Research involving human subjects must be performed with standards that promote protection of their rights. Several codes were developed and all are unanimous in the following ethic principles: respect for persons, beneficence and justice. The fulfilment of these principles will assure that the dignity, rights, safety and well-being of the participants in a biomedical research are guaranteed.

The need of a special protection to those with diminished autonomy is a common requirement to all codes. Pregnant women are defined as a vulnerable population because of the potential risk of harm of the foetus. Apart from the foetal risk, pregnant women in developing countries have additional potential for vulnerability, as in most of the cases they are economically or/and educationally disadvantaged individuals. Their generally low level of education may put them in a difficult situation to fully understand consent forms and the risk-benefit assessment provided by the researchers.

In recent years, the evidence-based medicine has been an important international goal. Mechanisms to support the use of research-based evidence in developing health policy are being promoted. The research must be sensitive to the potential for vulnerability when designing evidence-based trials and they have to assure that the vulnerable population is being protected.

With the intention to protect the vulnerable population, some groups such as pregnant women have being traditionally excluded from clinical trials. As a result of this exclusion, pregnant women are being exposed to medicines in situations where no evidence-based information is available on efficacy and safety. They are deprived of the benefits of treatment in order to be protected themselves and their offspring from an unknown risk. Although is very clear that protection of a vulnerable population, particularly the pregnant women, is mandatory in the design of any trial, the challenges of the implementation of this principle should be taken into account in order to not increase the gap between the introduction of a pharmaceutical product in the market and the availability of safety information for its use in pregnancy.

In developing countries maternal mortality is an important public health problem. Most maternal deaths occur in the poorest countries particularly in Sub-Saharan Africa. In Mozambique the rate could be between 408-1000 per 100 000 live births according to the source of information. Several studies have addressed different causes of maternal mortality in Africa, eclampsia and malaria being part of the five most reported.

There is strong evidence of the effectiveness of magnesium sulphate (MGSO4) in women with pre-eclampsia and eclampsia but the drug is not available in some countries. The example of MgSO4 has being used to describe failure in translating results of research into policy and/or practice.

For malaria treatment few drugs were considered effective and safe for use during pregnancy including chloroquine, sulphadoxine-pyrimethamine and quinine. The information on safety comes from a long experience of use. The emergency of resistance of P. falciparum to these drugs is limiting its use in most of endemic areas. The artemisinin derivatives are being strongly recommended to be used in combination with other antimalarial drugs. Preclinical studies have consistently shown that artemisinin and its derivatives are embryolethal and teratogenic in animals. Current available information is not adequate to extrapolate the results in animals to humans.

Limited data on the safety profile of antimalarial drugs during pregnancy is a challenge. Nowadays these drugs are used in pregnant women based on retrospective cumulative risk-benefit assessment. Mechanisms of prospectively monitoring the drugs use are required to protect pregnant women from the potential risk.

The aim of this thesis is to describe the (un)availability of drugs and their reasons, as well as, the (un)availability of safety information on drugs needed to be used during pregnancy in Southern Africa, and to propose mechanisms to effectively monitor drug safety in pregnancy in developing countries.

In order to achieve this objective four studies were performed. In the first study a qualitative case-study based on interviews and bibliographic review was performed in Mozambique and Zimbabwe. Factors affecting the (un)availability of MgSO4 were assessed. This study showed that research evidence regarding the effectiveness of MgSO4 for the treatment of eclampsia and pre-eclampsia, was widely known in the study countries. However, the registration, approval, acquisition and distribution of MgSO4, and hence its availability to clinicians, was affected by market and systems failures. With this study we concluded that the low cost of magnesium sulphate, as well as the mechanisms through which it is procured, means that market forces alone cannot be relied upon to ensure its availability. Governments and international organizations must be prepared to intervene to ensure the wide availability of low cost, effective drugs critical to improving public health in Africa.

The second and the third studies shown that antimalarial drugs are being used in pregnant women even without information on its safety profile in this particular group. Robust safety monitoring systems are clearly needed in developing countries to accompany the deployment of new drugs, especially those with a potential teratogenic risk.

The fourth study showed that spontaneous reporting system may be a tool for drug safety monitoring. This system could be used to increase health care providers' and patients' awareness of possible ADRs, and to develop a culture to report these reactions. Spontaneous reporting and pregnancy registries were presented as examples of mechanisms that could and should be in place.

Recommendations on how these systems could be effectively implemented in resource constrained countries were also presented.

All these studies suggest that the definition of vulnerability of pregnant women in developing countries should not be restricted to the potential risk of harm of the foetus (harm-based definition) or to the difficulty of fully understanding consent forms (consent-based definitions). Women are also vulnerable because of the high risk of dying from any cause related to or aggravated by the pregnancy. The implementation of these ethic principles need to take into account the urgent need to implement effective and safe drugs targeted to reduce the burden of maternal morbidity and mortality.