University of Malaya, Kuala Lumpur, Malaysia(Prof A Kamarulzaman); UN Special Envoy, HIV in Eastern Europe and Central Asia, Geneva, Switzerland (Prof. M Kazatchkine); Columbia University, New York City, USA (J. Csete PhD, Prof. C Hart); Yale University, New Haven CT, USA (Prof. F. Altice); Warsaw, Poland (M. Balicki); Central European University, Budapest, Hungary (Prof. J. Buxton); Center for Public Health and Human Rights, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA. (J. Cepeda, PhD, Prof. S. Sherman, Prof. C. Beyrer); RTI International, Washington, DC, USA (M. Comfort PhD); University of California – San Francisco, San Francisco, California, USA (Prof. E. Goosby); Ministry of Health, Lisbon, Portugal (J. Goulão); University of British Columbia, Center of Excellence in HIV/AIDS, Vancouver, BC, Canada (Prof. T. Kerr); Centro de Investigación y Docencia Económicas (CIDE), Mexico City, Mexico (Prof. A. M. Lajous); The Lancet, London, UK (R. Horton FMedSci); AIDS-Free World, Toronto, Canada (Prof. S. Lewis); University of California – San Diego, San Diego, California USA (N. Martin, DPhil); University of the Andes, Bogotá, Colombia (Prof. D. Mejía); Human Rights Watch, Yangon, Myanmar (D. Mathiesson); University of Uyo, Uyo, Nigeria (Prof. I Obot); Youth Rise—Nigeria, Lagos, Nigeria (A. Ogunrombi); Trivandrum Institute of Palliative Sciences, Trivandrum, India (N. Vallath, MBBS); University of Bristol, UK (P. Vickerman, PhD, J. Stone, MMathStat); Charles University, Prague, Czech Republic (Prof. T. Zábranský)
Corresponding Author: Chris Beyrer MD, MPH, Desmond M. Tutu Professor of Public Health and Human Rights, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe St, E 7152, Baltimore, MD, USA, 21205. Tel: 410 614 5247, Fax: 410 614 8371; ude.uhj@reryebc
The publisher's final edited version of this article is available at LancetIn September 2015, the member states of the United Nations endorsed sustainable development goals (SDG) for 2030 that aspire to human rights-centered approaches to ensuring the health and well-being of all people. The SDGs embody both the UN Charter values of rights and justice for all and the responsibility of states to rely on the best scientific evidence as they seek to better humankind. In April 2016, these same states will consider control of illicit drugs, an area of social policy that has been fraught with controversy, seen as inconsistent with human rights norms, and for which scientific evidence and public health approaches have arguably played too limited a role.
The previous UN General Assembly Special Session (UNGASS) on drugs in 1998 – convened under the theme “a drug-free world, we can do it!” – endorsed drug control policies based on the goal of prohibiting all use, possession, production, and trafficking of illicit drugs. This goal is enshrined in national law in many countries. In pronouncing drugs a “grave threat to the health and well-being of all mankind,” the 1998 UNGASS echoed the foundational 1961 convention of the international drug control regime, which justified eliminating the “evil” of drugs in the name of “the health and welfare of mankind.” But neither of these international agreements refers to the ways in which pursuing drug prohibition itself might affect public health. The “war on drugs” and “zero-tolerance” policies that grew out of the prohibitionist consensus are now being challenged on multiple fronts, including their health, human rights, and development impact.
The Johns Hopkins – Lancet Commission on Drug Policy and Health has sought to examine the emerging scientific evidence on public health issues arising from drug control policy and to inform and encourage a central focus on public health evidence and outcomes in drug policy debates, such as the important deliberations of the 2016 UNGASS on drugs.
The Johns Hopkins-Lancet Commission is concerned that drug policies are often colored by ideas about drug use and drug dependence that are not scientifically grounded. The 1998 UNGASS declaration, for example, like the UN drug conventions and many national drug laws, does not distinguish between drug use and drug abuse. A 2015 report by the UN High Commissioner for Human Rights, by contrast, found it important to emphasize that “[d]rug use is neither a medical condition nor does it necessarily lead to drug dependence.” The idea that all drug use is dangerous and evil has led to enforcement-heavy policies and has made it difficult to see potentially dangerous drugs in the same light as potentially dangerous foods, tobacco, alcohol for which the goal of social policy is to reduce potential harms.
The pursuit of drug prohibition has generated a parallel economy run by criminal networks. Both these networks, which resort to violence to protect their markets, and the police and sometimes military or paramilitary forces that pursue them contribute to violence and insecurity in communities affected by drug transit and sales. In Mexico, the dramatic increase in homicides since the government decided to use military forces against drug traffickers in 2006 has been so great that it reduced life expectancy in the country.
Injection of drugs with contaminated equipment is a well-known route of HIV exposure and viral hepatitis transmission. People who inject drugs (PWID) are also at high risk of tuberculosis. The continued spread of unsafe injection-linked HIV contrasts the progress that has been seen in reducing sexual and vertical transmission of HIV in the last three decades. The Commission found that that repressive drug policing greatly contributes to the risk of HIV linked to injection. Policing may be a direct barrier to services such as needle and syringe programmes (NSP) and use of non-injected opioids to treat dependence among those who inject opioids, known as opioid substitution therapy (OST). Police seeking to boost arrest totals have been found to target facilities that provide these services to find, harass, and detain large numbers of people who use drugs. Drug paraphernalia laws that prohibit possession of injecting equipment lead PWID to fear carrying syringes and force them to share equipment or dispose of it unsafely. Policing practices undertaken in the name of the public good have demonstrably worsened public health outcomes.
Amongst the most significant impacts of pursuit of drug prohibition identified by the Commission with respect to infectious disease is the excessive use of incarceration as a drug-control measure. Many national laws impose lengthy custodial sentences for minor, non-violent drug offenses; people who use drugs (PWUD) are over-represented in prison and pretrial detention. Drug use and drug injection occur in prisons, though their occurrence is often denied by officials. HIV and hepatitis C virus (HCV) transmission occurs among prisoners and detainees, often complicated by co-infection with TB and in many places multidrug-resistant TB, and too few states offer prevention or treatment services in spite of international guidelines that urge comprehensive measures, including provision of injection equipment, for people in state custody.
Mathematical modelling undertaken by the Commission illustrates that incarceration and high HCV risk in the post-incarceration period can contribute importantly to national HCV incidence amongst PWID in a range of countries with varying levels of incarceration, different average prison sentences, durations of injection, and OST coverage levels in prison and following release. For example, in Thailand where PWID may spend nearly half their injection careers in prison, an estimated 63% of incident HCV infection could occur in prison. In Scotland, where prison sentences are shorter for PWUD and OST coverage is relatively high in prison, an estimated 54% of incident HCV infection occurs in prison, but as much as 21% may occur in the high-risk post-release period. These results underscore the importance of alternatives to prison for minor drug offences, ensuring access to OST in prison, and a seamless link from prison services to OST in the community.
The evidence also clearly demonstrates that drug law enforcement has been applied in a discriminatory way against racial and ethnic minorities in a number of countries. The US is perhaps the best documented but not the only case of racial biases in policing, arrest, and sentencing. In 2014, African American men were more than five times more likely than whites to be incarcerated in their lifetime, though there is no significant difference in rates of drug use among these populations. The impact of this bias on communities of people of color is inter-generational and socially and economically devastating.
The Commission also found significant gender biases in current drug policies. Of women in prison and pretrial detention around the world, a higher percentage are detained because of drug infractions than is the case for men. Women involved in drug markets are often on the bottom rungs – as couriers or drivers – and may not have information about major traffickers to trade as leverage with prosecutors. Gender and racial biases have marked overlap, making this an intersectional threat to women of color, their children, families, and communities.
In both prison and the community, HIV, HCV and TB programmes for PWUD – including testing, prevention and treatment – are gravely underfunded at the cost of preventable death and disease. In a number of middle-income countries where large numbers of PWUD live, HIV and TB programmes for PWUD that were expanded with support from the Global Fund to Fight AIDS, TB and Malaria have lost funding due to changes in the Fund’s eligibility criteria. There is an unfortunate failure to emulate the example of Western European countries that have eliminated unsafe injection-linked HIV as a public health problem by sustainably scaling up prevention and care and enabling minor offenders to avert prison. Political resistance to harm reduction measures dismisses strong evidence of their effectiveness and cost-effectiveness. Mathematical modeling shows that if OST, NSP and antiretroviral therapy for HIV are all available, even if the coverage of each of them is not over 50%, their synergy can lead to effective prevention in a foreseeable future. PWUD are often not seen to be worthy of costly treatments, or they are thought not to be able to adhere to treatment regimens in spite of evidence to the contrary.
Lethal drug overdose is an important public health problem, particularly in light of rising consumption of heroin and prescription opioids in some parts of the world. Yet the Commission found that the pursuit of drug prohibition can contribute to overdose risks in numerous ways. It creates unregulated illegal markets in which it is impossible to control adulterants of street drugs that add to overdose risk. Several studies also link aggressive policing to rushed injection and overdose risk. People with a history of drug use, over-represented in prison because of prohibitionist policies, are at extremely high risk of overdose when released from state custody. Lack of ready access to OST also contributes to injection of opioids, and bans on supervised injection sites cut off an intervention that has proven very effective in reducing overdose deaths. Restrictive drug policies also contribute to unnecessary controls on naloxone, a medicine that can reverse overdose very effectively.
Though a small percentage of PWUD will ever need treatment for drug dependence, that minority faces enormous barriers to humane and affordable treatment in many countries. There are often no national standards for quality of drug dependence treatment and no regular monitoring of practices. In too many countries, beatings, forced labor, and denial of health care and adequate sanitation are offered in the name of treatment, including in compulsory detention centres that are more like prisons than treatment facilities. Where there are humane treatment options, it is often the case that those most in need of it cannot afford it. In many countries, there is no treatment designed particularly for women, though it is known that women’s motivations for and physiological reactions to drug use differ from those of men.
The pursuit of the elimination of drugs has led to aggressive and harmful practices targeting people who grow crops used in the manufacture of drugs, especially coca leaf, opium poppy, and cannabis. Aerial spraying of coca fields in the Andes with the defoliant glyphosate (N-(phosphonomethyl glycine) has been associated with respiratory and dermatological disorders and with miscarriages. Forced displacement of poor rural families who have no secure land tenure exacerbates their poverty and food insecurity and in some cases forces them to move their cultivation to more marginal land. Geographic isolation makes it difficult for state authorities to reach drug crop cultivators in public health and education campaigns and it cuts cultivators off from basic health services. Alternative development programmes meant to offer other livelihood opportunities have poor records and have rarely been conceived, implemented, or evaluated with respect to their impact on people’s health.
Research on drugs and drug policy has suffered from the lack of a diversified funding base and assumptions about drug use and drug pathologies on the part of the dominant funder, the US government. At a time when drug policy discussions are opening up around the world, there is an urgent to bring the best of non-ideologically-driven health science, social science and policy analysis to the study of drugs and the potential for policy reform.
Concrete experiences from many countries that have modified or rejected prohibitionist approaches in their response to drugs can inform discussions of drug policy reform. A number of countries, such as Portugal and the Czech Republic, decriminalised minor drug offenses years ago, with significant savings of money, less incarceration, significant public health benefits, and no significant increase in drug use. Decriminalisation of minor offenses along with scaling up low-threshold HIV prevention services enabled Portugal to control an explosive unsafe injection-linked HIV epidemic and likely enabled the Czech Republic to prevent one from happening.
Where formal decriminalisation may not be an immediate possibility, scaling up health services for PWUD can demonstrate the value to society of responding with support rather than punishment to people who commit minor drug infractions. A pioneering OST program in Tanzania is encouraging communities and officials to consider non-criminal responses to heroin injection. In Switzerland and the city of Vancouver, Canada, dramatic improvements in access to comprehensive harm reduction services, including supervised injection sites and heroin-assisted treatment, transformed the health picture for PWUD. Vancouver’s experience also illustrates the importance of meaningful participation of PWUD in decision-making on policies and programmes affecting their communities.
