Pramod K. Pathak

Practice-Based Referrals to a Tobacco Cessation Quit Line: Assessing the Impact of Comparative Feedback vs General Reminders

PURPOSE We undertook a study to assess the impact of comparative feedback vs general reminders on practice-based referrals to a tobacco cessation quit line and estimated costs for projected quit responses. METHODS We conducted a group-randomized clinical trial comparing the impact of 6 quarterly (18 months) feedback reports (intervention) with that of general reminders (control) on practice-based clinician referrals to a quit-line service. Feedback reports were based on an Achievable Benchmark of Care approach using baseline practice, clinician, and patient survey responses, and referrals per quarter. Comparable quit responses and costs were estimated. RESULTS Three hundred eight clinicians participated (171 family medicine, 88 internal medicine, 49 obstetrics-gynecology) from 87 primary care practices in Michigan. After 18 months, there were more referrals from the intervention than from the control practices (484 vs 220; P <.001). Practice facsimile (fax) referrals (84%, n = 595) exceeded telephone referrals (16%, n = 109), but telephone referrals resulted in greater likelihood of enrollment (77% telephone vs 44% fax, P <.001). The estimated number of smokers who quit based on the level of services utilized by referred smokers was 66 in the feedback and 36 in the gentle reminder practices. CONCLUSION Providing comparative feedback on clinician referrals to a quit-line service had a modest impact with limited increased costs.

A Monge property for the d -dimensional transportation problem

In 1963, Hoffman gave necessary and sufficient conditions under which a family of O(mn)-time greedy algorithms solves the classical two-dimensional transportation problem with m sources and n sinks. One member of this family, an algorithm based on the «northwest corner rule», is of particular interest, as its running time is easily reduced to O(m+n). When restricted to this algorithm, Hoffmans result can be expressed as follows: the northwest-corner-rule greedy algorithm solves the two-dimensional transportation problem for all source and supply vectors if and only if the problems cost array C={c[i,j]} possesses what is known as the (two-dimensional) Monge property, which requires c[i 1 ,j 1 ]+c[i 2 ,j 2 ] ≤ c[i 1 ,j 2 ]+c[i 2 ,j 1 ] for i 1 <i 2 and j 1 <j 2 . This paper generalizes this last result to a higher dimensional variant of the transportation problem. We show that the natural extension of the northwest-corner-rule greedy algorithm solves an instance of the d-dimensional transportation problem if and only if the problems cost array possesses a d-dimensional Monge property recently proposed by Aggarwal and Park in the context of their study of monotone arrays. We also give several new examples of cost arrays with this d-dimensional Monge property

Is liquid-based pap testing affected by water-based lubricant?

OBJECTIVE: To estimate the effect of water-based lubricant on the liquid-based Pap test. METHODS: Two hundred women each had 2 simultaneous Pap test collections. One specimen was randomly contaminated with 0.5 mL of water-based lubricant after collection. The physician was blinded as to which collection was contaminated. Both specimens were sent for cytology. Cytopathologists were blinded to the study. The rate of abnormal cytology and discordance between the control and the contaminated specimens was estimated. The discordance of secondary diagnoses, such as yeast infections and bacterial vaginosis, was also estimated. RESULTS: The incidence of abnormal cytology was similar in the contaminated specimens and the control specimens (6.5%, 95% confidence interval [CI] 3.5–10.9% versus 7.0%, 95% CI 3.9–11.5%). The rate of disagreement between the 2 specimens collected from each patient was 7.5% (95% CI 4.6–12.1%) and is similar to previously published estimates of discordance using conventional cytology. Secondary diagnoses, such as yeast infections or bacterial vaginosis, were similar in both groups, suggesting no interference from the lubricant. CONCLUSION: Water-based lubricant does not affect liquid-based cervical cytology Pap testing. Water-based lubricant does not affect secondary diagnoses such as bacterial vaginosis or yeast infections. LEVEL OF EVIDENCE: II-1

Bootstrap by sequential resampling

Abstract This paper examines resampling for bootstrap from a survey sampling point of view. Given an observed sample of size n, resampling for bootstrap involves n repeated trials of simple random sampling with replacement. From the point of view of information content, it is well known that simple random sampling with replacement does not result in samples that are equally informative (see Pathak (1964) Ann. Math. Statist. 35, 795–808; Biometrika 51, 185–193). This is due to the randomness in the number of distinct observations that occur in different bootstrap samples. We propose an alternative scheme of sampling sequentially (with replacement each time) until k distinct original observations appear. In such a scheme, the bootstrap sample size becomes random as it varies from sample to sample, but each sample has exactly the same number of distinct observations. We show that the choice of k = (1 − e−1)n∼0.632n has some advantage, stemming from the observation made by Efron (1983, J. Am. Statist. Assoc. 78, 316–331) that the usual bootstrap samples are supported on approximately 0.632n of the original data points. Using recent results on empirical processes, we show that main empirical characteristics of the sequential resampling bootstrap are asymptotically within the distance of order ∼n −3 4 of the corresponding characteristics of the usual bootstrap.

