Our study is the first study we are aware of to try to identify barriers to the availability and use of MgSO4 using a health system approach. We found that the current situation in Zambia shows that research knowledge regarding magnesium sulphate for the treatment of severe eclampsia and eclampsia has been translated into policy. However, implementation of policy into practice was found to be limited at the lower levels of care. Numerous barriers to the availability and use of MgSO4 were identified within the health system, including a lack of stock. There was a lack of dissemination of national standard treatment guidelines in health facilities and pharmacies and a general lack of in-service training for both midwives and pharmacists regarding the correct use of MgSO4. The demand for MgSO4 at the health centre level was reported to be low. However, as this assessment did not include an audit of clinical data it is not clear if this was because there were in fact few cases or that cases were being misdiagnosed. There can be no demand if there is failure to diagnose the condition. It should be noted that the health centres visited during the study period did have the necessary equipment and medical supplies to diagnose pre-eclampsia and eclampsia. At the health centre level, due to the low demand from the midwives, MgSO4 was not considered to be a priority medicine by the pharmacists in charge of procurement. As a result they were not putting pressure on the Central Medical Store to make it available.
While pre-eclampsia and eclampsia are clearly important causes of maternal mortality globally, the absolute frequency of cases presenting to an individual facility may be low. Although we did not assess clinical records, the anecdotes from staff suggested that cases of pre-eclampsia were infrequent at small hospitals. If this is correct, then lack of recent knowledge and experience would be another potential barrier to effective administration of MgSO4 and would highlight the need for frequent refresher courses and other educational reminders, to ensure appropriate diagnosis and treatment.
Poor availability of magnesium sulphate may also reflect limitations of procurement systems. Magnesium sulphate was not supplied to the lower levels of care because it was out of stock at the Central Medical Store. The demand for MgSO4 at the health centre level was apparently low, but it is not clear if the problem of availability was due to lack of demand leading to a lack of supply or vice versa.
During the development of the rapid assessment tool it was suggested that the lack of licensing of MgSO4 injection was a key potential barrier to the availability and use of MgSO4 and this may have contributed to it not being available at the district facilities. The University Hospital Pharmacy, following demand from obstetricians, was able to procure locally manufactured magnesium sulphate, but the lack of registration clearly raises uncertainty about the quality of the product. As an injectable medicine, the manufacture of magnesium suphate needs to be closely monitored to ensure that the ampoules are not contaminated during the manufacturing process.
In the walk through exercise at health facilities, apart from the quantity of MgSO4 required for the treatment of one patient for 24 hours, only the presence or absence of the other essential supplies and equipment required for the diagnosis, administration and monitoring of MgSO4 was evaluated, and not the quality or quantity of them. Clearly, the total stock and quality of supplies and equipment could have a bearing on the actual delivery of care and rational use of MgSO4 for the treatment of severe pre-eclampsia and eclampsia.
The main limitation of this case study is that it involved a small number of facilities in one district in Zambia, in the capital city and thus the findings may not reflect the current situation in other districts/provinces of Zambia. However, it is likely that MgSO4 availability in the capital city and surrounding district represents the most optimistic picture of the supply and use situation within Zambia. Follow-up studies should include public and private sector health facilities and pharmacies in other districts, including rural areas. Other limitations of our study include the lack of financial data regarding the cost of MgSO4 and its affordability, and the lack of information regarding the availability of MgSO4 injection in the private sector.
Ideally, qualitative methods could have been used to assess the knowledge and experience of staff at health facilities in relation to use of magnesium sulphate. However, our intention was to develop a rapid assessment instrument that might be applied in a variety of settings. Therefore our data sources needed to be publicly available archival materials and observations. Future assessments could be expanded to incorporate a qualitative method to uncover potential barriers to the use of MgSO4 injection by doctors and midwives in clinical practice. The role of community members, community health workers and traditional birth attendants in recognizing risk factors for pre-eclampsia and development of eclampsia was beyond the scope of this study. However, it should be noted that educational interventions to increase awareness of the risk factors for the development of severe pre-eclampsia/eclampsia at this level may have an impact on the number of referrals to health facilities and consequently on demand for MgSO4.
Since undertaking our study we have become aware of other frameworks that have been used to analyse access to health technologies in resource poor settings [20
]. There is still much to learn in this complex area and awareness of different approaches will enable the ultimate assessment instrument to be developed. Although our instrument does have limitations, it gathers useful information on barriers in an efficient way without disruption of health care provision.
The successful translation of research evidence into clinical practice is a complex process. Many studies [9
] have concentrated on mechanisms for translating research evidence into policy. What is suggested by this assessment is that although the policy makers - as reflected in essential medicines list and treatment guidelines - have been persuaded by the evidence, the gap in this case is the translation from policy into action plans. To close this gap requires an integrated approach throughout the pharmaceutical sector and pathways for clinical care that simultaneously responds to all of the factors identified in the fishbone diagram. Such a strategy would include acting on the recommendations listed in Table . The challenge is developing methods to implement such a strategy successfully in fragile health systems and resource limited settings. In order to determine if our recommendations are sufficient and appropriate for turning policy into practice further operational research is required. Our recommendations need to be implemented and evaluated using a rigorous methodology.
Recommendations for overcoming the barriers identified