Justification for the cervical cancer screening trials in India

What did policy makers need to know in the 1990s?

In the 1990s the established screening method for cervical cancer in developed countries was regular Pap smears. This method is not particularly sensitive, i.e. it misses a relatively high number of potential cancers. The Pap smears therefore need to be done at regular intervals, such as every two or three years. This method also requires highly specialized labs, and highly skilled personnel to interpret the microscopic images. Finally, the women need to come back again to get the results and to get treatment if necessary. If the screening program is carried out with appropriate facilities and personnel, and women come in for regular screening every two to three years between the ages of 20 to 60, the program is effective. Women in India who would follow the recommended procedures, and had access to high quality labs, would have access to an effective diagnostic tool. But there was, and still is, near consensus among experts that at the current stage of development of India, both in general and in its health system, it would not be appropriate to introduce regular Pap smears as a general population based screening program.

In the 1990s researchers were trying to identify a simpler screening tool, that requires fewer visits to clinics, less sophisticated clinics, and personnel that can get the necessary skills with less extensive training.  One such method is visual inspection after application of acetic acid (VIA). This requires less infrastructure both in terms of labs and personnel, and the screening and treatment (cryotherapy) can be done in one visit. Around 2000 VIA was seen as a promising method in low-income settings, but there were worries about its low specificity: it would lead to overdiagnosis and therefore unnecessary treatments. That is, doing VIA once was better at detecting cancers than Pap smears, but less specific, underscoring the previous emphasis on the need to do repeated tests conscientiously in order for a Pap smear screening program to be effective.

Policy makers who want to decide on any alternative screening programs, such as VIA, need to know the answers to two questions:

1. Do we know enough about alternative screening programs to recommend population wide implementation in India?
2.  If we do not know enough, what additional knowledge is needed, and how do we go about getting that knowledge? 

Was it already known in 2000 that cheaper screening methods were effective?

Ruth Macklin has attempted to provide arguments that address the two questions. Her first claim is that it was already known around 2000 that VIA was effective. Therefore, any clinical trial that investigated the effectiveness of VIA was unnecessary, and therefore unethical. Policy makers already know enough to implement VIA as a population based screening method. Her main argument to back up her claim is a report published by WHO in 2002:

The efficacy of VIA was already well established. According to a World Health Organisation (WHO) consultation report in 2002, “The test performance of VIA suggests that it has similar sensitivity to that of cervical cytology in detecting CIN, but has lower specificity. Further research is required to improve its specificity without compromising sensitivity” (2). The WHO report also pointed out the need for training personnel in the use of the method, as well as that for developing standard procedures for quality control. Also needed at the time was research on the development of a simple scoring system to objectively report the results of VIA. However, the important point is that the efficacy of VIA as a screening method had already been established when these trials were conducted in India.[My emphasis]

The problem is that Macklin seriously misrepresents what the WHO report actually says. Here is a statement that Macklin has decided not to refer to:

Information on test characteristics alone does not suffice for making decisions regarding whether VIA may be adapted for use in population-based organized screening programmes in routine public health practice. Evidence on its comparative efficacy and cost-effectiveness versus cytology screening in reducing incidence and/or mortality, as well as information on short-term and long-term consequences arising from decisions taken based on screening test outcomes, is essential for recommendations leading to public health policies. Such evidence may ideally be generated from randomized controlled trials.

According to Macklin, and against WHO, the only type of trial that what was needed at that time was either an implementation trial or a demonstration project. For Macklin, implementation research is

A study design that compares cancer rates following the introduction of VIA in urban wards or rural primary health centres with the past rates of cancer among women who used the same health facilities before VIA was introduced [and it] could provide results demonstrating that the implementation of the new technique was successful

Macklin gives an example of a demonstration project in several African countries where a screening program using VIA and cryotherapy was introduced. Quoting from the report from the project, Macklin goes on to say that

This demonstration project has shown that the ‘screen and treat’ approach can be introduced into existing reproductive health services in low-resource countries. Screening for precancerous lesions using VIA and treatment with cryotherapy is acceptable and feasible at low level health facilities in six African countries

The problem is again that Macklin seriously misinterprets the results of this demonstration project. Fewer than 10% of eligible women in the areas targeted were actually screened. Of the around 1000 who were scheduled for cryotherapy 600 were lost to follow up. Of the 326 women who were identified as having serious lesions requiring treatment, only 96 were followed up. The demonstration project showed that one could implement a screening program, but not a very successful one. The problems in the project were precisely those raised in a 2002 WHO report: There was no evidence that VIA could identify cancer cases reliably in a population wide screening program, because one does not know if such a program will achieve high enough follow up rates of the screened women, and therefore does not know if the program will achieve its goal, reducing the number of cervical cancer cases in the screened population.

As is evident in the WHO recommendations published in 2002 there was general agreement that the evidence was NOT sufficient for implementation of a policy of nationwide screening in a country such as India, and existing mentioned above confirms that. Contrary to what Macklin claim, the 2002 actually endorses the three Indian trials as necessary to obtain knowledge for policy, in addition to, and not instead of, demonstration projects. Here is the full quotation:

Currently, there are three large randomized trials ongoing in India (Dindigul district, Tamil Nadu; Mumbai (Bombay) city and Osmanabad district, Maharashtra) that are addressing the efficacy of VIA screening in reducing the incidence of and mortality from cervical cancer. The programme in Osmanabad district addresses the comparative efficacy of conventional cytology, VIA and HPV testing. The programmes in Dindigul district and Mumbai address the efficacy of VIA in reducing the incidence of cervical cancer. A three-arm, prospective randomized intervention trial in South Africa is currently addressing the comparative safety, acceptability and efficacy of screening women with VIA and HPV DNA testing. This trial is also investigating the effects of immediately treating women who screen positive with cryotherapy, performed by nurses in a primary health care setting. Outcome measures include reduction of high-grade cervical cancer precursors in treated versus untreated women, followed over a 12-month period. A large demonstration project of VIA, which is ongoing in the San Martin region of Peru, aims to investigate the effectiveness and acceptability of VIA integrated with the health services. Those screening positive by VIA are referred for magnified visual inspection after application of acetic acid (VIAM) and immediate biopsy and cryotherapy are provided if indicated by VIAM. A similar intervention involving VIA and VIAM has been initiated in Western Kenya. These studies are likely to provide the required evidence on the longitudinally-derived sensitivity and the cost-effectiveness of VIA in decreasing cervical cancer incidence and mortality.

It should therefore be abundantly clear that there is not enough evidence in 2000 to implement large scale population based screening methods in low- and middle-income countries such as India. It is only by selective quotations from a key 2002 WHO guidance document that Macklin can make her case.

In my next blog I will examine the second, and more important question of relevance for policy makers: If not enough is known about the value of alternative screening methods, what type of research is needed? Specifically, what should be the control group in a Randomized Clinical Trial?