Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “instrumental variable”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 163 records · Page 9Linked to original sources
PubMed1993

Testing hypotheses about direction of causation using cross-sectional family data.

We review the conditions under which cross-sectional family data (e.g., data on twin pairs or adoptees and their adoptive and biological relatives) are informative about direction of causation. When two correlated traits have rather different modes of inheritance (e.g., family resemblance is determined largely by family background for one trait and by genetic factors for the other trait), cross-sectional family data will allow tests of strong unidirectional causal hypotheses (A and B are correlated "because of the causal influence of A on B" versus "because of the causal influence of B on A") and, under some conditions, also of the hypothesis of reciprocal causation. Possible sources of errors of inference are considered. Power analyses are reported which suggest that multiple indicator variables will be needed to ensure adequate power of rejecting false models in the presence of realistic levels of measurement error. These methods may prove useful in cases where conventional methods to establish causality, by intervention, by prospective study, or by measurement of instrumental variables, are unfeasible economically, ethically or practically.

Adoption↗
PubMed2003

Competitive spillovers across non-profit and for-profit nursing homes.

The importance of non-profit institutions in the health care sector has generated a vast empirical literature examining quality differences between non-profit and for-profit nursing homes. Recent theoretical work has emphasized that much of this empirical literature is flawed in that previous studies rely solely on dummy variables to capture the effects of ownership rather than accounting for the share of non-profit nursing homes in the market. This analysis considers whether competitive spillovers from non-profits lead to higher quality in for-profit nursing homes. Using instrumental variables to account for the potential endogeneity of non-profit market share, this study finds that an increase in non-profit market share improves for-profit and overall nursing home quality. These findings are consistent with the hypothesis that non-profits serve as a quality signal for uninformed nursing home consumers.

Aged↗
PubMed2000

Clinical utilization of the international normalized ratio (INR).

The prothrombin time (PT) is one of the most important laboratory tests to determine the functionality of the blood coagulation system. It is used in patient care to diagnose diseases of coagulation, assess the risk of bleeding in patients undergoing operative procedures, monitor patients being treated with oral anticoagulant (coumadin) therapy, and evaluate liver function. The PT is performed by measuring the clotting time of platelet-poor plasma after the addition of calcium and thromboplastin, a combination of tissue factor and phospholipid. Intra- and interlaboratory variation in the PT was a significant problem for clinical laboratories in the past, when crude extracts of rabbit brain or human placenta were the only source of thromboplastin. The international normalized ratio (INR), developed by the World Health Organization in the early 1980s, is designed to eliminate problems in oral anticoagulant therapy caused by variability in the sensitivity of different commercial sources and different lots of thromboplastin to blood coagulation factor VII. The INR is used worldwide by most laboratories performing oral anticoagulation monitoring, and is routinely incorporated into dosage planning for patients receiving warfarin. Although the recent availability of sensitive PT reagents prepared from recombinant human tissue factor (rHTF) and synthetic phospholipids eliminated many of the earlier problems associated with the use of crude thromboplastin preparations, local instrument variability in the INR still remains a problem. Presently, the use of plasma calibrants seems the best solution to this problem. Standardizing the point-of-care instruments for INR monitoring is another dilemma faced by the industry. Ultimately, new generations of anticoagulant drugs may eliminate the need for laboratory monitoring of anticoagulant therapy.

Hematology↗
PubMed1996

Value of early variables as predictors of short-term outcome in patients with acute focal cerebral ischemia.

