Skip to main content

Table 3 Contraceptive sales negative binomial regression results by study group and time

From: Impact of a passive social marketing intervention in community pharmacies on oral contraceptive and condom sales: a quasi-experimental study

  Estimate SE LB UB t P value
Condoms       
intervention (vs. comparison) −1.43 0.73 −2.87 0.01 −1.96 0.051
time 2 (vs. 1) −0.05 0.07 −0.20 0.09 −0.74 0.461
time 3 (vs. 1) −0.46 0.07 −0.61 −0.32 −6.42 <.001
time 3 (vs. 2) −0.41 0.07 −0.55 −0.27 −5.69 <.001
intervention * time 2 (vs. 1) 0.16 0.14 −0.13 0.44 1.08 0.283
intervention * time 3 (vs. 1) 0.50 0.14 0.21 0.78 3.42 <.001
intervention * time 3 (vs. 2) 0.34 0.14 0.06 0.62 2.35 0.020
Oral Contraceptives       
intervention (vs. comparison) −0.74 0.22 −1.17 −0.32 −3.44 <.001
time 2 (vs. 1) −0.04 0.04 −0.11 0.04 −0.91 0.363
time 3 (vs. 1) 0.01 0.04 −0.07 0.09 0.02 0.842
time 3 (vs. 2) 0.04 0.04 −0.03 0.12 1.11 0.268
intervention * time 2 (vs. 1) 0.06 0.08 −0.10 0.21 0.73 0.465
intervention * time 3 (vs. 1) 0.03 0.08 −0.12 0.19 0.40 0.693
intervention * time 3 (vs. 2) −0.03 0.08 −0.18 0.13 −0.34 0.737
  1. Main effect and interaction estimates and p values are computed with contrast tests in the regression.
  2. Interaction terms reference category is always control.
  3. LB = Lower Bound, UB = Upper Bound for a 95% Confidence Interval.