Scarce Public Health Resources Research Articles

Assessment of Patient Safety in a Low-Resource Health Care System: Proposal for a Multimethod Study.

The high prevalence of adverse events (AEs) globally in health care delivery has led to the establishment of many guidelines to enhance patient safety. However, patient safety is a relatively nascent concept in low- and middle-income countries (LMICs) where health systems are already overburdened and underresourced. This is why it is imperative to study the nuances of patient safety from a local perspective to advocate for the judicious use of scarce public health resources. This study aims to assess the status of patient safety in a health care system within a low-resource setting, using a multipronged, multimethod approach of standardized methodologies adapted to the local setting. We propose purposive sampling to include a representative mix of public and private, rural and urban, and tertiary and secondary care hospitals, preferably those ascribed to the same hospital quality standards. Six different approaches will be considered at these hospitals including (1) focus group discussions on the status quo of patient safety, (2) Hospital Survey on Patient Safety Culture, (3) Hospital Consumer Assessment of Healthcare Providers and Systems, (4) estimation of incidence of AEs identified by patients, (5) estimation of incidence of AEs via medical record review, and (6) assessment against the World Health Organization's Patient Safety Friendly Hospital Framework via thorough reviews of existing hospital protocols and in-person surveys of the facility. The abovementioned studies collectively are expected to yield significant quantifiable information on patient safety conditions in a wide range of hospitals operating within LMICs. A multidimensional approach is imperative to holistically assess the patient safety situation, especially in LMICs. Our low-budget, non-resource-intensive research proposal can serve as a benchmark to conduct similar studies in other health care settings within LMICs. PRR1-10.2196/50532.

Does increasing the number of beds or health workers contribute to the rational use of scarce public health resources?

Turkey makes substantial investments to increase the number of qualified beds in hospitals, the shortage in health professionals remains one of the main obstacles of the health system in the country. To address this research gap, the study aims to formulate a rational solution for the dilemma on whether to invest in beds or health professionals contribute to the rational use of scarce public health resources. Data for testing the model were derived from the Turkish Statistical Institute across 81 provinces in Turkey. The path analytic approach was used to determine the associations among hospital size, utilization/facility, health workforce, and indicators of health outcomes. The results point to a strong link between quantity of qualified beds, utilization of health services, and facility indicators, and health workforce. Rational use of scarce resources, optimal capacity planning, and increased quantity of health professionals will be beneficial for the sustainability of health care services.

Design and Analysis of Hospital Throughput Maximization Algorithm under COVID-19 Pandemic.

Under the global pandemic of COVID-19, public health facilities, such as hospitals, are required to readjust, design, and plan a safe movement flow of people to meet the social distance rules and quarantine COVID-19 and the non-COVID-19 patients to prevent cross-infection. However, readjustments to separate patients have significantly reduced the maximum throughput of public health facilities, worsening already scarce public health resources. Therefore, this paper proposes throughput maximization algorithms based on the one-way street problem which meets the requirements of social distance rules. First, the floor plan of a hospital is transformed into a graph, each node is traversed by breadth-first search. Then, this paper considers patients' node pair sets as different set unions, the direction of edges, and the color of links based on DFS-XOR algorithm are designed to distinguish the paths of COVID-19 and non-COVID-19 patients. Finally, this paper utilizes minimum shared link algorithms to determine the minimized sharing links between paths linking different set unions and components. The throughput is maximized by reducing the number of shared links and alternating links. The results indicate that compared with the brute force algorithms, the algorithms proposed in this paper significantly improve the maximum throughput.

