Associations between the stringency of COVID-19 containment policies and health service disruptions in 10 countries

Table 3 Results from multi-level linear regression model for the association between the OxCGRT stringency index and relative service volumes

	Estimate	95% CI	p -value
Mean stringency index (per 10 percent)	-3.342	-3.942, -2.741	< 0.001
COVID-19 cases per million population	-0.001	-0.0026, 0.0002	0.087
Service type
Reproductive, maternal and newborn health services	Reference	N/A	N/A
Service use overall and injuries	-10.021	-14.970, -5.072	< 0.001
Child health services	-11.927	-19.109, -4.744	0.001
Antiretroviral therapy	13.318	-4.404, 31.040	0.141
Chronic disease services	-10.841	-22.742, 1.061	0.074
GDP per capita (per 1000 unit)	-0.581	-1.111, -0.051	0.032
Variance estimates
Country-level random effects	89.283
Residual	442.885

Mean stringency index (\({\beta }_{1}\)) is the mean stringency index value for the month in each country, rescaled to a factor of 10. COVID-19 cases (\({\beta }_{2}\)) is the COVID-19 cases per million monthly in each country. Service type categories (\({\beta }_{3}\)) for service volume included: (1) reproductive, maternal, or newborn, (2) service use overall and injuries, (3) child health services, (4) antiretroviral therapy and (5) chronic diseases. GDP (\({\beta }_{4}\)) is GDP per capita, rescaled to a factor of 1000

ISSN: 1472-6963