Integrating fragmented evidence by network meta-analysis: relative effectiveness of psychological interventions for adults with post-traumatic stress disorder.

BACKGROUND: To summarize the available evidence on the effectiveness of psychological interventions for patients with post-traumatic stress disorder (PTSD). METHOD: We searched bibliographic databases and reference lists of relevant systematic reviews and meta-analyses for randomized controlled trials that compared specific psychological interventions for adults with PTSD symptoms either head-to-head or against control interventions using non-specific intervention components, or against wait-list control. Two investigators independently extracted the data and assessed trial characteristics. RESULTS: The analyses included 4190 patients in 66 trials. An initial network meta-analysis showed large effect sizes (ESs) for all specific psychological interventions (ESs between -1.10 and -1.37) and moderate effects of psychological interventions that were used to control for non-specific intervention effects (ESs -0.58 and -0.62). ES differences between various types of specific psychological interventions were absent to small (ES differences between 0.00 and 0.27). Considerable between-trial heterogeneity occurred (τ²= 0.30). Stratified analyses revealed that trials that adhered to DSM-III/IV criteria for PTSD were associated with larger ESs. However, considerable heterogeneity remained. Heterogeneity was reduced in trials with adequate concealment of allocation and in large-sized trials. We found evidence for small-study bias. CONCLUSIONS: Our findings show that patients with a formal diagnosis of PTSD and those with subclinical PTSD symptoms benefit from different psychological interventions. We did not identify any intervention that was consistently superior to other specific psychological interventions. However, the robustness of evidence varies considerably between different psychological interventions for PTSD, with most robust evidence for cognitive behavioral and exposure therapies.

South African field strains of Haemonchus contortus resistant to the levamisole/morantel group of anthelmintics.

A strain of Haemonchus contortus from the Pietermaritzburg district of Natal was found to be resistant to levamisole (geometric mean efficacy 76.5%), morantel (41.9%), the benzimidazoles (oxfendazole: 33.7%) and rafoxanide (82.0%), but apparently fully susceptible to closantel and disophenol. In the case of ivermectin, a mean of 5.2% of the H. contortus was not removed at a dosage of 200 micrograms kg-1 live mass. A second strain of H. contortus, from Amsterdam in the south-eastern Transvaal, showed reduced susceptibility to levamisole (80.8%) and morantel (46.2%), the only 2 drugs tested. This is apparently the first report of resistance to the levamisole/morantel group of anthelmintics in sheep in South Africa.