Social Science & Medicine 133 (2015) 36e44

Contents lists available at ScienceDirect

Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed

Challenges to successful implementation of HIV and AIDS-related health policies in Cartagena, Colombia
mez b Nehla Djellouli a, *, María Cristina Quevedo-Go a b

School for Public Health and Primary Care (CAPHRI), Maastricht University, Duboisdomein 30, 6200 MD, Maastricht, The Netherlands
, Colombia Departamento de Medicina Preventiva y Social, Pontificia Universidad Javeriana, Carrera 7 No 40-62 Edificio Hospital San Ignacio, Bogota

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 28 March 2015

The Caribbean region presents the highest prevalence of HIV/AIDS worldwide after sub-Saharan Africa; leading to serious social, economic and health consequences at the local scale but also at the regional and global levels. In Colombia, a national plan to tackle the epidemic was formulated with little evidence that its implementation in the local context is effective. This study focused on Cartagena e one of Colombia's largest cities and an international touristic hub e that presents one of the highest HIV prevalences in the country, to investigate whether the national plan accounts for local specificities and what are the barriers to local implementation. Based on the Contextual Interaction Theory (CIT), this qualitative research relied upon 27 interviews and 13 life stories of local inhabitants and stakeholders, collected in a first fieldwork in 2006e2007. A follow-up data collection took place in 2013 with 10 participants: key policymakers and implementers, NGO representatives and local inhabitants. Barriers identified by the participants included: local population's understandings and beliefs on condom use; stigma and discrimination; lack of collaboration from the Church, the education sector and local politicians; corruption; high staff turnover; frequent changes in leadership; lack of economic and human resources; and barriers to health care access. The findings suggest that global influences also have an impact on the CIT framework (e.g. international organisations as a major financier in HIV prevention). The participants put forward several feasible solutions to implementation barriers. We discuss how several of the proposed solutions have been applied in other Latin American and Caribbean countries and yielded positive results. However, further research is needed to find possible ways of overcoming certain barriers identified by this study such as corruption, the lack of collaboration of the Church and barriers to health care access. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Cartagena Colombia HIV and AIDS Policy implementation Sexual tourism Machismo

1. Introduction The Caribbean region presents the highest prevalence of HIV/ AIDS worldwide after sub-Saharan Africa (The Lancet, 2008). This high prevalence was associated with the rise of sex tourism within the growing tourism industry, with large social inequalities and with the machismo culture (Cabezas, 2009; Inciardi, 2005; Padilla et al., 2010). Consequently, the main pattern of HIV transmission in the Caribbean remains heterosexual sex with vulnerabilities chiefly embodied by women and the youth (PANCAP, 2008). The Caribbean HIV/AIDS epidemic has serious social, economic and health consequences at the local scale but also at the regional

* Corresponding author. E-mail addresses: [email protected] (N. Djellouli), quevedo
mez). [email protected] (M.C. Quevedo-Go http://dx.doi.org/10.1016/j.socscimed.2015.03.048 0277-9536/© 2015 Elsevier Ltd. All rights reserved.

and global levels. Firstly, if ineffective efforts are made at all levels of governance to reduce poverty - one of the main determinants in the region - it would seem unlikely that public health efforts could be successful in decreasing HIV incidence (Grenade, 2008; HEU, 2009). The tourism industry also has a role to play in HIV prevention. Practice of unsafe sex in the context of sexual tourism places the locals and their clients at risk. Given that locals migrate to touristic areas in hopes of finding employment, and that sexual tourism takes place in international touristic hubs, there is a risk of spreading HIV beyond borders (Padilla et al., 2010). Moreover, there is a long-term risk of having the most economically active section of the population heavily burdened by the disease, since HIV/AIDS affects prominently the 15e44 age group and a rising number of children is involved in sex tourism (PANCAP, 2008). Besides the health consequences and the strain on health systems, the possible loss of labour force and productivity due to HIV morbidity and AIDS mortality is also relevant in this region (McDonald and Roberts,


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

2006). Finally, although the number of children exploited by sexual tourism is rising, few specific interventions target the youth (PANCAP, 2008). While it is recognised that HIV risk factors within the region are broad and numerous, public health efforts to tackle the epidemic have not been tailored to address the specific determinants of this region (PANCAP, 2008). Even if health policies were designed specifically to the Caribbean region, for local implementation to be effective, specificities of the local context need to be taken into account. While implementation research at the nexus of sexual tourism and HIV and AIDS in Asia is abundant, similar studies conducted within the Latin American context are sparse (Padilla et al., 2010; Quevedo-Gomez et al., 2011). To fill the knowledge gap, Cartagena, located on the Colombian Caribbean coast, was chosen as a case study to demonstrate whether the national plan was formulated taking into account the local context. Compared to Colombia's national HIV/AIDS prevalence of 0.57%, in Cartagena, one of the largest Colombian cities, average prevalence is 1.6%; reaching up to 10% amongst vulnerable
n Social, 2012). Howpopulations (Ministerio de Salud y Proteccio ever Arrivillaga et al. (2009) estimated that for each HIV case reported in Colombia, seven are not. Social and structural determinants of HIV in Cartagena are intertwined with social inequalities (Quevedo-Gomez et al., 2011). As a result, men, women and children turn to sexual tourism to survive; an activity that some locals do not consider prostitution (i.e. sex for ‘profit’, per-se) since it is for many the only means for subsistence (QuevedoGomez et al., 2012). The national plan to respond to the HIV/AIDS epidemic developed by the Colombian government focuses on targeting the following vulnerable populations: people living with AIDS, men who have sex with men, sex workers, persons deprived of freedom, people living in the streets, displaced youths and persons demo
n bilised from illegal armed groups (Ministerio de la Proteccio Social, 2008). The national plan thus leaves out the vulnerable populations of Cartagena (as identified by Quevedo-Gomez et al. (2011)): housewives, people engaging in sexual tourism and people with low socio-economic status (SES). With its four foci (promotion and prevention; comprehensive care; support and social protection; monitoring and evaluation of the response) the national plan seems exhaustive. Yet, there is little evidence that its implementation and monitoring are efficient at the local level (Moreno et al. 2012). Research into implementation performance is needed at a more local scale; particularly in cities such as Cartagena, with high HIV/AIDS prevalence. This research aims, through a qualitative approach, to identify the factors affecting the implementation of HIV/AIDS related activities in Cartagena, where health policies are crucial to curbing the HIV epidemic. This study further investigates potential solutions to overcome the barriers identified. 2. Methods 2.1. Theoretical framework This research is based upon the Contextual Interaction Theory (CIT), a framework originally developed by Bressers (1983) and since then continuously updated (O'toole, 2004; De Boer and Bressers, 2011). It was previously used to identify barriers to HIV/ AIDS policy implementation in Indonesia, Vietnam and China (Spratt, 2009). CIT is based on the assumption that the outcome of policy implementation depends on the policy instruments chosen, and - more importantly e characteristics (i.e. motivation, information and power) of the actors involved in the policy process (policymakers, implementers and populations targeted by the

