ABSTRACT
Yarkoni correctly recognizes that one reason for psychology's generalizability crisis is the failure to account for variance within experiments. We argue that this problem, and the generalizability crisis broadly, is a necessary consequence of the stimulus-response paradigm widely used in psychology research. We point to another methodology, perturbation experiments, as a remedy that is not vulnerable to the same problems.
ABSTRACT
This study assessed the cost-effectiveness of community-based management of acute malnutrition (CMAM) to prevent deaths due to severe acute malnutrition among children under-five. The analysis used a decision tree model to compare the costs and effects of two options to treat severe acute malnutrition: existing health services with CMAM vs existing health services without CMAM. The model used outcome and cost data from a CMAM programme in Dowa district, Malawi and a set of key assumptions regarding treatment-seeking behaviour and mortality outcomes. Under our 'base case' scenario, we found that CMAM cost US$42 per disability-adjusted life year (DALY) averted (2007 US$) and US$493 per DALY averted under an assumed 'worst case' scenario for each variable. The results suggest that CMAM was highly cost-effective in the 'base case' as defined by the World Health Organization, as the cost per DALY falls well below Malawi's 2007 gross national income (GNI) per capita of US$250, and is within the range of DALYs reported for other child health interventions. Under a hypothetical 'worst case' for all variables, the model indicates CMAM is still cost-effective. The results indicate the decision to scale-up CMAM within essential health services in Dowa was a cost-effective one and that scaling up CMAM in similar contexts is also likely to be cost-effective. However, several contextual and programmatic factors should be considered when generalizing to diverse contexts.
