Skip to main content
. 2015 May 27;10(5):e0125577. doi: 10.1371/journal.pone.0125577

Table 5. Univariate and multivariate analyses of associations of substandard, degraded and falsified samples with key risk factors (significant associations presented); N = 2824.

Variable Total samples Poor quality samples Crude odds ratios (95% CI) Adjusted odds ratios (95% CI)
The following are significant risk factors for poor quality (substandard, degraded and falsified) ACAs b
Generic type AM 1701 80 (4.7%) 1 1
DHA 501 72 (14.4%) 3.4 (2.4,4.9) 2.4 (1.6,3.4)
AS 622 43 (6.9%) 1.5 (1.0,2.2) 1.4 (0.9,2.2)
WHO prequalified/ QAACT not prequalified 2047 190 (9.3%) 1 1
Prequalified 777 5 (0.6%) 0.06 (0.02,0.15) 0.08 (0.02,0.3)
AMFm non AMFm drugs 2072 191 (9.3%) 1 1
AMFm drugs 752 4 (0.5%) 0.05 (0.02,0.13) 0.24 (0.1,0.8)
Region of stated country of manufacture Asia 1940 159 (8.2%) 1 1
Africa 546 30 (5.5%) 0.7 (0.4,1.0) 2.1 (1.3,3.2)
Europe 141 1 (0.7%) 0.08 (0.01,0.6) 0.04 (0.06,0.4)
North America/unknown 197 5 (2.6%) 0.3 (0.1,0.7) 12.5 (2.7,56.9)
Expired at time of analysis not expired 2537 132 (5.2%) 1 1
expired a 275 59 (21.5%) 5.0 (3.6,6.9) 6.4 (4.4,9.3)
The following are significant risk factors specifically for falsified ACAs c
Outlet type Pharmacies 1438 5 (0.4%) 1 1
PMVs 1332 24 (1.8%) 4.2 (2.0,12.0) 3.9 (1.5,10.1)
public health facilities 54 0 1 1
Generic type AM 1701 8 (0.5%) 1 1
DHA 501 18 (3.6%) 7.8 (3.4,18.3) 5.9 (1.9,18.1)
AS 622 3 (0.5%) 1.0 (0.3,4.1) 0.9 (0.2,3.5)
Region of stated country of manufacture Asia 1940 7 (0.4%) 1 1
Africa 546 17 (3.1%) 8.9 (3.5,22.4) 5.0 (1.9,13.2)
Europe 141 0 1 1
North America/unknown 197 5 (2.5%) 7.2 (2.4,21.1) 27.9 (5.2,149.4)

ACAs = artemisinin-containing antimalarials; AM = artemether; AMFm = Affordable Medicines Facility—malaria; AS = artesunate;

CI = confidence interval; DHA = dihydroartemisinin; LR = likelihood-ratio; PMVs = patent medicine vendors; QAACT = quality-assured

artemisinin combination therapy; WHO = World Health Organisation.

aof these 59 expired samples 58% were suspensions that were substandard, while only 8% were tablets that were degraded.

badjusted for generic type, WHO prequalification, AMFm status, manufacture region, expired at time of analysis.

cadjusted for outlet type, provider health related qualification, generic type, AMFm status, manufacture region.