TABLE 3.

Effects of one thousand US$ (at 2011 prices) of added GDP per capita on mortality (per 100,000 population), as estimated from regressions in which observations are weighted by the square root of the population size ¹ (Salud Urbana en América Latina Project, ten countries of Latin America, 1990–2015).

Dependent variable (death rate per 100,000 population)	Tertiles of socioeconomic status (SES) 2
	First tertile (low SES)		Second tertile (medium SES)		Third tertile (high SES)
	M1 3	M2 4	M1	M2	M1	M2
Total mortality	3.84	1.45	0.93	0.33	0.65	1.84
Female	3.49	1.71	4.70***	3.68**	1.32	2.33+
Male	4.51	1.30	−2.48	−2.55	0.05	1.52
0–9 years	5.13	3.14	2.08	0.95	1.75*	1.99**
10–29 years	1.17	0.76	−4.84	−7.08*	−1.07	−0.33
30–44 years	2.02	0.75	−5.57	−8.42+	−0.61	0.35
45–64 years	5.78	2.06	6.49*	4.77	1.95	3.81**
65+ years	50.16	0.36	44.25**	40.57**	12.14	23.50*
Cardiovascular diseases	0.66	−0.88	1.20*	0.82+	0.59	0.95
Respiratory diseases	−0.80	−0.23	0.09	0.13	0.14	0.27
Malignant neoplasms	0.86*	0.50	1.01*	0.72+	−0.08	−0.004
Diabetes mellitus	0.40	0.50	0.31	0.18	0.37*	0.22
Respiratory infections	−0.01	−0.21	0.35*	0.45**	0.22	0.34+
Infectious and parasitic diseases	−0.17	0.02	0.38*	0.45*	−0.28+	−0.16
Nutritional deficiencies	0.30	0.23	0.003	−0.03	−0.09	−0.02
Road traffic injuries	0.44	0.16	0.80**	0.76**	0.43***	0.48***
Suicide	0.13	0.09	0.07	0.10	0.07	0.06
Homicide	0.74	−0.01	−4.73	−4.58*	−0.67	−0.30

¹Regressions in which the dependent variable is an age-specific rate only include as demographic covariate an index of masculinity. All other regressions include the population proportions of children younger than 10 and elderly older than 64.

²Urban areas (n = 340) classified according to GDP per capita of 2014. Observations (years-city) = 4,383.

³Model including fixed effects for city and year, and city-specific linear trends.

⁴Model with series detrended with the Hodrick-Prescott filter with a smoothing parameter γ = 100.

***p < 0.001.

**p < 0.01.

*p < 0.05.

⁺ p < 0.10.