Fit Of Logistic Regression Model Research Articles

Pediatric nurses' cognitive representations of children's pain.

The aim of this mixed methods exploratory study was to describe pediatric nurses' cognitive representations (CRs) of the assessment and management of children's pain and to determine the relationships between their CRs and their choices about pain assessment and morphine administration. We recruited a convenience sample of 87 nurses caring for hospitalized children at 4 institutions. We measured the CRs with the Conceptual Content Cognitive Map (3CM) technique and pain assessment and morphine administration with smiling and grimacing child vignettes. We used content analyses for the 3CM data and fit logistic regression models to predict participants' analgesic choice for each vignette. Nearly all (91%) participants identified the child's behavior as an assessment approach; 48% indicated it as most important. Participants (92%) identified pharmacologic as a management approach; 47% indicated it as most important. Participants' CRs did not predict assessment or morphine administration choices. Significantly more participants chose the appropriate analgesic response for the grimacing child than they did for the smiling child. Nurses with more years of pediatric experience were less likely to select administration of the appropriate morphine dose. The 3CM method provided insights into nurses' thinking about pain that are indicative of gaps, which may be amenable to interventions. Findings are from an innovative, unique measure of nurses' knowledge and beliefs about the complex phenomenon of children's pain management. Extensive details about the thought processes of pediatric nurses regarding pain assessment and management surfaced through this analysis, which provide excellent information for direction of future research and practice innovations.

Effects of childhood body size on breast cancer tumour characteristics

IntroductionAlthough a role of childhood body size in postmenopausal breast cancer risk has been established, less is known about its influence on tumour characteristics.MethodsWe studied the relationships between childhood body size and tumour characteristics in a Swedish population-based case-control study consisting of 2,818 breast cancer cases and 3,111 controls. Our classification of childhood body size was derived from a nine-level somatotype. Relative risks were estimated by odds ratios with 95% confidence intervals, derived from fitting unconditional logistic regression models. Association between somatotype at age 7 and tumour characteristics were evaluated in a case-only analysis where P values for heterogeneity were obtained by performing one degree of freedom trend tests.ResultsA large somatotype at age 7 was found to be associated with decreased postmenopausal breast cancer risk. Although strongly associated with other risk factors such as age of menarche, adult body mass index and mammographic density, somatotype at age 7 remained a significant protective factor (odds ratio (OR) comparing large to lean somatotype at age 7 = 0.73, 95% confidence interval (CI) = 0.58-0.91, P trend = 0.004) after adjustment. The significant protective effect was observed within all subgroups defined by estrogen receptor (ER) and progesterone receptor (PR) status, with a stronger effect for ER-negative (0.40, 95% CI = 0.21-0.75, P trend = 0.002), than for ER-positive (0.80, 95% CI = 0.62-1.05, P trend = 0.062), tumours (P heterogeneity = 0.046). Somatotype at age 7 was not associated with tumour size, histology, grade or the presence or absence of metastatic nodes.ConclusionsGreater body size at age 7 is associated with a decreased risk of postmenopausal breast cancer, and the associated protective effect is stronger for the ER-negative breast cancer subtype than for the ER-positive subtype.

Importance of Assessing the Model Adequacy of Binary Logistic Regression

Logistic regression is a sophisticated statistical tool for data analysis in both control experimentation and observational studies. The goal of logistic regression is to correctly predict the category of outcome for individual cases using the most parsimonious model. To accomplish this goal, a model is created that includes all predictor variables that are useful in predicting the response variable. The logistic regression model is being used with increasing rate in various fields in data analysis. In spite of such increase, there has been no commensurate increase in the use of commonly available methods for assessing the model adequacy. Failure to address model adequacy may lead to misleading or incorrect inferences. Therefore, the goal of this study is to present an overview of a few easily employed methods for assessing the fit of logistic regression models. The summary measures of goodness-of-fit as Likelihood Ratio Test, Hosmer-Lemeshow goodness-of-fit test, Osius-Rojek large sample approximation test, Stukel test and area under Receiver Operating Characteristic curve indicate that the logistic regression model fits the data quite well. However, recommendations are made for the use of methods for assessing the model adequacy in different aspects before proceed to present the results from a fitted logistic regression model.

Latency Models for Analyses of Protracted Exposures

The effect of an increment of exposure on disease risk may vary with time-since-exposure. If the pattern of temporal variation is simple (eg, a peak and then a decline in excess risk of disease) then this may be modeled efficiently via a parametric latency function. Estimation of the parameters for such a model can be difficult because the parameters are not a function of a simple summary of the exposure history. Typically, such parameters are estimated via an iterative search that requires calculating a different time-weighted exposure for each combination of the latency function parameters. This article describes a simple approach to fitting logistic regression models that include a parametric latency function. This approach is illustrated using data from a study of the association between radon exposure and lung cancer mortality among underground uranium miners. This approach should facilitate fitting models to assess variation with time since exposure in the effect of a protracted environmental or occupational exposure.

