Graphs of study contributions and covariate distributions for network meta-regression.

The fixed-effect NMR reported a difference between artesunate and quinine at every displayed average age; no GRADE assessment was reported.

In the illustrative network meta-regression the log odds ratio for death differed between artesunate and quinine at every displayed average age; certainty was not graded.

The fixed-effect NMR reported a difference between artemether and quinine at ages 15 years or more; no GRADE assessment was reported.

In the illustrative network meta-regression the log odds ratio for death differed between artemether and quinine at average ages of 15 years or more; certainty was not graded.

In the illustrative network meta-regression no difference in the log odds ratio for death between artesunate and artemether was seen at any displayed average age; certainty was not graded.

The NMR reported an SMD by randomisation year; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between placebo and no treatment at randomization years 1970 to 1990; certainty was not graded.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in dentifrice and no treatment at every displayed randomization year; certainty was not graded.

The NMR reported a difference between fluoride in dentifrice and no treatment at every displayed randomisation year; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in rinse and no treatment at every displayed randomization year; certainty was not graded.

The NMR reported a difference between fluoride in rinse and no treatment at every displayed randomisation year; no GRADE assessment was reported.

The NMR reported a difference between fluoride in gel and no treatment at every displayed randomisation year; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in gel and no treatment at every displayed randomization year; certainty was not graded.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in varnish and no treatment at randomization years 1954 to 1990; certainty was not graded.

The NMR reported a difference between fluoride in varnish and no treatment at randomisation years 1954 to 1990; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in dentifrice and placebo at every displayed randomization year; certainty was not graded.

The NMR reported a difference between fluoride in dentifrice and placebo at every displayed randomisation year; no GRADE assessment was reported.

The NMR reported a difference between fluoride in rinse and placebo at randomisation years 1954 to 1990; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in rinse and placebo at randomization years 1954 to 1990; certainty was not graded.

The NMR reported a difference between fluoride in gel and placebo at randomisation years 1954 to 1980; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment differed between fluoride in gel and placebo at randomization years 1954 to 1980; certainty was not graded.

The NMR reported a difference between fluoride in varnish and placebo at randomisation years 1954 to 1980; no GRADE assessment was reported.

In the illustrative network meta-regression the SMD for caries increment showed a treatment-by-year interaction and differed between fluoride in varnish and placebo at randomization years 1954 to 1980; certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in rinse versus fluoride in dentifrice across randomization year; the review did not state whether the difference was significant and certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in gel versus fluoride in dentifrice across randomization year; the review did not state whether the difference was significant and certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in varnish versus fluoride in dentifrice across randomization year; the review did not state whether the difference was significant and certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in gel versus fluoride in rinse across randomization year; the review did not state whether the difference was significant and certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in varnish versus fluoride in rinse across randomization year; the review did not state whether the difference was significant and certainty was not graded.

The illustrative network meta-regression estimated an SMD for caries increment for fluoride in varnish versus fluoride in gel across randomization year; the review did not state whether the difference was significant and certainty was not graded.