Skip to main content

Table 1 County Characteristics and the Adjusted Association Between 2008 to 2016 Per Capita Local Public Health Department Spending and Percentage Point Differences in County-level (a) 2008 Sociodemographics and (b) 2008 to 2016 Sociodemographic Shifts

From: US local public health department spending between 2008 and 2016 did not increase for communities in need

 

County-level demographics

Changes in per capita spending

Sociodemographic characteristics

Characteristics of all US counties in 2008 (N = 3005), mean (SD)

Characteristics of sampled counties in 2008 (N = 793), mean (SD)

Based on 2008 sociodemographicsa

(95% CI)

Based on 2008 to 2016 sociodemographic shiftsb

(95% CI)

% over 65 years old

16.1 (4.2)

15.8 (4.2)

+ $0.53*

(+ $0.01 to + $1.06)

− $2.79**

(− $4.18 to − $1.40)

% Black

8.3 (14.0)

7.3 (12.1)

+ $0.00

(− $0.16 to + $0.17)

− $0.40

(− $2.39 to + $1.58)

% Hispanic

8.5 (13.5)

7.2 (10.3)

+ $0.02

(− $0.19 to + $0.22)

+ $0.56

(− $1.09 to + $2.21)

% in poverty

15.1 (6.0)

13.9 (5.2)

− $0.39

(− $1.10 to + $0.32)

− $0.08

(− $1.20 to + $1.04)

% unemployed

5.8 (2.1)

6.0 (1.8)

− $1.31*

(− $2.34 to − $0.27)

− $0.01

(− $1.26 to + $1.25)

% uninsured

14.3 (4.7)

12.9 (4.0)

− $0.14

(− $0.70 to + $0.41)

− $0.55

(− $1.20 to + $0.10)

  1. aAll reported values were from the same linear model, which included each of the 2008 county-level sociodemographic characteristics (percent over 65 years old, Black, Hispanic, in poverty, unemployed, and uninsured) and were additionally adjusted for baseline per capita spending in 2008, the median household income within a county, and a binary determination rurality
  2. bAll reported values were from the same linear model, which included 2008 to 2016 shifts in county-level sociodemographic characteristics (percent over 65 years old, Black, Hispanic, in poverty, unemployed, and uninsured) and were additionally adjusted for baseline per capita spending in 2008, shifts in the median household income within a county from 2008 to 2016, and a binary determination of rurality
  3. *p < 0.05; **p < 0.001