37

policy) (Bressers, 2004). Factors influencing policy implementation have an impact on the aforementioned key characteristics, which will in turn shape (and be shaped by) the interaction processes between the actors (De Boer and Bressers, 2011). Accordingly, determining the factors that impact the motivation, information and power of the implementation actors will lead to the identification of possible barriers to HIV/AIDS policy implementation. 2.2. Data collection The data collection was divided in two parts (Table 1). Part I aimed to identify the barriers to successful HIV/AIDS policy implementation in Cartagena using CIT. Data used to identify those barriers were based on primary research conducted by the second author during fieldwork in Cartagena for a 7-month period between 2006 and 2007. The data was originally collected for the purpose of identifying social and structural determinants of HIV/ AIDS in Cartagena. This original data presented opportunities for further analysis with regards to implementation barriers due to its qualitative richness and the diversity of stakeholders whose perspectives were captured during initial fieldwork. Part II was a follow-up data collection to track changes over time and to verify whether barriers identified in Part I were viewed by national pol
and local stakeholders in Cartagena as such. icymakers in Bogota With the collaboration of Part II participants, the aim was then to identify possible solutions to overcome the implementation obstacles identified. Data collection for Part II was conducted in a 3week period in June 2013 by the first author during fieldwork in
and Cartagena. Bogota For Part I, 13 life stories of persons living with HIV/AIDS (PLWHA) and 27 open-ended interviews with local inhabitants and other stakeholders (tested HIV negative or not diagnosed) conducted in Spanish were included in the data analysis. This sample included 18 men and 22 women, from ages 15 to 60. Participants were recruited by the second author through purposive sampling due to their knowledge of the social context of the epidemic, and came from an array of socio-economic backgrounds, ethnicities and sexual orientations. Part II entailed 9 semi-structured interviews, where one interview was conducted with two participants simultaneously. Interviews were conducted in Spanish and lasted approximately 1 h 10 participants, 6 men and 4 women, all university level educated, were recruited through quota and snowball sampling: - 2 policymakers at the national level that were involved in the formulation of the national plan; - 1 governmental implementer for the regional level; - 2 governmental implementers for the municipal level; - NGOs are also an important implementer identified in the national plan, thus 2 NGO employees, considered experts by their community regarding Cartagena's HIV situation, were further included; - To represent the target population, 3 local inhabitants affected by the epidemic (or one of its determinants) were interviewed ensuring that the policymaking vocabulary was adapted to their level of knowledge. Each participant was asked about the barriers identified in Part I and whether they were still perceived as barriers. Participants were also invited to expand on the barriers previously identified and to add any obstacle they deemed relevant. They were further asked to comment on the knowledge, motivation, interaction processes and power relationships between the policymaking chain's actors. Finally, participants were invited to think about and share possibilities to overcome the mentioned barriers to policy


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

38 Table 1 Overview of the data collection process.

Collection period Type of data Participants a

Part Ia

Part IIa

Secondary data collected in 2006e2007 27 open-ended interviews & 13 life stories of PLWHA 40 inhabitants and stakeholders: 18 men, 22 women including 13 PLWHA

Primary data collected in 2013 9 semi-structured interviews 10 participants, 6 men and 4 women, either policymakers, implementers or inhabitants

Both parts were analysed using codes based on the CIT framework.

implementation. Content saturation was reached as no new themes, barriers or solutions were identified by the participants.

strengthened by decentralisation, affects staff rotations, resources and implementers' motivation.

