In most industrialized countries, voluntary HIV counseling and testing services are a major component of HIV and AIDS control programs targeted toward pregnant women, although such programs are only now being advocated in developing countries [4,5]. Voluntary counseling and testing involves two basic components–an educational component on HIV/AIDS and an informed consent component related to an individual's choice to be HIV tested or not. The primary motivations for conducting voluntary HIV counseling and testing in lower risk populations such as pregnant women are that i) immediate knowledge of the woman's status can help reduce risk of HIV transmission to the unborn child; ii) if not infected, education at this point can help both the husband and wife to reduce risk behaviors to prevent transmission; and iii) testing of the partner can be encouraged [6]. Because voluntary HIV counseling and testing is the principal entry point for both prevention and care, it is critical that the content of the counseling is well understood and that procedures to ensure true informed consent are in place [7,8].

The basic ethical issues regarding consent may be universal to all countries. One of the potential limitations with the current approach is that consent has been reduced to a technical issue and is often removed from wider ethical considerations. The consent procedures required by international review boards include having to provide detailed scientific and legal information to participants. Although the intent is to increase knowledge, these messages may lose their meaning when administered within certain populations. The challenge then becomes one of how to balance the requirements of providing complete information while also obtaining true informed consent.

The eight major topics covered in the education component of the GEC included general transmission modes of HIV, sexual transmission, mother-to-infant HIV transmission, precautions to avoid HIV transmission, ways through which HIV does not transmit, identification of HIV in an individual, detection of HIV in the body, and symptoms of AIDS. In addition, nine major topics related to HIV testing and informed consent were covered, including procedural risks of HIV screening, social risks of HIV screening, availability of the report, the right to say no, repercussions of refusing to take the test, confidentiality of the testing procedures, the right to consult others, the benefits of HIV screening, what one's signature means, specific procedures of the HIV ELISA test and, if applicable, specific procedures of the rapid HIV test.

Many international health education and prevention programs have developed and tested creative and culturally appropriate communication strategies to provide information on issues as sensitive and diverse as family planning and changing defecation behaviors. However, innovation in, and evaluation of, the process of delivering the often complex information necessary for informed consent has been limited [10]. The use of simplified visuals and text as a means of communicating messages to populations with little or no literacy has commonly been used to convey information on family planning, health, nutrition and public health activities [11,12]. However, such visuals must be developed with the specific population in mind and by community members in order to ensure their positive impact on understanding [13].

There is very little evidence demonstrating that individuals in resource-poor countries cannot understand the basics of research design or biomedical treatment options just because they have little education or different views about health and illness. It may be difficult to communicate the purposes, conditions and risks of research, but the difficulty of doing so should not detract from the importance of obtaining individual informed consent. In fact, several researchers [10,14] have established that, with some effort, acceptable levels of information can be communicated. This study indicates that the full informed consent process (including both group and individual counselling), when combined with enhanced education and counselling materials, can lead to excellent comprehension of informed consent issues.

The dramatic improvement we found in the women's comprehension of informed consent issues, despite their varying socioeconomic and educational backgrounds, is encouraging. Caution should be exercised, however, in interpreting the improvements in comprehension as being a result of the individual counseling and visuals. The second interview was done within one-two hours of the first, and the women may have had time to reconsider their responses to the previous questionnaire.

The major limitation in this study was the experimental design, in that we did not have the opportunity to measure the impact of individual counseling without the use of visuals. This was mainly because during baseline data collection we were interested in comprehension immediately after the GEC session; interaction with the counselor for individual counseling directly following the GEC was very brief and, therefore, was not expected to have improved women's knowledge considerably. Moreover, by the time we tested the visuals we had developed, the counselors had gained a lot more experience in HIV counseling and the women might have received more information from various media about HIV.

Despite these limitations, we can assert from our data that simple didactic group education on HIV/AIDS and testing issues is not sufficient to help women in this setting to understand the complexities of informed consent for HIV testing. The use of visuals in the form of posters and flipcharts provided structure and uniformity to the GEC sessions, thereby reinforcing the messages for these women and enhancing the overall informed consent process.

