Medical research and clinical trials are necessary to allow the development of medicines and improve healthcare. Without doubt, ethical principles must be applied in all clinical trials. To guide trials involving human subjects in accordance with ethical principles, a number of guidelines (eg, the Declaration of Helsinki and those from Council for International Organisations of Medical Sciences (CIOMS)) and documents (eg, reports from the Nuffield Council on Ethics) have been published. The main theme of such guidelines and documents is always how to safeguard participants during the study period. However, in many circumstances, the diseases and conditions suffered by trial subjects, especially chronic illnesses, persist after research is over. Therefore, concerns about what should happen after research is over have been raised and debated. An important issue is whether or not experimental interventions (eg, administration of medicines, vaccines, procedures and care) that have been proven to be beneficial during the trial period should continue to be provided for trial participants after the study is concluded.

In developed countries, effective experimental interventions are usually integrated into the routine health system.1 If not, it is a general rule that beneficial new medicines will be freely supplied to all participants after trials are completed,2 except very expensive ones such as those that might be used for preventing or treating cancers (B. Greenwood, personal communication).

However, the situation in developing countries is quite different from that of the industrialised world. Firstly, patients in the developing world may have only limited or even no access to healthcare.1 3 Therefore, it generally seems unrealistic for participants from developing countries to enjoy the benefits of newly-developed interventions through the regular healthcare system. Secondly, participants from underdeveloped nations might lack adequate understanding of the research,1 partially due to the high rate of illiteracy and lack of common scientific knowledge. Inadequate understanding of the trials might be taken advantage for the purpose of avoiding post-trial provision. Thirdly, participants in poorer countries generally do not have a voice in negotiations about post-trial issues.

Nonetheless, with more and more studies sponsored by institutions, agencies and organisations from western countries being conducted in the developing world, post-trial provision has become a crucial issue. Fortunately, there is an increasing consensus that, in principle, participants in developing countries should continue to receive benefits originating from the studies in which they enlisted beyond the research period. The consensus is mainly based on basic ethical principles of beneficence and justice as reciprocity as well as concerns about exploitation and fair benefits. According to the Belmont report,4 two general rules of the beneficence principle should be followed in human research: (1) “do not harm”; (2) “maximise possible benefits and minimise possible harms”. Once the investigators withdraw beneficial treatments or prophylactic measures after the study is over, participants who still require these interventions will probably suffer deterioration in their health and experience feelings of frustration and helplessness. In my opinion, such withdrawal is not in accordance with the “do not harm” rule.3 5 Instead, post-trial supply of beneficial interventions is an important assurance to maximise possible benefits and minimise potential harm for trial participants. Furthermore, owing to the dedication of participants, researchers can acquire useful knowledge and develop new interventions to benefit the society as a whole.5 Thus, based on the principle of justice as reciprocity, since participants have contributed to others and society, they should be rewarded. It is obvious that their contribution does not end with the completion of their involvement in trials. Therefore, those who participate in a trial should have the opportunity to benefit from findings of the trial.6 In addition, due to limited resources, participation in clinical trials is an important way for patients in developing countries to access higher-quality healthcare.1 6 7 The interruption of necessary healthcare and interventions simply due to the completion of trials appears unfair and leads to the suspicion of exploitation.4 7 Some international trials aim to help people in poorer countries or are not driven by financial motives. However, most are conducted to develop medicines, vaccines, and procedures predominantly for use in developed countries and are for-profit, while the subjects are recruited from developing countries and may be exposed to known or unknown risks.1 There should be no double standards for ethical principles.6 Therefore, participants from developing countries should receive the same post-trial benefits as those in the industrialised world.

Despite the presence of a widespread consensus that trial subjects in developing countries should, in principle, be assured of post-trial provision, how far obligations of beneficence and reciprocity should be extended to trial participants from developing countries remains contentious. In other words, there is no agreement on whether or not post-trial provision should be mandatory in practice. The inconsistent and even contradictory guidance and recommendations from the bioethical guidelines and documents on trials involving human subjects make the situation more complicated.

GUIDELINES RELATING TO POST-TRIAL PROVISION

Guidelines supporting mandatory post-trial provision

Some international or national guidelines support mandatory post-trial provision of beneficial drugs,2 8 9 vaccines,10 proven prophylactic and therapeutic methods,11 or products12 13 for trials participants (table 1 and supplementary file). Among these guidelines, two clearly state that post-trial provision should be free of charge2 12 and two definitely require that the provision should be available for all participants.2 10

Guidelines and documents not supporting mandatory post-trial provision

Several guidelines or reports on bioethics in research point out that mandatory post-trial provision is unrealistic and impractical (table 2 and supplementary file). Instead of mandatory provision, these reports and guidelines require investigators and sponsors to make every14 or reasonable, good faith efforts7 15 16 to ensure that trial participants have access to proven interventions beyond the research period. They also recommended that any possible post-trial arrangements should be described in research proposal7 17 15 and should be discussed in advance with relevant parties usually including sponsors, researchers, national governments and health authorities, and representatives of the communities from which subjects were recruited7 17 15 16 and if the researchers fail to do so, they should justify the case to the ethics review committee (ERC) or institutional review board (IRB).

