Skip to main content

Table 3 Association between the dentists’ and patients’ characteristics and the identified clusters—Multivariable polytomous logistic analyses (n = 3,014 dentists)

From: Characteristics of dentists and patients associated with appropriate antibiotic prescriptions by French dentists: a cross-sectional study using Health Insurance databases

Characteristics in 2019*

Multivariable analysis

Cluster 2 “average practices”

VS Cluster 1 “better than average practices”

Cluster 3 “worse than average practices”

VS Cluster 1 “better than average practices”

Odds Ratio

(95% confidence interval)

p-value

Odds Ratio

(95% confidence interval)

p-value

Geographic area

Lorraine

1 (reference)

1 (reference)

Alsace

1.43 (1.16; 1.77)

0.0009

2.15 (1.72; 2.68)

 < 0.0001

Champagne-Ardenne

1.55 (1.18; 2.03)

0.002

2.83 (2.14; 3.73)

 < 0.0001

Dentist’s gender

Male

1 (reference)

1 (reference)

Female

1.06 (0.87; 1.29)

0.596

1.37 (1.12; 1.69)

0.003

Number of years of practice a

1.01 (1.00; 1.01)

0.219

1.02 (1.01; 1.03)

0.0002

Number of patients

1.000 (1.000; 1.001)

0.004

1.000 (1.000; 1.000)

0.532

Number of procedures per patient

1.16 (1.01; 1.34)

0.041

1.17 (1.01; 1.35)

0.037

% of patients with chronic diseases

1.04 (1.01; 1.07)

0.020

1.04 (1.01; 1.08)

0.023

  1. The characteristics included in this multivariable model were the following: geographic area, dentist’s gender, number of years of practice, number of patients, number of procedures per patient, percentage of patients with chronic diseases and percentage of patients with low income. The percentage of patients > 65 years old was not included because of high collinearity with the percentage of patients with chronic diseases (r = 0.74). The type of procedure was not included because of high collinearity with the number of procedures per patient (p < 0.001). The variable percentage of patients with low income was removed from the model because of its non-significance (p = 0.824). The reference class is Cluster 1 “better than average practices”
  2. Bold indicates a significant p-value (p-value < 0.05)
  3. aInterpretation of results: probability of belonging to cluster 2 (3) multiplied by 1.01 (1.02) as compared to cluster 1 for each additional year of practice