Policies meant to prohibit or greatly suppress drugs present a paradox. They are portrayed and defended vigorously by many policy-makers as necessary to preserve public health and safety, and yet the evidence suggests they have contributed directly and indirectly to lethal violence, communicable disease transmission, discrimination, forced displacement, unnecessary physical pain, and the undermining of people’s right to health.
Some would argue that the threat of drugs to society may justify some level of abrogation of human rights for protection of collective security, as is also foreseen by human rights law in case of emergencies. International human rights standards dictate that in such cases, societies still must choose the least harmful way to address the emergency and that emergency measures must be proportionate and designed specifically to meet transparently defined and realistic goals. The pursuit of drug prohibition meets none of these criteria.
Standard public health and scientific approaches that should be part of policy-making on drugs have been rejected in the pursuit of prohibition. The idea of reducing the harm of many kinds of human behavior is central to public policy in the areas of traffic safety, tobacco and alcohol regulation, food safety, safety in sports and recreation, and many other areas of human life where the behavior in question is not prohibited. But explicitly seeking to reduce drug-related harms through policy and programmes and to balance prohibition with harm reduction is regularly resisted in drug control. The persistence of unsafe injection-linked HIV and HCV transmission that could be stopped with proven, cost-effective measures remains one of the great failures of the global responses to these diseases.
Drug policy that is dismissive of extensive evidence of its own negative impact and of approaches that could improve health outcomes is bad for all concerned. Countries have failed to recognise and correct the health and human rights harms that pursuit of prohibition and drug suppression have caused and in so doing neglect their legal responsibilities. They readily incarcerate people for minor offenses but then neglect their duty to provide health services in custodial settings. They recognize uncontrolled illegal markets as the consequence of their policies, but they do little to protect people from toxic, adulterated drugs that are inevitable in illegal markets or the violence of organized criminals, often made worse by policing. They waste public resources on policies that do not demonstrably impede the functioning of drug markets, and they miss opportunities to invest public resources wisely in proven health services for people often too frightened to seek services.
To move toward the balanced policy that UN member states have called for, we offer the following recommendations:
Decriminalisation: Decriminalise minor, non-violent drug offenses – use, possession, and petty sale – and strengthen health and social-sector alternatives to criminal sanctions.
Reducing violence and discrimination in policing: Reduce the violence and other harms of drug policing, including phasing out the use of military forces in drug policing, better targeting of policing on the most violent armed criminals, allowing possession of syringes, not targeting harm reduction services to boost arrest totals, and eliminating racial and ethnic discrimination in policing.
Reducing harms: Ensure easy access for all who need them to harm reduction services as a part of responding to drugs, recognizing the effectiveness and cost-effectiveness of scaling up and sustaining these services. OST, NSP, supervised injection sites, and access to naloxone – brought to a scale adequate to meet demand – should all figure in health services and should include meaningful participation of PWUD in planning and implementation. Harm reduction services are crucial in prison and pretrial detention and should be scaled up in these settings. The 2016 UNGASS should do better than the UN Commission on Narcotic Drugs (CND) in naming harm reduction explicitly and endorsing its centrality to drug policy.
Treatment and care for PWUD: Prioritize PWUD in treatment for HIV, HCV, TB, and ensure that services are adequate to ensure access for all who need care. Ensure availability of humane and scientifically sound treatment for drug dependence, including scaled-up OST in the community as well as in prisons, rejecting compulsory detention and abuse in the name of treatment.
Access to controlled medicines: Ensure access to controlled medicines, establishing inter-sectoral national authorities to determine levels of need and giving the World Health Organization (WHO) the resources to assist the International Narcotics Control Board (INCB) in using the best science to determine the level of need for controlled medicines in all countries.
Gender-responsive policies: Reduce the negative impact of drug policy and law on women and their families, especially minimizing custodial sentences for women who commit non-violent offenses and developing appropriate health and social support, including gender-appropriate treatment of drug dependence, for those who need it.
Crop production: Efforts to address drug crop production must take health into account. Aerial spraying of toxic herbicides should be stopped, and alternative development programmes should be part of integrated development strategies, developed and implemented in meaningful consultation with the people affected.
Improve research: There is a need for a more diverse donor base to fund the best new science on drug policy experiences in a non-ideological way that, among other things, interrogates and moves beyond the excessive pathologising of drug use.
UN governance of drug control: UN governance of drug policy must be improved, including by respecting WHO’s authority to determine the dangerousness of drugs. Countries should be urged to include high-level health officials in their delegations to CND. Improved representation of health officials in national delegations to CND would, in turn, be a likely result of giving health authorities an important day-to-day role in multi-sectoral national drug policy-making bodies.
Better metrics: Health, development, and human rights indicators should be included in metrics to judge success of drug policy; WHO and UNDP should help formulate them. UNDP has already suggested that indicators such as access to treatment, rate of overdose deaths, and access to social welfare programmes for people who use drugs would be useful indicators. All drug policies should also be monitored and evaluated as to their impact on racial and ethnic minorities, women, children and young people, and people living in poverty.
Scientific approach to regulated markets: Move gradually toward regulated drug markets and apply the scientific method to their evaluation. While regulated legal drug markets are not politically possible in the short term in some places, the harms of criminal markets and other consequences of prohibition catalogued in this report are likely to lead more countries (and more US states) to move gradually in that direction, a direction we endorse. As those decisions are taken, we urge governments and researchers to apply the scientific method and ensure independent, multidisciplinary and rigorous evaluation of regulated markets to draw lessons and inform improvements in regulatory practices, and to continue evaluating and improving.
We urge health professionals in all countries to inform themselves and join debates on drug policy at all levels. True to the stated goals of the international drug control regime, it is possible to have drug policy that contributes to the health and well-being of humankind, but not without bringing to bear the evidence of the health sciences and the voices of health professionals.
Keywords: drug policy, prohibition, harm reduction, violence, crop eradication, HIV, viral hepatitis, mass incarceration, decriminalization, reform, public health
“We must consider alternatives to criminalization and incarceration of people who use drugs and focus criminal justice efforts on those involved in supply. We should increase the focus on public health, prevention, treatment and care, as well as on economic, social and cultural strategies.”
–Ban Ki-moon, UN Secretary-General, on World Day Against Drugs, June 26, 2015 1
In 2015, member states of the United Nations in the presence of more than 150 heads of state endorsed a set of Sustainable Development Goals (SDGs) that were formulated to embody founding principles of the UN, including universal human rights and justice for all. 2 The SDG resolution commits member states to addressing climate change and other large issues in ways that are informed by the best scientific research. The SDGs are also based on a notion of human security that is not confined to traditional public order authorities but in which health and social sectors play an important part. 2
In April 2016, the same UN member states in a General Assembly Special Session (UNGASS) will take on a social policy challenge that affects millions of lives – what the UN has called the “world drug problem”. As with the SDGs, addressing the use, production, and trafficking of drugs will challenge the UN to base its policies on the human rights norms that are the bedrock of the UN Charter as well as the best scientific evidence at hand. This challenge is significant because it is not only the “world drug problem” but also policy responses to drugs that can negatively affect human lives and human rights and contradict evidence-based public health approaches. As noted by former UN Secretary-General Kofi Annan, “…Drugs have destroyed many people, but wrong policies have destroyed many more.” 3
A 2015 report of the UN High Commissioner for Human Rights highlights some of the main ways in which drug control policies have been responsible for violations of human rights. 4 The High Commissioner concluded that drug policies, law, and law enforcement have resulted in arbitrary arrest, detention and ill treatment of people who use drugs; unjust use of the death penalty for drug offenses; cruel and inhuman treatment of PWUD in the guise of “treatment”; racial and ethnic discrimination in drug law enforcement; denial of life-saving care and prevention interventions to people who use drugs; excessive use of incarceration as a response to minor drug infractions; denial of the cultural rights of indigenous peoples; and poor access to opioids and other controlled medicines for pain management and other clinical uses, among other human rights violations.
The last General Assembly Special Session on drugs in 1998, under the theme “a drug-free world – we can do it!”, endorsed drug control policies based on the idea of elimination or prohibition of all use, possession, production, and trafficking of illicit drugs. 5 This idea is embodied in national law in many countries. The 1998 UNGASS declaration pronounced drugs a “grave threat to the health and well-being of all mankind.” 5 In this pronouncement it echoed the bedrock treaty of the global drug control regime, the widely ratified 1961 Single Convention on Narcotic Drugs, which states in its preamble that drug control is motivated principally by concern for “the health and welfare of mankind.” 6 Neither of these international agreements, however, refers to the negative health consequences of pursuing drug prohibition. The time is long overdue for a review of the health impacts of these drug policies. The disconnect between drug control policy and health outcomes is no longer tenable or credible.
The Johns Hopkins – Lancet Commission on Drug Policy and Health has sought to examine the scientific evidence on a broad range of public health issues arising from drug control policy to inform a focus on public health as a central consideration in drug policy discussions such as the important deliberations of the 2016 UNGASS.
The JHU-Lancet Commission is motivated partly by a concern that drug policies are often founded on ideas about drug use and drug dependence that are not scientifically grounded. Like the Single Convention, the declaration from the 1998 UNGASS on drugs, for example, does not distinguish between drug use and drug abuse; all use is referred to as abuse. 5 Indicating some evolution of thinking in the UN if not amongst member states, the UN High Commissioner for Human Rights in his 2015 report, by contrast, found it important to emphasize that “[d]rug use is neither a medical condition nor does it necessarily lead to drug dependence” or loss of dignity. 4 The UNODC 2015 annual report concluded that of an estimated 246 million people who used an illicit drug in the past year, some 27 million, or 11% experienced problem drug use, defined as drug dependence or drug use disorders. 7 The idea that all drug use is dangerous and evil has made it difficult to see potentially dangerous drugs in the same light as potentially dangerous foods, tobacco, alcohol, and other substances for which the goal of social policy is to reduce harms. Harm reduction, an essential element of public health policy, has too often been lost in drug policy-making amidst a dominant discourse on the overwhelming evil of drugs.
We hope that this review and analysis of evidence on the health consequences of pursuing prohibition of drugs and drug use can inform rights-based policy change. Because language is important to drug policy discussions, we include as an annex to this report a glossary of some policy-relevant terms.
The international drug control system itself has its origins in decades-old legal instruments framed by politics more than science. From the time of the 1912 Hague Opium Convention, minimizing the supply of some psychoactive drugs through policing has been the dominant strain in international drug law. 8 In the decades leading up to the 1961 Single Convention, international drug control agreements largely sidestepped issues of demand and consumption. 9 The eventual solution in the 1961 Single Convention to reserve some quantity of psychoactive substances for medical and scientific use did not resolve the issue of social, cultural, and recreational use that was not obviously harmful but was not “medical or scientific.” 9
In 1998 when the UN members states declared their commitment to a drug-free world, the UN estimated that there were an estimated 8 million people in the world who used heroin in the previous year, about 13 million who used cocaine, about 30 million who used amphetamine-type substances (ATS), and over 135 million “abusers” – that is, users – of cannabis. 10 When the countries came together after 10 years to review progress toward a “drug-free world” in 2008, the UN estimated that 12 million people used heroin, 16 million used cocaine, almost 34 million used ATS, and over 165 million used cannabis in the previous year. 11 The worldwide area under opium poppy cultivation was estimated at about 238,000 ha in 1998, and in 2008 was 235,700 ha, a small decline. 11 Prohibition as a policy had clearly failed.