2.3. Data analysis

3.2. Barriers to promotion of autonomous, responsible and pleasurable sexualities

Both sets of data were transcribed verbatim in Spanish and analysed using ATLAS.ti 7 qualitative analysis software. Predefined codes, based on the CIT (power, motivation, information & interactions), were used. Other codes were added to capture potential opportunities to overcome barriers. Further free codes were developed based on themes emerging from the data (such as global influences). 2.4. Ethical considerations The ethics committee of Los Andes University in Bogot
a approved Part I data collection. The ethical committee of Pontificia
approved Part II data collection. All Universidad Javeriana in Bogota participants gave written informed consent. 3. Results In this section, we present the elements that impact the motivation, knowledge and power of the actors involved in the formulation and implementation of the national plan. 3.1. Formulation versus implementation The national plan establishes five national goals: - Promotion of an autonomous, responsible and pleasurable sexuality, - Reduce HIV risk factors, focussing on vulnerable populations, - Improve comprehensive care coverage for HIV patients, - Reduce the socio-economic impact of the epidemic for the affected families, - Monitor the epidemic situation and the national response. Each of these goals has specific targets to reach, but no concrete activities are depicted. Implementers interviewed in the second phase of the research, expressed support for the formulation of the plan. However, not all actors involved in the implementation were included in the formulation process. Implementers also agreed that policymakers are aware of the high prevalence of HIV and AIDS in Cartagena, and of the determinants identified by Part I participants: extreme poverty, lack of education and social services, prevailing social inequalities, sexual tourism, and machismo culture. Yet, implementation in Cartagena is limited. Because of neoliberal health reforms promoted by the World Bank, Colombia is decentralised (Homedes and Ugalde, 2005). Consequently, municipalities are autonomous in implementing (or not implementing) the national plan; making local political will crucial for successful implementation. Moreover, participants explained how some implementers have little clarity of the plan, due to various elements discussed later in detail. We will see for example how corruption,

One of the targets of the plan's first goal is to increase access to, and adequate use of, condoms. Implementers in Cartagena attempted to reach this target via community-based condom distributions, together with awareness-raising about HIV. However, participants identified that the successful outcomes of those activities are hindered by the predominant machismo culture. They expressed that using condoms is difficult, for both men and women, as it conflicts with their compliance to traditional gender roles. The male sexual role within machismo encourages men to have extramarital affairs and unprotected sex. Their wives' vulnerability to HIV is further exacerbated by unprotected encounters with other men. Locals reported that machos believe only homosexuals can have HIV; since “they are penetrating, they are not homosexual and thus do not need protection” (Cartagena inhabitant). Complicated by economic dependency on their husbands, housewives have difficulties negotiating condom use, according to locals, since machos interpret it as a sign of infidelity or prostitution. “There are men who have sex with other men but they are also married, have children. And they don't protect themselves in the extra-marital affairs, and they don't protect themselves in their homes either. The wife doesn't have the ability to negotiate the condom with her husband. Because if she says, “Let's use a condom”, the man will reply “Why would we if you are with me? It means you are doing it with other men” [ … ] Often women accept because they are too dependent economically on their husbands.” e Expert/NGO 2 Financial matters worsen the situation: participants mentioned how expensive condoms are and that more money can be earned in sexual tourism by not using protection. Although sex workers' vulnerability is identified in the national plan, individuals involved in sex tourism are not. This discourse occurs in a context of poor sexual education at home and at school. The other target of the plan's goal to promote an autonomous sexuality is focused on adequate sexual and reproductive health education. The national plan is progressive and advocates for an education model centred on human, sexual and reproductive rights, gender equity and respect for sexual diversity. However, according to the participants, the education sector (identified as an implementer in the plan) has been passive and is sometimes opposed to implementing such programmes. Their position is facilitated by the decentralised Colombian governance system, which made them autonomous in their decisions. Three interviewees mentioned that some teachers or administrators forbid NGOs from distributing condoms and providing sexual education in schools or refuse to implement some of the activities suggested at the regional level by the Secretary of Health. Interviewees explained that these attitudes result from the educators'


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

beliefs that the epidemic does not affect children and teenagers; attitudes reinforced by their religious beliefs on contraception. In Colombia, the Catholic Church also raises barriers to implementation of activities linked to condom use and sexual education. Respondents commented on the Church's lack of collaboration and sometimes opposition - towards those activities. The Church is not identified as an actor in the national plan, whilst, according to participants, priests exert a great influence in communities where they are seen as leaders. The Catholic Church's position on contraception and sexual diversity is often adopted amongst the priests' congregations. “The Church, to some extent, controls (sometimes more than the town mayor) the municipalities. And what the priest thinks, it is law in almost all the communities. If they don't open to sexual diversity, the people too are very closed to accept people with differences.” e Regional implementer The barriers mentioned here illustrate how key informal actors (Church) have the power to influence the information characteristic of formal implementers (educators) but also of the target population.

39

comprehensive care for PLWHA, free HIV testing and confidentiality. Policymakers and implementers interviewed in the study all agreed that barriers to healthcare access are preventing successful and sustainable implementation of the national plan. The situation in Cartagena is further worsened by stigma and discrimination present within the communities and healthcare workers. Stigma and discrimination related to sexual diversity and HIV/AIDS persist among locals, to the extent that they have become an obstacle for the promotion activities related to HIV testing. Participants described HIV/AIDS as a taboo subject within families and communities across all social classes in Cartagena, where HIV/ AIDS is associated with social groups like homosexuals, transsexuals, and sex workers, and with “promiscuous” and “sinful” conduct. Consequently, people are afraid and ashamed to get tested or reveal their status. “Something very important and worrisome is happening here: stigma and discrimination are still present. They are due to ignorance; to the negative image we have of the association “AIDS equals death”. Within the community but also a lot in healthcare. Despite many guidelines for infection management and a health team assumedly sensitised. But inside the same health team, there will be a lot of stigma and discrimination, becoming a barrier to access.” e Expert/NGO 2

