Two Polar Considerations in Treatment System Planning: Infrastructure Development and Real-Time Management.

Private markets fail to achieve either a sufficient supply of substance use disorder (SUD) services overall or an appropriate allocation of the right services to the right individuals at the right time. Prototypically, low-income individuals do not wield sufficient means to incentivize a robust supply of SUD services. Another market failure is the mismatch of needs and services. The famous “drinker’s pyramid” serves as a useful model, with most of the population being low risk, some being moderate or high risk, and a small group having a severe disease (Babor et al., 1999; Rush et al., 2014). Different “services” are needed for these different risk groups, with prevention and intervention services being the least likely to be supported by private markets. Meanwhile, demographics, socioeconomics, and geography often yield underserved populations.

In the face of this market failure, public planners must take into account the fractured or non-existent market in their nations. Such planning can take many forms but must have clear, measurable objectives that are unique to their populations and culture, infrastructure, and resource constraints. Ritter et al. (2019), Rush and Urbanoski (2019), and Storbjörk and Stenius (2019) offer principles for planning systems with better quality, efficiency, a seamless continuum of care, and fully enfranchised populations.

In recent years, the Inter-American Drug Abuse Control Commission of the Organization of American States (OASCICAD) has made efforts to improve the specialty treatment systems in constituent countries by increasing the use of evidence-based practices and improving infrastructure. As part of an independent evaluation of the projects of OAS and several countries, I observed firsthand their successful efforts to institutionalize training and certifications of individual service providers by government departments. Taking El Salvador as an example, 88 organizations around the country saw at least one of their staff trained and certified by the new program in its first year of operation (2011–2012). These organizations represented almost the whole universe of treatment programs registered with the federal government. Two thirds of them were nonprofit, nongovernmental organizations (NGOs), and the governmental organizations were associated with either the national health system or the criminal justice system. The government directly maintained one residential treatment program in its National Psychiatric Hospital. Most of the NGOs were residential programs that had evolved from mutual help organizations, and many considered themselves faith based. Around half followed a therapeutic community model. They maintained themselves through peer staff, self-pay, local and international aid, and church affiliation. Despite their having a nonclinical background and a variety of education levels, the modern curriculum developed by OAS and local university partners was well received by the provider participants.

In terms of system planning in the case of El Salvador, many fundamental questions remain. Although providers at NGOs are pleased that government certification adds a sense of professionalism and respect to their service provider role, remuneration is unchanged. It is unclear how the government might set up its procurement model in a way that avoids some of the marketization pitfalls laid out by Storbjörk and Stenius (2019), particularly since some of the larger NGOs are franchises of larger international organizations such as Hogares CREA. Moreover, the government faces the complex task of financing expanded treatment services and increasing the size of the clinically trained workforce, all while not displacing its home-grown, self-organized supply side of the treatment system that serves rural areas. Finally, how does it balance investments in care for “severe cases” with those of prevention and early intervention?

In other nations, planners face different challenges. Estimating the potential demand for services to allocate resources is an ongoing, evolving challenge. Methods for predicting demand have room to improve beyond simply utilizing levels of service use in previous years. In North America, Europe, and Australia, many nations have nationally representative surveys that are an under-used source for estimating treatment demand along with unmet need (Chalmers et al., 2016). Also under-used are newer triangulation methods that can improve estimates by adjusting for underreporting in surveys, particularly by subpopulations (Parish et al., 2017). Combined, these methods allow for estimation of counterfactual increases in demand under scenarios in which the system engages all individuals who need treatment and successfully links them to service. Importantly, such methods also have the potential to estimate services need at multiple risk levels, thus informing resource allocation for prevention and early interventions.