In the 2014 statement from the high-level segment of the UN Commission on Narcotic Drugs (CND, the UN’s drug policy body), an important background document for the 2016 UNGASS, UN member states stressed the importance of drug policy that is consistent with human rights and acknowledged that “law enforcement measures alone” cannot achieve drug control. 12
In the lead-up to the 2016 UNGASS, UN agencies were asked to make statements about how drug-control policy intersects with their mandates and affects their work. These statements signal that high-level thinking in a number of UN agencies reflects some impatience with the pursuit of prohibition. The long list of human rights violations associated with drug-control measures led the High Commissioner for Human Rights to call for member states to consider “removing obstacles to the right to health, including by decriminalizing the personal use and possession of drugs.” 4 The UN Development Programme welcomed a change away from the dominant “prohibitionist, law enforcement-led and abstinence-based approach.” 13 The WHO executive board called for a stronger focus on prevention of drug use and treatment and care of PWUD in drug policy as well as on reducing the harms of drugs and drug use. 14
Outside the UN, debates have also evolved, including in regional bodies such as the Organization of American States (OAU) and the European Union. A 2013 OAU report raised the idea that dramatic departures from prohibition-based drug control might be the only way to reduce drug-related violence and criminality in the Americas. 15 The Global Commission on Drug Policy (GCDP) including a number of former heads of state and other prominent figures; the Latin American Commission on Drugs and Democracy, headed by former Brazilian president Ferdinand Cardoso; and the West Africa Commission on Drugs, convened by former UN Secretary-General Kofi Annan, all called for an end to strict prohibition-oriented policies and for decriminalisation of minor drug infractions, among other recommendations. 3,16,17 The Global Commission on HIV and the Law, convened by UNDP, also included former heads of state and other high-level officials, called on national authorities to “decriminalise the possession of drugs for personal use, in recognition that the net impact of such sanctions is often harmful to society” and to give priority to public health considerations in drug policy. 18 By contrast, the Association of Southeast Asian Nations (ASEAN) agreed on an objective of a “drug-free ASEAN” by 2015, 19 in spite of widespread criticism of the unrealistic nature of the goal. 20
North America continues to have by far the highest rates of drug consumption and drug-related death and morbidity of any region in the world, 7 and drug policy in this region tend to influence global debates heavily. Between 2002 and 2013, heroin-related overdose deaths quadrupled in the US, 21 and deaths associated with prescription opioid overdose quadrupled from 1999 to 2010. 22 Reactions to these trends have included calls for greater availability of naloxone, an opioid overdose antidote, for increased access to treatment for opioid dependence, 23 for greater restriction on prescription of opioids. 24 Opioid overdose has been found to be a major contributor to an almost unprecedented increase in mortality of middle-aged white, non-Hispanic persons in the US at a time when mortality in middle age has continued to decline in other populations. 25 Many observers have commented that opioid dependence is attracting policy attention in the US for the very reason that it is affecting whites in suburban and rural environments rather than only inner-city African Americans. 26 The policy challenge is to balance meeting the need to relieve pain and suffering with reasonable restrictions on controlled medicines, and without creating disparities – racial, economic, or otherwise – in care.
The world has also taken sharp notice of the cannabis legalisation experiences of the US states of Washington, Colorado, Oregon, Alaska and the District of Columbia in a country where opposition to drug legalisation has been deep, and of the nationwide cannabis legalisation experiment in Uruguay. 27 The fiscal imperative of reducing incarceration as well as the fear of adulterants in cannabis obtained illegally have been part of the debates in the US policy changes. 28 Though changes in the legal status of cannabis do not signal changes in prohibition-oriented policies with respect to other drugs in the US, concrete experiences with regulated cannabis markets at an important scale provide an opportunity for rigourous evaluations that will inform larger drug policy debates.
Since it published its first report on violence and health in 2002, 29 WHO has highlighted numerous forms of violence as health issues. 30 The Global Burden of Disease Study of 2013 found that interpersonal violence, including all types of violent assault, rose about 18.4 percent as a cause of mortality globally from 1990 to 2013. 31 The region most affected was Latin America where interpersonal violence was among the top five causes of death in 15 countries. 31 The 2014 WHO report on preventing violence discusses violence that is committed as a result of drug and alcohol use, but few investigators, including those at WHO, have investigated the violence resulting from drug policies. 30
A great deal of drug-related violence is associated with the effort by armed criminal groups to protect their illicit markets, often against armed police or military or paramilitary forces. Some experts have suggested that heavier crackdowns by drug police can lead to major increases in violence when disruption of a given criminal network leads rival groups to intensify their efforts to capture the territory of the weakened group. 32 Mexico and Central and South America have borne an enormous burden of drug-related violence. In 2013, the OAS asserted that the transit of illegal drugs through the Americas leaves persistent violence in its wake – “massacres, attacks by hired assassins, and cases of people being tortured to death”. 15 As the OAS noted, drug trafficking is so entwined with other criminal activity that it is not always possible to say that a given episode of extra-judicial killing is purely drug-related, but criminal networks dealing in drugs are plainly behind much of this carnage. 15 In its 2014 global analysis of homicides, UNODC noted that the 30% of homicides accounted for by “organized criminal groups and gangs” in the Americas, especially Central and South America, dwarf the corresponding percentages in other regions ( Figure 1 ). 33
From UNODC World crime trends, 2014. 33
In conventional wars, sexual violence is both a consequence of war and a weapon used to terrorise the enemy, and the war on drugs is no exception. UNODC asserts that organised criminal networks dominating drug trafficking in Central America regularly use rape with impunity as they defend their territories and routes. 34 Women and girls who may be hired as low-level couriers or smugglers experience sexual assault with no recourse. 34 There are numerous well documented accounts of rape of girls and young women fleeing gang violence in Central America and the severe injuries and post-traumatic stress suffered. 35 Some observers credit drug-related violence with increases in femicide in Mexico and Central America as brutal rape and killing of women are used to terrorise communities and rival gangs. 36,37
Intolerable levels of violence, insecurity, and corruption have led to mass displacement in Mexico and Central America, with displacement levels similar to those documented in war zones. 38 Displaced individuals, including children, are characterised by uncertain legal status and a dearth of services. By one estimate, about 2% of the population of Mexico, some 1.65 million people, were displaced because of violence or the risk of violence between 2006 and 2011. 38 In a publication of the London School of Economics endorsed by five Nobel Prize-winning economists and other experts, Atuesta refutes the idea that this migration is largely economic and not drug-related, showing that a majority of those leaving violence-ravaged communities in Mexico generally move to lower salaries and sometimes no employment opportunities at all. 39
The fateful decision of the Calderón government in Mexico in 2006 to use its military in civilian areas to fight drug traffickers ushered in an epidemic of violence in many parts of the country that also spilled over into Central America. 15 The increase in homicides in Mexico since 2006 is virtually unprecedented in a country not formally at war. The increases in homicides in some parts of the country were so dramatic as to contribute to a reduction in the country’s projected life expectancy. 42 Another analysis concluded that in the period 2008–2010 in the state of Chihuahua, one of the states most heavily affected by drug violence, about 5 years of life expectancy was lost for men. 43
In July 2015, the Mexican government reported that from 2007 to 2014, there were 164 345 homicides in the country, with a dramatic increase after 2006. Figure 2 shows a ‘join point’ analysis 44 conducted for this report using government data. 45 The increase in homicides after 2006 is highly significant statistically and notable, especially after a long downward trend in homicides. No other country in Latin America – and few elsewhere in the world – have seen such as rapid increase in mortality in so short a time. 46
Data from Mexican National Institute of Statistics and Geography (INEGI), 2014. 45
Not all of this increase in homicides can be attributed to drug-related violence, but much of it can be. One estimate suggested that the drug war-related deaths pushed the national homicide rate up by 11 per 100 000, resulting in an overall rate over 80 per 100 000 in heavily affected locations. 47 Eleven per 100 000 is 2.5 times the total homicide rate in the US in 2014. 47 Other observers suggest that the contribution of the drug war to overall mortality is readily quantifiable because drug-gang homicides bear tangible signatures, such as the use of identifiable weapons as well as torture, beheading and other dismemberment, group executions, and mass graves. 46 Though homicides have declined somewhat since 2012, by some estimates homicides perpetuated by organised crime continued to increase through 2014. 46
Drug-related violence in Mexico is not limited to killings and other armed incidents on the street. The Commission found that violence by state actors was also seen in the treatment of people in Mexico incarcerated for drug-related crimes. Analysis supervised by Commissioner Alejandro Madrazo used a probability sample of persons who were in prison for drug crimes (n=479) in Mexico during 2002–2012 – thus before and after the military campaign against drugs – from eight federal prisons. 48 About half of the detainees (n=241) reported having been beaten or tortured at some time in their imprisonment. Among these 241, having experienced an act of torture or abuse was 1.31 times more likely after the ‘war on drugs’ than before (p < 0.01). As shown in Figure 3 , being interrogated by the military in prison was also more likely for people serving drug sentences after the military involvement (p < 0.0001). Interrogation by the military, in turn, was significantly associated with reports of having been tortured or abused. In multivariate analysis controlling for sex, number of times interrogated, and geographical location, people who were detained after 2006 were 3.63 times more likely to have been interrogated by the military while detained (p < 0.0001). As Madrazo has noted, a deleterious outcome of the Mexican drug war is the government’s acquisition of special security powers that undermine fundamental principles of the country’s constitution and human rights responsibilities. 49
Proportion of people incarcerated for drug offenses and non-drug offenses who were interrogated by the military before and after the ‘war on drugs’ in Mexico
Analysis by Javier Cepeda, PhD
The costs, including health costs, of violence on citizens is vast and profound. Execution-style killings are clearly meant to terrorise the population. Living in fear of extreme violence is disruptive to the normal functioning of health and social services, education, and civic participation. The penetration of all aspects of society by drug trafficking organizations in Mexico, Colombia, and a number of Central American countries can corrupt everything from elections and local services to sports teams and other recreation. 50
Cannabis has been estimated to account for about US $2 billion per year of the revenue of Mexican drug cartels, almost as much as the estimated US $2.4 billion from cocaine. 51 It is not possible to know how the legalisation of cannabis in the US, if it were to spread beyond a few states, would affect drug trafficking in Mexico and Central America. Some observers think that even the modest legalisation enacted so far has cut into Mexican cartels’ trade and perhaps limited their capacity to disrupt security. 52
Mexico is far from alone in registering high rates of homicide linked to enforcement of drug prohibition. Colombia’s case is distinct from Mexico’s in that anti-drug efforts were superimposed on a lethal internal war, but homicides spiked when counter-narcotics activities were most intense. 50 Mejía and Restrepo estimate that about 25% of the current homicide rate in Colombia is explained by the thriving cocaine markets and the war on drugs in the country. In other words, absent the large increase in the size of cocaine markets, Colombia would have had a homicide rate in 2008 of about 27 per 100 000 population instead of the observed 37. 53
Mejía and Restrepo characterize these profound problems of homicide and other violence, corruption, and forced displacement as a package ‘outsourced’ from the major drug-consuming countries, mainly the US, to producer and transit countries. 50 That is, in return for a certain amount of foreign assistance for counter-narcotics activities, the US in particular keeps the worst of the heavy burden of violence, insecurity and displacement outside its borders. (See Box 2.) But, as these authors note, this exported pillar of the drug war is beginning to be questioned in earnest by some governments in Latin America, as demonstrated by statements criticizing the status quo in drug policy by the then presidents of Mexico, Colombia, and Guatemala in the UN General Assembly in 2012 which moved the 2016 UNGASS from 2019. 54
By Daniel Mejía and Pascual Restrepo 50
To illustrate the exportation of violence in the current situation in Mexico and Central America, consider the following scenario: suppose that cocaine consumption in the US disappears and is displaced to Canada, but cocaine continues to pass through the US. Because of its international treaty obligations, the US is obliged to do everything in its power to keep cocaine from passing through its borders to Canadian cities. Canada shares some of the cost of this effort, but the result of fighting the cocaine cartels is that the homicide rate in Seattle spikes from its current level of about 5 homicides per 100,000 population to over 100 to keep cocaine from reaching Vancouver. Similar violence seizes other border cities, and a massive wave of internal displacement in the northern US challenges social services and stability of governance. Even if the Canadian government shared the costs to the tune of billions of dollars per year, how long would such a situation be tolerated?