Obtaining proper informed consent in the case of HIV screening is not a discrete action, but a process that can be enhanced through effective communication, repetition and reflection. Although the intent is to increase knowledge, information regarding informed consent may lose its meaning when administered within certain populations. The time and the technicalities of the process itself may intimidate women in societies where literacy and awareness about medical and legal rights is low. It has been argued that complicated concepts conveyed in a consent form, often to fulfill the requirements of funding agency for institutional and policy purposes, may in themselves be unethical and, indeed, pose the biggest barrier to the informed consent process [15]. It should also be recognized that implementing counseling and informed consent procedures is considerably more difficult in certain settings in India where facilities, supplies, personnel, and time are at a premium. In addition, educational levels for women in this population are low, where 36% have a primary school education or less and another 33% are illiterate [16].

Generally, for HIV screening, an individual reads or is read a prepared statement that includes detailed scientific and legal information on the aims and biological significance of the test, the risks and benefits of testing, and the individual's rights. The participant is expected to understand the main components of what is written within the document and make an autonomous decision on whether or not to be tested for HIV. This is further complicated by the fact that, in some studies, the original consent document is composed in English and then translated nearly verbatim into the local language, making the communication of already complex topics even more difficult.

Typically, informed consent for pregnant women in most Indian hospitals and clinics is for operative procedures such as cesarean section or laparotomy. It is usual for the doctor or resident on duty to put down in his or her own handwriting the text of the consent on a patient's case papers. The content of this consent gives blanket permission to the hospital and doctors to undertake all procedures on the patient that are indicated in order to maintain the good health of the mother and her fetus, while at the same time absolving the attending physician and hospital of any blame in the event of a mishap. This is signed (or a thumbprint given) by the patient, her husband or an accompanying relative, and is generally considered to serve as legal consent; therefore, it is not interpreted as voluntary. Most often, due to time constraints, very little is explained to the patient about the procedure, risks, and benefits, or what her signature actually means. As found in other regions in India, there is a general perception by clinicians and other healthcare workers that women are "unable" to understand any of the procedures even if explained, because they are illiterate or have no medical background [17,18]. It is implicit in the physician-patient relationship that any treatment or procedures recommended by the physician will benefit the patient [19-21]. By de-mystifying the content and process of informed consent through standardization with structured visual cues and reiteration, we feel that these difficulties could be overcome.

In creating the modifications to the GEC, we focused on improving communication of those concepts that were the most unfamiliar to these women. For example, for most women in India, there is relatively little sense of autonomy [22]. For many, the woman's role is defined first by her father; after marriage, her husband's decision-making and the wishes of his parents or the elders in his home prevails. For such a woman, when an incurable disease like HIV/AIDS is presented to her in terms of her "autonomy" and "power" with reference to the disease, this could be confusing and even frightening. The prevailing practice of obtaining familial input on such decisions was demonstrated through women's knowledge of this topic despite the fact that it was hardly mentioned during the counseling sessions. On the other hand, our data show that by enhancing the GEC and reinforcing its messages through individual counseling, a significantly greater number of women can correctly understand the idea of their "right to refuse", indicating that even complex constructs such as autonomy can be conveyed. The understanding of the "meaning of the signature" was clearly enhanced with the use of visuals, because the improvements in women's knowledge directly followed the increase in coverage during the counseling sessions.

Clearly, some of the concepts related to informed consent may already be understood by these women. Women's understanding of "consequences of refusal" showed marked improvements in the second group of women studied. Although some of this improvement may be attributed to the individual counselling and visuals, the fact that only 55% of the GEC sessions actually adequately covered this topic indicates that other factors may have contributed to the women's knowledge of this topic.

This process of refining and evaluating the informed consent process can benefit the clinic and research settings in both developed and developing countries. Data from clinics in the USA, Belgium, and France report that even under well suited environments, informed consent for HIV screening was generally only obtained in about 70–85% of cases, and documentation of consent was substantially less [23,24]. In research settings, studies suggest that, despite having signed a consent form, participants may not fully understand critical aspects of research participation or their individual rights [25,26].

We have modified the standard model of informed consent by adapting it to suit the population that was served, and have documented subsequent improvements in patient understanding of the informed consent process. This study shows that culturally appropriate enhancement of the standard informed consent process moves the process towards its goal of being one that is truly informed and voluntary. Thus it not only fulfills the ethical requirements, but, most importantly, helps to assure that women's rights are preserved. This last point is critical because previous research has pointed out that, although women may be fully informed, they still may not feel their choice is fully voluntary [14]. We suggest that the current requirements of informed consent procedures are inadequate and that it should be a process that communicates information in an effective manner, allows for reiteration of information and includes an evaluation of the woman's knowledge prior to signing the informed consent document.