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 55 records · Page 3Linked to original sources
PubMed1999

Instrumental variables when evaluating screening trials: estimating the benefit of detecting cancer by screening.

When evaluating the benefit of detecting cancer by screening we try to answer the question, 'what would a screen detected subject's outcome have been if his/her cancer had progressed to clinical detection'. By 'outcome' we mean survival time, cancer size and stage, lead time effects and more. Because only an unethical study can answer it directly, researchers have attempted to answer the question indirectly using data from randomized cancer screening studies (subjects randomized to study (screened) or control (not screened)). Inferences are made by first selecting the cancer cohort (those subjects who are found to have cancer), then comparing subjects having screen detected cancers to subjects having clinically detected cancers. However, there are two difficulties with this approach: (i) because screening (intends to) detect cancers early, at the trial's end the study group contains more cancer cases than the control group and so the cancer cohort has some unidentified control subjects missing (that is, subjects having cancer during the screening period that have not yet been clinically detected); (ii) because screen detected cancers (may) differ from clinically detected cancers, the comparison group should include only a (non-identified) subset of the cancer cohort's control subjects (that is, only those control subjects having cancers that would have been screen detected). Statistical literature acknowledges these difficulties and attempts to solve them separately, but without success; those methods do not yield meaningful causal inferences and admit substantial bias. Recently, Angrist, Imbens and Rubin and Imbens and Rubin provide a framework for instrumental variable methods that we interpret as allowing us to make causal inferences with incompletely identified comparison groups. We apply their framework to evaluating cancer screening trials and find that we may simultaneously accommodate both difficulties while giving a meaningful answer to the question posed above. Using data from a breast cancer screening trial we demonstrate the general method with a variety of outcome measures and extensions.

Adult↗
PubMed2005

Correlations between estimated and true dietary intakes: using two instrumental variables.

PURPOSE: We describe a new application of the method of triads that allows an estimate of the correlation between a dietary questionnaire measure (Q) and true intake (T). METHODS: Three surrogate variables Q, M, and P are observed where M and P are both instrumental (often biological) variables. A reference dietary method (R) is not required. The variables M and P may be concentration rather than recovery biomarkers. Estimating equations produce Corr(Q,T), Corr(M,T), Corr(P,T), conditional on assumptions about error correlations. Correlations between errors in both Q and a reference dietary measure can also be estimated if R is available. A small validation study of California Seventh-day Adventists provided food frequency, repeated 24-hour dietary recalls (R), and biological data (blood, overnight urines, and subcutaneous fat). RESULTS: Values of Corr(Q,T) ranged between 0.40 and 0.66. Values of Corr(R,T) were higher, between 0.48 and 0.83. Estimated correlations between errors in R and Q were all positive. CONCLUSIONS: When carefully chosen, M and P, rather than M and R, should better satisfy assumptions about error correlations. Food frequency data and repeated 24-hour recalls both provide estimates of T, but the latter has greater validity. Standard errors suggest that for good precision Corr(Q,T) requires large validation studies (2000-3000 subjects).

Adult↗
PubMed1998

Is instrumental variability abnormally high in children exhibiting ADHD and aggressive behavior?

To test whether instrumental behavior of some children with attention deficit hyperactivity disorder (ADHD) is more variable than control subjects, the sequences of responses by three groups of children were compared: (i) those diagnosed with ADHD who lived in a residential treatment facility for children demonstrating aggressive behavior; (ii) age-matched children from the same facility who did not have an ADHD diagnosis but who had also demonstrated abnormal levels of aggression; and (iii) normal children from a local school. The experiment consisted of rewarding children for responding in a computer 'game' on two keys of a keyboard during each of three phases. In phase 1, rewards were provided independently of sequence variability (IND). In phase 2, rewards depended upon highly variable sequences of left and right responses. Phase 3 was a return to the IND contingencies. The results showed that sequence variability was higher in the sequence variability (VARY) phase than in the preceding IND phase, but remained high when the contingencies returned to IND, thus replicating previous findings. In none of the phases did the ADHD children respond more variably than the controls. However, ADHD subjects made more off-task responses than did controls. Thus, although the present research showed differences between ADHD and control subjects, there was no evidence to support higher behavioral variability in the ADHD subjects.

Aggression↗
PubMed2006

Biomechanical assessment of variable instrumentation strategies in adolescent idiopathic scoliosis: preliminary analysis of 3 patients and 6 scenarios.