At a time when gains in reduction of sexual transmission of HIV are evident worldwide, HIV transmission linked to injection of drugs with unsterile equipment continues to drive HIV incidence in many regions, including Eastern Europe and Central Asia (EECA) and much of Asia, in spite of the availability of proven interventions to stop it. 55,56 The prevalence of HIV among PWID is many times higher than in the general population in many countries ( Figure 5 ). 57 Outside sub-Saharan Africa an estimated 30% of HIV transmission is linked to unsafe injection. 57 Drug injection is a more important determinant of HIV transmission in EECA than in any other region. 58 While HIV incidence declined by 35% globally from 2000 to 2014, new infections increased by 30% in that period in EECA, where unsafe drug injection accounts for over 65% of cumulated cases. 58
HIV Prevalence among PWID compared to the general population in countries with >30,000 PWID, 2009 – 2014
From UNAIDS, Gap report, 2014. 57
WHO estimates that about two thirds of PWID in the world are living with hepatitis C virus (HCV) infection, a much higher percentage than the estimated 13% living with HIV. 59 WHO notes that EECA, sub-Saharan Africa and East Asia are particularly affected, 59 though data are not regularly kept in some countries. In high-income and upper-middle-income countries generally a high percentage of new HCV infections are among PWID. 59 A seminal study in the US found that over half of PWID were infected with HCV in their first year of injecting. 60 An estimated 20–30% of people living with HIV have HCV coinfection, but the coinfection rate amongst PWID is estimated at 90%. 61
An extensive body of research has demonstrated that there are effective tools for preventing HIV and HCV among people who use drugs by injection and other means. Rigorous reviews of this research have informed strong recommendations by WHO, UNAIDS, and UNODC for comprehensive services for PWUD, 62 which include these elements:
Needle and syringe programmes including other injection equipment; Opioid substitution therapy and other drug dependence treatment; HIV testing and counselling; Antiretroviral therapy; Prevention and treatment of sexually transmitted infections; Condom programmes for PWUD and their sexual partners; Targeted information, education and communication for PWUD and their sexual partners; Vaccination, diagnosis and treatment of viral hepatitis; and Prevention, diagnosis and treatment of tuberculosis.Programmes that provide PWID with sterile injection equipment – often in the form of exchange programmes in which used equipment is traded for sterile equipment – are a crucial part of prevention services and decreasing circulation time of contaminated syringes. WHO found that NSP, particularly low-threshold (easy-access) exchange programmes, were effective in reducing HIV transmission and were not associated with increased injection frequency or initiation of new injection among persons not already injecting drugs. 63 A recent meta-analysis estimated that NSP reduced HIV transmission by about 58%, though with caveats about the quality of some studies and the difficulty of disentangling the effect of NSP from that of other services. 64
As the high HCV prevalence among PWID indicates, HCV is transmitted more efficiently than HIV through unsafe injection. Evidence from controlled trials on the effectiveness of NSP in HCV prevention is more equivocal than for HIV. 65 Part of the challenge is that some people new to drug injection will be HCV-infected even before they begin to take advantage of NSP services. NSP is most effective in HCV prevention when coverage is very high, reaching people from a time close to when they first inject. 66
There is repeated reference in this report to the opioid agonists methadone and buprenorphine, which are the oral medicines most commonly used in medication-assisted treatment of opioid dependence, referred to as OST. OST plays a dual role as treatment for opioid dependence, which can help stabilize lives with all of the attendant benefits, and as HIV and HCV prevention because when effective, it eliminates injection. There is arguably no form of treatment of any drug dependence that has as vast a scientific evidence base or as long a successful clinical experience as OST. 67 In both its treatment and harm reduction roles, OST faces drug policy impediments because the medicines used are heavily regulated in most countries. Countries do not always allocate adequate quantities of these oral opioid medicines for OST, and doctors in some countries are reluctant to prescribe them for fear of prosecution if there is diversion of these medicines to non-medical use.
A 2012 meta-analysis from Europe, North America, and Asia concluded that oral OST, methadone maintenance in particular, reduces risk of HIV transmission among people who inject opioids by about 54%. 67 A 2014 ‘review of reviews’ concluded that the evidence is strong for OST’s HIV prevention impact, particularly where doses of opioid agonists are adequate. 65 Observational studies from San Francisco, the UK, Scotland, Vancouver and Sydney found that OST use was associated with dramatically reduced risk of HCV acquisition among PWID, 66,68–70 with the data from Scotland, Amsterdam and the UK also concluding that combined OST with NSP further reduces the risk of HCV acquisition. 66,71,72 A model analysis based on data from the UK illustrates that if enough people can get access to OST and to sufficient sterile injection equipment for virtually every injection, HCV transmission could decline substantially ( Figure 6 ). 73
Impact on hepatitis C virus (HCV) prevalence among people who inject drugs (PWID) over time for different coverage levels of opiate substitution therapy (OST) and high coverage needle and syringe programmes (NSP) (100%NSP) for three epidemic scenarios with a baseline chronic HCV prevalence among PWID of 20, 40 or 60% with no OST or 100%NSP. 100% NSP is defined as obtaining one or more sterile syringes from a NSP for each injection reported per month
From Vickerman et al., 2012. 73
In spite of the very large body of evidence of the effectiveness and cost-effectiveness of opioid agonist therapy, some countries insist that generating new research in their settings is necessary before scaling up OST. For these and other reasons, OST has remained in a “perpetual pilot” mode in a number of countries. 74
Access to OST in Western Europe is a positive contrast to most other regions, with a number of Western European countries virtually eliminating HIV from unsafe injection as a public health concern by scaling up NSP and OST as well as treatment for HIV. 75 Unlike their counterparts in Western Europe, EECA countries generally have inadequate coverage, quality, and accessibility of NSP and limited or no access to OST. 76,77
Gains have been made in harm reduction policy and practice in some of the Asian countries with large populations of PWUD. In China, Malaysia, and Vietnam, zero tolerance of harm reduction has given way to government-supported OST and sometimes NSP. 56 China was estimated in 2015 have been serving about 200 000 OST patients, 58 but this still represented a small percentage of people who might benefit, and the problems of high drop-out rates and low dosages remained challenging. 56 By a 2015 estimate, Vietnam was reaching 32 000 OST patients in 44 provinces in a country with an estimated 130 000 PWID. 78 Though coverage may be relatively low, the existence and continued growth of these programmes are important achievements.
While it is advantageous with respect to HIV prevention that coverage levels of these measures be as high as possible, an important body of research demonstrates that if OST, NSP, and HIV treatment are all present their synergistic effect can compensate for partial coverage. Figure 7 illustrates this point using data from Dushanbe, Tajikistan. In this case, if needle exchange and ART alone are available, a 50% decrease in HIV incidence over a 10-year period requires that the coverage of both programmes be about 30%. 79 But if ART, NSP, and OST are all available, a 50% decline in HIV incidence over this period can be achieved with 20% coverage of these interventions. 79 Similar results have been found in other settings. 55,80 Therefore partial coverage of OST, NSP, and ART may provide effective prevention if it is not possible to attain very high ART coverage, which may be especially challenging where people who use drugs are criminalised.
Coverage required of ART, NSP, and OST to achieve 30% and 50% decreases in HIV incidence and prevalence among PWID over 10 years
From Vickerman et al., 2014. 79
HIV testing with a link to treatment is important for all people. For PWUD as for other populations, ART can suppress viraemia and lower transmission risks. ART coverage for PWUD is high in Western Europe, North America and Australasia, but it was not always so. In the early years of ART availability, HIV-positive people who used drugs had to battle scientifically unfounded ideas that excluded them from treatment programmes. One was that the lives of people who use drugs are too chaotic to allow them to adhere to daily multi-pill treatment regimens, 81 although research had shown that PWUD can adhere to ART and achieve viral suppression. 82 It took more research in several settings and the experience of successfully expanded treatment programmes for PWUD to dispel these ideas. 83
Studies from a number of settings have reported that agonist treatment for opioid dependence improves ART adherence among PWUD. 83 In Vancouver, Canada, several longitudinal studies showed not only that OST continuation improved ART adherence over time, 84 but also the converse – that OST discontinuation significantly increased the risk of ART non-adherence 85 – and that OST patients with higher opioid agonist doses had the strongest adherence to ART. 86 In China, the understanding of the importance of the OST-ART link led to an effort to integrate ART services in methadone clinics. 87 While a number of practical challenges were encountered, the effort showed an appreciation for the value of integrating these areas of care. In Ukraine, patients having access to integrated and co-located ART and OST services had greater access to ART than those receiving OST in non-integrated facilities. 88
In places where there is significant HIV transmission linked to unsafe injection, denying treatment to HIV-positive PWUD both ensures that they and their injection and sex partners will be at risk of HIV and violates the rights of all concerned. Nonetheless, PWUD seem to be systematically excluded from ART in many parts of the world. A 2014 review of HIV services in the countries with the largest number of people who inject drugs estimated that in both China and Malaysia, less than 5% of PWUD living with HIV had access to treatment, and in the Russian Federation about 1%. 56 There is relatively poor access to HIV testing and ART in EECA, due at least in some places to fear of police harassment or arrest, as well as to systematic exclusion of PWUD from treatment programmes. 89 UNAIDS’ 2014 report on gaps in the global HIV response summarises the crisis of inaccessibility of ART for PWID, also noting that less than 1% of HIV-positive PWID in Africa receives ART ( Figure 8 ). 57
From UNAIDS, Gap report 2014. 57
People who use drugs in many parts of the world have no access to HCV screening and treatment. Unlike HIV, HCV infection can be cured and cleared from the body. Interferon-based therapies as the treatment of choice are giving way to direct-acting antivirals (DAA) marketed since 2013. The cost of the DAAs, however, is orders of magnitude greater than interferon-based therapy. 90 There is an urgent need for measures to reduce the price of the new generation of hepatitis C medicines and to ensure that PWUD can benefit from these treatments. In this regard, there may be many applicable lessons from the well documented efforts that succeeded in bringing down the prices of HIV medicines. 91
For PWUD, cost is far from the only barrier to being able to benefit from DAA therapies (see Figure 9 ). Policy-making on HCV treatment is replaying a number of misinformed tropes from the HIV past, including the idea that PWUD – or even those with any history of drug use – do not adhere well to treatment and are not worthy of expensive care. 92 This non-adherence myth was disproven with respect to HIV treatment, which usually requires a lifetime regimen of several medicines, and it has been researched and equally disproven with respect to HCV therapies, which are much shorter in duration. 92 Requiring abstinence from drugs or alcohol before initiating HCV treatment, a condition already established in many US states, is also not scientifically justified and excludes underserved and needy persons from care. 92 It has also been suggested that active drug users are poor candidates for HCV treatment because they are likely to be reinfected, but several studies disprove this assertion. 92,93 None of these claims should stand in the way of comprehensive HCV prevention and treatment for PWID. WHO along with many professional liver and infectious disease associations urge HCV screening and treatment of PWUD as a public health priority. 92,93 Modeling analyses have indicated that HCV treatment for PWID could be an effective and cost-effective means of HCV prevention, 94,95 and that combination prevention strategies incorporating OST, NSP and HCV treatment could dramatically reduce HCV incidence and prevalence among PWID in a range of settings. 96
From Wolfe et al., 2015. 97
Unsafe injection-linked transmission of HIV sometimes overshadows sexual transmission in programme priorities for PWUD, but both are essential. UN reports and research in many settings have for years highlighted the importance of condom programmes for all men who have sex with men (MSM), and particularly those who use drugs either to enhance sexual pleasure, to lower sexual inhibitions, to escape or cope with situations of discrimination, persecution, or uncertainty about sexuality, or for other reasons. 98,99 Many studies have demonstrated a link between drug use at the time of sexual activity (‘sexualized drug use’) and lower condom use, resulting in a higher prevalence of HIV and other sexually transmitted diseases and lower incidence of condom use. 99–101 But more work is needed in many settings to understand the complex motivations for sexual decision-making that would inform effective condom promotion programmes. 102
The UN recommendations do not include a number of interventions for which an evidence base exists to justify their contribution to an HIV or HCV response. Supervised injection sites are an example. In a number of European countries, Australia and Canada, there are legally sanctioned indoor locations where people can inject (and sometimes smoke and inhale) illegal drugs under medical supervision, obtain clean equipment, be referred to OST and learn HIV and overdose prevention education. The harm reduction intended by these facilities is not only with respect to HIV transmission but also to prevent mortality and other adverse outcomes of overdose, as well as to reduce unsafe disposal of syringes. 75 A recent meta-analysis reported a 69% reduction in syringe sharing resulting from use of supervised injection sites. 103 In the case of Insite, the supervised injection facility in Vancouver, Canada, a conservative estimate indicates that on the sole grounds of HIV cases averted, Insite more than pays for itself, and savings are even greater if one takes into account behavioral change leading to use of sterile syringes outside Insite. 104
As noted by Coffin and colleagues, research on pre-exposure prophylaxis (PreP) using tenofovir, an important newer HIV prevention measure, has often excluded PWUD. 105 Nonetheless, an important PreP trial reported in the Lancet amongst PWID in Bangkok demonstrated an HIV prevention effect for both men and women who inject drugs. 106
Preventable outbreaks of HIV in recent years have constituted graphic real-life demonstration of the value of ready access to harm reduction services and the cost of impeding access to them. EECA bear a heavy burden from the neglect of harm reduction measures. Harsh anti-drug policies and moral judgments against PWUD contribute to making health services for this population a low political priority. 89 In the first decade of its work, financial support from the Global Fund helped to overcome these difficult political environments and supported the expansion of harm reduction services, especially NSP and OST, in a number of EECA countries as well as in East and Southeast Asia. 107 However, with changes in Global Fund policy that have eliminated or reduced funding for middle-income countries, some of these services have been cut (see Box 3 below). 108
In pure fiscal terms, preventing HIV through harm reduction measures should be an easy sell. Cost-effectiveness is high, and start-up costs for these services are low. But harm reduction continues to be resisted as a funding priority in too many countries. Support by the Global Fund in its first decade, however, inspired some countries that had not previously scaled up NSP and OST to do so, particularly in EECA. 107 The Global Fund encouraged the inclusion of HIV prevention services for PWID and other “key populations” in country proposals. 108
In the first ten funding rounds of the Global Fund plus a special transitional funding period, some $620 million in grant support went to programmes for PWIUD in 55 countries, an unprecedented wave of life-saving support for a politically unpopular population. 108 When the official country proposal to the Global Fund in Thailand, for example, excluded programmes for PWUD in spite of high HIV prevalence in that population, the Global Fund made a special grant to NGOs that were able to bring services directly to the community. 109
In 2013, the Global Fund unveiled a “new funding model” that, unlike its previous processes, assigned ceiling amounts to countries and significantly limited funding to most middle-income countries, even those with severe injection-linked epidemics where it was unlikely that governments would pick up the costs of the newly scaled-up programmes. 110 Romania lost funding at a key moment (see main text); Serbia’s harm reduction programmes are operating on a shoestring; 111 programmes in Ukraine – a country with over 350 000 PWID – are gravely threatened; 110 and Vietnam may suffer a similar fate. 108 Thailand is no longer eligible for support.