3.3. Barriers to comprehensive care for HIV patients Despite the national plan's goal to increase coverage of comprehensive care for HIV patients, efforts to achieve this objective are hindered by the existing barriers to healthcare access: red tape, stigma and discrimination. Although there are universal healthcare access laws and the right to comprehensive care for PLWHA in Colombia, research participants explained that, in order to access care, patients must first undergo a myriad of paperwork referred to as tramitología best translated as ‘red tape’. This paperwork maze results from the fragmentation of the Colombian healthcare system, following the World Bank's recommendations (Abadia and Oviedo, 2009). Unfortunately for the low SES population, the paperwork is extremely difficult to comprehend; consequently, many give up their access to comprehensive care or end up getting treatment intermittently. Those who continue to seek care face months of bureaucracy before gaining access. This theme came up often in Part I interviews. Participants referred to it as the paseo de la muerte (‘walk of death’) since patients are denied health services for lack of appropriate paperwork and sent away to another healthcare entity until it is too late. “The walk of death is coming to a clinic and they can't attend to you, because you don't have the necessary paperwork or because you are not registered. And they tell you to go to another clinic and so they keep you walking until you die.” e Cartagena inhabitant Additionally, patients have to pay an out-of-pocket payment for medical visits and drugs, which becomes a barrier for financially vulnerable individuals. One NGO representative explained that the fee for service is small, but in addition to travel costs, the total amount adds up to almost a day's worth of food for low-income families. Those families cannot necessarily afford expensive treatment drugs, and when they can, they might have to take them on an empty stomach. Given that over a third of Cartagena's population lives under the poverty line, these obstacles to care are considerable (Perez and Mejia, 2007). Participants further acknowledged that locals are unaware of their rights as patients; particularly with regards to rights to healthcare access,

According to local participants, it is not uncommon for medical and other clinical staff to violate patient confidentiality and reveal the patient's status to a third party, leading to delayed healthseeking behaviours for fear of being judged or revealed in the community. Despite the existence of ‘whistle-blower’ mechanisms that encourage patients to report such misconduct, most patients abstain from using them - either because they are unaware of their rights, or because they seek to avoid further exposing their HIV status. Some participants described that many locals prefer to get tested in another city and one inhabitant explained that some students get tested under the pretence of donating blood, since donated blood is screened and confidentiality is assured. Therefore, lack of confidentiality, stigma and discrimination have been identified by local participants as major barriers to healthcare access and, consequently, to policy implementation.

3.4. Impact of corruption Given the administrative decentralisation, local political will is vital for the implementation of the plan in Cartagena. Yet, local inhabitants spoke of a political class with little interest in problems faced by Cartagena's inhabitants. Local participants described a lack of political will from mayors and politicians to implement the national plan due to lack of knowledge and interest for public health and due to politiquería - the abuse of political power in order to gain economic power, or corruption. Corruption is a theme that emerged in all interviews conducted in Cartagena. For interviewees, corruption is present in every aspect of the city's life “where the political class only considers profits that can be made immediately and lacks a coherent vision for sustainability in the future” (Cartagena inhabitant). Participants described how many projects, city works, and policy implementations started in Cartagena but were never seen through, usually because the economic resources were embezzled before project completion. “It seems that the city is mortgaged for the benefit of third parties and there is no continuity in policies.” e Expert/NGO 2 The most significant consequence of corruption for all

40


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

implementers is the constant rotation of staff involved in implementation. With public health agents and health institution administrators “hired based on political favour instead of competence and experience” (Regional implementer), a majority of policy implementers have limited knowledge of HIV/AIDS, the national plan and communication patterns between organisations involved in implementation. “But it's also true there are administrations, for purely political matters, giving contracts only for three or four months. You hire a civil servant for three months and there is a lapse of time before making the new contract. But between a contract and another, a week can go by … Up to two months can go by. From there, the personnel change according to political favours and many times the staff hired is not the most appropriate or the most prepared for the different posts or activities. [ … ] In everything you come across, unfortunately it's always like that. Then, when you don't encounter corruption, you find someone who only wants to favour himself, his political group. [ … ] In a few words, I tell you, this is corruption: to handle things in order to get a benefit that instead of benefiting the whole community, it's only a benefit for a minority group.” e Local implementer 2 Participants commented that even if the person hired is eventually trained and capable, another political party might take over and this person will be replaced by someone closer-affiliated to the new power in place, regardless of job performance. As outlined by one of the policymakers, this leads to a waste of resources spent on training and a loss of capacity and vision. High-ranking staff are also affected: for example, a chief decides to lead the prevention strategy at grass-root level and is then replaced, a year later, by someone who chooses to shift the strategy to the institutions. This also explains the lack of policy continuity and unsuccessful implementation. “I've found that here in the health sector and especially public health, everything is handled as politics too. Many positions are treated as politics. If you are a partisan of the candidate who won and you voted for him, you have a guaranteed position while the candidate is in charge. But if this candidate leaves and another comes that is not from your party, you are immediately removed and they place instead someone from that party. Many times it doesn't matter if you're good or bad at your job, just because there was a change in government. So, often the training that was given to this individual is practically lost because the person who works is replaced rapidly.”e Expert/ NGO 1 By coincidence, the policymakers and implementers interviewed in Part II were all working in their field for a long time and had the competence and experience required. However, job security is an ongoing concern for them. Another consequence of corruption that turns into an implementation obstacle is a lack of economic resources. Low resources are resultants of a lack of political will to address HIV/AIDS, corruption and other health programmes competing for the same resources. Furthermore, one policymaker explained that, usually, only cost-effective activities are implemented, not necessarily meaning they are needed in the local context. Low economic resources translate into a lack of human resources, which are frequently rotated and sometimes lack capacity and experience. One policy implementer linked the lack of monitoring and evaluation at municipal and departmental levels with limited human resources. Low economic and human resources are also an issue in

