Antimicrobial Resistance: Is Health Technology Assessment Part of the Solution or Part of the Problem?

Antimicrobial resistance is a serious challenge to the success and sustainability of our healthcare systems. There has been increasing policy attention given to antimicrobial resistance in the last few years, and increased amounts of funding have been channeled into funding for research and development of antimicrobial agents. Nevertheless, manufacturers doubt whether there will be a market for new antimicrobial technologies sufficient to enable them to recoup their investment. Health technology assessment (HTA) has a critical role in creating confidence that if valuable technologies can be developed they will be reimbursed at a level that captures their true value. We identify 3 deficiencies of current HTA processes for appraising antimicrobial agents: a methods-centric approach rather than problem-centric approach for dealing with new challenges, a lack of tools for thinking about changing patterns of infection, and the absence of an approach to epidemiological risks. We argue that, to play their role more effectively, HTA agencies need to broaden their methodological tool kit, design and communicate their analysis to a wider set of users, and incorporate long-term policy goals, such as containing resistance, as part of their evaluation criteria alongside immediate health gains.

Quantifying uncertainty about future antimicrobial resistance: Comparing structured expert judgment and statistical forecasting methods.

The increase of multidrug resistance and resistance to last-line antibiotics is a major global public health threat. Although surveillance programs provide useful current and historical information on the scale of the problem, the future emergence and spread of antibiotic resistance is uncertain, and quantifying this uncertainty is crucial for guiding decisions about investment in antibiotics and resistance control strategies. Mathematical and statistical models capable of projecting future rates are challenged by the paucity of data and the complexity of the emergence and spread of resistance, but experts have relevant knowledge. We use the Classical Model of structured expert judgment to elicit projections with uncertainty bounds of resistance rates through 2026 for nine pathogen-antibiotic pairs in four European countries and empirically validate the assessments against data on a set of calibration questions. The performance-weighted combination of experts in France, Spain, and the United Kingdom projected that resistance for five pairs on the World Health Organization's priority pathogens list (E. coli and K. pneumoniae resistant to third-generation cephalosporins and carbapenems and MRSA) would remain below 50% in 2026. In Italy, although upper bounds of 90% credible ranges exceed 50% resistance for some pairs, the medians suggest Italy will sustain or improve its current rates. We compare these expert projections to statistical forecasts based on historical data from the European Antimicrobial Resistance Surveillance Network (EARS-Net). Results from the statistical models differ from each other and from the judgmental forecasts in many cases. The judgmental forecasts include information from the experts about the impact of current and future shifts in infection control, antibiotic usage, and other factors that cannot be easily captured in statistical forecasts, demonstrating the potential of structured expert judgment as a tool for better understanding the uncertainty about future antibiotic resistance.

Health and economic benefits of scaling up a home-based neonatal care package in rural India: a modelling analysis.

Approximately 900 000 newborn children die every year in India, accounting for 28% of neonatal deaths globally. In 2011, India introduced a home-based newborn care (HBNC) package to be delivered by community health workers across rural areas. We estimate the disease and economic burden that could be averted by scaling up the HBNC in rural India using IndiaSim, an agent-based simulation model, to examine two interventions. In the first intervention, the existing community health worker network begins providing HBNC for rural households without access to home- or facility-based newborn care, as introduced by India's recent programme. In the second intervention, we consider increased coverage of HBNC across India so that total coverage of neonatal care (HBNC or otherwise) in the rural areas of each state reaches at least 90%. We find that compared with a baseline of no coverage, providing the care package through the existing network of community health workers could avert 48 [95% uncertainty range (UR) 34-63] incident cases of severe neonatal morbidity and 5 (95% UR 4-7) related deaths, save $4411 (95% UR $3088-$5735) in out-of-pocket treatment costs, and provide $285 (95% UR $200-$371) in value of insurance per 1000 live births in rural India. Increasing the coverage of HBNC to 90% will avert an additional 9 (95% UR 7-12) incident cases, 1 death (95% UR 0.72-1.33), and $613 (95% UR $430-$797) in out-of-pocket expenditures, and provide $55 (95% UR $39-$72) in incremental value of insurance per 1000 live births. Intervention benefits are greater for lower socioeconomic groups and in the poorer states of Chhattisgarh, Uttarakhand, Bihar, Assam and Uttar Pradesh.