Inter-annual variation in prevalence of Borrelia burgdorferi sensu stricto and Anaplasma phagocytophilum in host-seeking Ixodes scapularis (Acari: Ixodidae) at long-term surveillance sites in the upper midwestern United States: Implications for public health practice

The geographic range of the blacklegged tick, Ixodes scapularis, and its associated human pathogens have expanded substantially over the past 20 years putting an increasing number of persons at risk for tick-borne diseases, particularly in the upper midwestern and northeastern United States. Prevention and diagnosis of tick-borne diseases rely on an accurate understanding by the public and health care providers of when and where persons may be exposed to infected ticks. While tracking changes in the distribution of ticks and tick-borne pathogens provides fundamental information on risk for tick-borne diseases, metrics that incorporate prevalence of infection in ticks better characterize acarological risk. However, assessments of infection prevalence are more labor intensive and costly than simple measurements of tick or pathogen presence. Our objective was to examine whether data derived from repeated sampling at longitudinal sites substantially influences public health recommendations for Lyme disease and anaplasmosis prevention, or if more constrained sampling is sufficient. Here, we summarize inter-annual variability in prevalence of the agents of Lyme disease (Borrelia burgdorferi s.s.) and anaplasmosis (Anaplasma phagocytophilum) in host-seeking I. scapularis nymphs and adults at 28 longitudinal sampling sites in the Upper Midwestern US (Michigan, Minnesota, and Wisconsin). Infection prevalence was highly variable among sites and among years within sites. We conclude that monitoring infection prevalence in ticks aids in describing coarse acarological risk trends, but setting a fixed prevalence threshold for prevention or diagnostic decisions is not feasible given the observed variability and lack of temporal trends. Reducing repeated sampling of the same sites had minimal impact on regional (Upper Midwest) estimates of average infection prevalence; this information should be useful in allocating scarce public health resources for tick and tick-borne pathogen surveillance, prevention, and control activities.

Exploring young people's preferences for STI screening in the UK: A qualitative study and discrete choice experiment

IntroductionStigma remains a key issue for many health screening interventions such as screening for sexually transmitted infections (STIs). Young people continue to experience the greatest burden of STI infection. In order to increase uptake, screening services need to be more patient-focused. This study sought to examine young people's preferences for sexual health screening to understand how scarce public health resources can optimise screening uptake. MethodsThis study involved both qualitative and quantitative components. Focus groups and individual interviews were undertaken with young people aged 16–24 recruited from community settings and a specialist clinic. Themes which emerged from the focus groups were used to inform the design of a discrete choice experiment (DCE). A questionnaire survey (incorporating the DCE) was conducted with members of an internet panel, with over-sampling of black, Asian and minority ethnic groups. ResultsOverall, 41 participants took part in eight focus groups and two in individual interviews. Six major themes emerged as important when making decisions about STI screening - stigma and embarrassment; knowledge about STIs and risk; where to get tested; how staff would treat them; what STIs to be tested for; and convenience (waiting times). Overall, 1946 participants took part in the survey. The DCE results revealed that the most important factors for young people are that all STIs are tested for, and that staff attitude is non-judgemental. The results also suggest that there is a preference for screening in specialist clinics and for full appointments over limited ones. Although respondents preferred shorter time periods for appointments and results, other ‘process’ factors were also important. ConclusionThis study demonstrates that by combining qualitative and quantitative methods, a richer understanding of STI screening preferences is possible. The findings show that comprehensive testing and a perceived ‘non-judgemental’ attitude are particularly important to young people, as well as convenience.

Geographical Variation in Traumatic Brain Injury Mortality by Proximity to the Nearest Neurosurgeon