healthcare leading to declining quality of care and poorly qualified staff. Finally, the ramifications of corruption also affect the local implementers who pointed out high demotivation among their teams due to frequent staff rotations. Those with expertise have to collaborate with individuals lacking necessary competencies, making their own tasks more difficult. Demotivation also stems from constant change of leadership within their organisation and city government. Additionally, they are expected to implement national guidelines in a context of low resources. Newly rotated staff might feel demotivated if they lack competencies to carry out their duties. Incentives to learn new skills are limited given the temporary appointment period and the possibility of being rotated to another field. Consequently, implementers spoke of ‘the desire to work for the community’ as their sole motivator because they feel their communities are greatly affected by HIV. Some feel, as civil servants, that it is their responsibility to implement the existing national guidelines and policies the best they can. Those barriers illustrate how corruption influences the CIT's power characteristic in terms of human and economic resources available for implementation. This will also affect the motivation of implementers and the information made available to them. 3.5. Impact of global actors Part II interviews revealed the importance of international organisations in three areas: the formulation of the national plan, the funding of HIV prevention activities in Cartagena, and the decentralised system in Colombia. Some participants mentioned that assemblies organised by international organisations such as the World Health Organisation had a positive impact and helped, for example, bringing focus and awareness to rights of PLWHA, eventually influencing policy formulation. For instance, The Joint United Nations Programme on HIV/AIDS and the United Nations Population Fund office in Colombia have been actively cooperating with the MoH, offering consultancy as well as technical assistance in the formulation and creation of the object of this study: the national plan to tackle the HIV epidemic (and other related documents). This collaboration also explains why the national plan is standardised, progressive and not centred around religious tenets. However, given the gap between this progressive standpoint and the local population's beliefs, implementation barriers have arisen regarding condom use and sexual education - barriers that could have been foreseen if key local stakeholders had been involved in the formulation of the national plan. Additionally, international organisations - in particular the Global Fund (GF) e provide major funding for HIV prevention activities in Cartagena. The local policy implementers and NGOs outlined how the GF's financial contribution steered the focus of programming toward vulnerable populations, stigma and discrimination. The GF is also subsidising several of the projects in which these participants are currently involved. However, three participants, including one policymaker, pointed out that international funding is decreasing. According to them, Colombia is now perceived as a “rich” middle-income country and is subsequently receiving less international aid. A local inhabitant argued that although the government has been internationally promoting Colombia as a ‘safer country with better conditions of life’ to attract foreign investments, Colombia is still greatly affected by extensive social inequalities that leave part of the society in continuing need of international aid. The two examples above show the direct impact of international organisations on tackling Cartagena's HIV epidemic. But the actions of international actors can also have an indirect effect on the HIV


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

status in Colombia. As mentioned before, the World Bank's neoliberal reforms led to the fragmentation of the health system creating new barriers to healthcare access for HIV patients. These reforms also led to a decentralised system in Colombia reinforcing local corruption, which culminates in major barriers to the successful implementation of the national plan. Based on these results, global influences can potentially become facilitators or barriers to the implementation of the national plan and were therefore added to the CIT framework (Fig. 1). 3.6. Solutions put forward by the participants All participants agreed that educating and raising awareness about the virus, as well as sexual diversity, is vital. Participants believed this awareness-raising to be the optimal way to reduce stigma and discrimination and break the taboo around the disease and sexual diversity in families. One policymaker emphasised the need to deliver public health messages without inadvertently increasing stigma and discrimination. Other respondents suggested educating and sensitising simultaneously through community work and mass media to reach the greatest number of individuals affected by HIV. Given the levels of illiteracy, visual and audio campaigns should mainly be used. When asked about the receptivity to such activities by the locals, NGO actors and local implementers who worked at grass-root level affirmed that locals were open and interested in learning about the disease. They also observed more tolerance and less discrimination in the communities they worked with (compared to those where no implementation had occurred).

41

“The work has been noticed. Trustfully, the work is noticed. I observe more tolerance and less discrimination against diversity in general.” e Local implementer 2 For NGO actors, appropriate training should also be provided to healthcare workers to sensitise them on issues of stigma and discrimination, including education in ethics. They further stressed the importance of showing the staff that there are serious repercussions to violating patient rights. To break the influence of machismo on condom use, participants believe that education has to be provided at an early age with the introduction of the theme of gender equity, and sensitisation to sexual diversity. The participants' argument is that customs of machismo culture are not yet anchored in the children's frame of reference, and thus it will be easier to have an impact (in years to come) with children than with adults. Women should also be given focus, as according to a female inhabitant, they are responsible for keeping alive the machismo culture via their role in children's education. Furthermore, some teachers should be sensitised to the teenagers vulnerability to HIV/AIDS and improve the quality of sexual education. To raise political will, NGO participants suggested continuing the work they have been doing: “go over the head of local politicians straight to the governor (head of department) to keep the epidemic on the agenda” (Expert/NGO 2). By doing so, they expect that pressure will be exerted on local actors to implement the national plan. They further suggested, given the employment conditions, training staff according to their rotations. Policy implementers advised, in light of low resources, that

Fig. 1. Map of barriers to HIV/AIDS policy implementation in Cartagena using the CIT framework. Global influences play a role in policy formulation and implementation and therefore have an impact on the CIT framework.

