Meta-analyses of data from human studies are invaluable resources in the life sciences and the methods to conduct these are well documented. Similarly there are a number of benefits in conducting meta-analyses on data from animal studies; they can be used to inform clinical trial design, or to try and explain discrepancies between preclinical and clinical trial results. However there are inherit differences between animal and human studies and so applying the same techniques for the meta-analysis of preclinical data is not straightforward. For example preclinical studies are frequently small and there is often substantial heterogeneity between studies. This may have an impact on both the method of calculating an effect size and the method of pooling data. Here we describe a practical guide for the meta-analysis of data from animal studies including methods used to explore sources of heterogeneity.

The reliability of experimental findings depends on the rigour of experimental design. Here we show limited reporting of measures to reduce the risk of bias in a random sample of life sciences publications, significantly lower reporting of randomisation in work published in journals of high impact, and very limited reporting of measures to reduce the risk of bias in publications from leading United Kingdom institutions. Ascertainment of differences between institutions might serve both as a measure of research quality and as a tool for institutional efforts to improve research quality.

Introduction: We conducted a meta‐analysis of the conflicting epidemiologic evidence on the association between midlife body mass index (BMI) and dementia.Methods: We searched standard databases to identify prospective, population‐based studies of dementia risk by midlife underweight, overweight, and obesity. We performed random‐effects meta‐analyses and meta‐regressions of adjusted relative risk (RR) estimates and formally explored between‐study heterogeneity.Results: We included 19 studies on 589,649 participants (2040 incident dementia cases) followed up for up to 42 years. Midlife (age 35 to 65 years) obesity (BMI ≥ 30) (RR, 1.33; 95% confidence interval [CI], 1.08–1.63), but not overweight (25 < BMI < 30) (RR, 1.07; 95% CI, 0.96–1.20), was associated with dementia in late life. The association with midlife underweight (RR, 1.39; 95% CI, 1.13–1.70) was potentially driven by residual confounding (P from meta‐regression = .004), selection (P = .046), and information bias (P = .007).Discussion: Obesity in midlife increases the risk of dementia. The association between underweight and dementia remains controversial.

Pain can significantly decrease the quality of life of patients with advanced cancer. Current treatment strategies often provide inadequate analgesia and unacceptable side effects. Animal models of bone cancer pain are used in the development of novel pharmacological approaches. Here we conducted a systematic review and meta-analysis of publications describing in vivo modelling of bone cancer pain in which behavioural, general health, macroscopic, histological, biochemical, or electrophysiological outcomes were reported and compared to appropriate controls. In all, 150 publications met our inclusion criteria, describing 38 different models of bone cancer pain. Reported methodological quality was low; only 31% of publications reported blinded assessment of outcome, and 11% reported random allocation to group. No publication reported a sample size calculation. Studies that reported measures to reduce bias reported smaller differences in behavioural outcomes between tumour-bearing and control animals, and studies that presented a statement regarding a conflict of interest reported larger differences in behavioural outcomes. Larger differences in behavioural outcomes were reported in female animals, when cancer cells were injected into either the tibia or femur, and when MatLyLu prostate or Lewis Lung cancer cells were used. Mechanical-evoked pain behaviours were most commonly reported; however, the largest difference was observed in spontaneous pain behaviours. In the spinal cord astrocyte activation and increased levels of Substance P receptor internalisation, c-Fos, dynorphin, tumor necrosis factor-α and interleukin-1β have been reported in bone cancer pain models, suggesting several potential therapeutic targets. However, the translational impact of animal models on clinical pain research could be enhanced by improving methodological quality.