BackgroundTrauma mortality disproportionately affects populations farther from potentially lifesaving trauma care, and traumatic brain injury (TBI) is no exception. Previous examinations have examined proximity to trauma centers as an explanation for trauma mortality, but little is known about the relationship between proximity to neurosurgeons specifically in TBI mortality. Materials and methodsIn this cross-sectional study, county-level TBI mortality rates from 2008 to 2014 were examined in relation to the distance to the nearest neurosurgeon and trauma facility. The locations of practicing neurosurgeons and trauma facilities in the United States were determined by geocoding data from the 2017 Medicare Physician and Other Supplier and Provider of Services files (respectively). The association between TBI mortality and the distance from the population-weighted centroid of the county to a closest neurosurgeon and trauma facility was examined using multivariate negative binomial regression. ResultsA total of 761 of the 3108 counties (24.5%) in the continental United States were excluded from the analysis because they had 20 or fewer TBI deaths during this time, producing unstable estimates. Excluded counties accounted for 1.67% of the US population. Multivariate analysis revealed a county's mortality increased 10% for every 25 miles from the nearest neurosurgeon (adjusted incident rate ratio: 1.10 [95% confidence interval: 1.08-1.12]; P < 0.001). The distance to the nearest trauma facility was not found to be significantly associated with mortality (adjusted incident rate ratio: 1.01 [95% confidence interval: 0.99-1.03]; P = 0.36). ConclusionsThese findings suggest that proximity to neurosurgeons may influence county-level TBI mortality. Further research into this topic with more granular data may help to allocate scarce public health resources.

Geographic clustering and region-specific determinants of obesity in the Netherlands.

As a leading cause of morbidity and premature mortality, obesity has become a major global public health problem. It is therefore important to investigate the spatial variation of obesity prevalence and its associations with environmental and behavioral factors (e.g., food environment, physical activity), to optimize the targeting of scarce public health resources. In this study, the geographic clustering of obesity in the Netherlands was explored by analyzing the local spatial autocorrelation of municipal-level prevalence rates of adulthood obesity (aged ≥19 years) in 2016. The potential influential factors that may be associated with obesity prevalence were first selected from five categories of healthrelated factors through binary and Least Absolute Shrinkage and Selection Operator (LASSO) regressions. Geographically Weighted Regression (GWR) was then used to investigate the spatial variations of the associations between those selected factors and obesity prevalence. The results revealed marked geographic variations in obesity prevalence, with four clusters of high prevalence in the north, south, east, and west, and three clusters of low prevalence in the north and south of the Netherlands. Lack of sports participation, risk of anxiety, falling short of physical activity guidelines, and the number of restaurants around homes were found to be associated with obesity prevalence across municipalities. Our findings show that effective, region-specific strategies are needed to tackle the increasing obesity in the Netherlands.

Abstract P154: Disparities in Geographic and Temporal Trends of Myocardial Infarction Hospitalizations in Florida, 2005-2014

Background: Knowledge of the extent of geographic disparities in the burden of myocardial infarction (MI) is useful for allocation of scarce public health resources to reduce health disparities and improve population health, regardless of whether MI is the primary or secondary cause for hospitalization. The objectives of this study were to identify: (a) geographic disparities in hospitalization risks for MI in Florida; and (b) the temporal changes in these disparities from 2005 to 2014. Methodology: We aggregated county-level data for principal and secondary inpatient MI-related hospital discharges in Florida between 2005 and 2014 by 2-year intervals and calculated age- and sex- adjusted MI hospitalization risks for each time interval. We identified spatial clusters of low- and high-risk MI hospitalization risks using circular spatial scan statistics and tracked MI risks in clusters that persisted throughout the 10-year study period. We also assessed health disparities between persistent high- and low-risk clusters at the end of the study (2013-2014) compared to the beginning of the study (2005-2006) periods. Results: MI hospitalization risks decreased by 15% in Florida overall during the 10-year study period. However, we found persistent disparities in MI risks by geographic location, with high-risk clusters occurring in north-, west-, south-central and southeast Florida, and low-risk clusters occurring in southeast and southwest Florida. A low-risk cluster that transitioned to high-risk status in the last four years of study was identified in northwest Florida. We also found substantial differences in the magnitude of decline in MI risks amongst clusters, with risks decreasing by 5%, 16% and 31% in high-risk clusters in west central, south central and north central Florida, respectively, and by 6.5% and 26% in low-risk clusters in southwest and southeast Florida, respectively. Furthermore, the risks only decreased during the first 6-8 years of study, after which they leveled off or ticked upwards. Consequently, health disparities between high- and low-risk clusters at the end of the study compared to the beginning of the study period decreased by 57% and 31.5% in north- and south-central Florida, respectively, but they remained relatively unchanged in west central Florida. Moreover, MI hospitalization risks in high-risk clusters lag behind those in low-risk clusters by at least a decade. Conclusion: Myocardial infarction hospitalization risks declined modestly during 10-year study period. However, persistent disparities continue to exist across space and time. Addressing these disparities will require targeting intervention efforts to counties with persistently high risks.

