Peter J. Fos

Data mining a diabetic data warehouse

Diabetes is a major health problem in the United States. There is a long history of diabetic registries and databases with systematically collected patient information. We examine one such diabetic data warehouse, showing a method of applying data mining techniques, and some of the data issues, analysis problems, and results. The diabetic data warehouse is from a large integrated health care system in the New Orleans area with 30,383 diabetic patients. Methods for translating a complex relational database with time series and sequencing information to a flat file suitable for data mining are challenging. We discuss two variables in detail, a comorbidity index and the HgbA1c, a measure of glycemic control related to outcomes. We used the classification tree approach in Classification and Regression Trees (CART) with a binary target variable of HgbA1c >9.5 and 10 predictors: age, sex, emergency department visits, office visits, comorbidity index, dyslipidemia, hypertension, cardiovascular disease, retinopathy, end-stage renal disease. Unexpectedly, the most important variable associated with bad glycemic control is younger age, not the comorbiditity index or whether patients have related diseases. If we want to target diabetics with bad HgbA1c values, the odds of finding them is 3.2 times as high in those <65 years of age than those older. Data mining can discover novel associations that are useful to clinicians and administrators [corrected].

Evaluación del estado de salud con la Encuesta SF-36: resultados preliminares en México

Objective. To establish a multidimensional profile of the health status in a population of southeastern Mexico and analyze the psychometric properties of a translation of the SF-36 survey authorized by the International Quality of Life Assessment Project. Material and methods. The SF-36 was administered to 257 volunteers of a clinic and a control group in a governmental institution. The methodology proposed by the author of the survey to build the health scales was used without modifications. The validity and reliability of the SF-36 adaptation to Mexico was psychometrically analyzed. Results. Eight scales or health concepts related to physical function, physical role, body pain, general health, vitality, social function, emotional role and mental health were constructed. In the health service volunteers, the scale with the lowest mean was general health (63), and that with the highest (89) was physical role. In the control population, the lowest mean was general health (73) and the highest (94.6) was physical function. A comparison of the health scales of both groups showed significant differences in physical function, physical role, body pain, general

Assessing Health-Related Quality of Life in Cataract Patients: The Relationship between utility and Health-Related Quality of Life Measurement

The relationship between health-related quality of life (HQL) measures and patient preference for their health status was studied. Study subjects consisted of 132 patients at four hospitals who were scheduled for cataract surgery. Generic and disease-specific health status measures were determined in study subjects. The Medical Outcomes Study Short-form 36 (SF-36) item health status instrument was used to measure generic health status and the Visual Function 14 (VF-14) item visual health status instrument was used as the disease-specific health measure. Preference for general health and visual health was measured by assessing utilities assigned by patients to certain health states. Utilities assigned for general health were correlated with all categories of the SF-36 and VF-14 scores. Utilities assigned for visual health were correlated with four categories of the SF-36 (role limitation due to emotional health, general health, physical functioning, and vitality) and VF-14 scores. Utilities assigned for visual health were more strongly correlated with VF-14 scores than generic measures of health. Verbal ratings for visual health were correlated with Snellen visual acuity (SVA) (r=0.20), utilities assigned for visual health (r=0.58), VF-14 scores (r=0.74), all categories of the SF-36 (r values ranging from 0.21 to 0.28), utilities assigned for general health (r=0.19), and verbal ratings for general health (r=0.29). Utility measures and verbal ratings for general and visual health were shown to be appropriate HQL measures. These measures were strongly correlated with other established generic and disease-specific health measures and should be included in the array of health status measures.

Body mass index and health related quality of life in elementary school children: a pilot study

BackgroundWe investigated the relationship between Body Mass Index (BMI) and health-related quality of life (HRQOL) indicated by baseline health status in elementary school children.MethodsData were obtained via parents whose children enrolled in an elementary school, kindergarten to fourth grade, in southern Mississippi in spring 2004. Parents completed the SF-10 for Children™, a brief 10-item questionnaire designed to measure childrens HRQOL on a voluntary basis.ResultsA total of 279 parents completed the questionnaires for their children. On average, physical and psychosocial summary scores, major indicators for HRQOL, were significantly higher among the elementary school children in our study relative to those from U.S. children overall (p < 0.0001 and p = 0.0007, respectively). Males tended to have better physical functioning than their female classmates, whereas females had better psychosocial health. Overall, except for third graders, the physical summary scores increased as grade level increased. The means for psychosocial score fluctuated without a clear pattern over the five grade levels. High level of BMI was significantly associated with childrens physical summary scores below 50, a norm used for U.S. children (p = 0.003). Gender and grade were not significant predictors of childrens physical and psychosocial scores.DiscussionThis study can be used as baseline information to track changes over time, in BMI and health status among the elementary school children. In addition, this study can be used to investigate relationships between BMI, health status, intellectual ability, and performance in school.ConclusionThe findings suggest that programs designed to encourage children to lose weight in a healthy manner, thus reducing their BMI, could improve the physical and psychosocial health, and subsequently increase HRQOL.

An analysis of program and family costs of case managed care for technology-dependent infants with bronchopulmonary dysplasia

Caring for technology-dependent infants with bronchopulmonary dysplasia requires a wide range and intense level of services. Case management programs can offer comprehensive services to patients with complex needs. The Neonatal Pulmonary Program at Tulane University Medical Center is a case management program for infants with chronic pulmonary problems. The purpose of this study was to describe the costs of providing care for technology-dependent infants with bronchopulmonary dysplasia and to include the direct and indirect costs to families as well as the Program costs. The study population included 89 infants enrolled in the program from September 1987 through February 1992. Outpatient, inpatient, and professional staff costs were derived from hospital and clinic billing information; all other costs were determined through family interviews. Total costs for outpatient services were

Decision science: a scientific approach to enhance public health budgeting.

59,627 (89), professional team members time

Health-related quality of life of cataract patients: cross-cultural comparisons of utility and psychometric measures

185,539 (89); inpatient services

Racial variations in cesarean section rates: an analysis of Medicaid data in Louisiana.

1,144,930 (89), professional costs for inpatient services

Assessment of primary health care access status: an analytic technique for decision making

88,946; direct health costs

Combining the benefits of decision science and financial analysis in public health management: a county-specific budgeting and planning model.

32,543 (37) for home health care, equipment, medications, special diets; direct non-health costs