Civil society advocacy continues for governments to provide the funding no longer available from the Global Fund, 112 but it is clear that when it comes to politics harm reduction will remain a hard sell in many places.
In 2010–2012, of the 27 European Union member states plus Norway, Iceland and Turkey, it was estimated that Romania and Greece accounted for one third of all the HIV incidence among PWID, the two countries together having seen a 20-fold increase in new HIV diagnoses linked to drug injection. 113 In Romania the reduction in external support for harm reduction services coincided with the availability of relatively inexpensive amphetamine-type “legal highs” – new psychoactive substances (NPS) not yet under legal control. Some people who previously injected heroin shifted their consumption to these new stimulants. But heroin is injected two or three times a day, and the stimulants six to ten times. 113 NPS use was found to be associated with syringe sharing and high-risk sexual practices to a greater degree than heroin use. The number of persons injecting drugs is estimated to have risen from about 17 000 in 2008 to about 20 000 in 2011 – and with riskier and more frequent injection 114 – and harm reduction services were largely curtailed in 2010. NGOs ran effective NSP and OST services that kept HIV low until then, but funding from the Global Fund was lost when Romania joined the EU. 108 The dramatic rise in HIV cases through 2013 is shown in Figure 10 , representing cases at a major hospital in Bucharest, which practitioners judged to mirror the national situation. Among the newly infected PWID, about 20% were estimated to be injecting heroin, 20% NPS, and 20% a combination of the two. 115 As UNAIDS has noted, HIV outbreaks among PWID tend to be dramatically fast-growing. 56
From Dr Christina Oprea, Victor Babes Hospital, Bucharest, 2015. 115
In Greece, even before the severe economic recession of 2008–09, harm reduction services for PWUD were provided at a relatively low level of coverage. 116 The recession was associated with impoverishment and dramatic increases in homelessness among PWID, which separated some people even more from existing services and funding to existing NSP services was cut significantly. 117 After years of fewer than 20 new cases of HIV transmission among PWID in the country, in 2011 the number of new cases of HIV linked to injection was 260 and in 2012 it jumped to 522. 116 With assistance from the European Union, Greece scaled up low-threshold harm reduction services, including in cities that had not had them previously, and existing services got support to distribute low-dead-space syringes, which reduce HIV transmission risk. 118
For most of the period since the emergence of HIV as a public health problem, the US government banned the use of federal funds for NSP, though some states and municipalities supported them. 119 In January 2016, the US Congress lifted that ban for all NSP program costs other than needles and syringes, a move seen by many as a response to an increasingly visible opioid injection problem even outside major urban areas. 120 In 2015, a rural county in the US state of Indiana experienced a dramatic increase in HCV infection followed by a linked outbreak of HIV cases linked to injection of oxymorphone, a synthetic opioid. 121 Some 135 persons were infected with HIV in a short time in a district that previously had reported very few HIV cases. Almost half the new infections were among women, and they spanned a wide age range as it was found that injection took place in multi-generational groups. 121 Indiana did not permit needle and syringe programmes before the outbreak–non-medical use of syringes was a felony punishable by up to three years in prison. 119 The governor of Indiana changed the state’s policy to allow NSP services for a year on the basis of a public health emergency. 119 Similar outbreaks of HCV among PWID have been reported across this region, including in Kentucky, and West Virginia, all states with limited or no NSP, and with limited access to OST. 119
For policy-makers interested in hard numbers on the value of comprehensive HIV and HCV prevention, the cost savings associated with these services are considerable. The government of Australia, for example, which has invested significantly in harm reduction from the early years of HIV, estimated that for every dollar spent on NSP, more than $4 was gained in short-term (10-year) savings on health care costs, and over a longer period where coverage has been maintained, as much as $27 is gained. 122 A World Bank study in Malaysia, where about two thirds of HIV transmission is related to unsafe injection, concluded that over the long term NSP in that country even at a relatively low rate of coverage would give a more than threefold return on investment. 123 Other studies have shown that needle and syringe programmes can also help refer people to treatment for HIV and drug dependence and other services. 124, 125 A 2015 review suggests that the low cost of these programmes and the high cost of the HIV suffering and treatment that NSP can avert means that NSP are “one of the most cost-effective interventions ever funded.” 124
Commenting on the cost-effectiveness of OST, WHO, UNODC and UNAIDS UN asserted both the effectiveness and the cost-effectiveness of OST with respect to HIV, noting that for every dollar spent on it, a return of $4–7 could be expected due to crime reduction alone and a return of about $12 if health-care savings are included. 126 Although OST is more expensive per person than NSP, Wilson and colleagues assert in their review that OST is highly cost-effective not only in HIV prevention terms but because of health savings linked to less relapse, reduced incarceration, and a wide range of quality of life improvements. 124
Evidence from a number of countries indicates that drug law, policy, and law enforcement practices can be barriers to provision and use of harm reduction and other HIV prevention services. These barriers take many forms, a few related to the letter of the law in force in a country – the “law on the books” – but many more related to the way in which law is enforced in practice, or what Burris calls the “law on the street”. 127 Some of these are the following:
In some cases, there are legal prohibitions against or poor legal grounding for harm reduction services for PWID. The case of the Russian Federation is extreme: OST is prohibited by law though opioid injection is widespread, and NSP have been allowed only sporadically and are generally not supported by the state. 128 The official estimate of Russians living with HIV rose to 907 000 by end 2014, up almost 7% from 2013 figures and up from 500 000 in 2010. 129 Over 57% of new cases were attributed to unsafe drug injection. NSPs are banned by law or effectively blocked by policy, including zoning restrictions, in many jurisdictions. 76
While OST may not be banned outright or explicitly, in some countries methadone and buprenorphine, the medicines used most in OST, may not be registered or authorized for this use. 76 This problem persists in spite of the inclusion of methadone and buprenorphine on the WHO Model List of Essential Medicines and strong support from WHO for OST. There are many other ways in which drug-control laws or regulations limit the use or usefulness of OST, including arbitrary restrictions on numbers of patients; arbitrary limitation of dosages and the duration of treatment; prohibition of take-home doses; periods of drug or alcohol abstinence or having to try other kinds of treatment as a prerequisite to starting OST; limiting the neighborhoods or geographical zones where OST services can be offered; lack of integration with accessible community health services so that people have to make special trips for OST; and lack of access to OST in prison and pretrial detention. 76,130 In a number of countries there is relatively good access to OST in the community but none offered to people in prison or other detention. 76
As with OST, NSP not banned outright can be undermined by various laws and policies. According to a 2014 estimate by Harm Reduction International, significant drug injection is reported in 158 countries, but only 90 have functioning NSP, most of which have very low coverage. 76 Laws, policies or local ordinances may limit NSP to remote or unpleasant neighborhoods; limit the hours of operation or permitted geographic coverage; limit the number of needles or syringes that can be exchanged or require one-to-one exchange in every transaction (that is, getting clean syringes requires producing the same number of used syringes); limit the age of NSP participants; and limit or forbid provision of clean injection equipment in prison and pretrial detention. 76 In the US, the 50 states have a dizzying array of laws and regulations regarding needle exchange. In some jurisdictions, local health authorities have to declare emergencies periodically to continue to justify NSP; some states simply ban these services. 76
In many countries, drug paraphernalia laws undermine NSP, often prohibiting the possession of syringes. In the Global Fund-supported project known as CHAMPION (2008–2013), which was meant to help address the high prevalence of HIV among PWID in Thailand, evaluators reported that an important impediment to scaling up NSP was that PWID feared carrying syringes. 131 They reported that being caught with syringes could lead to arrest, detention, forced drug treatment, and obligatory urine testing.
In some countries health workers are required or strongly encouraged to register PWUD, and registries are turned over to the police (see web appendix I).