Performance and image enhancing drug injectors’ access to needle syringe programs: Responding to a public policy dilemma

BackgroundThe number of people who inject performance and image enhancing drugs (PIEDs) attending Australian needle syringe programs (NSPs) has increased recently with cost and capacity implications for this already stretched public health program. The need to prioritise populations’ NSP access poses dilemmas for a program that has always sought to minimise entry barriers. MethodsTo assess their injecting-related risk of HIV and HCV, the Kirketon Road Centre (KRC) surveyed PIEDs injectors attending its two NSPs in inner Sydney in late 2013. Demographic, injecting risk, HIV and HCV testing, and NSP access data were analysed and then compared with similar data collected in the Australian Needle Syringe Program Survey (ANSPS) and the NSW NSP Enhanced Data Collection (NNEDC) survey conducted in the state of NSW in the same time period. ResultsPIEDs injectors surveyed (n=103) were predominantly male (99%) and only one reported ever receptive needle syringe sharing any injecting equipment. KRC participants were similar to the other survey populations in having rarely ever injected intravenously (6%); being tested for HIV and HCV in the last year (44% and 32% respectively), and reporting low rates of HCV (0%). But they were much more likely to be gay or bisexual (42% vs 2% and 4% in the NSW surveys, p<0.001). Compared to their heterosexual counterparts at KRC, this subgroup was older (35 vs 31 years, p=0.011), and more likely to: have been injecting for <3 years (70% vs 44%, p=0.025); to have ever been tested for HIV (88% vs 46%, p<0.001); to have been tested for HIV and HCV in the last year (72% vs 24%, p<0.001 and 51% vs 19%, p=0.001 respectively); and to report being HIV positive (9% vs 0%, p<0.001). ConclusionsPIEDs injectors with no other risk factors were at low risk of HIV and HCV, informing KRC's more targeted approach to their health needs. This included providing PIEDs-focused health information and promoting more frequent BBI testing, while encouraging injecting equipment access through other legal sources. This is a case study of how evidence can help resolve public policy dilemmas at the local level, thereby ensuring that scarce public health resources continue to be directed towards those people who inject drugs most at risk of BBIs.

Predictive tools for severe dengue conforming to World Health Organization 2009 criteria.

BackgroundDengue causes 50 million infections per year, posing a large disease and economic burden in tropical and subtropical regions. Only a proportion of dengue cases require hospitalization, and predictive tools to triage dengue patients at greater risk of complications may optimize usage of limited healthcare resources. For severe dengue (SD), proposed by the World Health Organization (WHO) 2009 dengue guidelines, predictive tools are lacking.MethodsWe undertook a retrospective study of adult dengue patients in Tan Tock Seng Hospital, Singapore, from 2006 to 2008. Demographic, clinical and laboratory variables at presentation from dengue polymerase chain reaction-positive and serology-positive patients were used to predict the development of SD after hospitalization using generalized linear models (GLMs).Principal findingsPredictive tools compatible with well-resourced and resource-limited settings – not requiring laboratory measurements – performed acceptably with optimism-corrected specificities of 29% and 27% respectively for 90% sensitivity. Higher risk of severe dengue (SD) was associated with female gender, lower than normal hematocrit level, abdominal distension, vomiting and fever on admission. Lower risk of SD was associated with more years of age (in a cohort with an interquartile range of 27–47 years of age), leucopenia and fever duration on admission. Among the warning signs proposed by WHO 2009, we found support for abdominal pain or tenderness and vomiting as predictors of combined forms of SD.ConclusionsThe application of these predictive tools in the clinical setting may reduce unnecessary admissions by 19% allowing the allocation of scarce public health resources to patients according to the severity of outcomes.

