Description of HASP

HASP was the first systematic HIV surveillance system in Pakistan. Prior to HASP, HIV surveillance in Pakistan had been largely passive, with disease trends monitored through clinical samples and occasional ad hoc studies. HASP enabled Pakistan to institutionalise true SGS allowing for better understanding of the distribution of and interactions among risk behaviours. At the time, most HIV surveillance systems outside high-income countries had been developed for the generalised epidemics in sub-Saharan Africa or used convenience samples such as antenatal clinic attendees, so there was a need to develop a new framework appropriate for the concentrated epidemic of Pakistan's unique situation. In addition to the unique social and political context of Pakistan and HIV transmission dynamics within the country, the framework was developed to incorporate sampling methodologies and questionnaires appropriate for subpopulations unique to South Asia. For example, hijra sex workers, transgender individuals who perform at ceremonies and practice sex work through large established networks, were included.

To establish the project, staff were assigned to a central data coordinating unit, a national field epidemiologist was placed within National AIDS Control Program (NACP), and surveillance positions were designated at the provincial level to support Government of Pakistan counterparts. Provincial data coordination units were established in each of the provinces for local data entry, to be transferred later to HASP staff for cleaning and analysis. HASP mapping methodologies were developed, piloted, evaluated and standardised as the official, government-approved approach. After pilot testing, four rounds of mapping and surveillance took place among IDUs, MSWs, FSWs and hijra sex workers over a period of 7 years. The numbers of cities varied each round, with certain key cities and key sub-populations included every round to allow for comparable data. The SGS activities included mapping, questionnaire administration and dried blood spot collection.

The first step of HASP methodology (figure 1) is geographic mapping to obtain a better understanding of the size of key populations and patterns of risk behaviour. The approach involves characterising high-risk activities such as sex work and injection drug use, and estimating the number of individuals involved in these activities. In addition, mapping identifies and characterises locations where high-risk activities take place. Finally, mapping provides information about sub-types of high-risk activities (eg street-based sex work and brothel-based sex work). This information is critical for developing sampling strategies that can ensure selection biases are minimised and information collected is representative of the sub-population studied. After many discussions, it was decided that recruitment of key populations should take place through offices dedicated specifically to HASP rather than pre-existing sexually transmitted infection clinics and primary health centres. Non-governmental organisations (NGOs) were engaged as implementation partners in many aspects of surveillance planning and implementation, including mapping, the recruitment of key population members and data collection. Through the use of questionnaires and dried blood spot collection at regular intervals, information about HIV prevalence and key behavioural and structural vulnerability indicators were then obtained from IDUs, MSWs, FSWs and hijra sex workers. Mapping and integrated behavioural and biological surveillance were conducted in four separate rounds and in many cities across Pakistan, allowing for geographic comparisons and the analysis of temporal trends.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

HASP methodology.

To provide a comprehensive understanding of the local HIV epidemic, in addition to HIV prevalence, HASP also performed HIV molecular subtype analysis. At times, qualitative approaches including in-depth interviews and focus groups were used to obtain a better understanding of the risk environment. Through this process, various protocols have been developed, including surveillance methodologies, ethics and gender guidelines, and supervision and monitoring protocols. Data management and reporting systems were developed, with the ability to integrate data from diverse sources to provide a more complete understanding of the epidemic. As a result of training and experience gained through this SGS project, expertise in data management and analysis, surveillance and epidemiology have been built in Pakistan, which now has a more comprehensive and centralised surveillance system than Canada or the USA. Although knowledge use has depended, to some extent, on the availability of funding and on political will to implement relevant programmes, national surveillance reports have been released and used to inform the development of HIV programme plans, donor funding proposals and subsequent disbursements, Joint United Nations Programme on HIV/AIDS (UNAIDS) HIV estimates, and improvements to and scale-up of service delivery and prevention programming. Through this collaborative process, Pakistani counterparts participated in supporting and administering large-scale surveillance, developing evidence-based policy and programmes and improving HIV prevention and care infrastructure in terms of both reach and quality.

Funded by the Canadian International Development Agency, HASP involved multilateral partnerships between Canadian and Pakistani organisations. The Canadian partners included the University of Manitoba, ProAction: Partners for Community Health, the Public Health Agency of Canada, and Agriteam Canada. The partners in Pakistan included the Ministry of Health, the National AIDS Control Program, provincial AIDS control programs, and a national technical advisory group for HIV surveillance and research within Pakistan.

HASP knowledge translation initiatives

Knowledge translation initiatives have been used to disseminate knowledge generated in Pakistan over the course of HASP and have informed HASP methodologies from the very beginning. Key HASP staff travelled to India to learn from the large-scale key population mapping and surveillance projects that have taken place in the state of Karnataka over the past decade, informing the design of HIV prevention programmes being delivered with high coverage of target populations, particularly FSWs and MSM.10 The ability to adapt methodologies used in India to the context of Pakistan provided confidence in the ability to adapt these methodologies to specific settings. Much learning took place over the course of HASP, allowing the methodologies used in Pakistan to be improved and refined from round to round. Also, it was essential for methodologies to be flexible so that the surveillance system could be responsive to changes in the local context, communities and epidemics.