Bayes estimation of hazard and acceleration in accelerated testing

In accelerated life testing, the time transformation function theta (t) is often unknown, even if that function is assumed to be linear. If theta (t) is known, data in the accelerated condition can be adjusted to provide information about the failure time distribution in the use condition. If theta (t) is unknown, the usual estimation procedures require data from the use condition as well as data from the acceleration condition. In this work it is assumed that the uncertainty about theta can be modeled by a prior distribution, chosen from the truncated Pareto family of distributions, and that the uncertainty in lambda , the failure rate, can be modeled by a prior distribution from the gamma family. Under these assumptions, the posterior distributions and their first two moments are provided for both lambda and theta . Thus, this complete Bayes approach to accelerated life testing with the assumed model allows the adjustment of data taken in the accelerated condition to provide the user with the important estimates in the use condition. The results are illustrated by examples. >

Prognosis Based on Creatine Kinase Isoenzyme MB, Cardiac Troponin I, and Right Ventricular Size in Stable Patients With Acute Pulmonary Embolism

Prognosis of stable patients with acute pulmonary embolism (PE) has been assessed with cardiac troponin I (cTnI) and right ventricular (RV) function or size. Whether creatine kinase-MB isoenzyme (CK-MB) would add to the prognostic assessment is uncertain. We retrospectively assessed in-hospital mortality from PE in 392 stable patients to test the hypothesis that CK-MB would be of greater prognostic value than cTnI or RV size and we assessed whether combinations would increase prognostic value. CK-MB was high in 29 patients (7.4%); cTnI was high in 76 patients (19%) and intermediate in 78 patients (20%). The right ventricle was dilated in 128 patients (33%). Trends showed highest in-hospital mortality from PE in 4 of 29 (14%) with high CK-MB compared to 6 of 76 (7.9%) with high cTnI and 8 of 128 (6.3%) with RV dilatation (differences NS). High CK-MB and high cTnI provided added prognostic information only in patients with RV dilatation. Mortality with high CK-MB plus RV dilatation (4 of 19, 21%) tended to exceed mortality with high cTnI plus RV dilatation (5 of 39, 13%, NS). When CK-MB and cTnI were high and the right ventricle was dilated, PE mortality tended to be highest (4 of 14, 29%, NS). In conclusion, cardiac biomarkers contributed to prognosis only in patients with RV dilatation. CK-MB was the strongest predictor of death from PE but its prevalence was low, thus limiting its value as a single prognostic indicator. The combination of high CK-MB, high cTnI, and RV dilatation tended to indicate the highest mortality.

Systematic and nonrandom sampling in the presence of polynomial trends

In the case of finite populations with low-order polynomial trends present, the use of the least squares regression estimator of the mean is discussed. A sampling scheme, which optimizes the efficiency of the regression estimator over a particular class of schemes, is presented.

Log-Odds Rate and Monotone Log-Odds Rate Distributions

Since the 1960s, reliability models for time to failure based on monotone failure rate models have become important models of failure time for reliability practitioners. Bounds for monotone increasing failure rates (IFR) have been developed and are esp..

Limits in motor control bandwidth during stick balancing

Why can we balance a yardstick but not a pencil on the tip of our finger? As with other physical systems, human motor control has constraints, referred to as bandwidth, which restricts the range of frequency over which the system can operate within some tolerated level of error. To investigate control bandwidth, the natural frequency of a stick used during a stick-balancing task was modified by adjusting the height of a mass attached to the stick. The ability to successfully balance the stick with the mass positioned at four different heights was determined. In addition, electromyographic signals from forearm and trunk muscles were recorded during the trials. We hypothesized that 1) the probability of successfully balancing would decrease as mass height decreased; and 2) the level of muscle activation in both agonist and antagonist would increase as the natural frequency of the stick increased. Results showed that as the mass height decreased the probability of successfully balancing the stick decreased. Changes in the probability of success with respect to mass height showed a threshold effect, suggesting that limits in human control bandwidth were approached at the lowest mass height. Also, the level of muscle activation in both the agonist and antagonist of the forearm and trunk increased linearly as the natural frequency of the stick increased. These changes in muscle activation suggest that the central nervous system adapts muscle activation to task dynamics, possibly to improve control bandwidth.

Empirical Bayesian estimation of mean life from an accelerated life test

Abstract In accelerated life testing, the time to failure of an item is observed under a high stress level, and the estimates are needed for the normal stress levels. This paper is concerned with empirical Bayes estimation of the hazard rate constant for exponentially distributed life times. We have assumed that the effect of acceleration is to scale up the hazard rate, and that the hazard rate has the natural conjugate prior with a known mean and unknown variance. We have considered two smooth empirical Bayes estimates of the hazard rate, and computed their Bayes risks by simulation.