Since the introduction of modern multi-segmental instrumentation systems, disagreement exists about the appropriate instrumentation strategies for the "optimal" correction of scoliotic deformities, and the difference between alternative scenarios is difficult to predict a priori. The purpose of this study is to evaluate the effect of different instrumentation strategies using a computer assisted surgery simulator (S3). We obtained from 32 experienced Fellows of the Scoliosis Research Society and members of the Spinal Deformities Study Group the detailed preoperative planning for three AIS patients with Lenke curve types 1A, 3A and 5C. Their scenarios were individually simulated using a computer model implemented in a spine surgery simulator (S3). The resulting Cobb angles varied for the 3 cases (e.g.: main thoracic: 6-17 degrees; 16-29 degrees; 16-30 degrees). The variability of correction remained important when sub-classifying the results according to the instrumentation strategies: A- "Pedicle Screws Constructs"; B- "Hooks Constructs"; C- "Hybrid Constructs". But overall, the average correction was better in group A (71%) than in groups B (55%) and C (54%). For the first time the effect of various instrumentation strategies can be assessed preoperatively thanks to S3. A large variability of instrumentation strategies exist within experienced surgeons and these produce rather different results. This study also questions the criteria for optimal configuration and standards to objectively design the best surgical construct.

Adolescent↗
PubMed2004

Reliability of three commercially available heart rate variability instruments using short-term (5-min) recordings.

AIM: To assess the reliability of heart rate variability (HRV) measures made by three commercially available analysers in healthy subjects. METHODS: Twenty-nine volunteers (20 males, mean age 35 +/- 13 years and nine females, mean age 29 +/- 11 years) underwent repeated HRV measures under three conditions: lying supine, standing, lying supine with controlled breathing. HRV was measured simultaneously by three instruments. Reliability was assessed statistically by calculating coefficient of variation (CV), intraclass correlation coefficient (ICC) and limits of agreement (LoA). RESULTS: A wide range of values were found for CV (1-235%) and ICC (R = 0.16-0.99) dependent on the HRV measure assessed and the position in which the measurement was made. For the most part the analysers gave similar values in each condition. The values for CV and ICC were high but within the range reported in the literature. Values for LoA were also high and showed a wide range of values. CONCLUSIONS: The similarity in measures between systems indicates that biological variation and experimental error play a major role in determining the repeatability of HRV measurements. It is therefore recommended that population-specific reliability coefficients should be published where possible and that authors should take into account the reliability of measures when making sample size calculations.

Adult↗
PubMed1997

Instrumental variability of respiratory blood gases among different blood gas analysers in different laboratories.

The aim of this study was to test the hypothesis that differences in oxygen tension (PO2) and carbon dioxide tension (PCO2) values from measurements performed on different blood gas analysers in different laboratories are clinically insignificant. Samples of fresh whole human tonometered blood (PO2 8.1 kPa (60.8 mmHg); PCO2 5.3 kPa (39.9 mmHg)) were placed in airtight glass syringes and transported in ice-water slush. Blood gas analysis was performed within 3.5 h by 17 analysers (10 different models) in 10 hospitals on one day. The mean of the differences between the measured and target values was -0.01+/-0.19 and 0.21+/-0.13 kPa (-0.06+/-1.45 and 1.55+/-1.01 mmHg) for PO2 and PCO2, respectively. The mean of the differences between two samples on one analyser was 0.06+/-0.06 and 0.04+/-0.03 kPa (0.47+/-0.48 and 0.29+/-0.24 mmHg), respectively. For PO2 and PCO2 the interinstrument standard deviations (s(b)) were 0.18 and 0.13 kPa (1.38 and 0.99 mmHg), respectively, whereas the intra-instrument standard deviations (s) were 0.06 and 0.03 kPa (0.47 and 0.26 mmHg), respectively. Both for PO2 and PCO2 the ratios of s(b)2 and s2 were statistically significant (analysis of variance (ANOVA) p<0.001). The standard deviations of a random measurement on a random analyser were 0.19 and 0.14 kPa (1.46 and 1.02 mmHg) for PO2 and PCO2, respectively. We conclude that the variability in measurement of blood gas values among different blood gas analysers, although negligible, depends much more on inter- than intra-instrument variation, both for oxygen tension and carbon dioxide tension. Technical improvements and adequate quality control programmes, including tonometry, may explain why the variability in blood gas values depends mainly on errors in the pre-analytical phase.

Blood Gas Analysis↗
PubMed2003

Weighting in instrumental variables and G-estimation.

We propose here a simple scheme to use information on compliance and prerandomization covariates to improve analysis of randomized trials with non-compliance. We use the data to determine the effect of randomization on treatment received among various strata defined by pretreatment covariates. When the effect of treatment received on the outcome of interest is the same across strata and pretreatment covariates predict non-compliance, weighting the estimating functions by the effect of randomization on treatment received can improve the precision of explanatory estimates of treatment effect and can increase the power of intent-to-treat tests of the null hypothesis. Efficiency gains under the weighting scheme are a simple increasing function of the variability of these weights. Such weighting schemes will often lead to improvements even when these conditions are not met. We use a randomized trial of cholestyramine to illustrate these points.

Cholesterol↗
PubMed1975

Pre-instrumental variables in coagulation testing.