Reliable, simple and safe criteria are needed for the early prediction of short-term outcome in patients with acute ischemic stroke. The aim of our study was to evaluate, in terms of their individual and combined power, the prognostic value of a few widely available clinical and instrumental variables obtained during the acute phase. The study involved 351 consecutive patients who were examined within 48 hours of their first ischemic stroke. Eight variables were chosen: age, initial level of consciousness, limb paresis, arterial blood pressure, glycemia, the results of electrocardiography and electroencephalography, and the infarct size revealed by computed tomography. Mortality and disability were evaluated on Day 30, when the variables that significantly correlated with disability were the severity of limb paresis, electroencephalographic abnormalities, infarct size and (less significantly) the level of consciousness and hyperglycemia. There was no statistical correlation with blood pressure. Logistic analysis confirmed only infarct size, the severity of limb paresis and electroencephalographic abnormalities as independent variables. The variables that significantly correlated with early death were the severity of limb paresis, infarct size, electrocardiographic abnormalities, the level of consciousness, electroencephalographic abnormalities and hyperglycemia. More intriguingly, logistic analysis confirmed only the electroencephalographic and electrocardiographic abnormalities as independent variables. The predictive prognostic value of limb paresis, infarct size, the level of consciousness and hyperglycemia is well known, but we would like to stress the fact that only a few independent variables are predictive of early death (electroencephalographic and electrocardiographic abnormalities) and poor recovery (infarct size, the severity of limb paresis, electroencephalographic abnormalities). The prognostic value of electroencephalography may express the potential involvement of dynamic non-structural phenomena, such as penumbra ischemica and diaschisis.

Adult↗
PubMed1999

The influence of social programs in source countries on various classes of U.S. immigration.

"This article uses a unique set of pooled cross-sectional and time series data to examine the annual rate of U.S. immigration during 1972-1991 from 60 source countries. One distinguishing feature of the article is that it breaks out and cross-classifies various classes of immigrants--numerically limited versus numerically exempt and new immigrant versus adjustment of status. A second distinguishing feature is that it utilizes a unique vector of variables relating to the presence and characteristics of various social programs in source countries. The models developed here emphasize the importance of both differential economic advantage and the ease with which a prospective migrant can transfer skills to the U.S. labor market. Hausman-Taylor instrumental variable estimates of the coefficients indicate that in addition to other factors, social programs in source countries are significant determinants of immigration to the USA." Data are from the Immigration and Naturalization Service's Public Use Files.

Americas↗
PubMed1992

Relation between clinical and instrumented measures of motor coordination in traumatically brain injured persons.

Right and left upper extremity motor coordination was evaluated in 40 traumatically brain injured subjects using two different measurement techniques--a conventional clinical evaluation and an instrumented evaluation. The relationship between the results from the clinical evaluation (finger to nose test) and the instrumented evaluation (timed visual-arm coordinated lateral reach) was determined using canonical correlation analysis (Rc). Clinical and instrumented test scores were significantly correlated (Rc = .685, p = .002; Rc = .629, p = .008 for the right and left sides, respectively). The scores obtained by the clinical test had little variance; instrumented scores varied greatly among subjects. Significant differences did not exist between the right and left sides for the instrumented variables of average movement speed, accuracy, and index of coordination (paired t-tests, p = .550, p = .548, p = .627, respectively) or for the clinical variables of time of execution (paired t-test, p = .468) and tremor (Wilcoxon matched-pairs, p = .228). Although meaningful correlations were obtained, they do not conclusively indicate that the two tests measure the same concept of coordination. Therefore, it is suggested that the instrumented evaluation be used to complement the traditional clinical evaluation to increase the degree of objectivity in clinical measurement.

Adolescent↗
PubMed2006

Insuring children or insuring families: do parental and sibling coverage lead to improved retention of children in Medicaid and CHIP?

Recent research indicates that 3 million children leave Medicaid or the Children's Health Insurance Program (CHIP) each year and become uninsured, despite continuing eligibility. This paper explores the effect of family coverage on drop-out among children in these two programs, using instrumental variables to address the endogeneity of parental and sibling coverage in Medicaid/CHIP. Using the Current Population Survey March Supplement (1999-2004), I find that having a parent (primarily a mother) or sibling in public insurance is associated with a significantly lower risk of drop-out; after instrumenting with parental and sibling eligibility, only parental coverage remains a significant predictor of retention. I conclude that policy attempts to expand health insurance to the 8.5 million uninsured children in the U.S. would be much more effective if they covered parents and children in the same program.

Adolescent↗
PubMed2003

Medical student indebtedness and the propensity to enter academic medicine.

