Skip to main content
. 2022 Apr 16;399(10334):1489–1512. doi: 10.1016/S0140-6736(22)00172-6

Table 2.

Factors associated with variation in cross-country cumulative infections per capita, IFR, and hypothetical levels of trust and prevalence of risk factors

Variation in infections per capita explained by each factor, % (95% UI) Variation in IFR explained by each factor, % (95% UI) Reduction in global infections each country's level of trust had exceeded 75th percentile across countries, % (95% UI) Reduction in global IFR if each county's mean BMI was less than the 25th percentile across all countries, % (95% UI)
Seasonality 2·1% (1·7–2·7)* .. .. ..
Age structure .. 46·7% (18·1–67·6)* .. ..
GDP per capita 4·2% (1·8–6·6)* 3·1% (0·3–8·6)* .. ..
Population density 1·8% (0·8–3·2) 1·7% (0·3–5·6) .. ..
Altitude 5·4% (4·0–7·9)* .. .. ..
Pre-exposure to betacoronavirus 2·1% (1·1–3·1) 0·7% (0·1–2·1) .. ..
Body-mass index .. 1·1% (0·2–2·6)* .. 11·1% (2·1–20·6)*
Smoking prevalence .. 0·3% (0·1–3) .. ..
Air pollution .. 0·3% (0·1–2·1) .. ..
COPD prevalence .. 0·2% (0·0–0·7) .. ..
Cancer prevalence .. 1·6% (0·1–4·8) .. ..
Trust in government 7·4% (5·4–9·6)* .. 12·9% (5·7–17·8)* ..
Interpersonal trust 16·5% (12·3–19·5)* .. 40·3% (24·3–51·4)* ..
Unexplained variation 60·6% (55·6–65·2) 44·4% (29·2–61·7) .. ..

BMI=body-mass index. COPD=chronic obstructive pulmonary disease. IFR=infection-fatality ratio. UI=uncertainty interval.

*

Estimated parameters that are statistically different from zero.

These covariates are assumed to be independent from each other and all other covariates. Further, a few countries had incomplete reporting of these covariates. Corresponding figures reflect those countries where the respective covariate was present.