Handgrip strength and balance in older adults following withdrawal from long-term use of temazepam, zopiclone or zolpidem as hypnotics

Table 4 **Balance* by success of short-term (“ONE MONTH”) and long-term (“SIX MONTHS”) withdrawal**

ONE MONTH (short-term withdrawal)	Measurement point
	Baseline	1 week	2 weeks	3 weeks	1 month	2 months	6 months
	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	P ¹	P ²	P ³
short-term withdrawers under 33 points ^†	27 (40)	19 (28)	17 (25)	14 (21)	13 (19)	9 (13)	9 (13)	0.474	0.997	<0.001
short-term non-withdrawers under 33 points ^†	7 (35)	5 (26)	4 (22)	2 (11)	5 (26)	4 (24)	1 (6)	0.474	0.997	<0.001
*SIX MONTHS (long-term withdrawal)*	*Measurement point*
	*Baseline*	*1 week*	*2 weeks*	*3 weeks*	*1 month*	*2 months*	*6 months*
	*n (%)*	*n (%)*	*n (%)*	*n (%)*	*n (%)*	*n (%)*	*n (%)*	P ¹	P ²	P ³
long-term withdrawers under 33 points ^†	10 (30)	8 (25)	4 (12)	4 (12)	3 (9)	3 (9)	2 (6)	0.054	0.165	<0.001
long-term non-withdrawers under 33 points ^†	24 (44)	16 (30)	17 (32)	12 (23)	15 (27)	10 (20)	8 (15)	0.054	0.165	<0.001

*Using Short Berg’s Balance Scale (BBS-9).
Note: Number of withdrawers and non-withdrawers under 33 points (and their percentage from all participants) at different time points are given.
P ¹ = Statistical significance for group × time interaction effect; logistic regression analysis using GEE estimation; adjusted for gender.
P ² = Statistical significance for group effect; logistic regression analysis using GEE estimation; adjusted for gender and time.
P ³ = Statistical significance for time effect; logistic regression analysis using GEE estimation. After adjustment for gender and group at one month’s time point balance was better at 1 week (P = 0.003), 2 weeks (P = 0.003), at 3 weeks (P < 0.001), at 1 month (P < 0.001), at 2 months (P < 0.001), and at 6 months (P < 0.001) compared to baseline. After adjustment for gender and group at six months’ time point balance was better at 1 week (P = 0.003), at 2 weeks (P = 0.003), at 3 weeks (P < 0.001), at 1 month (P < 0.001), at 2 months (P < 0.001), and at 6 months (P < 0.001) compared to baseline.
^†Cut off -point (<33) suggests increased fall risk [24].
Men and women combined (n = 89).

ISSN: 1471-2318