This paper considers the potential impact of medical school indebtedness and other variables on the propensity of US doctors to enter academic medicine. Probit models provide some evidence that indebtedness reduces the likelihood that physicians will choose academic medicine as their primary activity. Nevertheless, the magnitude of this effect is not large. As indebtedness may be endogenous, the probits are rerun using an instrumental variables approach. These estimates imply that over time indebtedness may have an important impact on the propensity of physicians to enter academic medicine.

Academic Medical Centers↗
PubMed2005

Socioeconomic status, depression disparities, and financial strain: what lies behind the income-depression relationship?

Prior studies have consistently found the incidence and persistence of depression to be higher among persons with low incomes, but causal mechanisms for this relationship are not well understood. This study uses the National Longitudinal Survey of Youth 1979 cohort to test several hypotheses about the robustness of the depression-income relationship among adults. In regressions of depression symptoms on income and sociodemographic variables, income is significantly associated with depression. However, when controls for other economic variables are included, the effect of income is considerably reduced, and generally not significant. Employment status and the ratio of debts-to-assets are both highly significant for men and for women both above and below the median income. Fixed-effects estimates suggest that employment status and financial strain are causally related to depression, but income is not. Instrumental variable estimates suggest that financial strain may not lead to depression.

Adult↗
PubMedDate not supplied

[Income inequality, corruption, and life expectancy at birth in Mexico].

OBJECTIVE: To ascertain if the effect of income inequality on life expectancy at birth in Mexico is mediated by corruption, used as a proxy of social capital. MATERIAL AND METHODS: An ecological study was carried out with the 32 Mexican federative entities. Global and by sex correlations between life expectancy at birth were estimated by federative entity with the Gini coefficient, the Corruption and Good Government Index, the percentage of Catholics, and the percentage of the population speaking indigenous language. Robust linear regressions, with and without instrumental variables, were used to explore if corruption acts as intermediate variable in the studied relationship. RESULTS: Negative correlations with Spearman's rho near to -0.60 (p < 0.05) and greater than -0.66 (p < 0.05) between life expectancy at birth, the Gini coefficient and the population speaking indigenous language, respectively, were observed. Moreover, the Corruption and Good Government Index correlated with men's life expectancy at birth with Spearman's rho -0.3592 (p < 0.05). Regressions with instruments were more consistent than conventional ones and they show a strong negative effect (p < 0.05) of income inequality on life expectancy at birth. This effect was greater among men. CONCLUSION: The findings suggest a negative effect of income inequality on life expectancy at birth in Mexico, mediated by corruption levels and other related cultural factors.

Aged↗
PubMed1994

Depression and parental bonding: cause, consequence, or genetic covariance?

It is shown how information on the direction of causation between variables may be obtained from a cross-sectional study of pairs of relatives. This method is applied to the study of the relationship between ratings of parents' rearing style and depression in their offspring. Adult female twins ascertained from a population-based registry in Viroffia completed the Center for Epidemiological Studies--Depression Scale (CESD) and a 7-item short form of the Parental Bonding Instrument (PBI) about each of their parents. Two dimensions of parental behavior, overprotectiveness and coldness, were analyzed jointly with depression data in both genetic factor and directional genetic models. Models that specify ratings of parents as a cause of depression in the offspring fit the data significantly better than models that specify depression as a cause of ratings of parents. A still better fit is obtained with models that specify common genetic variance to depression and ratings, though causal models with error variance perform almost as well. In general, ratings of fathers show more genetic and less shared environmental variance than ratings of mothers, which might arise from more consistent treatment of offspring by mothers than by fathers. No effect of children eliciting parental rearing style was detected with these data. The relative merits of instrumental variable, longitudinal, and family approaches to testing causal models are discussed.

Adult↗
PubMed2005

Simplex optimization of the variables affecting the micelle-stabilized room temperature phosphorescence of 6-methoxy-2-naphthylacetic acid and its kinetic determination in human urine.

