Skip to main content
. 2017 Sep 5;177(9):1326–1333. doi: 10.1001/jamainternmed.2017.2649

Table 2. Association of Patient- and Physician-Level Characteristics With Having a Potentially Low-Value ECG Within 30 Days After an Annual Health Examination in 2 873 357 Patientsa.

Fixed Effects Odds Ratio (95% CI)b
Patient characteristics
Age group, y
45-64 vs 18-44 3.46 (3.44-3.48)
≥65 vs 18-44 4.73 (4.67-4.79)
Male vs female 1.10 (1.09-1.11)
Rural 0.78 (0.77-0.80)c
Neighborhood income quintile
2 vs 1 (Lowest) 1.07 (1.06-1.08)
3 vs 1 (Lowest) 1.11 (1.10-1.12)
4 vs 1 (Lowest) 1.14 (1.13-1.15)
5 vs 1 (Lowest) 1.12 (1.11-1.13)
Hospital admission in past 3 years 1.02 (1.00-1.03)
Cancer 1.45 (1.44-1.47)
COPD 1.07 (1.05-1.08)
Asthma 0.86 (0.85-0.87)
Mental health 1.06 (1.05-1.07)
Dementia 0.86 (0.81-0.91)
Rheumatologic disease 1.27 (1.25-1.29)
Physician characteristics
Male vs female 1.11 (1.10-1.12)
IMG 1.13 (1.12-1.14)
Years since graduation (mean)
21-30 vs 0-20 1.04 (1.04-1.05)
>30 vs 0-20 1.15 (1.14-1.16)
Organizational structure
Family health group vs FFS 1.55 (1.54-1.57)
Family health network vs FFS 1.22 (1.06-1.40)
Family health organization vs FFS 1.31 (1.28-1.35)
Family health team vs FFS 0.99 (0.97-1.02)
Other vs FFS 1.64 (1.59-1.70)
Random effect
Practicec 2.50 (2.40-2.60)

Abbreviations: COPD, chronic obstructive pulmonary disease; ECG, electrocardiograms; FFS, fee for service.

a

All reported values based on SAS (version 9.4) PROC GLIMMIX output; model estimation method = RSPL; denominator degrees of freedom estimation method = between and within; covariance structure = standard variance. P < .05 for all variables except family health team vs FFS (P=.71)

b

All odds ratios presented are adjusted for all other factors in the table.

c

Median odds ratio reported with 95% CI in parentheses.