Linkage of primary care prescribing records and pharmacy dispensing Records in the Salford Lung Study: application in asthma

Table 1 Cox Proportional hazards model risk factors associated with time to collecting a prescribed medication

	Hazard Ratio (95% Confidence Interval)	Statistical significance (p-value)
Season
Spring	{reference}
Summer	0.967 (0.944–0.991)	0.008 ^*
Autumn	0.981 (0.958–1.005)	0.123
Winter	1.003 (0.979–1.028)	0.791
Drug Class
SABA	1.433 (1.387–1.479)	< 0.001 ^*
LABA	0.938 (0.890–0.990)	0.019 ^*
ICS	{reference}
ICS + LABA	1.067 (1.033–1.102)	< 0.001 ^*
Cromoglicate	0.778 (0.389–1.558)	0.479
Immuno-suppressants	1.244 (1.100–1.408)	< 0.001 ^*
LAMA	1.349 (1.161–1.567)	< 0.001 ^*
LTRA	1.350 (1.289–1.414)	< 0.001 ^*
Theophylline	1.040 (0.897–1.205)	0.604
Oral steroids	1.839 (1.743–1.940)	< 0.001 ^*
Previously unclaimed medications
Low tertile	{reference}
Mid tertile	0.565 (0.553–0.577)	< 0.001 ^*
High tertile	0.198 (0.193–0.204)	< 0.001 ^*
Quantity of doses prescribed	1.000 ** (1.000–1.000)	< 0.001 ^*

Statistically significant variables (using a threshold of p = 0.05) are denoted by a star (^*)
^** Coefficient 0.9999 to four decimal places, and therefore lower than the reference value

ISSN: 1471-2288