Perhaps more than most development matters, debates over the appropriate approach to the epidemic are heated. In most developing countries HIV/AIDS is fatal because of the non-availability or non-affordability of anti-retroviral therapies; prevention of HIV/AIDS requires behavioural change in the most intimate aspects of human experience and adequate care and support is premised on breaking down entrenched social prejudices. While in some quarters, the devastation brought by the rapidly spreading epidemic calls for strong, centralised measures, others argue that many HIV/AIDS interventions run a high risk of creating ethical and human rights violations, as programmes in both developed and developing countries have shown.3 Both sides of this debate agree that high level political commitment to addressing HIV/AIDS is critical, but differences emerge in approach in that the latter would attach more importance than the former to equipping individuals and communities with the information, tools and ultimately power to reduce vulnerability.

3 For an argument that HIV/AIDS should not be granted special status and should be confronted with more of the “traditional tools of public health” in the context of the US, see Chandler Burr’s “The AIDS Exception: Privacy vs. Public Health”, Atlantic Monthly, June 1997.

This somewhat exaggerated dichotomy also reflects differences in the broader health arena as to whether one adopts a medical or social model of health (Evans 1999). A medical model would assume that programmes should focus on particular interventions or technologies that are known to affect the limited range of specific behaviours that have an impact on the evolution of the epidemic. By contrast, the social model of health would focus more on the broader social determinants of the health problems concerned. In the case of HIV/AIDS, therefore, this would entail addressing the social processes and inequalities driving the epidemic and intensifying its effects. For example, structural changes in gender inequalities and in the differential power in sexual negotiation between men and women, and the elimination of sexual coercion are considered by many to be critical to reducing vulnerability to HIV transmission. This approach would argue that without greater control by people over the circumstances governing their health behaviour, long-term vulnerability to the epidemic would not be addressed. In practice, many non-governmental organisations active in HIV/AIDS may derive elements of their programmes from both perspectives, but in general the social model is likely to predominate.4

4 This debate echoes that about child health in the 1980s when there were calls in some quarters for the implementation of selective technical interventions known to be effective in combating childhood diarrhoeal disease, for example. These were countered by others who argued that the social factors leading to disease such as poverty and poor sanitation must be addressed – see Rifkin and Walt 1988.

Following from these debates, definitions of objectives of scaling up and the question of impact depend on whether one adopts a national or community perspective, a public health/epidemiological or development approach, or whether one is referring to prevention or caring for, supporting and improving the quality of life of those infected by HIV. National governmental authorities generally want to reach a scale that represents a significant proportion of that country’s population. They would share with those supporting a public health/epidemiological perspective a major concern in preventing the epidemic, and therefore reaching a sufficient scale to have an impact on the evolution of the epidemic. While care and support of those infected or those affected by it may also be a national priority, it is typically not given as much priority by governments as prevention until a relatively high level of prevalence is reached. Similarly, economists (whose perspective is presented more on pages 20-21) would analyse scale in terms of the coverage reached or impact attained in relation to a finite level of resources 5 - a concern about resource allocation shared by governments and international donors. To those subscribing to the above perspectives, therefore, “scale” is likely to denote an absolute level of coverage (usually of services, or interventions) in order to have impact (usually defined in terms of spread of the epidemic).

5 As a recent World Bank paper on costing HIV/AIDS scaling up states this concern: “How can we reach the greatest number in the cheapest manner?” AIDS Campaign Team for Africa, World Bank, “The Costs of Scaling HIV Program Activities to a National Level in Sub-Saharan Africa” p. 2.

By contrast, a community-based or non-governmental organisation, however, might share the above long-term goals but be more immediately concerned about changing the underlying social parameters in which HIV/AIDS activities are carried out, such as reducing stigma, passing legislation which makes discrimination in HIV/AIDS less likely or helping households and communities to cope when their members are hit by HIV/AIDS. Impact in these cases is clearly difficult – although not impossible -- to measure. Their notion of “scaling up” is likely to be a more relative one, evaluated in relation to the capacity of their own organisation, the need in the particular population group they aim to reach or the social trends they are confronting. While many may focus on services or interventions, their emphasis lies more on the process of engaging their respective constituents and eliciting a response, a component of which is likely to be greater demand for services.

According to a social model of health, the need to scale up may focus on inducing long-term social change that would enable societies or particular communities to come to terms with the HIV/AIDS epidemic.6 An individual is more likely to change behaviour if those around him or her – and particularly peers in similar social contexts - are also changing (Hughes 1993), and thus change on a collective level acts to reduce the risk factors conducive to the spread of HIV between individuals. The heart of behavioural change, however, lies deeper within the norms and values perpetuated by society across generations. Modifying expectations, beliefs and attitudes at a societal level is necessary in the long-term for sustained behavioural change to become the norm.

6 Or, in the words of the local response team at UNAIDS, making “AIDS-competent societies”.

Such changes, however, are only likely to come about and be sustained when there is corresponding progress in the policy and legislative context underpinning HIV/AIDS efforts. This, in turn, is most often induced when there is sufficient public debate and civic engagement concerning HIV/AIDS to demand it. Thus the issue of scale becomes a prerequisite to broader change. In many cases, such pressures have been brought about through community organising at a local level by groups disproportionately affected by the epidemic, who individually often lack political power but collectively are able to have an impact. The organisation of gay groups in the United States is of course a classic example. Yet at the same time, activism based on identity politics can also have its dangers in impeding efforts to expand interventions more broadly so that they permeate the whole society. As Richard Parker has expressed: “The very effectiveness of local-level politics has made it difficult to build a broader coalition to address HIV/AIDS as a global issue and the kinds of social movements that have emerged around other (similarly global) issues such as environment, or reproductive health and rights.” (Parker 1998).

Disparities of definitions of scaling up, therefore, reflect more than semantics but rather underlying differences in philosophy, approach and objectives. To clarify from the outset, the focus of this publication, and the workshop on which it draws, is on increasing the scale of NGOs which, although they themselves vary enormously in initial scale, largely share the second perspective outlined above. That is, the process of scaling up at the micro or community level is given more attention than reaching a specific level of scale in terms of the epidemiological picture at the national level.7 Emphasis is placed on the social change needed to both prevent HIV/AIDS and improve the lives of those affected by it.

7 For this reason, preference is shown here for “scaling up” rather than “going to scale” because of the implication in the latter expression that scale is a particular size and that the diverse perspectives on scaling-up could agree on the optimal absolute level of scale.

It should be noted, however, that not all NGOs start from a similar initial scale of activity or the same micro level frame of reference. Indeed, examples in this publication range from small community-based operations (such as that by Family AIDS Caring Trust/Family, Orphans and Children Under Stress (FACT/FOCUS)), to national (such as Kenya AIDS NGO Consortium (KANCO)), to regional (such as the Programme Support Group (PSG)) and even to international initiatives led by NGOs (such as Private Agencies Co-operating Together (PACT) or the International HIV/AIDS Alliance). Moreover, many NGOs, particularly those that assume a “catalysing role” in supporting other smaller organisations both technically and financially (see Strategy 2 in the section on Strategies) may begin with a national frame of reference. And NGOs acting together can clearly combine to create a national force, as is the case with the national coalition of HIV/AIDS organisations comprised by KANCO, which does not operate local field-offices. Indeed, Myers’ term association described below, which he argued might be particularly useful in the field of early childhood education, may similarly be an effective strategy in HIV/AIDS. As noted below, he uses the term to mean expanding programme size through common efforts and alliances among a group of organisations, each of whom develop similar programmes but tailor them to the needs of specific communities or populations.