The bacterial nature of infected foot lesions presenting to a diabetic unit during the course of twelve months was studied. Fifty-six swabs were obtained from 25 patients. Swabs were processed within ten minutes of being collected. An average of 3.3 organisms per swab were isolated, Staphylococcus aureus was the most common bacterial species isolated, while anaerobic bacteria comprised 10% of the isolates. Sensitivity testing of the isolates against commonly used antibiotics showed that the antimicrobial regimen that we have previously used empirically on such patients may not always be optimal.

This article argues that processes of disconnect and capture have affected Nepal's efforts to decentralise its education system, leading to a failure to engage the very stakeholders Á parents and communities Á that the reforms sought to reach. Specifically, disconnect occurred in the development and implementation of the latest 'decentralisation' reforms because they were formulated via a highly centralised policy-making process and then implemented from the top down. As a result, they had little impact on community-level school leadership or ownership, and led to a continued exclusion of the very community-level stakeholders that the reforms sought to engage. Furthermore, the policy reforms have also led to an increased central legitimisation and empowerment of school management committees, which in turn has facilitated the capture and politicisation of these bodies by locally established leaders who are often not motivated to engage parents and community members in school reform. While these processes are not necessarily new phenomena to Nepal, they illustrate the extent to which globalised policy agendas have been ineffective in engaging community-level stakeholders. IntroductionDecentralisation of public education management, from central to local, community or even school levels, is often viewed as a panacea to the problems faced by struggling education systems. It is commonly argued that localised school management results in increased efficiency, greater accountability and equity, and more democratic decisionmaking processes; all outcomes that also conveniently align with prevalent discourses in the international development sector. For these reasons, educational decentralising measures and the notion of 'school-based management' have been widely implemented in recent decades in diverse contexts across the globe. But what happens when policy-makers attempt to operationalise these powerful global discourses at the most local of levels? As with many such reforms, results may be quite different from those that are planned. This article explores the ways that globalised discourses around education decentralisation are mediated by contact with local societies, in particular the actual sites of policy implementation: schools and communities. How do endogenous social orders interact with and transform the discourses typically associated with modernity and development? And how do these dynamics shape the relationship between decentralisation policy and practice? I propose a two-part

During a 32-month period 94 foot ulcers in 54 diabetic patients aged 38-90 years (mean 64 years) were managed in a specialist foot clinic. Fifty-six percent were men, and they were significantly younger than women; 46% were taking insulin. Mean duration of diabetes was 13.4 years. Comparison with controls revealed a higher prevalence (p less than 0.01) of retinopathy (60% vs 23%), neuropathy (89% vs 31%), vasculopathy (71% vs 34%), arterial calcification (31% vs 20%) and previous lesions (54% vs 4%). There was no difference in quality of diabetic control, or smoking habit. A simple classification of lesions was used. All types yielded mixed cultures of microorganisms (average 2.1 per swab); the flora obtained was affected by systemic antibiotics. Abnormal pressure was judged to have contributed to all lesions occurring in areas of callus. In addition definable trauma precipitated the event in up to 60% of all other types. Lesions in areas of callus were more likely to have healed by the end of the study period, but average time to healing was significantly longer than other lesions. Despite intensive outpatient support, 33 patients spent a total of 1188 days in hospital during the 974 day period, an average of 36 days per patient and 1.2 beds per day. Further research is urgently required to define optimal methods of prevention and treatment of diabetic foot ulcers.

No abstract