This article reports the kinetic determination of 6-methoxy-2-naphthylacetic acid (6-MNA), the major metabolite of nabumetone, from micelle-stabilized room temperature phosphorescence (MS-RTP) measurements made by using the stopped-flow mixing technique. This methodology allows one to determine analytes in complex matrices without the need for a tedious separation process. It also shortens analysis times substantially. The proposed method uses simplex methodology to optimize the chemical and instrumental variables affecting the phosphorescence. It was applied to the determination of 6-MNA in human urine. The maximum phosphorescence signal is obtained within only 10 s after the sample is prepared. The maximum slope of the kinetic curve, which corresponds to the maximum rate of the phosphorescence development, is measured at lambda(ex)=273 nm and lambda(em)=516 nm. Least-squares regression was used to fit experimental data, and the detection limit, repeatability, and standard deviation for replicate samples were determined.

Enzyme Inhibitors↗
PubMed2000

High endogenous estradiol is associated with increased venous distensibility and clinical evidence of varicose veins in menopausal women.

OBJECTIVE: The purpose of this study was to determine if there is an association between elevated sex hormones (ie, serum estradiol, sex hormone binding globulin [SHBG], testosterone) and increased venous distension and clinical evidence of varicose veins in menopausal women. METHODS: Participants were 104 healthy volunteer menopausal women, aged 48 to 65 years, who were not undergoing hormonal treatment. Of these 104, 14 were excluded from analyses because their estradiol levels were compatible with a premenopausal condition (4), because they had missing values for insulin concentration (5), and because they did not show up at venous vessel examination (5). Patients underwent a physical examination to determine the presence of varicose veins; a venous strain-gauge plethysmographic examination to compute instrumental measures of venous distensibility; and laboratory analyses of blood so serum testosterone, estradiol, SHBG, glucose, and insulin could be measured. There were also prevalence ratios and odds ratios used to test the presence of an association between biochemical and instrumental variables. RESULTS: Serum levels of estradiol in the upper tertile of the frequency distribution were significantly associated with clinical evidence of varicose veins (prevalence odds ratios 3.6; 95% CI 1.1-11.6) and with increased lower limb venous distensibility (prevalence odds ratios 4.4; 95% CI 1.2-15.5). No association was found for SHBG and testosterone. CONCLUSIONS: Our finding that high serum levels of estradiol are associated with clinical evidence of varicose veins and instrumental measurements indicating increased venous distensibility in menopausal women suggests that endogenous estrogens may play a role in the development of this very common venous vessel abnormalities.

Aged↗
PubMed2002

Modeling treatment effects on binary outcomes with grouped-treatment variables and individual covariates.

During evaluation of treatment effects in observational studies, confounding is a constant threat because it is always possible that patients with a better prognosis, not adequately characterized by measured covariates, are chosen for a specific therapy. Ecologic analyses may avoid confounding that would be present in analysis at the individual level because variations in regional or hospital practice may be unrelated to prognosis. The authors used simulated data with an excluded confounder to evaluate the reliability and limitations of the grouped-treatment approach, a method of incorporating an ecologic measure of treatment assignment into an individual-level multivariable model, similar to the instrumental variable approach. Estimates based on the grouped-treatment approach were closer to the true value than those of standard individual-level multivariable analysis in every simulation. Furthermore, confidence intervals based on the grouped-treatment approach achieved approximately their nominal coverage, whereas those based on individual-level analyses did not. The grouped-treatment approach appears to be more reliable than standard individual-level analysis in situations where the grouped-treatment variable is unassociated with the outcome except via the actual treatment assignment and measured covariates.

Confounding Factors, Epidemiologic↗
PubMed2005

Is health care a luxury? New evidence from OECD data.

Most research using cross-country data find income elasticities equal to or exceeding unity with respect to health expenditure. These conclusions might be confounded due to omitted variables bias and the presence of unobserved country and year specific determinants of per capita health expenditures. I obtain results supporting these hypotheses using recent (1990-98) data from fifteen OECD countries. Specifically, OLS coefficient estimates drop by more than 50% with the use of two-way fixed effects models and the inclusion of various demand and supply based determinants of per capita health expenditures, implying income elasticities of between 0.21 and 0.51. Weighted Least Squares (WLS), Generalized Least Squares (GLS) and Instrumental Variables (IV) estimation yield similar results.

