PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “hurdle model”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources

On the use of zero-inflated and hurdle models for modeling vaccine adverse event count data.

We compared several modeling strategies for vaccine adverse event count data in which the data are characterized by excess zeroes and heteroskedasticity. Count data are routinely modeled using Poisson and Negative Binomial (NB) regression but zero-inflated and hurdle models may be advantageous in this setting. Here we compared the fit of the Poisson, Negative Binomial (NB), zero-inflated Poisson (ZIP), zero-inflated Negative Binomial (ZINB), Poisson Hurdle (PH), and Negative Binomial Hurdle (NBH) models. In general, for public health studies, we may conceptualize zero-inflated models as allowing zeroes to arise from at-risk and not-at-risk populations. In contrast, hurdle models may be conceptualized as having zeroes only from an at-risk population. Our results illustrate, for our data, that the ZINB and NBH models are preferred but these models are indistinguishable with respect to fit. Choosing between the zero-inflated and hurdle modeling framework, assuming Poisson and NB models are inadequate because of excess zeroes, should generally be based on the study design and purpose. If the study's purpose is inference then modeling framework should be considered. For example, if the study design leads to count endpoints with both structural and sample zeroes then generally the zero-inflated modeling framework is more appropriate, while in contrast, if the endpoint of interest, by design, only exhibits sample zeroes (e.g., at-risk participants) then the hurdle model framework is generally preferred. Conversely, if the study's primary purpose it is to develop a prediction model then both the zero-inflated and hurdle modeling frameworks should be adequate.

Adolescent↗

Identifying Single-Cell Expression Quantitative Trait Loci Using a Bootstrap Penalized Hurdle Model.

BACKGROUND: Expression quantitative trait loci (eQTL) analysis links genetic variants to gene expression levels, helping to uncover how genetic variation contributes to gene regulation. While traditional eQTL analyses rely on bulk RNA-seq data, recent advances in single-cell RNA sequencing (scRNA-seq) have made it possible to detect cell-type-specific eQTLs. However, the inherent sparsity and heterogeneity of scRNA-seq data present major challenges for standard modeling approaches. METHODS: In this paper, we propose a novel statistical framework, Bootstrap Penalized Hurdle regression model (BPHurdle), designed specifically for scRNA-seq data. BPHurdle employs a hurdle modeling framework, where a logistic component accounts for the excess zeros in single-cell expression data, and a Poisson component jointly evaluates the effects of multiple SNPs on positive gene expression levels. RESULTS: Through simulation studies, we show that BPHurdle achieves high accuracy and robustness in identifying regulatory variants. We further demonstrate its utility on a real dataset through a case study focusing on a subset of differentially expressed genes, where it successfully identifies reliable cell-type-specific eQTLs. CONCLUSIONS: Overall, BPHurdle offers an advanced and flexible approach for single-cell eQTL mapping, providing deeper insight into the genetic regulation of gene expression at cellular resolution.

Quantitative Trait Loci↗

Equity in health care utilization: further tests based on hurdle models and Swedish micro data.

This paper tests the null hypothesis of no horizontal inequity in delivery of health care by use of count data hurdle models and Swedish micro data. It differs from most earlier work in three principal ways: First, the tests are carried out separately for physician and hospital care; second, the tests are carried out separately for the probability of seeking care and the amount of care received (given any use); and third, the tests are based on a model that includes several socioeconomic variables, e.g. income, education and size of community of residence. The paper rejects the hypothesis of no inequity because socioeconomic factors also have significant effects on utilization, e.g. income and size of community of residence. Size of community of residence has a positive significant effect on the frequency of physician visits but not on the probability of visiting a physician.

Adolescent↗

Demand for traditional medicine in Taiwan: a mixed Gaussian-Poisson model approach.

Hurdle count models are used to examine the participation and consumption decisions in Chinese medicine use. Motivated by a household production model, a second censoring mechanism is introduced into existing single-hurdle models, and the resulting specification accommodates conscientious abstainers, as well as economic non-consumers, and admits excessive zeros in the sample. In contrast to previous studies that found few predictors, empirical results based on a Taiwanese national sample suggest that Western medicine is a gross substitute to Chinese medicine, and both time price and money price play more important roles than income. Insurance, lifestyle and demographics also determine the use of Chinese medicine.

Adult↗

Latent class models for utilisation of health care.