Unique considerations for advancing genomic medicine in African populations.

Genomic medicine presents many opportunities for improved health outcomes in the future. African countries, however, face many challenges in harnessing these opportunities for the benefit of African patients. Unique aspects that fuel these challenges include the enormous genetic diversity found in African individuals and communities across the continent; a high burden of infectious diseases and prioritized commitment of scarce public health resources to primary healthcare; limited economic resources for genomic health research and translation; and a history of the one-way transfer of samples, human resources and research translation off the continent. While these challenges are significant, there are opportunities for African countries to harness the economic and health benefits of genomic medicine for people in Africa. An active and supported biotechnology sector can provide an avenue for translating the benefits of African genomic research to African patients and populations.

Informing the Gestalt: An Ethical Framework for Allocating Scarce Federal Public Health and Medical Resources to States During Disasters

During catastrophic disasters, government leaders must decide how to efficiently and effectively allocate scarce public health and medical resources. The literature about triage decision making at the individual patient level is substantial, and the National Response Framework provides guidance about the distribution of responsibilities between federal and state governments. However, little has been written about the decision-making process of federal leaders in disaster situations when resources are not sufficient to meet the needs of several states simultaneously. We offer an ethical framework and logic model for decision making in such circumstances. We adapted medical triage and the federalism principle to the decision-making process for allocating scarce federal public health and medical resources. We believe that the logic model provides a values-based framework that can inform the gestalt during the iterative decision process used by federal leaders as they allocate scarce resources to states during catastrophic disasters.

Maybe not perfect—but surely good enough?

Although many countermeasures have been implemented in Europe, young drivers continue to have a high rate of involvement in car crashes. Their crash rate is higher in presence of peer passengers than when driving alone. Peer presence could contribute toward explaining this involvement, especially regarding speeding. Peers are known to often influence young drivers’ risky behaviors through proximal (direct and indirect active pressures) and distal (passive pressure) forms of intervention (Allen & Brown, 2008). This present study examined the effect of peer pressure and peer risk taking on the estimated speeding behavior of a scenario driver (ESBSD) as well as on the estimated speeding intention of a scenario driver (ESISD). The scenario driver was the main character in an online-scenario based questionnaire regarding speeding. The main character is a fictional young male depicted driving with his best friends in all the experimental scenarios in which the variables type of pressure and peer risk taking were manipulated. One hundred and eighty young French drivers (18–25 years, 50% men) took part in the study. The key findings were that (1) only direct and indirect active pressure, but not passive pressure, increased the ESBSD and that (2) high (vs. low) peer risk taking lead to a higher ESISD. However, no interaction between the pressure type and peer risk taking on ESBSD and ESISD was found. Implications of the findings are discussed.

Optimizing Influenza Sentinel Surveillance at the State Level

Influenza-like illness data are collected via an Influenza Sentinel Provider Surveillance Network at the state level. Because participation is voluntary, locations of the sentinel providers may not reflect optimal geographic placement. The purpose of this study was to determine the "best" locations for sentinel providers in Iowa by using a maximal coverage model (MCM) and to compare the population coverage obtained with that of the current sentinel network. The authors used an MCM to maximize the Iowa population located within 20 miles (32.2 km) of 1-143 candidate sites and calculated the coverage provided by each additional site. The first MCM location covered 15% of the population; adding a second increased coverage to 25%. Additional locations provided more coverage but with diminishing marginal returns. In contrast, the existing 22 Iowa sentinel locations covered 56% of the population, the same coverage achieved with just 10 MCM sites. Using 22 MCM sites covered more than 75% of the population, an improvement over the current site placement, adding nearly 600,000 Iowa residents. Given scarce public health resources, MCMs can help surveillance efforts by prioritizing recruitment of sentinel locations.