42


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

during policy formulation, feasible actions are planned and resources focused accordingly. Furthermore, a re-evaluation of resource needs and allocations is required at the national level in order to prioritise the most needed programmes that lack funding. Local participants suggested that regular workshops with all formal and informal actors would improve communication between actors and unite the response across all levels of governance. For example, it could help identify effective solutions to improve healthcare access and to restore patients' trust in the healthcare system. Suggestions were also made to include NGOs in the monitoring and evaluation process since they already have the capacity to perform this work. Meanwhile, educating the population about their patient rights is essential, considering their vulnerability is heightened by their lack of knowledge. One NGO representative called for the establishment of users' committees in health entities to monitor the respect of patient rights. Concurrently, local governmental implementers and NGOs have already been helping patients navigate through the paperwork maze. They further argue that, in the current structural context, prevention activities should not be exclusively delivered in health centres, given the difficulties accessing them. Instead, community work seems more appropriate for prevention and health promotion activities. 4. Discussion Barriers to policy implementation in Cartagena are intertwined and found in all characteristics of the CIT (information, power, motivation) and can be influenced by global actors. Overcoming those barriers will thus require a complex multilevel approach involving a plethora of actors. Condom use in the context of sexual tourism is difficult since the economic incentive of practising unsafe sex is extremely attractive for financially vulnerable individuals. Thailand, known for its sexual tourism industry, successfully implemented the 100% Condom policy nationwide (Ainsworth et al., 2003). Could the 100% Condom policy be implemented in the Caribbean context? In the Dominican Republic, condom use perceptions are anchored in the machismo culture and women experience difficulties negotiating condom use as it is associated with infidelity (Haddock, 2007). Implementation of an adapted 100% Condom policy in the Dominican Republic encountered successes in increasing condom use among individuals engaging in commercial sex (Halperin et al., 2009; Rojas et al., 2011). However, perceptions of infidelity remained, and women negotiated condom use with their clients more easily than with partners (Rojas et al., 2011). Nonetheless, in cities where the government mandated the use of condoms, protected encounters outside the commercial sex sphere increased (Haddock, 2007). These findings seem transferable to Cartagena if political support for policy implementation is provided. It is necessary nevertheless to also address the drivers of sexual tourism - financially vulnerable individuals will still not be in a position to refuse the financial incentive of unsafe sex if their basic needs are not met. Our participants suggested community-based work in combination with mass media interventions to overcome barriers linked with condom perceptions. This approach proved efficient in increasing condom use and changing perceptions in Colombia (Vernon et al., 1990). In other Caribbean and Latin American countries, mass media interventions combined with community work had an impact on social norms, improved knowledge and awareness of the youth and increased the use of condoms and health services (UNAIDS, 2006). Brazil is a great example of behavioural change with massive media campaigns including messages tailored to each population group targeted by the government (Berkman et al., 2005). Participants also advocated for the

introduction of gender equity awareness in schools considering gender and social norms are internalised early in life. Other authors agree (Pulerwitz et al., 2010) and a study conducted in Latin American and Caribbean countries (Demaria et al., 2009) corroborate the success of sexual education programmes in school that incorporates gender equity themes. Those programmes should, however, be backed by legislation and implemented by the Ministry of Education. This could prove a difficult task, given the lack of involvement from the education sector in the national plan's implementation. Stigma and discrimination were also identified as implementation obstacles in Cartagena. Although, as discussed earlier, mass media campaigns had an impact on reducing stigma and discrimination, attention should be given to how the messages are delivered. In Cartagena, population exposed to public health interventions create their own “bricolage” by incorporating familiar elements of the interventions to their own frames of reference (Quevedo-Gomez et al., 2012). Therefore they might incorporate the “risk groups” conception of public health to their own understandings and associate it with sexuality-related stigma, creating the concept of “AIDS carriers”. In Brazil, universal free access to antiretroviral therapy had an impressive impact on reducing stigma and discrimination. People's understandings changed: as patients looked healthier thanks to medication, HIV/ AIDS was interpreted as a manageable chronic disease instead of a death sentence (Abadía-Barrero and Castro, 2006). Similar findings were denoted in the Dominican Republic and Haiti (Castro and Farmer, 2005). Introducing free treatment in Cartagena, at least for those with low SES, would likely yield similar results and, more importantly, would help reduce social inequalities in health. In fact, the results of the study demonstrate that barriers to healthcare access potentially become barriers to policy implementation. In particular, the health system fragmentation led to a paperwork maze difficult for most people to navigate through. Such challenges have been previously described in the literature as “Bureaucratic Itineraries”, entailing the citizens' struggles to access care and the detrimental consequences on their lives (Abadia and Oviedo, 2009). As a result, access to healthcare is dependent on one's capability to deal with this maze or one's ability to pay, rather than one's need for healthcare (Abadia and Oviedo, 2009; Arrivillaga et al., 2009). A new law to reform the healthcare system in Colombia was adopted in June 2013, however it is still premature to predict if its application will improve healthcare access and have an impact on HIV/AIDS policy implementation. As a product of this study, global influences were added to the CIT framework. Our findings show the effects of these influences in mediating policymaking efforts and financing projects at the local scale. The impact of global influences on local implementation substantiates previous work on the CIT that described how the three actors' characteristics are also influenced by the multilayered external context (De Boer, 2012). In the case of Cartagena, global influences are the external factor most greatly impacting local implementation, and thus, were deemed relevant as part of the CIT framework. However, some participants pointed out that HIV prevention activities are dependent on international donors. Since international aid in Colombia is decreasing, the question to raise is: will the government increase its budget allocated to HIV/ AIDS? If not, how could Cartagena tackle its epidemic? Furthermore, while some of the global initiatives contribute financing to local projects that change the lives of few; other global initiatives such as the World Bank adjustment programmes set at risk the lives of many. Indeed, the fragmentation of the healthcare system is the result of recommendations given by the World Bank in the nineties, leading Colombia to fully reform its health system to a managedcare system model, with detrimental consequences for many


