The review by Embleton and colleagues  points to the same situation analysis and gaps in research and knowledge identified in the early 1990s, when the World Health Organization (WHO) Program on Substance Abuse (PSA) started the WHO/PSA Street Children Project [2, 3]. The problems with definitions, difficulties in collecting standardized information, following-up ‘hidden populations’, the role of substance use in the lives of street children and the need for standardized and tested methods to assess and intervene with these groups were raised, discussed and researched and a comprehensive training package was published in 2000 . Thousands of street children were interviewed in the study, hundreds received training to conduct assessments and provide interventions according to the principles of the training package developed for the project, governmental and non-governmental agencies from more than 20 countries participated, and projects were implemented mainly in developing nations. One can easily guess that millions of dollars were raised directly and indirectly in the last 20 years with the goal of helping street children to reduce their substance use problem.
The WHO also published a guide to drug abuse epidemiology in 2000  which included a chapter on special population studies, describing some findings from the Street Children Project and recommending ethnographic and quantitative methods of assessing and understanding substance use among these and other vulnerable groups.
More than 20 years later, I feel disheartened to read a systematic review that does not include any reference to WHO and the large program of work it conducted on substance use among street children; it covers only the English literature and excludes work with these populations in developed countries or among indigenous groups (where substantial research has been carried out). Health outcomes are poorly researched, but there are also significant difficulties with researching these groups in an ethical manner.
At the same time, the issue of street children seems to have disappeared altogether from the spotlight, from the news, from new research funding initiatives and from aid programs as well. Did the children or their problems disappear all of a sudden? Estimates in the early 1990s indicated that between 10 and 100 million street children existed around the world, depending on their definition. Expecting that a proportion survived into the 21st century, what would their lives look like today? Where can they be found? How many more or less are there now?
Street children, no matter what they are called and where they are, exist largely as a result of rapid urbanization, migration, violence and war, exclusion policies and social inequities that continue to exist around the world. I wonder if, under the umbrella of the social determinants of health, these groups are now dealt with under a different name, and with that one can no longer see the faces of those affected, or their ills. Compared to the lack of access to education, health services or security, substance use has again become a minor issue which can be overlooked easily in the big health and social agenda. Volatile solvents were identified as a prevalent problem a long time ago which very few governments ever cared about addressing, unless it affected middle-class or otherwise integrated youth. I believe street children somehow became lost again, and along with that their problems from the eyes of researchers and policy makers. Countries which have invested in social development may have had an impact upon the number of street children they have, if they were counting, but the problems of those who are still there are likely to be the same, and will require more resources to eliminate.
Making these populations and their health outcomes visible again should be one of the priorities of the agenda on social determinants of health. Longitudinal studies are difficult to conduct, not only because these children are difficult to follow-up, have many other health risks which are difficult to assess and control for and probably die early from non-substance-use-related problems, such as violence and lack of access to health services. It would also be unethical to observe a situation without interfering to modify it. Providing health, education and social support to these children should be a duty of the state, and changing their present and future would also make the assessment of the impact of solvent use or other drugs more meaningful, because it could be linked to a variety of interventions.
Finally, what is not counted often does not count. Volatile solvents are legal, so they fall out of the mandates of the UN Conventions on illicit drugs. They are used by people who fall out of the general or student population surveys and from vital statistics at country level, so their impact is difficult to estimate and therefore their importance is minimized without any evidence of their actual impact. Solvent use among street children and other vulnerable groups seems to be lost in the 21st-century health and research agendas. It might be time for ‘affirmative’ action to change this situation.
Declaration of interests