Skip to main content

Table 5 Segmented logistic regression results examining impact of transition from tFFS to eFFS on FPSC ED Visits

From: The impact of the adoption of a patient rostering model on primary care access and continuity of care in urban family practices in Ontario, Canada

Parameter Unadjusted Model Adjusted Model
Estimatea 95% CI P-Value Estimate 95% CI P-Value
Intercept (baseline ED) −3.51 −3.53 to −3.50 < 0.0001 −3.28 −3.33 to −3.23 < 0.0001
Pre-intervention slope (secular trend, per year) 0.015 0.013 to 0.019 < 0.0001 0.018 0.015 to 0.021 < 0.0001
Change in intercept (immediate impact) −0.011 −0.014 to −0.0070 0.0098 −0.010 −0.018 to − 0.0020 0.0128
Change in slope (gradual effect, per year) −0.010 − 0.014 to − 0.0068 < 0.0001 −0.011 − 0.014 to − 0.0080 < 0.0001
Female physician     −0.041 − 0.089 to − 0.022 0.0123
Physician panel size
  < 500     0   
 500–999     −0.021 −0.055 to − 0.060 0.015
 1000–1999     − 0.030 −0.067 to − 0.015 0.0018
 2000–2999     − 0.070 −0.098 to − 0.042 < 0.0001
  > 3000     − 0.074 − 0.110 to − 0.039 < 0.0001
Foreign Trained     − 0.23 −0.27 to − 0.19 < 0.0001
Years since graduation     0.0058 0.0040 to 0.0080 < 0.0001
Patient age     −0.021 −0.022 to − 0.021 < 0.0001
Female patient     − 0.047 − 0.051 to − 0.044 < 0.0001
Adjusted Clinical Group (ACG)c
 0     0   
 1–4     0.20 0.19 to 0.21 < 0.0001
 5–9     0.79 0.77 to 0.80 < 0.0001
 10+     1.53 1.52 to 1.56 < 0.0001
Income Quintileb
 1     0   
 2     −0.12 −0.12 to 0.11 0.88
 3     −0.194 −0.200 to − 0.189 < 0.0001
 4     − 0.256 − 0.261 to − 0.251 < 0.0001
 5     − 0.333 −0.338 to − 0.327 < 0.0001
Patient rurality
 Urban     0   
 Suburban     0.65 0.64 to 0.65 < 0.0001
 Rural     1.31 1.29 to 1.31 < 0.0001
  1. aEstimates represent the log odds of an FPSC ED visit
  2. bincome quintile represents the rank of the patient’s total household income based on the aggregate census data derived from postal code. The first quintile represents the highest incomes
  3. cAdjusted Clinical Groups (ACG) quantifies morbidity by grouping patients based on age and gender and all medical diagnoses in a given year. Those in group three represent represents those with the greatest morbidity