Regarding Neil's request for comments on "how we can move towards a world where abstracts are routinely available in the language of the country where the research was undertaken", I think journals should put it as a prerequisite for publishing. I do not know to what extent authors can demand the publication of translated abstracts to certain journals (mainly in paper format). Nevertheless, both authors and journal editors should be aware of the importance of this issue. Therefore, actions to create that awareness should target both sides.
Background: Community health workers (CHWs) are increasingly being used in low-income countries to address human resources shortages, yet there remain few effective, evidence-based strategies for addressing the enduring programmatic constraints of worker motivation, retention and performance. This paper describes how two interventions were designed by the Innovations at Scale for Community Access and Lasting Effects (inSCALE) project to address these constraints in Uganda and Mozambique drawing on behavioural theory and formative research results.
Re Justus Krabshuis' comment below that we should name the homeopathic treatments which do not work: it would be a far shorter list and more scientifically justified to state which homeopathic remedies HAVE been shown to work. A list of those which 'do not work' (by which, I suppose he means 'have been studied in randomized controlled trials and found no treatment effect' would also be a good idea.
I am an American-trained family doctor, currently teaching in a medical school in Mozambique. I do not believe that using mid-level providers in any way corresponds to lesser or inferior care. The Chinese use surgical technicians extensively, even to reattach severed limbs. In the US, mid-level providers (nurse practitioners and physicians’ assistants) are used extensively in primary care.