Costs of fall injuries in the STRIDE study: an economic evaluation of healthcare system heterogeneity and heterogeneity of treatment effect

Table 4 Omnibus (Type 3) Wald tests for cost of FIMA, in Tweedie and negative binomial models

All healthcare systems	Tweedie	negative binomial
Effect	p value	p value
Treatment arm (intervention vs. control)	0.340	0.982
Healthcare system (dummy-coded)	0.035	0.002
Healthcare system by treatment arm interaction	0.090	0.065
Tertile of practice size (dummy-coded)	0.043	0.053
Study participants in practice were majority white race (vs. not)	0.429	0.318
Urban practice (vs. rural)	0.337	0.698
Sensitivity analysis: healthcare system A excluded	Tweedie	negative binomial
Effect	p value	p value
Treatment arm (intervention vs. control)	0.954	0.453
Healthcare system (dummy-coded)	0.043	0.004
Healthcare system by treatment arm interaction	0.683	0.343
Tertile of practice size (dummy-coded)	0.014	0.022
Study participants in practice were majority white race (vs. not)	0.421	0.342
Urban practice (vs. rural)	0.115	0.309

This table shows results for the main analysis that includes all 10 healthcare systems, and a sensitivity analysis where healthcare system A (which showed a confidence interval that excluded zero in favor of the intervention) is excluded
Abbreviations: FIMA, fall injuries with medical attention

ISSN: 1478-7547