Health Economics in Public Health

Economic analysis is an important tool in deciding how to allocate scarce public health resources; however, there is currently a dearth of such analysis by public health researchers. Public health researchers and practitioners were surveyed to determine their current use of health economics and to identify barriers to use as well as potential strategies to decrease those barriers in order to allow them to more effectively incorporate economic analyses into their work. Data collected from five focus groups informed survey development. The survey included a demographic section and 14 multi-part questions. Participants were recruited in 2006 from three national public health organizations through e-mail; 294 academicians, practitioners, and community representatives answered the survey. Survey data were analyzed in 2007. Despite an expressed belief in the importance of health economics, more than half of the respondents reported very little or no current use of health economics in their work. Of those using health economics, cost-benefit and cost-effectiveness analysis and determination of public health costs were cited as the measures used most frequently. The most important barriers were lack of expertise, funding, time, tools, and data, as well as discomfort with economic theory. The resource deemed most important to using health economics was collaboration with economists or those with economic training. Respondents indicated a desire to learn more about health economics and tools for performing economic analysis. Given the importance of incorporating economic analysis into public health interventions, and the desire of survey respondents for more collaboration with health economists, opportunities for such collaborations should be increased.

An exploratory spatial analysis of overweight and obesity in Canada

Objective The identification of spatial clusters of overweight and obesity can be a key indicator for targeting scarce public health resources. This paper examines sex-specific spatial patterns of overweight/obesity in Canada as well as investigates the presence of spatial clusters. Methods Using data on Body Mass Index (BMI) from the 2005 Canadian Community Health Survey (20 years and older) cycle 3.1, a cross-sectional ecological-level study was conducted. Sex-specific prevalence of overweight and obesity were first mapped to explore spatial patterns. In order to assess the degree of spatial dependence, exploratory spatial data analysis was performed using the Moran's I statistic and the Local Indicator of Spatial Association (LISA). Results Results revealed marked geographical variation in overweight/obesity prevalence with higher values in the Northern and Atlantic health-regions and lower values in the Southern and Western health-regions of Canada. Significant positive spatial autocorrelation was found for both males and females, with significant clusters of high values or ‘hot spots’ of obesity in the Atlantic and Northern health-regions of Alberta, Saskatchewan, Manitoba and Ontario. Conclusions Findings reveal overweight/obesity clusters and underscore the importance of geographically focused prevention strategies informed by population-specific needs.

Using Epidemiological Data to Value the Positive Externality of Vaccination

Introduction: Although it is well-known that many vaccines against infectious diseases confer a positive economic externality via indirect protection (herd immunity), this fact has not influenced the analysis of vaccination programs in a very sophisticated way. Policy analysts have typically ignored possible herd immunity effects or at best made overly-simplistic ad hoc assumptions (often without an epidemiological basis). They often then cite the existence of a (generic) herd immunity effect to justify the universal provision of vaccines without user charges, even though public health budgets are severely limited in many poor countries and private demand may be substantial. Recent research in spatial epidemiology and economics has the potential to allow for more nuanced policy analysis. Objectives: The first objective of the paper is to review the microeconomics of vaccines' positive externalities, pointing out several policy implications of herd immunity effects not commonly observed in the literature. The second objective is to develop and demonstrate an approach for combining economic and epidemiological data to better inform policy advice for vaccines. Methods: We approach this second objective by deconstructing it into three separate tasks. The first task, an economic one, is to estimate the shape of the marginal private benefit curve, typically using stated preference data (contingent valuation or choice experiment). The second task, an epidemiological one, is to calculate the probability of protection as a function of the vaccine's coverage. The final task is to calculate the economic value of this increased protection to the unvaccinated in order to construct the marginal social benefits curve. We demonstrate this approach using recently-published epidemiological work on the indirect effects of cholera vaccination in Matlab, Bangladesh (Ali et al, 2005 Lancet 366: 44-49). We then use synthetic data on private demand for cholera vaccines to illustrate how policy advice changes as herd immunity effects are accounted for. Conclusions: It is becoming possible for the analysis of vaccines' positive externalities to move beyond generic diagrams in health economics textbooks. Increased collaboration between economists and epidemiologists could pay large dividends in better informing the use of scarce public health resources, especially as the infusion of donor funding into global public health research increases the number of vaccines competing for government support.