A number of variables thought to affect measurement of prothrombin time (PT) and partial thromboplastin time (PTT) were examined in an effort to determine more precisely their effects on these measurements. On the basis of these studies, it is proposed that blood specimens be anticoagulated with one part 3.8% (w./v.) sodium citrate solution to 19 parts whole blood to avoid excessive anticoagulation of blood samples drawn from patients with polycythemia. Because of the smaller amount of plasma in such samples, relatively larger amounts of anticoagulant are used, and spuriously prolonged PT and PTT measurements commonly result. No deleterious effect on anemic specimens is evident when the smaller amount of citrate is used. Studies of the stability of these specimens indicate that unopened, vacuum-drawn specimens do not noticeably deteriorate for as long as 6 hours, even when kept at room temperature. Prothrombin time measurements remain constant for as long as 24 hours. However, a 10--15% lengthening of the partial thromboplastin time is evident after 24 hours of storage.

Anticoagulants↗
PubMed1997

Plaque removal with variable instrumentation.

The purpose of this study was to evaluate dental plaque removal in a normal healthy mouth, during routine oral hygiene appointments using different techniques and without the use of any disclosing agents. 12 dental hygienists, randomly selected from a continuing education course, were asked to perform oral hygiene on the same patient to remove all the supra-gingival plaque without any time restriction and without the use of a disclosing agents. The plaque index score (O'Leary) was assessed before and after each session with the use of fluorescine and UV light source by an independent examiner. 3 groups of instruments were utilized: group A: ultrasonic scalers + prophy cups; group B: ultrasonic scalers + prophy cups + dental floss; group C: Gracey curettes + prophy cups. While no group was able to remove all the plaque, groups B and C performed significantly better.

Adult↗
PubMed2002

Outcomes and sample selection: the case of a homelessness and substance abuse intervention.

In the likely event that some clients refuse to participate in a psychosocial field experiment, the estimates of the effects of the experimental treatment on client outcomes may suffer from sample selection bias, regardless of whether the statistical analyses include control variables. This paper explores ways of correcting for this bias with advanced correction strategies, focusing on experiments in which clients refuse assignment into treatment conditions. The sample selection modelling strategy, which is highly recommended but seldom applied to random sample psychosocial experiments, and some alternatives are discussed. Data from an experiment on homelessness and substance abuse are used to compare sample selection, conventional control variable, instrumental variable, and propensity score matching correction strategies. The empirical findings suggest that the sample selection modelling strategy provides reliable estimates of the effects of treatment, that it and some other correction strategies are awkward to apply when there is post-assignment rejection, and that the varying correction strategies provide widely divergent estimates. In light of these findings, researchers might wish regularly to compare estimates across multiple correction strategies.

Adult↗
PubMed1998

Controlling for selection bias in the evaluation of Alcoholics Anonymous as aftercare treatment.

OBJECTIVE: The purpose of this research was to control for self-selection bias in the evaluation of Alcoholics Anonymous (AA) as aftercare treatment. Observational studies of alcoholism aftercare treatment are subject to selection bias whenever the self-selection process results in important differences in unobserved casemix dimensions between treatment groups. METHOD: The sample included 118 male veterans discharged from inpatient alcohol treatment, 85% of whom were followed-up at 3 months. Drinking outcomes were measured by self-reported abstinence in the third month after discharge. The aftercare treatment effect of AA was estimated using standard logistic regression analysis and instrumental variables analysis. Instruments included the subject's ability to drive oneself to AA meetings and the presence/absence of an AA meeting in the subject's town of residence. A Hausman exogeniety test was used to determine whether the standard logistic regression results were subject to self-selection bias. RESULTS: Estimates from the standard logistic regression yielded a positive (OR = 3.7) and significant (p = .018) treatment effect for AA attendance. However, the instrumental variables analysis yielded a smaller (OR* = 1.7) and insignificant treatment effect estimate (p = .782). The Hausman exogeniety test indicated that the treatment effect estimate from the standard logistic regression was subject to significant self-selection bias (chi2 = 83.9, 1 df, p <.01). CONCLUSIONS: The AA aftercare treatment effect observed in this sample was due to differences in unobserved casemix factors between the treatment groups. Results suggest that previous AA aftercare research may have also been subject to self-selection bias. Researchers of substance abuse outcomes should consider analyzing nonexperimental data using instrumental variables methodologies.

Adult↗
PubMed1981

A time series of instrumental fertility variables.

Temporal variations in conventional fertility measures reflect the operation of instrumental variables: quantitative and temporal intentions; success in achieving intentions; and reproductive conditions. A set of such variables is described, using data from the 1975 National Fertility Study. There was a large decline in the number of intended conceptions, a recent large rise in the extent of their delay, a very large decline in rates of failure to delay or terminate fertility, and a very large recent rise in sterilization. But one problem proved important and intractable: When the data source is a cross-sectional survey, the length of open interval is inherently different for real and for synthetic cohorts, it is strongly related to reproductive intention, and that affects the classification of exposure to risk in the open interval.

Data Collection↗