Towards a rational response

It is confusing for research sponsors, investigators and members of ERCs or IRBs to deal with post-trial provision, since no consensus is present in the guidelines and documents to which they refer. In addition, beyond the guidelines, many doubts about the ethical basis and practical feasibility of mandatory post-trial provision in the developing world have been raised. Therefore, the ethical basis of mandatory post-trial provision needs to be reviewed and practical difficulties associated with implementation should be discussed in order to make a rational response.

Post-trial provision is unnecessary in some situations

It is clear that not all studies can yield beneficial interventions and not all participants need post-trial provision. Firstly, phase I clinical trials, epidemiological studies, and research focusing on basic scientific knowledge do no immediately result in beneficial treatments or translate into development of new medicines and vaccines.17 Secondly, as the NCOB 2002 report7 argued, in certain circumstances, even if new treatments (eg, antimicrobial agents) are highly efficacious and safe, they may be reserved by policy-makers while current therapies are still effective in order to put off the development of resistance to new treatments. Thirdly, not all trial participants will have a positive response to the experimental intervention and so would not benefit from post-trial provision. Fourthly, patients with acute or sub-acute diseases usually do not need post-trial treatment9 and some participants may not be at the appropriate age or stage of diseases to benefit from the intervention once the study is over. Finally, sometimes, trial participants are not the eventual target group of research. For instance, trials testing new medicines against severe fatal malaria recruited patients with moderately severe disease.17 Therefore, post-trial provision of the beneficial experimental intervention beyond the research period seems unnecessary in all situations. Correspondingly, a clause such as “if need be” should be added to guidelines that require mandatory post-trial provision, as in the Indian guideline.8

To make “if need be” clear, a set of criteria of access to post-trial provision should be established. The following criteria are proposed in the present paper to assure those who still need and can continue to benefit from experimental interventions but have no alternative access of post-trial provision.

1. The trial directly leads to the development of new medicines, vaccines, surgical procedures, or commercial products, establishment of new diagnostic methods, or improvement of healthcare.

2. Only those who make a positive response to the intervention during the trial period should be considered.

3. After the research is over, discontinuation of the intervention may cause harm or losses for trial subjects.

4. The intervention is better than or at least equal to those that are available through local regular health system in validity and safety.

5. There are no alternative ways to access to the beneficial experimental intervention beyond the study period.

6. Post-trial access is approved by ERCs or IRBs including local ones.

7. Participants agree to receive the intervention after they fully understand potential long-term harm and benefits of it. For this purpose, an additional informed consent is necessary.

Post-trial provision should be mandatory, as long as participants can benefit from the experimental intervention and have no alternative access

Based on ethical principles, once studies do yield beneficial interventions and trial participants can continue to benefit from them but have no alternative access such as through the regular health system, post-trial provision for participants from developing countries should be mandatory. Generally, according to the Belmont Report, three universal ethical principles should be followed in research involving human subjects—respect for persons, beneficence and justice.4 As described above (see introductory paragraph), in my opinion, withdrawal of beneficial interventions from participants who still need them but have no alternative access does not accord with the “do not harm” rule of the beneficence principle. Conversely, the continuous provision of beneficial medicines, vaccines, and procedures is consistent with the “maximise possible benefits and minimise possible harms” rule. As for the principle of justice, the Belmont Report4 clearly stated that “justice demands both that these not provide advantages only to those who can afford them and that such research should not unduly involve persons from groups unlikely to be among the beneficiaries of subsequent applications of the research.” Therefore, it is clear that if participants from developing countries won’t benefit from the findings, they should not be recruited for the research.18

Nonetheless, some arguments on the ethical basis, the boundary of research and benefits from research against mandatory post-trial provision, if need be, have been presented. Firstly, the respect for a person’s autonomy is also a basic ethical principle.4 Therefore, some commentators may argue that an individual has the right to make decisions about his or her health and welfare and participants’ autonomy has been protected by the informed consent procedure.19 Therefore, in their opinion, if no post-trial provision is assured in the informed consent and participants decide to take part in a trial after they signed the informed consent form, no one can interfere with post-trial issues. However, the informed consent procedure for trial participants from developing countries is significantly undermined by their inadequate understanding of the implications of what they are consenting to and the inability of many investigators to explain the trial effectively. Therefore, the informed consent procedure should not be taken as an excuse to deprive participants from developing countries of the right to access post-trial interventions.20

Furthermore, it is argued that a clear distinction between research and healthcare should be established.19 According to this viewpoint, access to interventions beyond the research period is not research itself but healthcare and the application of research funds to cover healthcare seems misguided.19 However, the post-trial provision of beneficial interventions is actually an extension of research and is not simply equal to healthcare provided by the regular health system, because health status of the participant may be altered by trials and post-trial provision is helpful to observe long-term efficacy and side-effects of the experimental intervention.