This paper explores different approaches to econometric modelling of count measures of health care utilisation, with an emphasis on latent class models. A new model is proposed that combines the features of the two most common approaches: the hurdle model and the finite mixture negative binomial. Additionally, the panel structure of the data is taken into account. The proposed finite mixture hurdle model is shown to fit the data substantially better than the existing models for a particular application to data from the RAND Health Insurance Experiment. The estimation results indicate a higher price effect for low users of health care. It is furthermore found that this results mainly from the difference of the price effects on the probability to visit a doctor, while the price effect on the conditional number of visits does not differ significantly between high and low users.

Health Services↗

Accessing a new medication in Germany: a novel approach to assess a health insurance-related barrier.

PURPOSE: Knowledge of the existence, position, and magnitude of barriers to equitable healthcare is an important consideration in shaping healthcare policies. METHODS: The authors developed a novel three-dimensional person-time-related hurdle model and followed a cohort of 7358 statutorily (SHI) and 457 privately health insured (PHI) patients with migraine headaches at 377 primary-care practices (MediPlus, IMS Health) in the second to fourth year of the HealthCare Structural Reform Act in Germany. RESULTS: For SHI compared with PHI migraine patients, there was a hurdle to receiving sumatriptan at all (2.4-fold lower hazard, 95% confidence interval, 1.8-3.2). Among patients who received sumatriptan, frequency and intensity of use differed only minimally between SHI and PHI patients. CONCLUSIONS: These findings have implications for healthcare researchers and healthcare policies. The framework extends traditional hurdle models and can serve as a model for quantifying barriers to receipt of services under different funding policies.

Adolescent↗

Assessing household health expenditure with Box-Cox censoring models.

In order to assess the combined presence of zero expenditures and a heavily skewed distribution of positive expenditures, the Box-Cox transformation with location parameter is used to define a set of models generalising the standard Tobit, Heckman selection and double-hurdle models. Extended flexibility with respect to previous specifications is introduced, notably regarding negative transformation parameters, which may prove necessary for medical expenditures, and corner-solution outcomes. An illustration is provided by the analysis of household health expenditure in Switzerland.

Family Characteristics↗

Quantifying the hurdle concept by modelling the bacterial growth/no growth interface.

The hurdle concept described eloquently over many years by Professor Leistner and his colleagues draws attention to the interaction of factors that affect microbial behaviour in foods. Under some circumstances these effects are additive. Under others the implication is that synergistic interactions lead to a combined effect of greater magnitude than the sum of constraints applied individually. Predictive modelling studies on the combined effects of temperature and water activity and temperature and pH suggest that the effect of these combinations on growth rate is independent. Where the effect of the two factors is interactive rather than independent is at the point where growth ceases--the growth/no growth interface. An interesting and consistent observation is that a very sharp cut off occurs between conditions permitting growth and those preventing growth, allowing those combinations of factors to be defined precisely and modelled. Growth/no growth interface models quantify the effects of various hurdles on the probability of growth and define combinations at which the growth rate is zero or the lag time infinite. Increasing the stringency of one or more hurdles at the interface by only a small amount will significantly decrease the probability of an organism growing. Understanding physiological processes occurring near the growth/no growth interface and changes induced by moving from one side of the interface to the other may well provide insights that can be exploited in a new generation of food preservation techniques with minimal impact on product quality.

Bacteria↗

Use of Bayesian Markov Chain Monte Carlo methods to model cost-of-illness data.

It is well known that the modeling of cost data is often problematic due to the distribution of such data. Commonly observed problems include 1) a strongly right-skewed data distribution and 2) a significant percentage of zero-cost observations. This article demonstrates how a hurdle model can be implemented from a Bayesian perspective by means of Markov Chain Monte Carlo simulation methods using the freely available software WinBUGS. Assessment of model fit is addressed through the implementation of two cross-validation methods. The relative merits of this Bayesian approach compared to the classical equivalent are discussed in detail. To illustrate the methods described, patient-specific non-health-care resource-use data from a prospective longitudinal study and the Norfolk Arthritis Register (NOAR) are utilized for 218 individuals with early inflammatory polyarthritis (IP). The NOAR database also includes information on various patient-level covariates.

Arthritis↗

[Social inequality in health and the utilization of health services among the elderly in Latin America].

