Impact of joined-up HIV harm reduction and multidrug resistant tuberculosis control programmes in Estonia: System dynamics simulation model
Introduction
In Eastern Europe and Central Asia (ECA) the control of tuberculosis, multidrug resistant tuberculosis (MDRTB) and human immunodeficiency virus (HIV) poses important public health challenges [1]. Damage inflicted by HIV on cell-mediated immunity is a major risk factor for progression of latent tuberculosis infection to active disease. Death rates in patients with MDRTB are often high, particularly in individuals co-infected with HIV [2].
At a population level, the impact of HIV on tuberculosis rates will, amongst other factors, depend on: (a) the prevalence of latent tuberculosis infection; (b) the prevalence of active tuberculosis and pulmonary tuberculosis; (c) the prevalence of drug resistant tuberculosis; (d) the prevalence of HIV and the populations infected; (e) degree of immune suppression in HIV-infected individuals; (f) effectiveness of interventions aimed at controlling and treating tuberculosis, other opportunistic infections, and HIV.
Estonia is a Baltic State that regained independence from the USSR in 1991 and, in 2004, joined the European Union (EU). The country is witnessing an explosion in new cases of HIV where the cumulative number of newly identified cases of HIV has risen from 8 new cases in 1996 to a peak of 1474 new cases in 2001, a rate of increase that exceeds that of all other states in the former Soviet Union (FSU) [3]. By the end of 2004, there were almost 4500 registered HIV positive individuals [4]. Concurrent with this increase in the number of HIV infections, tuberculosis notifications have also increased, from 26 per 100,000 in 1992 to 52 per 100,000 in 1999, although since 2000, when the WHO-recommended directly observed therapy strategy (DOTS) was introduced, notifications of tuberculosis have stabilized [5]. In 1998, Estonia had one of the highest MDRTB rates in the world—with 14% of newly diagnosed pulmonary tuberculosis cases being MDRTB [5]. The tuberculosis global surveillance program identified Estonia as one of the “hot spots” for high MDRTB prevalence [6], [7], [8]. Although there is an ongoing programme to manage and control MDRTB, cure rates were initially estimated to be 20% but unpublished results indicate the cure rates may be increasing to around 50–60% (personal communication, Dr. Manfred Danilovits, National TB Programme, March 2005). Weaknesses in MDRTB control are of particular concern where HIV prevalence is also high or rising [2], [3]. Given the current immaturity of the HIV epidemic in Estonia, the epidemiological impact of the linkage between HIV and tuberculosis is not yet evident, but is likely to become so as populations at risk of HIV acquisition and progression mix with those who have active tuberculosis. As immune function deteriorates in those co-infected, the number of cases of active HIV-associated tuberculosis is likely to rise.
Although many factors impact upon HIV spread in the FSU, epidemics have been principally driven by injecting drug use [3], [9], [10], [11], [12]. In Estonia, injecting drug use remains the main driver of the epidemic with approximately 90% of new HIV infections associated with drug use [12].
This paper explores the potential epidemiological impact of different policy choices in communicable disease control in a setting where control of MDRTB and HIV are both priority public health challenges.
Section snippets
Methodology
We developed a system dynamics (SD) simulation model [13], [14], [15], described in detail elsewhere [16], to represent the transmission dynamics of tuberculosis and HIV. The transmission of infectious diseases is complex, involving interconnected systems with feedback loops, time delays and non-linear relationships. The behaviour of these complex systems is generally difficult to predict. Formal modelling approaches, such as SD, may be used to represent the transmission dynamics and simulate
Results
Possible outcomes for each of the five scenarios are summarised in Table 3 and are presented graphically in Fig. 2, Fig. 3. Under the baseline scenario, which represents the current situation with respect to the HIV/AIDS transmission drivers (PSH and DIF) and MDRTB cure rates, the cumulative number of AIDS deaths will reach 10,522 by the year 2023. With respect to tuberculosis deaths, the total cumulative tuberculosis-associated deaths will be 1543—of which 1066 will be deaths from DSTB (70%)
Discussion
Our findings have important policy implications and inform decisions developing communicable disease programme in Europe. We demonstrate close interrelationships between HIV, DSTB and MDRTB. In settings of high MDRTB prevalence, with immature, yet explosive HIV epidemics among IDUs, effective HIV harm reduction and DSTB/MDRTB control programmes must be established concurrently if substantial numbers of deaths are to be averted. Early in an HIV epidemic centred on the IDU population, inadequate
References (56)
- et al.
Health care systems frailties and public health control of communicable disease on the European Union's new eastern border
Lancet
(2004) - et al.
HIV in central and Eastern Europe
Lancet
(2003) - et al.
Modelling emerging HIV epidemics: the role of injecting drug use and sexual transmission in the Russian Federation, China and India
International Journal of Drug Policy
(2003) - et al.
The role of injection drug use in the emergence of human immunodeficiency virus infection in Estonia
International Journal of Infectious Diseases
(2002) Misperceptions of feedback in dynamic decision making
Organizational Behaviour and Human Decision Processes
(1989)- et al.
Effects of feedback complexity on dynamic decision making
Organizational Behaviour and Human Decision Processes
(1995) AIDS and the health crisis of the US urban poor: the perspective of critical medical anthropology
Social Science and Medicine
(1994)The time table of tuberculosis
Tubercle
(1948)- et al.