Quantifying uncertainty in intervention effectiveness with structured expert judgement: an application to obstetric fistula.

OBJECTIVE: To demonstrate a new application of structured expert judgement to assess the effectiveness of surgery to correct obstetric fistula in a low-income setting. Intervention effectiveness is a major input of evidence-informed priority setting in healthcare, but information on intervention effectiveness is generally lacking. This is particularly problematic in the context of poorly resourced healthcare settings where even efficacious interventions fail to translate into improvements in health. The few intervention effectiveness studies related to obstetric fistula treatment focus on the experience of single facilities and do not consider the impact of multiple factors that may affect health outcomes. DESIGN: We use the classical model of structured expert judgement, a method that has been used to quantify uncertainty in the areas of engineering and environmental risk assessment when data are unavailable. Under this method, experts quantify their uncertainty about rates of long-term disability in patients with fistula following treatment in different contexts, but the information content drawn from their responses is statistically conditioned on the accuracy and informativeness of their responses to a set of calibration questions. Through this method, we develop best estimates and uncertainty bounds for the rate of disability associated with each treatment scenario and setting. PARTICIPANTS: Eight experts in obstetric fistula repair in low and middle income countries. RESULTS: Estimates developed using performance weights were statistically superior to those involving a simple averaging of expert responses. The performance-weight decision maker's assessments are narrower for 9 of the 10 calibration questions and 21 of 23 variables of interest. CONCLUSIONS: We find that structured expert judgement is a viable approach to investigating the effectiveness of medical interventions where randomised controlled trials are not possible. Understanding the effectiveness of surgery performed at different types of facilities can guide programme planning to increase access to fistula treatment.

Analysis of the Universal Immunization Programme and introduction of a rotavirus vaccine in India with IndiaSim.

BACKGROUND AND OBJECTIVES: India has the highest under-five death toll globally, approximately 20% of which is attributed to vaccine-preventable diseases. India's Universal Immunization Programme (UIP) is working both to increase immunization coverage and to introduce new vaccines. Here, we analyze the disease and financial burden alleviated across India's population (by wealth quintile, rural or urban area, and state) through increasing vaccination rates and introducing a rotavirus vaccine. METHODS: We use IndiaSim, a simulated agent-based model (ABM) of the Indian population (including socio-economic characteristics and immunization status) and the health system to model three interventions. In the first intervention, a rotavirus vaccine is introduced at the current DPT3 immunization coverage level in India. In the second intervention, coverage of three doses of rotavirus and DPT and one dose of the measles vaccine are increased to 90% randomly across the population. In the third, we evaluate an increase in immunization coverage to 90% through targeted increases in rural and urban regions (across all states) that are below that level at baseline. For each intervention, we evaluate the disease and financial burden alleviated, costs incurred, and the cost per disability-adjusted life-year (DALY) averted. RESULTS: Baseline immunization coverage is low and has a large variance across population segments and regions. Targeting specific regions can approximately equate the rural and urban immunization rates. Introducing a rotavirus vaccine at the current DPT3 level (intervention one) averts 34.7 (95% uncertainty range [UR], 31.7-37.7) deaths and $215,569 (95% UR, $207,846-$223,292) out-of-pocket (OOP) expenditure per 100,000 under-five children. Increasing all immunization rates to 90% (intervention two) averts an additional 22.1 (95% UR, 18.6-25.7) deaths and $45,914 (95% UR, $37,909-$53,920) OOP expenditure. Scaling up immunization by targeting regions with low coverage (intervention three) averts a slightly higher number of deaths and OOP expenditure. The reduced burden of rotavirus diarrhea is the primary driver of the estimated health and economic benefits in all intervention scenarios. All three interventions are cost saving. CONCLUSION: Improving immunization coverage and the introduction of a rotavirus vaccine significantly alleviates disease and financial burden in Indian households. Population subgroups or regions with low existing immunization coverage benefit the most from the intervention. Increasing coverage by targeting those subgroups alleviates the burden more than simply increasing coverage in the population at large.