OBJECTIVE: To learn if there is social inequality in health and in the utilization of health services among the elderly in six cities of Latin America: Buenos Aires, Argentina; São Paulo, Brazil; Santiago, Chile; Havana, Cuba; Mexico City, Mexico; and Montevideo, Uruguay. METHODS: This study used data from a project called Health, Well-being, and Aging in Latin America and the Caribbean (known as the "SABE project"). To investigate the presence of social inequality in health an ordinal probit model was used, with health status indicated by the following dependent variables: activities of daily living, instrumental activities of daily living, physical mobility, and self-rated health. To measure social inequality in the utilization of health services two characteristics were assessed: one for outpatient medical services (whether or not individuals received outpatient care, and how many times), and one for hospital admissions (whether or not individuals were hospitalized). For outpatient services, we estimated a negative binomial hurdle model, and for hospital admissions, a logit model was constructed. RESULTS: Our main results suggest the presence of social inequality in health in all the cities, with better-off socioeconomic groups having better health. The difference in health was less marked in Havana, Buenos Aires, and Montevideo, and it was more pronounced in São Paulo and Mexico City. With respect to the utilization of health services we found inequalities in the use of outpatient services in Santiago, Mexico City, and São Paulo. In Santiago and Mexico City, more schooling was associated with a higher number of expected outpatient medical visits; however, the opposite was found in São Paulo. For hospital admissions, inequality was found only in São Paulo, with individuals with more years of schooling being much more likely to be hospitalized. CONCLUSION: To a certain extent, our results reflect the socioeconomic and demographic characteristics of the countries in which each of the six cities is located. The cities in the countries with the worst social indicators (high income inequality and low human development index) tended to have the greatest inequalities in health and in the utilization of health services.

Age Factors↗

A needs index for mental health care in England based on updatable data.

BACKGROUND: Mathematical models relating rates of mental health care use to population characteristics such as social deprivation are widely used in both planning and researching mental health services. The models currently in wide use in England are based on data mostly derived from the 10-yearly population censuses. These are perceived to be out of date many years before new census data are available for their replacement. A new set of government deprivation monitoring statistics based mainly on annually updatable data has recently been developed. This study set out to produce a mental illness needs index based on these new data. METHODS: A series of regression models were tested using individual domain scores from the DETR Index of Multiple Deprivation and the Office of National Statistics area-type classification as independent variables to predict 1998/9 psychiatric admission rates for broad diagnostic groups for 8251 of the 8414 electoral wards in England as dependent variables. RESULTS: The distribution of admission numbers in wards showed a pattern of over-dispersion with an excessive number of zero values for conventional regression approaches. A two-stage 'hurdle' model was, thus, adopted, predicting first the likelihood that wards would produce any admissions and second the probable number. This produced satisfactory predictive power, with residual variance showing strong geographical patterns associated with administrative areas, probably arising from differential resourcing or idiosyncratic clinical practice. CONCLUSIONS: A website providing data on the various indicators has been provided and its uses are indicated.

Adolescent↗

The decision-making process of health care utilization in Mexico.

Using individual-level data from the 2000 Mexican Survey of Satisfaction with Health Services we estimate a two-part negative binomial hurdle model to evaluate the decision-making process of health care utilization in Mexico. We find that there are income-related differences in utilization associated with the first visit to a physician, as well as substantial utilization differences by region, employment, insurance and financial status. There are also income-related differences in the first visit to a specialist but not in the number of days hospitalized. The results suggest that increasing initial access to services via income and insurance coverage and providing financial resources to underserved regions can substantially improve access to care and, ultimately, population health.

Adult↗

An analysis of private health insurance purchasing decisions with national health insurance in Taiwan.

The Taiwanese health insurance industry is just over 30 years old. Originally private and domestic, the industry underwent substantial institutional changes when it opened to foreign competition between 1987 and 1994 and when the Taiwanese government established national health insurance (NHI) coverage in 1995. Congruent with these changes, rapid growth occurred in the Taiwanese demand for private health insurance. In order to better understand the recent performance of the Taiwanese health insurance industry, the structure of the NHI system is described and then household decisions to purchase private health insurance are analyzed using a two-part (hurdle) model on 1998 Survey of Family Income and Expenditure data. Logistic and OLS regressions are used to examine the factors influencing the probability and amount of private health insurance purchased. Generally, factors affecting the probability of having insurance also influence the amount of insurance coverage purchased. Higher income and education levels are associated with increased probabilities and larger quantities of private insurance purchases. Married females, the employed, and household heads working in state-run enterprises are more likely to purchase private insurance than their counterparts. The probability of private insurance purchases varies by region, with northern Taiwanese households having higher odds of owning private insurance than non-northern households. Compared to those in rural villages, households in cities and towns are more likely to have private insurance. The likelihood of private insurance purchase also tends to rise with advancing age and larger family sizes. In addition, one important implication in the private health insurance market is highlighted. There is no complementarity between the public and private systems.

Adult↗

HIV testing and retesting for men and women in Switzerland.