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

Colombians (Abadia and Oviedo, 2009; Molina and Spurgeon, 2007). Stemming from these recommendations, Colombia adopted a decentralised system, which also gave more autonomy to local corrupt politicians and decision makers. Corruption and its consequences are another considerable hindrance to policy implementation. Two other studies (Molina and Spurgeon, 2007; Vargas and Molina, 2009) support our findings regarding corruption and staff rotation. However, more research on corruption in the health sector is needed to identify ways to tackle this issue. This study presents some limitations. Although the authors carried out data and method triangulations, investigator triangulation was not performed as only the first author coded the data. Due to time and access constraints, implementers in healthcare insurance companies and service provider institutions; as well as members of the Church, education sector representatives and local politicians could not be interviewed. Given the findings, their perspective on the matter is needed; thus, they should be included in future research. For example, the education sector is reluctant in implementing sexual education programmes. Marrugo et al. (2004) further described the following hindrances encountered in such implementation on the Colombian Caribbean Coast: gap between formulation and implementation, little interest expressed from educational institutions and low resources. On the other hand, our findings revealed the importance of the Catholic Church in HIV policy implementation; unfortunately, little research was conducted on its role in HIV policy implementation in Latin America, except in Brazil. Surprisingly, the Brazilian government response was well-articulated amongst the Church's own network in the country, possibly because many priests were caught between the Church's ideologies and the reality of HIV devastating their communities (García et al., 2009; Murray et al., 2011). However, the Catholic Church in Brazil is decentralised, allowing for more discretion and tolerance (especially for condom use) at the local level compared to the macro level (Murray et al., 2011). Further research on ways to achieve cooperation in Cartagena as well is also needed. Regarding the Colombian situation, the gap between health policy formulation and implementation stems from a lack of inclusion of all stakeholders in the early stages of policymaking. As a result, the national plan does not take into account local specificities. Consequently, implementation in Cartagena, one of Colombia's largest cities, fails to curb the HIV epidemic. Participants advocated for the need to include all stakeholders at all levels of governance during policy formulation to identify feasible actions and to allocate funding accordingly. By involving all actors during formulation and giving them the opportunity to be actively and democratically involved, there is a possibility not only to overcome the implementation gap, but also to get uncooperative stakeholders interested in curbing the HIV epidemic.

5. Conclusions The participants have put forward several solutions to the barriers of implementation identified (such as corruption, stigma and discrimination, lack of collaboration from some key stakeholders). Some solutions they suggested have proven to be effective in other Latin American and Caribbean countries. However, barriers identified in this study are all intertwined; therefore, a complex, multilevel approach, with active participation of all stakeholders, is needed to overcome them. Finally, any effort to overcome barriers to policy implementation should also be accompanied by efforts to tackle the social determinants of HIV and AIDS in Cartagena, stemming from large social inequalities present in this context.

43

Acknowledgements The authors would like to thank all the participants. Their contribution was greatly appreciated and made this research possible. The authors also thank Professor Hans Maarse for his insights and guidance. References Abadia, C.E., Oviedo, D.G., 2009. Bureaucratic Itineraries in Colombia. A theoretical and methodological tool to assess managed-care health care systems. Soc. Sci. Med. 68 (6), 1153e1160. Abadía-Barrero, C.E., Castro, A., 2006. Experiences of stigma and access to HAART in children and adolescents living with HIV/AIDS in Brazil. Soc. Sci. Med. 62 (5), 1219e1228. Ainsworth, M., Beyrer, C., Soucat, A., 2003. AIDS and public policy: the lessons and challenges of “success” in Thailand. Health Policy 64 (1), 13e37. Arrivillaga, M., Ross, M., Useche, B., Alzate, M.L., Correa, D., 2009. Social position, gender role, and treatment adherence among Colombian women living with HIV/AIDS: social determinants of health approach. Pan Am. J. Public Health 26 (6), 502e510. ~ oz-Laboy, M., Paiva, V., Parker, R., 2005. A critical Berkman, A., Garcia, J., Mun analysis of the Brazilian response to HIV/AIDS: lessons learned for controlling and mitigating the epidemic in developing countries. Am. J. Public Health 95 (7), 1162e1172. Bressers, H., 1983. Beleidseffectiviteit en waterkwaliteitsbeleid. Universiteit Twente, Enschede. Bressers, H., 2004. Implementing sustainable development: how to know what works, where, when and how. In: Lafferty, W.M. (Ed.), Governance for Sustainable Development: The Challenge of Adapting Form to Function. Edward Elgar, Cheltenham, pp. 284e318. Cabezas, A.L., 2009. Economies of Desire. Sex and Tourism in Cuba and the Dominican Republic. Temple University Press, Philadelphia (PA). Castro, A., Farmer, P., 2005. Understanding and addressing AIDS-related stigma: from anthropological theory to clinical practice in Haiti. Am. J. Public Health 95, 53e59. De Boer, C., Bressers, H., 2011. Complex and Dynamic Implementation Processes. Universiteit Twente, Enschede. De Boer, C., 2012. Contextual Water Management. A Study of Governance and Implementation Processes in Local Stream Restoration Projects. Universiteit Twente, Enschede.
rraga, O., Campero, L., Walker, D.M., 2009. Educacio
n sobre Demaria, L.M., Gala
n del VIH: un diagno
stico para Ame
rica Latina y el sexualidad y prevencio Caribe. Rev. Panam. Salud Pública 26 (6), 485e493. García, J., Street, W., York, N., Parker, R., 2009. Local impacts of religious discourses on rights to express same-sex sexual desires in Peri-Urban Rio de Janeiro. Sex. Res. Soc. Policy 6 (3), 44e60. Grenade, W., 2008. Balancing economic development and security: tourism and HIV/AIDS (Grenada). Soc. Econ. Stud. 57 (2), 27e60. Haddock, S., 2007. Policy empowers: condom use among sex workers in the Dominican Republic. Popul. Action Int. 2 (1).
rez-Then, E., Pappas, G., Garcia Calleja, J.M., 2009. Halperin, D.T., de Moya, E.A., Pe Understanding the HIV epidemic in the Dominican Republic: a prevention success story in the Caribbean? J. Acquir. Immune Defic. Syndr. 51 (Suppl. 1), S52eS59. HEU Centre for Health Economics, 2009. Poverty & HIV/AIDS in the Caribbean. CARICOM & PANCAP, St. Augustine (Guyana). Homedes, N., Ugalde, A., 2005. Why neoliberal heath reforms have failed in Latin America. Health Policy 71 (1), 83e96. Inciardi, J., 2005. HIV/AIDS in the Caribbean Basin. AIDS Care 17, 9e25. Marrugo, M., Reason, F., Daguer, D., De La Hoz, C., Bolivar, L., 2004. Salud sexual y
stico en la costa Caribe Colombiana. Rev. Psicogente 12, reproductiva: diagno 35e44. McDonald, S., Roberts, J., 2006. AIDS and economic growth: a human capital approach. J. Dev. Econ. 80 (1), 228e250.
n Social, 2008. Plan Nacional de Respuesta ante el VIH y el Ministerio de la Proteccio
n Social & ONUSIDA, SIDA Colombia 2008-2011. Ministerio de la Proteccio
. Bogota
n Social, 2012. Resumen de situacio
n de la epidemia Ministerio de Salud y Proteccio por VIH/SIDA en Colombia 1983 a 2011. Observatorio Nacional en VIH/SIDA,
. Bogota Molina, M.G., Spurgeon, M., 2007. La decentralizacion del sector salud en Colombia. Gest. Polit. Publica 16 (1), 171e202.