In some places, there is no legal prohibition of possession of drug paraphernalia, but police nonetheless use possession of injection equipment as grounds for stop-and-search, arrest, and detention. For example, among nearly 600 Russians living with HIV surveyed in 2014, over 50% reported having been arrested for possessing a syringe (or having a syringe planted on them by the police), though it is not against the law in the Russian Federation to possess a syringe. 128 Those reporting such arrests were more likely to have shared needles with others and to have suffered an overdose than those not arrested. 128 This quantitative study corroborates qualitative accounts suggesting that repressive policing in the Russian Federation in many ways raises HIV risk and discourages seeking out and using the few HIV prevention services that may exist. 132 In a number of countries, even if syringe possession is legal, police routinely seize injection equipment that they find, further undermining protection of health. 75 Police presence was associated with unsafe rushed injection among PWID in Bangkok in a multivariate analysis, 133 and a small sample of PWID in Hai Phong, Vietnam, reported greater likelihood of needle sharing and other risky practices when police were present or their presence was feared. 134
The performance of drug police in many countries is judged by the number of arrests they make, and PWUD are likely to be easier to find to help bolster arrest totals than major drug traffickers. It is perhaps for this reason that police may target facilities providing health and harm reduction services to PWUD. 135 A 2015 study involving over 500 methadone patients conducted by NGO service providers in New York City found that 38% of the patients reported being stopped and searched by police outside the clinics where they received methadone, and 70% reported witnessing someone else being searched in these locations. 136 In some countries, extortion of bribes from PWUD may be an important source of income for poorly paid police. 137
Crackdowns and other intensive policing, often targeting low-income, minority, or marginalized persons, can undermine harm reduction and add to drug-related risk. During a crackdown on drug use known as “Operation 24/7” in Vancouver, Canada in 2003, researchers documented a significant decline in access to sterile injection equipment as police actions drove PWID away from the only NSP open at night. 138 During police crackdowns in Australia, PWUD reportedly switched from inhalation or smoking of substances to injection, which is much riskier, partly because during crackdowns drugs became scarcer and injection could be accomplished with lower quantities of drugs as well as more quickly and less visibly than smoking. 139 Other studies have shown that crackdowns lead to rushed injections, more vascular accidents, and the likelihood that steps such as disinfecting the injection site will be skipped. 140 In Malaysia, rushing an injection because of police presence was found to be linked to risk of overdose. 141
According to WHO, tuberculosis is the single most important cause of death among people living with HIV, responsible for one in four deaths. 142 People living with HIV have a 30-fold higher risk of TB infection than HIV-negative persons. 142 But WHO emphasizes that PWUD are at very high risk of both infection with Mycobacterium tuberculosis and TB disease even if they do not live with HIV; the risk of TB linked independently to drug injection – and even to non-injection drug use – was well established before HIV was in the picture. 143
WHO estimates that people who both live with HIV and inject drugs are two to six times more likely than HIV-positive people who do not inject drugs to contract TB. 144 But the role of drug use in the epidemiology of TB is complex and, as noted by Deiss and colleagues, the existing literature does not always distinguish drug injection from other drug use. 145 Many elements of the risk environment of at least some people who use drugs – homelessness or sub-standard housing, heavy alcohol and tobacco use, and incarceration, for example – are risk factors for TB. Some studies suggest that people who use drugs present later than other persons to seek TB testing or care. 145 Deiss and colleagues also raise the possibility that use of opioids may inhibit the cough reflex and thus mask symptoms of TB that might otherwise lead to seeking care. 145
Multi-drug-resistant TB (MDR-TB) has threatened to undermine progress in TB control in many parts of the world. 146 The region with the highest documented proportion of MDR-TB among TB cases is EECA, also home to major unsafe injection-linked HIV and HCV epidemics. 142 Remarkably, though HIV-HCV coinfection is high in the region, HIV-TB coinfection is reportedly low, but experts warn that with the combination of sparse harm reduction services, low ART coverage among people who use drugs, high rates of incarceration of people who use drugs, non-integrated vertical health services, and substandard housing and social support, a perfect storm of HIV-MDR-TB coinfection may be brewing. 147
WHO recommends that PWUD be included systematically in anti-TB efforts and especially that HIV, HCV and TB services be integrated and low-threshold for people who use drugs. 144 The reality, however, is that recommended services remain out of reach for many people who use drugs worldwide. Diagnosis of TB with microscopy identifying acid-fast TB bacilli and molecular DNA detection using geneXpert systems is recommended, 144 but in Central Asia, for example, TB diagnosis is still made mostly using chest radiography 147 though x-ray results are compromised by the presence of HIV. 144
WHO has compiled detailed guidance for integrated treatment of TB and HIV as well as TB and HCV, including ensuring sustained access to ART for all who need it. 148 The exclusion of PWUD from ART, which persists in many parts of the world, undermines the effectiveness of TB treatment as well as HCV treatment.
The importance of integrated and sustained care cannot be overstated. Deiss and colleagues report cases in which TB treatment was integrated with treatment for drug dependence but was lost after people left drug treatment. 145 The NGO Partners in Health addressed the challenge of keeping people who use drugs in sustained care for MDR-TB in a programme called Sputnik in Tomsk, Russian Federation, through a strategy of intensive accompaniment of patients. 149 Trained teams of nurses, drivers, and others worked with patients to ensure delivery of treatment in places and circumstances that the patient could maintain to minimize missed appointments. Family, friends, and neighbors were helped to understand the importance of treatment and to provide support to patients. 149 Over 70% of high-risk patients completed treatment. The cost compared to hospitalization was small. A study in Malaysia demonstrated that TB screening and care in drug rehabilitation centres and facilities offering OST was a very effective targeting strategy. 150
TB and drug use experts at WHO, writing in 2013 in the WHO Bulletin, asserted that it was urgent to address the undermining role of “punitive drug policies and laws in fueling the tuberculosis epidemic among people who use drugs”. 151 Not only do punitive laws drive PWUDs away from health services, they may also contribute to stigmatising or disrespectful treatment in health services. 151 For these reasons, in its 2014 guidance on HIV services for “key populations” including PWUD, WHO recommended decriminalisation of drug use as well as training and protections for health workers to reduce fear of treating PWID. 152
In 2014, UNODC estimated that persons convicted of drug crimes make up about 21% of incarcerated persons worldwide. Possession defined, by UNODC as possession of drugs for individual use, was the most frequently reported crime globally ( Figure 11 ). 32 Based on data from 2011 annual country reports, UNODC estimated that drug possession offenses constituted 83% of drug offenses reported worldwide. 32 While not all of the crimes reported by the police result in incarceration, mandatory prison sentences are attached to possession of even a small amount of drugs in many countries. In some countries that have decriminalised drug use, possession for individual use remains an offense, or the amount defined for non-criminalised individual use is so low that possession is effectively a crime. 153
From UNODC, World crime trends, 2014. 33
UNAIDS estimates that in places where drug use and small-scale drug possession are criminal offenses, the majority of people who use drugs may wind up in the custody of the state at some time in their lives. 57 In Central Asia, one estimate suggests that more than 50% of PWID have been arrested at least one time. 77 Though there have been some reform efforts, many countries have drug laws that impose extended custodial sentences on people convicted of non-violent offenses including drug use alone, possession of amounts of drugs intended only for individual use, and sale of very small amounts of drugs. 76 The over-representation of PWUD in prison and the lack of essential care and support for them while in state custody are amongst the most devastating health legacies of pursuing drug prohibition. There is, moreover, no evidence that incarceration is an effective deterrent for drug use either in prison or afterward. 154 Indeed, the Vancouver Injection Drug User Study (VIDUS), a long-running cohort study, found that recent incarceration was negatively associated with cessation of injection. 155
Several studies conclude that criminal prosecution of minor use and possession infractions does not have the deterrent effect with respect to drug use, possession, or minor crimes that supporters of these sanctions claim. A classic study comparing cannabis use in San Francisco and Amsterdam – cities with very different approaches to cannabis regulation – found that the partial decriminalisation of cannabis in Amsterdam was not associated with increased use or possession, and the rigorous criminalisation in San Francisco was not associated with reductions in use or possession. 156
The OAS in its landmark 2013 report on drugs and drug policy in the Americas lamented the dramatic rise in prison populations especially linked to prosecution of minor offenses because they are less likely than major traffickers to be able to afford legal assistance in attaining “access to justice”. 15 This increase, at least in some Latin American countries, is a detrimental outcome of steady increases in legislated penalties for drug offenses since the 1950s ( Figure 12 ). 157 (See also web appendix II.)
From Uprimny et al., 2013. 157
Table 1 shows the most recent information for selected countries on the proportion of people incarcerated for drug offenses among all incarcerated persons. UNODC’s data on the prominence of possession offenses and the data informing Table 1 do not distinguish the proportion of drug-related offenders who are incarcerated for minor, non-violent offenses. But, as noted by Penal Reform International in a 2015 report, mandatory prison sentences are attached to possession of even a small amount of drugs in many countries. 160
Incarceration for drug offences as percentage of all incarceration in selected countries
| Proportion imprisoned for drug offences overall | Proportion of women imprisoned for drug offences | |
|---|---|---|
| Argentina | 33 | 68.2 |
| Australia | 12 | 17 |
| Bolivia | 45 | |
| Brazil | 24.8 | 53.9 |
| Canada (fed.) | 26.3 | |
| Canada (provincial) | 15.7 | |
| Colombia | 17 | 45 |
| Ecuador | 33.5 | 77 |
| Ireland | 19.6 | |
| Italy | 38.8 | |
| Latvia | 14.3 | 68 |
| Mexico* | 57 | 80 |
| New Zealand | 10 | |
| Peru | 23.8 | 68.4 |
| Russian Federation | 20 | |
| Thailand | 68 | |
| US (federal) | 49 | 59.4 |
| US (states) | 16.8 | 25.1 |
From Penal Reform International; 158 Giacomello, 2014; 159 OAS-CIM report, 2014; 160 US Bureau of Justice Statistics, 2015; 161 Perez Correa and Azeola, 2012 48
The US has the highest rate of incarceration in the world at about 707 persons per 100 000 population, about 50% higher than in the Russian Federation and more than five times higher than in China. 169 As Table 1 shows, drug-related offenses account for a significant percentage of this incarceration. Aggressive prosecution of drug offenses along with mandatory minimum sentences for certain infractions helped to make drug-related mass incarceration a major engine for growth in US state and federal prison populations beginning in the 1980s ( Figure 13 ). 170
From Snyder and Mulako-Wangota, 2014. 170
The racially disparate application of drug-related incarceration in the US is a prominent feature of mass incarceration. Persons of color, particularly African Americans, have been disproportionately affected by drug-related mass incarceration. In 2011, amongst men aged 30–34, 1 in 13 African Americans were in prison, 1 in 36 Hispanic Americans, and 1 in 90 whites, though prevalence of drug use in these populations was similar. 171 The Sentencing Project, an NGO focused on criminal justice, calculated in 2014 that African American men had a 32% probability of being in prison or other state custody at some time in their lives, compared to 17% for Hispanic men and 6% for whites. 172 Figure 14 shows the racial disparity in drug-related incarceration at the federal and state level in 2013. 173
From Carson (Bureau of Justice Statistics), 2014 173
This pattern reflects documented racial disparities at all stages of US law enforcement, from “stop and search” and arrest to sentencing and incarceration. Beginning in the late 1990s, New York City undertook to clamp down on cannabis infractions, resulting eventually in nearly a half million arrests through 2013 – of young people for the most part – for minor cannabis infractions. 174 There was consistent evidence that marijuana use was higher among whites than among African Americans or Hispanic Americans. In the decade beginning in 2004, African Americans comprised 25% of the population of the City but accounted for 54% of cannabis arrests; Hispanics made up 27% of the population but accounted for 33% of arrests. 174 Arrests for drug-related infractions amongst teenagers across the US from 1980 to 2012, the large majority for cannabis, show a similar racial disparity ( Figure 15 ). 177
The striking racial disparity in arrest and incarceration in the US parallels racially disparate patterns of HIV, and some investigators conclude that the two phenomena are closely related. Though African Americans comprise 14% of the US population, about 40% of new HIV cases and about half of AIDS cases in the US occur in this population. 176 A number of studies show that a history of incarceration is associated with HIV incidence and prevalence among African American men and women. 176–178
Racial and ethnic minorities are over-represented in prison and in arrest figures in countries other than the US, including aboriginal people in Canada and Australia and people of African origin in Brazil, but the contribution of drug-related arrests and convictions to these patterns is not clear. 179 In Canada, aboriginal persons accounted for 3% of the adult population but 20% of adults sentenced to prison in 2013–14. 180 Afro-Brazilians reportedly receive longer sentences for all categories of crime than Brazilians of non-African origin, and they are disproportionately targeted in drug policing and crackdowns. 181
In October 2015, the US government announced that it would release 6,000 federal prisoners incarcerated for minor drug offenses, meant to be the first tranche in a release of a possible 46 000 federal prisoners of the 100 000 convicted of federal drug offenses. 182 This unprecedented release is occasioned by a decision by the executive branch to reduce federal mandatory minimum sentences for minor drug offenses and make the reduction retroactive. 182 The great majority of people serving prison sentences for drug offenses are in state – not federal – prisons, which are not affected by this change. 173
Table 1 illustrates a striking gender disparity in drug-related imprisonment. While in any given drug market there are likely to be many more men than women involved in use, possession, and sale of drugs, a higher percentage of women than men are imprisoned because of drug-related convictions in virtually all countries where data exist on this point. 158 The unanimously endorsed UN Bangkok Rules urge governments to find alternatives to incarceration for women convicted of non-violent offenses – the vast majority of incarcerated women – and to ensure protections from violence and other human rights abuses for those who are in state custody. 183 But these rules seem to be commonly honoured in the breach.
