Search

Objective: Spatio and/or temporal surveillance systems are designed to monitor the ongoing appearance of disease cases in space and time, and to detect potential disturbances in either dimension. Patient addresses are sometimes reported at some level of geographic aggregation, for example by ZIP code or census tract. While this aggregation has the advantage of protecting patient privacy, it also risks compromising statistical efficiency. This paper investigated the variation in power to detect a change in the spatial distribution in the presence of spatial aggregation., Methods: The authors generated 400,000 spatial datasets with varying location and spread of simulated spatial disturbances, both on a purely synthetic uniform population, and on a heterogeneous population, representing hospital admissions to three community hospitals in Cape Cod, Massachusetts. The authors evaluated the power of the M-statistic to detect spatial disturbances, comparing the use of exact spatial locations versus twelve different levels of aggregation, where the M-statistic is a comparison of two distributions of interpoint distances between locations., Results: When the spread of simulated spatial disturbances was contained to a small portion of the study region or affects a large proportion of the population at risk, power was highest when exact locations were reported. If the spatial disturbance was a more modest signal, the best power was attained at an aggregated level., Conclusions: The precision at which patients' locations are reported has the potential to affect the power of detection significantly.

The past decade has brought major new developments in the psychopharmacologic management of generalized anxiety disorder and social phobia. We examined medication-prescribing patterns for the treatment of these anxiety disorders for 12 years to assess changes in patients' anti-anxiety psychotropic medication usage during that period of evolving practice guidelines. We examined psychotropic medication use in 305 patients with generalized anxiety disorder and 232 with social phobia enrolled in the Harvard/Brown Anxiety Disorders Research Project (HARP), a prospective, longitudinal study of anxiety disorders. Psychotropic treatment patterns seem to have remained relatively stable over 12 years with benzodiazepines the medications most commonly used for both generalized anxiety disorder and social phobia. Comparatively, selective serotonin reuptake inhibitor (SSRI) and venlafaxine usage as stand-alone medications for these disorders remained low throughout the follow-up period. At the 12-year follow-up, 24% of patients with generalized anxiety disorder and 30% of patients with social phobia were utilizing neither an SSRI/selective norepinephrine reuptake inhibitor (SNRI) nor a benzodiazepine. Treatment recommendations for use of SSRIs and venlafaxine in the management of generalized anxiety disorder and social phobia initially promulgated in 1998 had a modest impact on changes in psychopharmacologic practice 4-5 years later. Difficulties in the implementation of treatment guidelines are discussed.

Objective: To explore the hypothesis that variation in respiratory management among newborn intensive care units (NICUs) explains differences in chronic lung disease (CLD) rates., Design: Case-cohort study., Setting: NICUs at 1 medical center in New York (Babies' and Children's Hospital [Babies']) and 2 in Boston (Beth Israel Hospital and Brigham and Women's Hospital [Boston])., Study Population: Four hundred fifty-two infants born at 500 to 1500 g birth weight between January 1991 and December 1993, who were enrolled in an epidemiologic study of neonatal intracranial white matter disorders., Case Definition: Supplemental oxygen required at 36 weeks' postmenstrual age., Results: The prevalence rates of CLD differed substantially between the centers: 4% at Babies' and 22% at the 2 Boston hospitals, despite similar mortality rates. Initial respiratory management at Boston was more likely than at Babies' to include mechanical ventilation (75% vs 29%) and surfactant treatment (45% vs 10%). Case and control infants at Babies' were more likely than were those at Boston to have higher partial pressure of carbon dioxide and lower pH values on arterial blood gases. However, measures of oxygenation and ventilator settings among case and control infants were similar at the 2 medical centers in time-oriented logistic regression analyses. In multivariate logistic regression analyses, the initiation of mechanical ventilation was associated with increased risk of CLD: after adjusting for other potential confounding factors, the odds ratios for mechanical ventilation were 13.4 on day of birth, 9.6 on days 1 to 3, and 6.3 on days 4 to 7. Among ventilated infants, CLD risk was elevated for maximum peak inspiratory pressure >25 and maximum fraction of inspired oxygen = 1.0 on the day of birth, lowest peak inspiratory pressure >20 and maximum partial pressure of carbon dioxide >50 on days 1 to 3, and lowest white blood count <8 K on days 4 to 7. Even after adjusting for white blood count <8 K and the 4 respiratory care variables, infants in Boston continued to be at increased risk of CLD, compared with premature infants at Babies' Hospital., Conclusion: In multivariate analyses, a number of specific measures of respiratory care practice during the first postnatal week were associated with the risk of a very low birth weight infant developing CLD. However, after adjusting for baseline risk, most of the increased risk of CLD among very low birth weight infants hospitalized at 2 Boston NICUs, compared with those at Babies' Hospital, was explained simply by the initiation of mechanical ventilation.

The authors developed and cross-validated prediction models for newly diagnosed cases of liver disorders by using logistic regression and neural networks. Computerized files of health care encounters from the Fallon Community Health Plan were used to identify 1,674 subjects who had had liver-related health services between July 1, 1992, and June 30, 1993. A total of 219 subjects were confirmed by review of medical records as incident cases. The 1,674 subjects were randomly and evenly divided into training and test sets. The training set was used to derive prediction algorithms based solely on the automated data; the test set was used for cross-validation. The area under the Receiver Operating Characteristic curve for a neural network model was significantly larger than that for logistic regression in the training set (p = 0.04). However, the performance was statistically equivalent in the test set (p = 0.45). Despite its superior performance in the training set, the generalizability of the neural network model is limited. Logistic regression may therefore be preferred over neural network on the basis of its established advantages. More generalizable modeling techniques for neural networks may be necessary before they are practical for medical research.

This article is a report on different methods of estimation of the number of injection drug users (IDUs) in the Boston standard metropolitan statistical area (SMSA) in 1993. Because the sharing of needles is a means of transmission for HIV, an estimate of the prevalence of injection drug use is essential for acquired immunodeficiency syndrome (AIDS) prevention programs. The data for this study come from the Massachusetts Department of Public Health Bureau of Substance Abuse, the Massachusetts AIDS Surveillance Unit, the Drug Abuse Warning Network (DAWN), the National Household Survey on Drug Abuse: Population Estimates 1993, and the Massachusetts State Police Crime Reporting Unit. The estimates in this study can be assessed with regard to the different time frames that are involved, noting that some estimates are useful only to establish upper or lower bounds. The estimate that best meshes with the findings of this report is that there were between 13,500-16,500 injection drug users in greater Boston and 45,000-60,000 IDUs in Massachusetts in 1993.