A cross-sectional study using a face to face structured interview schedule was conducted amongst purposively sampled UK-resident adult Yemeni male khat chewers, aged 18 years and above, selected during random visits to places of khat sale. A factor analysis with principal components extraction was conducted to explore the construct validity of the proposed SDS-khat. Reliability of the proposed scale was assessed using test–retest and internal reliability tests. The concurrent validity of the proposed SDS-khat was assessed in relation to individual measures and a composite index of khat chewing behaviours using univariate analyses.
Two hundred and four Yemeni male adult khat chewers were interviewed. The mean score of the proposed SDS-khat was 5.52 (SD ±4.03). Forty nine percent of respondents (95 % CI = 43–55 % ) with ≤5 scores were more likely not psychologically dependent. A single factor, uni-dimensional solution identifying the five items making up the scale accounted for 52.33 % of variance. The internal reliability was good (Cronbach's alpha coefficient = 0.76) and the test retest intraclass correlation coefficient (ICC) was 0.93 (95 % CI = 0.80–0.97). Many individual aspects of the khat chewing were significantly related to the scale. Assessment of the proposed SDS-khat's concurrent validity with individual items of khat chewing identified several significant relationships (p ≤ 0.05) whilst the composite index of khat behaviour also identified a significant relationship (OR = 14.40, 95 % CI = 6.71–30.89). The proposed SDS-khat also correlated with self-reported reasons for khat chewing (p ≤ 0.001, OR = 3.54; 95 % CI = 1.80–6.96).
In this sample of Yemeni khat chewers the SDS-khat is recommended as a valid and reliable research tool for measuring psychological dependence upon khat. Further validation in other samples is indicated.