Giacomello asserts that a large proportion of women convicted for small-scale sale and other non-violent offenses in Latin America are uneducated women living in poverty who had limited opportunities to earn licit income. 159 A report by the OAS and the Inter-American Commission (IAC) on women echoes this view, estimating that most women imprisoned for drug-related offenses in the Americas are engaged in ‘micro-trafficking’ though they may be sentenced for long periods under harsh anti-trafficking statutes. 160 The OAS-IAC report also notes that in many countries in the Americas, many women are convicted for bringing drugs into a prison or pretrial detention facility for a spouse or family member and that women’s low level in the drug market power chain means they have little leverage in plea bargaining or sentence reduction. 160 In Mexico, CIDE researchers found that virtually all of the women imprisoned for drug-related crimes in 2012 were first-time offenders, and 92% were convicted of non-violent offenses. 48 Of women accused of drug infractions in Argentina in 2013, almost 30% had been detained without trial for one to two years and about 12% for more than two years. 160
Overall in Europe and Central Asia, about one quarter of women in state custody are convicted drug offenders. 184 In the US, there was a doubling of drug-related arrests of women, mostly for drug possession, from about 400 per 100 000 population in 1990 to a peak of about 800 per 100 000 in 2006, after which the rate declined somewhat. 175
Women who use drugs in prison are also at risk of HIV from sexual violence or unprotected coercive sex as well as from drug use. 185 As much as HIV services, including access to condoms, and drug dependence services are inadequate in men’s prisons, they are even more so in women’s prisons. 185 While incarceration of women has increased in many countries in the last 20 years, women are still a small percentage of the prison population in most countries, and developing specialised HIV, HCV or TB programmes for them is rarely a political, public health, or budgetary priority. 185
Detention of children and young people and the effect on children of parents’ incarceration are too little studied (web appendix III). Pretrial detention of children and adults for drug offenses also poses health risks (web appendix IV).
The over-reliance on incarceration as a response to drug use may have a profound effect on the well-being of relatives and partners of people imprisoned for drug offenses. Many studies document that incarceration of a family member imposes unique forms of financial strain, psychological distress, and logistical hardship on the family and is associated with deleterious health outcomes. 186–196
Caring for a family member who uses drugs has its own challenges, 197 but incarceration may generate further difficulties by increasing geographical distance between PWUD and their families, erecting barriers to communication, and subjecting family members to correctional surveillance and regulations when they maintain contact with their incarcerated loved one. 198–200 Parole and probation conditions may sometimes be incompatible with resources family members have to offer (e.g., housing outside of a district of parole, or in government-subsidised housing). 201 In the US the impact of all of these factors falls disproportionately on people of color. (See web appendix V.)
A 2014 survey of people visiting family members in Mexican prisons indicated similar kinds of challenges in that setting. Of those visiting relatives, mostly women, over 50% said that because of the imprisonment of a spouse or family member they had had to get a job or an additional job; 41% on the other hand said they lost a job; over 18% said they had to move house; and almost 40% said the imprisonment impeded their ability to care for their children or grandchildren. 48 Spouses of incarcerated persons in this study also reported suffering disproportionately from a range of health problems, including high blood pressure and depression. 48
Prisons and pretrial detention facilities worldwide are high-risk environments for infectious disease transmission. UN agencies estimate that prevalence of HIV, other sexually transmitted diseases, hepatitis C and hepatitis B infection, and TB is from 2 to 10 times higher in prisons than in the community. 57 Coinfections among these infectious diseases are also likely in prison. In Argentina, for example, people living with TB who had a history of incarceration were 6 times more likely to be HIV-positive and 18 times more likely to have HCV infection than the general population. 76 Since most people in prison and pretrial detention return to the community, what happens behind prison walls in addressing infectious disease has ramifications for the whole population.
As UNAIDS notes, excessive criminalization of drug-related offenses is one factor that contributes both to prison overcrowding and to the over-representation in prisons of people who are likely to have been exposed to HIV and, in the case of PWID, HCV and TB. 57 PWUD are likely to be over-represented in prison particularly in countries where laws allow for lengthy custodial sentences for minor drug use, possession, and sale, and many may be imprisoned repeatedly. 202 These factors figure in the 2014 recommendation of WHO to decriminalise drug use – and thus reduce incarceration of PWUD – as a critical step to enabling optimal HIV prevention, treatment, and care. 152
Drug injection does take place in prison, even where very restrictive measures are in place. Over 90% of men surveyed in a 2015 study in Indonesia said they shared injection equipment while injecting drugs in prison, and 78% said they shared equipment with ten or more other prisoners. 203 UNODC in 2015 summarized reports from 43 UN member states that had estimated or surveyed lifetime, annual, and past-month drug use while in prison among people in custody ( Figure 16 ). 7 A study of drug use in prison in the European Union, found that reported rates of ever having injected drugs in prisons among the countries providing data were in the range of 15–30%. 204 While some PWUD before serving a prison sentence will discontinue or reduce their use in prison or change their method of use, some people will seek to maintain drug use, including drug injection, or will begin using drugs while in prison. 205 In addition to drug-related risk, PWUD in prison face HIV risks associated with unprotected sex, sexual violence, and unsafe tattooing. As noted above, risk of sexual transmission of HIV may persist after prison if incarceration destabilises existing sexual relationships. 176
From UNODC country reporting, World Drug Report 2015 7
Numerous studies have documented HIV and HCV transmission in prison linked to drug injection, and others have demonstrated high prevalence of HIV and HCV among formerly incarcerated persons compared to other populations. 202 One study estimated that about 10% of adults in the penitentiary system of the Russian Federation inject drugs with two thirds of those sharing syringes. 202
There is not a recent comprehensive accounting of HIV prevalence in prison worldwide. In its 2014 report on gaps in the global HIV response, UNAIDS noted results for selected countries: HIV prevalence among adult prisoners is 15 times higher than in the general adult population in Ukraine, 10 times higher in Argentina, and 2.4 times higher in both South Africa and the US. 57 In 2007, 11 sub-Saharan African countries reported HIV prevalence in prisons at least twice that in the general population. 206 Data on HIV transmission in prison are rarer. An evaluation of attributable risks among PWID in a long-term study in Vancouver concluded that 21% of the HIV infections in this population were likely acquired in prison. 207
People living with HCV infection are also over-represented in prison in many countries. Based on data from 39 countries, a 2013 review found an average prevalence of 26% positivity to HCV antibody among people in prison and about 65% among prisoners with a history of drug injection. 208 Overall prevalence among women in this analysis was 32% compared to 24% for men. 208 A number of the reviewed studies presented evidence of HCV transmission in prisons. HCV prevalence in prisons may be high even where harm reduction services are available in the community. In Australia, for example, the prevalence of HCV infection among people entering prison in 2010 was 22% and among those with a history of drug injection 51%. 209 Phylogenetic and spatial analysis in Australia located a number of clusters of in-prison HCV transmission and suggested high transmission risk when people move from between prisons or from prison to the community. 210
The Commission sought to investigate through mathematical modelling the contribution of incarceration to HCV transmission among PWID in several countries. Given the high incarceration rate amongst PWID 211–214 and association between HCV infection or high risk behavior and a history of incarceration, 213, 215–224 it is unsurprising that incarceration could play an important role in driving HCV transmission among PWID.
According to our results, interventions that aim to reduce transmission risk in prison (such as OST and possibly HCV treatment) 225,226 or post-release could substantially reduce HCV incidence among PWID. Figures 17 and and18 18 illustrate this point, using modelling undertaken by Commissioner Peter Vickerman and colleagues Jack Stone and Natasha Martin to consider the role of incarceration on HCV transmission among PWID in four illustrative scenarios (similar to Scotland, Australia, Ukraine and Thailand), chosen to mimic important PWID incarceration characteristics of varied global settings. Using a model calibrated to a Scotland-like scenario, where moderate levels of incarceration (61% of PWID ever incarcerated, 12.7% in the last 6 months), and short sentence lengths for PWID (average 7 months) 71,220,227 means that PWID spend 16% of their injecting career being incarcerated (1.1 years). Despite lower HCV incidence amongst incarcerated PWID than amongst PWID in the community in Scotland 227 (likely due to high levels of prison OST– 57% – among incarcerated PWID), 71,227 modelling suggests incarceration still has a negative impact on the overall HCV epidemic due to the elevated risk of HCV acquisition among recently released PWID (threefold greater in first 6 months following release). 220 Indeed, prison contributes only 5% of incident infections, whereas 24% of all incident infections occur in the period of elevated risk post-release. Because of the heightened HCV incidence post release, the HCV incidence amongst PWID in our Scotland-like scenario could be 47% lower if this risk was not present with OST maintained ( Figure 18 ), but only 20% lower if incarceration had no effect on HCV transmission during or after prison.
Modelled overall endemic HCV incidence amongst PWID resulting from various effects of incarceration in several illustrative global settings
From Prof. Vickerman and colleagues
Modelled relative reduction in the overall endemic HCV incidence among PWID for each illustrative scenario if there was no elevated risk associated with incarceration (light gray bars) or if there was high scale-up of prison OST and no elevated risk post-release (dark gray bars)
From Prof. Vickerman and colleagues
Although Australia has similar incarceration rates and durations to those of Scotland, a lower level of prison OST (19% PWID receiving OST in prison) 228 correlates with higher HCV incidence among incarcerated PWID, 70,229 such that 22% of incident infections may occur in prison. In a setting such as Australia, modelling indicates HCV incidence amongst PWID could possibly be 49% lower if incarceration had no effect on HCV transmission and 66% lower with high coverage of prison OST and no elevated risk following release ( Figure 18 ). By comparison, in a setting with similar incarceration dynamics to those of Ukraine, where PWID receive longer sentences on average (14 months) compared to Scotland or Australia and inject for much longer (25 years on average compared to 7 years in Scotland), the lower proportion of PWID recently or ever incarcerated (52% of PWID ever incarcerated, 9.7% in last 6 months) results in a similar proportion of their injecting career in prison (18% of injecting career or 4.4 years) as in Scotland and Australia. Here, possibly due to longer durations of injecting, the overall contribution of prison to the epidemic amongst PWID could be far less than the other two settings if the pattern of transmission risk in and out of prison were similar; the model projects that HCV incidence amongst PWID could only be 14% lower if incarceration had no effect on HCV transmission and 26% lower if there was higher coverage of prison OST and no elevated risk following release.
