25073151
75–79	male	17	643	2.6439	7334	19390	22116112
75–79	female	29	660	4.3939	9865	43346	64789451
80+	male	9	388	2.3196	7791	18072	36288203
80+	female	17	509	3.1434	15327	48179	145077055
Total		128	5881	2.1765	84541	197547	351373093

ISSN: 1472-6963