This study was conducted to describe voluntary HIV testing in the general population in Switzerland and to estimate yearly HIV test incidence. In 1994, a representative telephone survey of individuals aged 17 to 45 years obtained self-reported information on HIV testing. In addition to describing cumulative HIV test incidence, yearly HIV test incidence over time was estimated by a Bayesian hurdle model allowing for the plausible scenario of test consumption differing between first test and subsequent retests. Overall, 33% of the Swiss population (age 17 to 45 years) has been tested at some time for HIV on a voluntary basis (30% men, 36% women). For the time period 1990-1994, the result showed for 35-year-old individuals with supposedly low risk behavior, that 1) annual test incidence (first test or retest) showed a greater increase for men (4.2 to 5.9%) than for women (5.0 to 6.0%); 2) annual first test incidence increased moderately and differed for men and women (2.9 to 3.4% for men, 4.6 to 5.2% for women), and 3) annual retest incidence was twice as high for men (17.6%) as for women (8.6%). In conclusion, a substantial part of the Swiss population has been tested at some stage for HIV on a voluntary basis, and differences exist for testing and retesting between men and women.

Adolescent↗

The relationship between the stages of change for exercise and health insurance costs.

OBJECTIVE: To determine the relationship between the stages of exercise participation and health insurance costs. METHODS: A hurdle model was used to examine health survey and health insurance costs data by stage of exercise participation. RESULTS: Employees classified in the maintenance stage (regular exercisers) of exercise adoption had lower costs and a lower probability of being classified in the high-cost group than did employees classified in the other stages of change for exercise participation. CONCLUSION: This study offers evidence that the health insurance of individuals classified in the maintenance stage of exercise costs less than does that of individuals classified in other stages of exercise adoption.

Adult↗

Screening, prevention and socioeconomic costs associated with the treatment of colorectal cancer.

Colorectal cancer (CRC), the third most prevalent cancer worldwide, imposes a significant economic and humanistic burden on patients and society. One study conservatively estimated the annual expenditures for colorectal cancer to be approximately dollars US 5.3 billion in 2000, including both direct and indirect costs. However, other investigators estimated inpatient costs alone incurred in the US in 1994 to be around dollars US 5.14 billion. Therefore, the economic burden of colorectal cancer in the US could be projected to be somewhere in the range of dollars US 5.5-6.5 billion by considering that inpatient costs approximate 80% of total direct costs. No worldwide data have been published, but assuming that the US represents 25-40% of total expenditures in oncology, as seen for breast and lung cancers, a rough estimate for colorectal cancer would be in the range of dollars US 14-22 billion. Screening helps increase patient survival by diagnosing colorectal cancer early. The ideal method among the four tests most used (faecal occult blood test, flexible sigmoidoscopy, colonoscopy and double contrast barium enema) has not been identified. Economic studies of colorectal cancer screening are complex because of the many variables involved, as well as the fact that the outcomes must be followed for many years, and the lack of consensus on screening guidelines. Intuitively, modelling colorectal cancer is one way to overcome these hurdles; published modelling studies predict colorectal cancer screening programs to be within the threshold of dollars US 40000 per life-year saved. The faecal occult blood test appears to be the only clearly effective test, both from a clinical and an economic viewpoint. Important limitations are the invasiveness and inconvenience of the screening procedures, except faecal occult blood test. Patients' comfort and satisfaction are essential in improving compliance with screening recommendations, which appears to be low even in the US (35% of the general population aged over 40 years and 60% of the high-risk population), the country with the highest awareness and compliance in the world. Since colorectal cancer is generally a disease of the elderly, its economic burden is expected to grow in the near future, mainly due to population aging. Potential avenues to pursue in order to contain or reduce the economic burden of colorectal cancer would be the design and implementation of efficient screening programmes, the improvement of patient awareness and compliance with screening guidelines, the development of appropriate prevention programs (i.e. primary and secondary), and earlier diagnosis.

Antineoplastic Agents↗

Application of a hurdle negative binomial count data model to demand for bass fishing in the southeastern United States.

This paper identifies factors that influence the demand for a bass fishing trip taken in the southeastern United States using a hurdle negative binomial count data model. The probability of fishing for a bass is estimated in the first stage and the fishing trip frequency is estimated in the second stage for individuals reporting bass fishing trips in the Southeast. The applied approach allows the decomposition of the effects of factors responsible for the decision to take a trip and the trip number. Calculated partial and total elasticities indicate a highly inelastic demand for the number of fishing trips as trip costs increase. However, the demand can be expected to increase if anglers experience a success measured by the number of caught fish or their size. Benefit estimates based on alternative estimation methods differ substantially, suggesting the need for testing each modeling approach applied in empirical studies.

Adult↗