Finally, some commentators may argue that trial subjects have already received many benefits from the studies in which they enrolled even without post-trial provision.1 7 17 In their opinion, since these benefits (see below) might be more valuable, important, and meaningful for trial participants and their communities than post-trial provision, the trials are already ethically sound and so mandatory post-trial provision is not always necessary.1 7 17 These benefits usually include improving the quality and infrastructure of healthcare, developing the skills and expertise of local scientists and health professionals, increasing job opportunities, promoting economic development and promulgating health knowledge.1 7 17 In addition, some commentators point out that provision of short-term interventions during the research period is better than nothing for those in poor nations who have no access to necessary healthcare.21 However, it is quite clear that basic ethical principles such as those described by the Belmont Report should be followed in all trials involving human subjects no matter where they are carried out. In other words, no double ethical standards are acceptable for trials recruiting subjects from developing countries6 and provision of interventions during the research period and other benefits resulting from the trials do not rule out the necessity of post-trial provision.

Besides the compliance with ethical principles, another extremely important issue is implementation. It is true that implementation of post-trial provision could be very complex and may encounter many difficulties. Some guidelines and documents as described above used the difficulties associated with implementation to justify their non-support of mandatory post-trial provision. However, it is also clear that such difficulties are not unmanageable.

An important concern about implementation is that mandatory post-trial provision may result in a heavy financial burden for providers. Indeed, trial sponsors and some analysts considered compulsory post-trial supply as a major disincentive for new drug development3 7 22 23 and argued that provision beyond the research period, especially in the long term, might divert limited funding from valuable studies4 7 and would subsequently jeopardise crucial research in resource-limited settings.3 24 The recent suspension of a planned trial of a new antiretroviral drug in Cambodia when the community of participants required a 30-year provision of post-trial treatment added powerful evidence for this consideration.3 25 However, since only a limited number of participants are recruited for a single trial, it seems relatively easy for participants to receive post-trial interventions.26 For for-profit studies, the cost of post-trial provision can be compensated by profits from marketing. On the other hand, for not-for-profit research, there still are ways to tackle financial difficulties. For instance, one report from Thailand demonstrated that it was possible to assure all participants of the access to post-trial interventions without introduction of a heavy financial burden to sponsors.24 In this example, trial sponsors created a fund from revenue derived from the research (overhead costs) and modest profits from training activities.24 The programme obtained some drugs at low prices by bulk purchasing and the fund provided different levels of financial support based on the financial status of the participant. As a result, all participants were able to access post-trial treatment at affordable rates. To further deal with the potential financial burden resulting from post-trial provision, special foundations or grant schemes funded by the governments of industrialised countries and donations from the pharmaceutical industry should be established. It would then be possible for sponsors of not-for-profit studies and small companies to obtain funds to support post-trial provision.

Besides financial concerns, many practical difficulties have also been mentioned with regard to implementation, mainly involving who should take responsibility to provide, deliver, and monitor post-trial treatment (sponsors, investigators, or the local health system?) and how long should the interventions be provided after research is over. Although it seems difficult to come to a consensus, these issues are not unsolvable. To establish a comprehensive partnership among sponsors, investigators, local healthcare authorities and providers, patient advocacy groups, ERCs or IRBs, and, of course, participants appears to be a good way to tackle the problems. In the model of the comprehensive partnership, the ERC or IRB approves the post-trial arrangements, the sponsor organises post-trial provision, the local healthcare system delivers and monitors it with the help of investigators, the patient advocacy group provides consultation, and participants are actively involved in post-trial interventions by complying with the post-trial protocols, attending regular follow-ups, reporting possible adverse events and so on. As King21 pointed out, the partnership model can be established and concerns raised by different sides can be overcome with careful planning. In my opinion, as for the duration of post-trial provision, it should be continued as long as the participant still needs it but have no access to it or equivalent alternatives from the regular health system.

Conclusions

It is increasingly believed that trial participants should continue to benefit from the results of studies after the research is over. However, current guidelines provide different recommendations about post-trial provision. It is clear that mandatory post-trial provision is not necessary in certain situations. Therefore, guidelines that support mandatory post-trial supply should add “if need be”. Nonetheless, in my opinion, owing to basic ethical principles (beneficence and justice), as long as participants from the developing world need and are still able to benefit from experimental interventions but have no alternative access, post-trial provision should be mandatory. The informed consent procedure for participants from developing countries may be incomplete and should not be taken as an excuse to escape post-trial responsibility. Implementation of mandatory post-trial provision usually encounters many difficulties, but they may be tackled by careful advanced planning and establishing a partnership among various relevant parties, usually including research sponsors, investigators, local healthcare authorities and providers, ERCs or IRBs, patient advocacy groups, and participants.