Prospects for worldwide tuberculosis control under the WHO DOTS strategy
Lancet
(1998) Multidrug resistant tuberculosis: public health challenges
Tropical Medicine & International Health
(2004)
Global trends in resistance to antituberculosis drugs.
New England Journal of Medicine
Global surveillance for antituberculosis drug resistance
New England Journal of Medicine
Estonia races to halt multi drug resistant TB before HIV takes hold
Bulletin of the World Health Organisation
Explosive spread and high prevalence of HIV infection among injecting drug users in Togliatti City, Russian Federation
AIDS
HIV infection associated with drug injecting in the Newly Independent States, Eastern Europe, The social and economic context
Addiction
Industrial dynamics
Introduction to system dynamics with DYNAMO
Business dynamics. Systems thinking and modeling for a complex world
Impact of an effective multidrug resistant tuberculosis control programme in the setting of an immature HIV epidemic: system dynamics simulation model
International Journal of STD & AIDS
Modeling management behavior: misperception of feedback in a dynamic decision making experiment
Management Science
Boom, bust and failures to learn in experimental markets
Management Science
Alternative conceptions of feedback in dynamic decision environments: an experimental investigation
Management Science
Communities with multiple afflictions: A System Dynamics Approach to the study and prevention of syndemics
A model of HIV transmission through needle sharing
Interfaces
Formal aspects of model validity and validation in system
System Dynamics Review
The treatment of multidrug-resistant tuberculosis in Turkey.
New England Journal of Medicine
Cited by (38)
A predictive decision analytics approach for primary care operations management: A case study of double-booking strategy design and evaluation
2023, Computers and Industrial EngineeringInsights from system dynamics applications in addiction research: A scoping review
2022, Drug and Alcohol DependenceCitation Excerpt :This review illuminated novel uses and applications of SD models in studying addiction-related processes and policies, as well as key gaps in the addiction processes and substances studied using SD. We found that SD applications in this review provided a better understanding of underlying addiction-related intervention processes (e.g., (Arantes Do Amaral and Hess, 2018; Brown et al., 2019)); provided virtual simulation test beds for policy implementation, including understanding potential pros and cons of hypothetical policies (e.g., (Chalmers and Ritter, 2012; Hill and Camacho, 2017; Houle and Siegel, 2009; Tengs et al., 2004)); pinpointed data needs, challenges, and variable relationships lacking evidence (e.g., (Koehler et al., 2019; McGregor et al., 2019)); helped understand syndemic substance use-related phenomena, or intertwined health outcomes acting synergistically (e.g., related to HIV, tuberculosis, drug crime, and substance use disorders) (Atun et al., 2007; Batchelder et al., 2015; Batchelder and Lounsbury, 2016; Nyabadza and Coetzee, 2017)); and provided an interactive game-like platform to learn about substance use-related effects (Shi et al., 2015). While this review highlighted the benefits and novel uses of SD modeling for addiction researchers, we also identified areas that warrant future attention in addiction research using an SD approach.
Dynamic modeling approaches to characterize the functioning of health systems: A systematic review of the literature
2017, Social Science and MedicineCitation Excerpt :They explored the impact of two policy scenarios, with different disease cure rates, on cumulative deaths. Another paper used the same transmission model to explore the impact of varying population coverage rates of highly active antiretroviral therapy (HAART) in addition to the two MDR-TB treatment scenarios (Atun et al., 2007a,b). Dangerfield et al. (2001) developed a HIV model, fitted to reflect empirical data in the United Kingdom at the time, to explore how the HIV epidemic changes with HAART.
Implications of long-term care capacity response policies for an aging population: A simulation analysis
2014, Health PolicyCitation Excerpt :SD offers a practical approach to engaging stakeholders in understanding the behavior of real-world systems over time in a way that allows alternative policies and scenarios to be tested systematically in order to answer both “what if” and “why” questions [22]. A substantial and expanding academic literature supports SD models as being useful in assisting policy formulation in healthcare and addressing the dynamic complexity that characterizes many public health issues [22–27]. SD modeling has been applied to issues of population health since the 1970s [22] and, more recently, has also been applied to human resource issues in healthcare.
Tuberculosis, injecting drug use and integrated HIV-TB care: A review of the literature
2013, Drug and Alcohol DependenceCitation Excerpt :Multidrug-resistance poses a major threat to TB control, particularly in Eastern Europe and Central Asia. Modelling data from this region suggest that effective harm reduction could decrease HIV and TB mortality by 30% and 50%, respectively, when combined with good MDR-TB control (Atun et al., 2005, 2007). Given the high prevalence of HCV among PWID and the risk of hepatotoxicity associated with TB prevention and treatment, incorporation of HCV care is also vital to improving the quality of integrated healthcare for PWID (Altice et al., 2010; Getahun et al., 2012).
The unequal health of Europeans: Successes and failures of policies
2013, The LancetCitation Excerpt :One of the few successes of the Soviet health system was control of infectious diseases. Unfortunately, the systems that many of the newly independent states inherited in 1991 rapidly collapsed, and diseases that had once been controlled (eg, diphtheria) re-emerged69 and new problems, most notably multidrug-resistant tuberculosis and HIV/AIDS, arose.70 Indeed, 15 of the 27 countries noted to have a high burden of multidrug-resistant tuberculosis are in central and eastern Europe or the former Soviet Union.