Older people’s attitudes towards deprescribing cardiometabolic medication

BMC Geriatrics

Table 4 Comparison of the perceived appropriateness and the concerns about stopping of insulin, sulfonylurea, antihypertensive and statin

	Number of participants (n)	Mean	SD	P value	Effect size (Cohen’s D)
Appropriateness of the medication^d
Insulin^a	32	3.81	0.84	0.026*	0.51
Sulfonylurea^e	50	3.43	0.68	0.026*	0.51
Insulin^a	32	3.81	0.84	0.002**	0.59
Statins	209	3.38	0.71	0.002**	0.59
Sulfonylureas^a	64	3.56	0.75	0.20	0.21
Antihypertensive	182	3.69	0.61	0.20	0.21
Statin^b	170	3.38	0.69	< 0.001**	0.47
Antihypertensive	170	3.68	0.63	< 0.001**	0.47
	(n)	Median	IQR	P value	Effect size
Insulin^c	32	3.8	2.2–3.4	0.53	0.19
Antihypertensive	182	3.8	2.4–3.0	0.53	0.19
Sulfonylurea^c	64	3.6	2.0–3.0	0.090	0.24
Statin	209	3.4	1.8–3.0	0.090	0.24
Concerns about stopping medication
	(n)	Mean	SD	P value	Effect size
Insulin^a	31	2.61	0.83	0.88	0.040
Sulfonylurea	46	2.59	0.51	0.88	0.040
Insulin^a	32	2.63	0.82	0.86	0.034
Antihypertensive	183	2.65	0.64	0.86	0.034
Insulin ^a	32	2.63	0.82	0.86	0.041
Statin	204	2.60	0.60	0.86	0.041
Sulfonylurea ^a	58	2.59	0.60	0.52	0.097
Antihypertensive	183	2.65	0.64	0.52	0.097
Sulfonylurea^a	58	2.59	0.62	0.89	0.021
Statin	204	2.60	0.60	0.89	0.021
Statin ^b	169	2.57	0.58	0.19	0.079
Antihypertensive	169	2.63	0.65	0.19	0.079

^aUnpaired student t-test, between participants comparison
^bPaired student t-test, within participant comparison
^cWilcoxon rank-sum, between participant comparison
^dQuestions of appropriateness are scored inversely to obtain the factor score (means and median), a higher appropriateness score thus means that participant considered their medication more appropriate
^eWhen comparing insulin to sulfonylureas there were 14 participants who used both medication groups. For the comparison for these participants only the insulin answers were used to prevent a mixture of within and between participants comparison
*p-value of ≤0.05 is considered significant before correction for multiple testing
**p-value of ≤0.0083 is considered significant after correction for multiple testing

ISSN: 1471-2318