The impact of penicillin allergy labels on antibiotic and health care use in primary care: a retrospective cohort study

Table 2 Total number of antibiotic prescriptions and GP contacts in primary care per patient year (absolute effect)

Outcomes	Total n = 5010	Pen-A patients n = 1254	Non-Pen-A patients n = 3756	Unadjusted B	95% CI	Adjusted B^a	95% CI
No. of AB per PY (%)^d	2110 (100.0%)	1051 (100.0%)	1059 (100.0%)	0.556	0.377 to 0.736	0.392	0.183 to 0.600
No. of 2nd choice AB per PY^b (%)^d	269.4 (5.4%)	93.7 (7.5%)	175.7 (4.7%)	0.039	−0.005 to 0.084	0.024	0.008 to 0.111
Various types of antibiotics: No. of AB per PY(%)^d
Tetracyclins	332 (15.7%)	136 (12.9%)	195 (18.4%)	0.057	0.036 to 0.078	0.029	0.005 to 0.053
BL, penicillins	501 (23.7%)	142 (13.5%)	372 (35.1%)	0.014	−0.008 to 0.037	−0.023	−0.049 to 0.002
BL, others	21 (1.0%)	7 (0.7%)	14 (1.3%)	0.002	−0.010 to 0.013	0.004	−0.010 to 0.018
Sulfonamides/trimethoprim	140 (6.6%)	70 (6.7%)	71 (6.7%)	0.037	0.022 to 0.051	0.027	0.010 to 0.043
Macrolides^c	537 (25.4%)	458 (43.6%)	78 (7.4%)	0.345	0.176 to 0.514	0.263	0.066 to 0.60
Quinolones	195 (9.2%)	78 (7.4%)	116 (11.0%)	0.032	0.000 to 0.063	0.034	−0.002 to 0.007
Other antibacterials	373 (17.7%)	159 (15.1%)	214 (20.2%)	0.070	0.045 to 0.096	0.058	0.029 to 0.087
No. of GP contacts per PY	6.3	16.0	3.0	3.732	3.089 to 4.375	3.052	2.316 to 3.788

Pen-A penicillin allergy label, B regression coefficient per patient year, No. number, AB antibiotic prescriptions, PY patient year, BL beta-lactam, GP general practitioner, OR odds ratio, CI confidence interval
Italic values represent significant differences between Pen-A and non-Pen-A
^aAdjusted for gender, age >12 years, comorbidities (R96, R97, S87, F71, A12), GP practice, ≥4 GP contacts/year
^b2nd choice AB based on Dutch guidelines for primary care. Data obtained by matching AB prescriptions with corresponding ICPC-codes
^cLincosamides and streptogramins included
^dPercentage of total number of AB prescriptions per year