Table 3.
Estimates of the fitted GLMM using a fractional polynomial of degree 1 for age and a logit-link function
| Effect | Parameter | log OR (s.e.) | t-value | P | OR | |
|---|---|---|---|---|---|---|
| Intercept | β0 | −3·04 (0·38) | −8·09 | <0·001 | ||
| Study site (Reference = Walukuba) | Kihihi | β1 | 0·86 (0·43) | 2·01 | 0·045 | 2·36 (1·02–5·49) |
| Nagongera | β2 | 2·19 (0·40) | 5·45 | <0·001 | 8·94 (4·08–19·57) | |
| Infection status at the previous visit (Ref = Negative and No AL treatment in past) | Negative + AL | β3 | −0·01 (0·10) | −0·05 | 0·956 | 0·99 (0·82–1·21) |
| Symptomatic | β4 | −0·24 (0·10) | −2·30 | 0·022 | 0·78 (0·64–0·97) | |
| Asymptomatic | β5 | 1·23 (0·12) | 9·94 | <0·001 | 3·43 (2·69–4·37) | |
| Age−1 | Walukuba | β6 | −0·05 (0·83) | −0·06 | 0·948 | 0·95 (0·19–4·82)b |
| Kihihi | β7 | −4·01 (0·87) | −4·62 | <0·001 | 0·02 (0·003–0·10)b | |
| Nagongera | β8 | −1·75 (0·45) | −3·89 | 0·001 | 0·17 (0·07–0·42)b | |
| Shifted year of birtha | Walukuba | β9 | −0·13 (0·06) | −2·00 | 0·045 | 0·88 (0·78 –1·00) |
| Kihihi | β10 | 0·11 (0·04) | 2·58 | 0·010 | 1·12 (1·13–1·22) | |
| Nagongera | β11 | 0·04 (0·03) | 1·33 | 0·184 | 1·04 (0·98–1·10) | |
| Variance components | Variance | z-value | ||||
| Variance for random intercepts for subjects | d11 | 0·24 (0·07) | 3·32 | <0·001 | ||
| Variance for random intercepts for households | Walukuba | d22 | 2·80 (0·88) | 3·20 | 0·001 | |
| Kihihi | d33 | 1·16 (0·28) | 4·21 | <0·001 | ||
| Nagongera | d44 | 0·21 (0·08) | 2·48 | 0·007 | ||
a
Birth year – min (birth year).
b
Note that the OR here should be interpreted at the Age−1 level.