The use of dried blood spots for HIV testing and molecular typing has been demonstrated to be safe and cost-effective and is now accepted as an important component of a SGS system. Significant laboratory capacity was built in Pakistan through HASP, both in terms of technical ability to test dried blood spots and in the ability to handle large volumes of specimens. In the fourth round of SGS, a new specimen tracking system was developed, allowing the team to reliably monitor the quality of the specimens and to archive the specimens centrally under ideal conditions, while also permitting decentralised serologic testing. This was a significant advance and has the potential to also inform how similar systems can be improved in Canada. WHO guidelines for the use of dried blood spots for HIV surveillance were informed by experiences from HASP.11

HASP staff held workshops to train provincial and NGO staff in HASP methodologies and developed their skills in the use of surveillance data for program planning and monitoring.

To allow others to capitalise on the lessons learned over the course of HASP, ongoing knowledge translation has been taking place through discussions, consultations and capacity-building initiatives with countries in the Middle East and North Africa (MENA), Central Asia, subSaharan Africa, Colombia and Canada. Given the societal and religious similarities between Pakistan and countries in the Middle East, methodologies and approaches developed through HASP are particularly applicable to the context of the MENA region. Mapping, venue-based sampling and population size estimation methodologies which were originally developed in Karnataka, India, have been adapted and optimised for difficult-to-reach networks in Pakistan and have since been implemented in other Islamic countries such as Maldives, Afghanistan and Yemen, as well as in Sri Lanka, China, Nigeria and Kenya. These methodologies have been presented at several meetings, including an inter-country consultation on preventing HIV among IDUs held in Kolkata, India (April 2007), a Pan American Health Organization (PAHO) meeting on designing SGS among MSM in Colombia (May 2007) and at the Second Global HIV/AIDS Surveillance Meeting held in Bangkok, Thailand (March 2009). After these presentations, the WHO Eastern Mediterranean Regional Office expressed interest in learning more about this methodology. A regional workshop on size estimation techniques for key populations was conducted by the Regional Knowledge Hub for HIV/AIDS Surveillance in Kerman, Iran, in October 2009 with representatives from over ten countries in attendance. Following this meeting, a mapping fact sheet was prepared for use in this context.12 the WHO Eastern Mediterranean Regional Office published guidelines for HIV surveillance in the MENA region which incorporates numerous examples from HASP.13 The methodologies were presented again at the MENA Regional Expert Group Meeting on Strategic Information and Surveillance in Cairo in May 2011. Pakistan's SGS was regarded as one of the best in the region and strong interest was expressed in further learning from its successes.

In 2010, HASP methodologies were used to map and estimate the size of MSM populations in Yemen. Due to the political and security situation in Yemen at the time, the team was not able to complete this study, but initial spot validation for recruitment of MSM for integrated behavioural and biological surveillance was completed and there are plans to continue this work when the situation changes.

Importantly, the donor country, Canada, has benefited in significant ways from this partnership. The Public Health Agency of Canada, one of the project partners, performs mapping, HIV behavioural and biological surveillance among MSM, IDUs and street youth across Canada in collaboration with provincial and local partners.14–16 Many of the same techniques used in these Canadian studies were refined and implemented on a very large scale in Pakistan as part of HASP. Subsequently, a number of operational and logistical lessons from HASP have, in turn, improved how surveillance is performed in Canada, especially with respect to scaling up such studies. The lessons learned from HASP have also provided the leads of the Canadian studies with additional confidence and experience in these methods and provided institutional review boards and partners across Canada with increased confidence in the safety, acceptability and effectiveness of these methods.

Donor countries benefit greatly from international capacity building projects. A systematic approach to document and publicise the ways in which donor countries benefit from international capacity building projects would be useful to ensure ongoing support in foreign countries, document the importance of international collaborations and highlight the value for lessons learned to feed back to donor countries to improve their own programmes.

Challenges for mutual learning

Generally, the recognition of the strengths of the approaches and methodologies used by HASP has ensured an appetite for opportunities of mutual learning. However, this is not to suggest that challenges have not been encountered. The approaches and methodologies of HASP can be put in place without knowledge translation occurring.17 ,18 In fact, different organisations and individuals need to take ownership of the knowledge generated in order to ensure sensible and practical knowledge translation takes place.19 As to be expected, these organisations and individuals have different agendas and priorities and, at times, these must be negotiated to ensure the success of collaborative activities.

It is well documented that the use of new knowledge results from complex interactions between knowledge producers and knowledge users.20 One barrier to the use of new knowledge may be misconceptions on the part of both the knowledge producers and the knowledge users. These misconceptions may relate to the type of information necessary to develop interventions21 or the objectives themselves.22 In addition, the acquisition and diffusion of knowledge is all too often a unidirectional process, which does not favour knowledge use.23 ,24

The political context in Pakistan had made progress difficult at times. For example, heightened political attention on injection drug use as a criminal activity made it difficult for government partners to adopt and act upon knowledge translation messages which employed a public health approach to harm reduction. Furthermore, the political nature of government appointments meant that there was little continuity in the HIV program staff, who are, ideally, important users of the knowledge generated through HASP.

The changing priorities of Canadian funding agencies also threaten the sustainability of projects. Although a second phase of HASP was planned, the Canadian International Development Agency reconsidered the scope of this investment and supported an 18-month extension instead to allow HASP to conduct a final round of surveillance, build additional capacity and hand its SGS model over to the Pakistani government. The government of Pakistan has been an essential HASP partner since its inception and now has the opportunity to ensure the survival and sustainability of the surveillance infrastructure that has been built.