Moreno, A., Alvarez-rosete, A., Moreno, C., Rodriguez-garcía, R., Moreno, L.A., Gaillard, M.E., 2012. Evidence-based Implementation Efficiency Analysis of the HIV/AIDS National Response in Colombia. The World Bank, Washington (DC). ~ oz-laboy, M., Parker, G., 2011. Strange bedfellows: the Murray, L.R., Garcia, J., Mun catholic church and Brazilian national AIDS program in the response to HIV/ AIDS in Brazil. Soc. Sci. Med. 72 (6), 945e952. O'toole, L.J., 2004. The theory-practice issue in policy implementation research. Public Adm. 82 (2), 309e329. Padilla, M.B., Guilamo-Ramos, V., Bouris, A., Reyes, A.M., 2010. HIV/AIDS and

44


mez / Social Science & Medicine 133 (2015) 36e44 N. Djellouli, M.C. Quevedo-Go

tourism in the Caribbean: an ecological systems perspective. Am. J. Public Health 100 (1), 70e77. Pan-Caribbean Partnership on HIV/AIDS (PANCAP), 2008. Caribbean Regional Strategic Framework for HIV/AIDS 2008-2012. CARICOM, Georgetown (Guyana). Perez, G.J., Mejia, I.S., 2007. La pobreza en Cartagena: un analisis por barrios. Banco de la República e Sucursal Cartagena, Cartagena de Indias. Pulerwitz, J., Michaelis, A., Verma, R., Weiss, E., 2010. Addressing gender dynamics and engaging men in HIV programs: lessons learned from horizons research. Public Health Rep. 125 (2), 282e292. Quevedo-Gomez, M.C., Krumeich, A., Abadia-Barrero, C., Pastrana-Salcedo, E., van den Borne, H., 2011. Structural actions toward HIV/AIDS prevention in Cartegena, Colombia: a qualitative study. Pan Am. J. Public Health 30 (1), 65e73. Quevedo-Gomez, M.C., Krumeich, A., Abadia-Barrero, C., Pastrana-Salcedo, E., van den Borne, H., 2012. Machismo, public health and sexuality-related stigma in Cartagena. Cult. Health Sex. 14 (2), 223e235.

Rojas, P., Malow, R., Ruffin, B., Rothe, E.M., Rosenberg, R., 2011. The HIV/AIDS epidemic in the Dominican Republic: key contributing factors. J. Int. Assoc. Physicians AIDS Care 10 (5), 306e315. Spratt, K., 2009. Policy Implementation Barriers Analysis: Conceptual Framework and Pilot Test in Three Countries. Futures Group, Health Policy Initiative, Task Order 1, Washington (DC). The Lancet, 2008. HIV/AIDS in Latin America and the Caribbean (editorial). Lancet 372 (9635), 263. UNAIDS, 2006. Preventing HIV/AIDS in Young People. A Systematic Review of the Evidence from Developing Countries, vol. 21. WHO, Geneva. Vargas, J., Molina, G., 2009. Acceso a los servicios de salud en seis cuidades de Colombia: limitaciones y consecuencias. Rev. Fac. Nac. Salud Publica 27 (2), 121e130. Vernon, R., Ojeda, G., Murad, R., 1990. Incorporating AIDS prevention activities into a family planning organization in Colombia. Stud. Fam. Plan. 21 (6), 335e343.