By contrast, in a setting with similar incarceration patterns to those of Thailand, the combination of high incarceration rates (80% of PWID ever incarcerated, 17% in the last 3 months) and long prison sentences for PWID (12 months on average) 230,231 means PWID are likely to be incarcerated for a substantial proportion of their injecting career (estimated at 46%), and to experience numerous periods of elevated HCV transmission post-release. For this Thailand-like scenario, the model estimates 53% of incident infections could occur in prison; HCV incidence could be 60% lower if incarceration had no effect on HCV transmission, and 94% lower with high coverage of prison OST and no increased risk following release ( Figure 18 ). This analysis, although illustrative, highlights that incarceration could contribute substantially to HCV transmission among PWID and supports a growing body of evidence that interventions to reduce HCV risk among PWID in prison and post-release (such as OST and possibly HCV treatment) 226 could result in substantial benefits to the community and reduction in HCV transmission.
TB in prison and other closed settings has long been a public health concern, but TB risks increase in the presence of drug injection in closed settings. Overcrowding, poor sanitation, inadequate ventilation, the relatively high prevalence of HIV, and the insufficiency of basic services all contribute to TB transmission in prison. 232 The significant representation in prisons in many countries of people with HIV, PWUD, people living in poverty, and formerly incarcerated people means that many people in custodial settings have been exposed to TB before they are incarcerated. 144 Biadglegne and colleagues in 2015 indicated sub-Saharan Africa and Eastern Europe and Central Asia as the regions of greatest concern for TB transmission in prison, though data from Africa are sparse. 232 Central Asia has the highest estimated prevalence of TB and MDR-TB of any region. 147 A widely cited study by Stückler and colleagues estimated that in Eastern Europe and Central Asia from 1991 to 2002, increases in the rate of incarceration accounted for about 60% of the increase in TB in the general population. 233 In the WHO-Europe region, which includes Eastern Europe, it was estimated in 2010 that the relative risk for TB in prison was 145 times higher than in the community. 234 WHO cites the estimate that worldwide about 1 in 11 cases of TB transmission in high-income countries occurs in prison, and about 1 in 16 in low- and middle-income countries. 144
Containing MDR-TB is crucial to national TB responses. Both MDR-TB and XDR-TB – extensively drug-resistant TB – have been reported in prisons at high levels in some cases. 146 A 2015 review of MDR-TB in prison found, for example, about 19% of all TB patients in Thai prisons were classified as MDR-TB, 13–55% in the Russian Federation, 52% in Azerbaijan, and almost 10% in Zambia. 232 One study in the Russian Federation found XDR-TB cases to be 11% of TB patients tested in the prison. 232
Infectious disease in prisons is a heavy burden in EECA. Central Asia is estimated to have the highest rate of HCV prevalence among prisoners of any region. 235 Ukraine, with the next-highest prison population in Eastern Europe after the Russian Federation, reported an HIV prevalence among prisoners of 14.5% in 2008 and 13.6% in 2011, 235 as against HIV prevalence in the general population in that period of 1.2%. 236 The Russian Federation has prison population of about 800 000 annually – the second highest in the world after the US – of which about 20% is estimated to have been convicted of drug offenses. 237 It has not participated in reporting on HIV in prisons in the Dublin Declaration process, but it has experienced HIV outbreaks in prison in recent years. 202 In 2002 it was estimated that the 36 000 HIV-positive persons in Russian prisons at the time accounted for about 20% of all HIV cases in the fast-growing epidemic in the country. 205
It is an international norm that people in prison and other custodial settings have a right to health services at the level of those offered in the community in their jurisdictions. 238 When it comes to HIV, HCV, and TB services, that norm is far from being respected. UNODC and WHO recommend a comprehensive package of measures for HIV prevention, care, and support for incarcerated persons, including NSP and OST (web appendix VI). 239 These measures are also important for HCV prevention and care. Making these measures a reality, however, is proving challenging.
OST has been shown in many countries to be very effective in custodial settings where people can be directly observed in taking medicine and can be followed if they have problems with dosage. 205 But, according to a 2014 estimate by Harm Reduction International, of the 80 countries where OST is available in the community, only 43 provide the services in at least one prison. 76 In all of East and Southeast Asia, only Malaysia and Indonesia provide OST in prison. 76 Even in the European Union, which has high OST coverage in the community, OST in prison lags behind OST in the general population ( Figure 19 ), though it is offered in prison in the majority of EU countries. 240 Eight countries in Western, Central and Eastern Europe allow a people in prison to benefit from OST only if they were already OST patients before incarceration. 235 In Central Asia where the need for HIV and HCV prevention services is so great, only Kyrgyzstan has OST and NSP in prison. 76 OST is generally absent from US prisons but is available in most Canadian prisons. 76 Resistance to OST in prisons is motivated partly by the belief, also found outside corrections systems, that any drug treatment in prison should be abstinence-based. As was noted by authorities in Scotland in the 1990s, however, it is as unrealistic to aspire to a drug-free prison as it is to aspire to a drug-free society. 205
From EMCDDA, Prisons and drugs in Europe, 2012. 240
As noted by Kopak, the failure to provide effective treatment and care to people with problem drug use in the enormous US prison population perpetuates crime when people are released and returned to their previous circumstances. 241 In the European Union, EMCDDA judged that in 2013 the availability of drug treatment programmes tailored especially to people in prison was “extensive.” 75 Most EU countries report a variety of treatment options for drug dependence in prison, including “low-intensity” counseling, therapeutic community-type interventions, detoxification using various methods, abstinence-based Narcotics Anonymous, and group sessions, in addition to OST. 240
Provision of sterile injection equipment in prison is even rarer than OST, having been established and sustained in prisons in only eight countries, mostly in Western Europe. 240 Several countries in Eastern Europe had prison NSP but were unable to sustain the programmes, which are always politically challenging. 76 In a few countries, prison staff have resisted these programmes, and advocates for the programmes have faced the argument that providing injecting equipment encourages drug use. But in the case of Germany, when closure of NSP in prisons was proposed, prison workers protested, knowing that the programme protected them from injuries with contaminated needles as well as protecting the prison population. 205
HCV diagnosis and treatment services are limited in prisons in many countries. Diagnosing HCV is not a good investment if treatment cannot be provided, and the cost of HCV medicines as well as the need to ensure treatment over a long period are likely to have impeded treatment as a priority in prisons, especially in Western Europe where some drug sentences are relatively short. 240 In the US from 2000 to 2012, only 12 of 50 state prison systems did any systematic HCV antibody testing. 241 The much shorter duration of DAA therapies may make them more attractive in prisons, and a recent UK-based cost-effectiveness analysis indicates that HCV testing and treatment with short-course DAA therapy is cost-effective. 226 One middle-income country that has made a breakthrough is Georgia, which struck a deal with Gilead Sciences for concessionary pricing on its DAA sofosbuvir and decided to include free treatment for people in prison who need it. 244
HIV and HCV services other than these harm reduction measures are equally important and frequently lacking in prisons and pretrial detention settings. Availability of voluntary HIV testing at any time during incarceration is recommended by WHO and UNODC, 239 and some jurisdictions routinely offer HIV tests to people entering prisons. 76 Nonetheless, UNAIDS reports have consistently concluded that people in prison have poor access to HIV testing and treatment. 57 While research from North America has shown that optimal outcomes from ART can be achieved in prisons, 245 a large body of work from a range of settings shows that among PWID, incarceration is often strongly associated with poor access and adherence to ART, premature discontinuation of ART, and low rates of viral suppression. 246–249 It appears that problems ensuring access to ART and related care within North America occurs throughout the incarceration process (e.g., in detention, during transfers, at discharge), and low access and adherence is shaped by high rates of HIV-related stigma and concerns about breaches of privacy within prison systems. 246,250 (see web appendix VII).
WHO’s 2007 global review of prison HIV services found virtually no ART in prison in countries with significant populations of PWID outside the Global North. 251 Reviews in 2010 and 2014 of ART availability in prisons in the five countries outside the US with the largest number of PWID – the Russian Federation, China, Malaysia, Vietnam and Ukraine – indicated very limited ART overall for PWID in the community and virtually none at scale in prisons. 56,83 Indonesia, a country with significant representation of PWID in prison, has provided ART to prisoners incarcerated for drug offenses. A 2015 study of randomly sampled prisoners found that HIV-positive prisoners with a history of drug use were more likely to be receiving ART partly because they had been incarcerated for longer periods than other prisoners. 252 Work from a range of settings, including Zambia, 253 Namibia, 254 India, 255 Argentina, 256 Brazil, 257 and Thailand, 258 reveals low rates of engagement in HIV care, which often reflect structural and social barriers, including suboptimal health systems, privacy concerns, and violence. Whether for PWUD or others, WHO has recommended that ART in prison be given in a way that ensures confidentiality of the prisoner’s HIV status and that treatment efforts take care to ensure continuity of treatment for prisoners who are transferred or released. 152
Prisons are an extremely high-risk environment for tuberculosis, but prison TB services remain inadequate in many countries, making TB a risk of incarceration. The significant representation in prisons in many countries of people with HIV, people who use drugs, people living in poverty, and formerly incarcerated people means that many people in custodial settings have been exposed to TB before they even face the TB risks of prisons. 144
WHO and UNODC recommend a range of measures to control TB in prison, including:
“active case finding” including systematic offer of TB tests to all people in custody as well as monitoring of respiratory symptoms;
case reporting to a central health authority;isoniazid preventive therapy for people living with HIV in prison, even in the absence of a positive TB test;
treatment of TB and reliable linking to care in the community if the course of treatment is of longer duration than the custodial sentence;
improvements in ventilation and sanitation; provision of TB information to people in custody; and offer of HIV testing for people testing positive for TB. 239,259TB testing does not take place systematically in many prisons. 144 Among the many barriers to TB services is the fact that prison health services are often not managed by health ministries, which may compromise the quality and coverage of care in prisons and may impede ensuring continuity of care between prison and the community. 234 Loss to follow-up of people receiving TB treatment in prison is also a major challenge. An estimated 60–70% of prisoners testing positive for TB in Eastern Europe are not referred to any care in the community upon release. 260 Dara and colleagues also note that prisons in many countries have not invested in laboratory capacity to use the GeneXpert assay that WHO recommends for diagnosis but rather rely on less accurate tools. 260 Collaboration between TB and HIV authorities and integration of interventions in the two areas is crucial for TB control in prisons but lacking in many places. 259
Capacity to address HIV, HCV and TB in prisons, using the best medicines and diagnostic tools available in the community, is obviously dependent on financial resources. The Global Fund has been an important source of funding for HIV and TB interventions in prison, enabling previously unavailable services to be expanded in prisons, especially in EECA and sub-Saharan Africa. 261 But many EECA countries have already become or soon will be ineligible for Global Fund support, 108 and it remains to be seen whether governments or other donors will fund these services.
Even where services are available to people in state custody, delivering them in a patient-centred way is a particular challenge given the coercive nature of incarceration. Another central challenge is ensuring continuity of care upon release. Previously unpublished work in northern California by Commissioner Megan Comfort and colleagues illustrates that a lack of discharge planning and coordination of services virtually ensures the disruption of care. Among the 60 persons living with HIV in an in-depth qualitative study, many described being released from county jail around midnight. Although it was standard practice to provide a 30-day supply of medications at release, if people were discharged when the jail pharmacy was closed, they left with no medications at all. Furthermore, participants characterised leaving jail in the middle of the night as generally destabilising for them, especially when public transportation was not running. The feeling of being sent back out onto the streets without even the most basic means of “reentering” the community encouraged people to immediately seek comfort in familiar activities, such as drug use, rather than wait for daylight to take the uncomfortable steps of seeking services on their own.
The importance of continuity of care is illustrated quantitatively using data from the US and Canada reported by Iroh and colleagues ( Figure 20 ). 262 These authors conclude that testing and treatment can be achieved for people in prison, even at higher rates than in the general population, but that without attention to links to care after release, treatment interruptions are likely and may have serious health consequences. 262 A PAHO report on HIV services in the Caribbean also concludes that for the large prison populations in that region, ensuring HIV services in prison is less challenging than making reliable links to care in the community for those leaving prison. 263
From Iroh et al., 2015. 262