Prioritization MICA

Although setting priorities is an important step in making public health policy, the benefit of using epidemiology to prioritize scarce public health resources has not been fully recognized. This situation is mostly due to the complexity of proposed models for setting priorities. We describe a public health priority setting model, Missouri Information for Community Assessment Priority Setting Model (Priority MICA), which uses epidemiologic measures available in most surveillance systems across the United States. Priority MICA uses data from birth and death certificates, hospital discharges, emergency departments, risk factors from the Behavioral Risk Factors Surveillance System, and eight epidemiologic measures to construct six priority criteria: size (the number of emergency department visits, hospitalizations, and deaths), severity (number of deaths of people younger than 65), urgency (trends in deaths and hospital morbidity), preventability (evidence-based score), community support (score of social support for preventive action), and racial-disparity (race comparison through death and morbidity rate ratio). Priority MICA is part of a Web-based interactive tool that makes available data from a wide variety of surveillance systems (http://www.dhss.mo.gov/MICA). The top 10 priority diseases determined by Priority MICA were compared to a more traditional method of ranking diseases by mortality rates. Using the additional criteria in Priority MICA identified four more priority diseases than were identified using just mortality while the ranking of the other six priority diseases differed between methods.

Health and Economic Impacts of an HIV Intervention in Out of Treatment Substance Abusers: Evidence from a Dynamic Model

A community-based intervention program found that the high-risk target population interacts with its surrounding community as a source of drugs and prostitution, creating a measure of co-dependence in the health status of each group. The intervention collected extensive data on sexual and drug use practices in the target population. A dynamic compartment model estimates the epidemiological impact of the intervention, which serves as the basis for the economic assessment comparing intervention costs and lifetime HIV treatment costs. Approximately 2/3 of the new infections arise in the surrounding community. Intervention spillover benefits in the surrounding community are sufficient to make the intervention cost-saving in the first year--a savings of approximately 534,000 dollars. Conducting the intervention results in health benefits and cost-savings not only for the risk group, but for the entire community in which it resides. Quantifying the spillovers is vital to policymakers attempting to allocate scarce public health resources.

Response to Drs. Corso, Grosse, and Finkelstein

Dear Sir: We welcome the part-recognition and part-rebuttal by Corso et al. of the arguments put forward in our recent article. As Corso et al. have done with our article, we also acknowledge the validity of some of the arguments they make, thus providing fertile common ground for debate about the usefulness of cost-of-illness (COI) studies and economic evaluation more generally. Corso et al. contend that COI studies “can be valuable tools for promoting attention to the economic burden imposed by specific illnesses.” We are not sure how much progress can be made on whether this is the case without, perhaps, some empirical evidence on their added value over and above estimates of morbidity and mortality. This would require some work not only on the perceived usefulness (by decision makers) of such studies but also on decision makers’ interpretations of them. Corso et al. also state that COI studies can be a useful first step in economic evaluation. Here, we suppose the emphasis should be on can be, the approach taken to any economic evaluation largely depending on circumstances and the expert judgments of the research team as to how best to approach such circumstances. However, for both promoting attention to diseases and as a first step in economic evaluation, we would doubt that COI estimates are always necessary. If, as Corso et al. imply, COI estimates are being used just as pieces of information by the government in a complex decision-making process, we feel more comfortable. Obviously, many pieces of information will feed into such decisions about research and care priorities. If anything, our concern is not to place too much emphasis on the size of the problem, but rather, the payoff from interventions vis-à-vis other uses of resources. These will not always be in congruence. Finally, we also welcome the cautions highlighted by Finkelstein and Corso (1) with respect to the potential flaws of both economic evaluation and COI studies (1,2). Our only addition, or proposed substitution, would be to say that we need to debate further the state of the science of economics more generally, rather than just COI studies, in terms of their contributions to the allocation of scarce public health resources.