Delivery of Health Care↗
PubMedDate not supplied

[Methodology of the National Seroepidemiologic Survey, Mexico].

This paper briefly describes the methodology employed in the National Seroepidemiologic Survey (NSS). Among the aspects discussed are: previous studies of this kind; the master sampling frame; the sample design, the study population and sample size; the selection of diseases for study; the variables; instruments for data collection; operational design for the survey; as well as data processing and analysis. The overall response rate of the NSS was 78.4 per cent of the surveyed homes, 40 per cent of which were from rural areas (communities of less than 2,500 persons). The non-response rate was 41.3 per cent with the highest level in Baja California Sur (RNR = 26%) and the lowest in Puebla (RNR = 50.9%). The poor response in Puebla may have been due to the fact that it was the first state surveyed and served as the pilot project. The variable estimates from the NSS were similar to those found in other national surveys and the General Population Census. Therefore, these is reason to believe that the results of the NSS were representative and comparable. The final part of the paper describes the NSS projects already published, national seroprevalences, considerations concerning inference, and the laboratory procedures utilized in each project.

Adolescent↗
PubMed2006

Overcoming bias in estimating the volume-outcome relationship.

OBJECTIVE: To examine the effect of hospital volume on 30-day mortality for patients with congestive heart failure (CHF) using administrative and clinical data in conventional regression and instrumental variables (IV) estimation models. DATA SOURCES: The primary data consisted of longitudinal information on comorbid conditions, vital signs, clinical status, and laboratory test results for 21,555 Medicare-insured patients aged 65 years and older hospitalized for CHF in northeast Ohio in 1991-1997. STUDY DESIGN: The patient was the primary unit of analysis. We fit a linear probability model to the data to assess the effects of hospital volume on patient mortality within 30 days of admission. Both administrative and clinical data elements were included for risk adjustment. Linear distances between patients and hospitals were used to construct the instrument, which was then used to assess the endogeneity of hospital volume. PRINCIPAL FINDINGS: When only administrative data elements were included in the risk adjustment model, the estimated volume-outcome effect was statistically significant (p=.029) but small in magnitude. The estimate was markedly attenuated in magnitude and statistical significance when clinical data were added to the model as risk adjusters (p=.39). IV estimation shifted the estimate in a direction consistent with selective referral, but we were unable to reject the consistency of the linear probability estimates. CONCLUSIONS: Use of only administrative data for volume-outcomes research may generate spurious findings. The IV analysis further suggests that conventional estimates of the volume-outcome relationship may be contaminated by selective referral effects. Taken together, our results suggest that efforts to concentrate hospital-based CHF care in high-volume hospitals may not reduce mortality among elderly patients.

Aged↗
PubMed2000

The marginal benefits of invasive treatments for acute myocardial infarction: does insurance coverage matter?

This paper applies instrumental variable (IV) techniques and estimates the average benefits of invasive surgical treatments for marginal acute myocardial infarction (AMI) patients by insurance coverage. The study uses data from the Agency for Healthcare Research and Quality's Healthcare Cost and Utilization Project (HCUP), State Inpatient Databases for the state of Washington, for years 1988-1993. We observed differences in average benefits for marginal patients across insurance subpopulations that cannot be explained by differences in measured clinical circumstances. Our empirical results show that the insurance subpopulations with the greatest estimated marginal benefits are those with the lowest expected payment generosity to providers. Because of the relatively weak explanatory power of our instruments for three insurance subpopulations, and because many of the parameter estimates across the insurance subpopulations are not statistically different from each other, policy recommendations should be approached cautiously. However, our application of IV techniques to AMI patients demonstrates the usefulness of this approach to estimate treatment effects across patient populations (e.g., across types of insurance coverage) for medical conditions of interest to policymakers. These estimates may help determine whether treatments are overutilized or underutilized.

Adult↗