Risk prediction for perioperative mortality of endovascular vs open repair of abdominal aortic aneurysms using the Medicare population

The impact of risk factors upon perioperative mortality might differ for patients undergoing open vs endovascular repair (EVAR) of abdominal aortic aneurysms (AAA). In order to investigate this, we developed a differential predictive model of perioperative mortality after AAA repair. A total of 45,660 propensity score matched Medicare beneficiaries undergoing elective open or endovascular AAA repair from 2001 to 2004 were studied. Using half the dataset we developed a multiple logistic regression model for a matched cohort of open and EVAR patients and used this to derive an easily evaluable risk prediction score. The remainder of the dataset formed a validation cohort used to confirm results. The derivation cohort included 11,415 open and 11,415 endovascular repairs. Perioperative mortality was 5.3% and 1.8%, respectively. Independent predictors of mortality (relative risk [RR], 95% confidence interval [CI]) were open repair (3.2, 2.7-3.8); age (71-75 years 1.2, 0.9-1.6; 76-80 years 1.9, 1.4-2.5; >80 years 3.1, 2.4-4.2); female gender (1.5, 1.3-1.8); dialysis (2.6, 1.5-4.6); chronic renal insufficiency (2.0, 1.6-2.6); congestive heart failure (1.7, 1.5-2.1); and vascular disease (1.3, 1.2-1.6). There were no differential predictors of mortality across the two procedures. A simple scoring system was developed from a logistic regression model fit to both endovascular and open patients (area under the receiver operator curve [ROC] curve of 72.6) from which low, medium, and high risk groups were developed. The absolute predicted mortality ranged from 0.7% for an EVAR patient </=70 years of age with no comorbidities to 38% for an open patient >80 with all the comorbidities considered. Although relative risk was similar among age groups, the absolute difference was greater for older patients (with higher baseline risk). Mortality after AAA repair is predicted by comorbidities, gender, and age, and these predictors have similar effects for both methods of AAA repair. This simple scoring system can predict repair mortality for both treatment options and thus may help guide clinical decisions.

Head and Neck Squamous-Cell Cancer and its Association with Polymorphic Enzymes of Xenobiotic Metabolism and Repair

Tobacco smoking, alcohol drinking, and occupational exposures to polycyclic aromatic hydrocarbons are the major proven risk factors for human head and neck squamous-cell cancer (HNSCC). Major research focus on gene–environment interactions concerning HNSCC has been on genes encoding enzymes of metabolism for tobacco smoke constituents and repair enzymes. To investigate the role of genetically determined individual predispositions in enzymes of xenobiotic metabolism and in repair enzymes under the exogenous risk factor tobacco smoke in the carcinogenesis of HNSCC, we conducted a case-control study on 312 cases and 300 noncancer controls. We focused on the impact of 22 sequence variations in CYP1A1, CYP1B1, CYP2E1, ERCC2/XPD, GSTM1, GSTP1, GSTT1, NAT2, NQO1, and XRCC1. To assess relevant main and interactive effects of polymorphic genes on the susceptibility to HNSCC we used statistical models such as logic regression and a Bayesian version of logic regression. In subgroup analysis of nonsmokers, main effects in ERCC2 (Lys751Gln) C/C genotype and combined ERCC2 (Arg156Arg) C/A and A/A genotypes were predominant. When stratifying for smokers, the data revealed main effects on combined CYP1B1 (Leu432Val) C/G and G/G genotypes, followed by CYP1B1 (Leu432Val) G/G genotype and CYP2E1 (–70G>T) G/T genotype. When fitting logistic regression models including relevant main effects and interactions in smokers, we found relevant associations of CYP1B1 (Leu432Val) C/G genotype and CYP2E1 (–70G>T) G/T genotype (OR, 10.84; 95% CI, 1.64–71.53) as well as CYP1B1 (Leu432Val) G/G genotype and GSTM1 null/null genotype (OR, 11.79; 95% CI, 2.18–63.77) with HNSCC. The findings underline the relevance of genotypes of polymorphic CYP1B1 combined with exposures to tobacco smoke.

Serotonin transporter binding as a possible predictor of one-year remission in major depressive disorder

Objective Lower serotonin transporter (5-HTT) binding ( BP P = f P B avail/ K D) is reported during a major depressive episode (MDE) compared to healthy controls. Higher 5-HTT binding in the diencephalon has previously been associated with acute response to antidepressant treatment. We assessed baseline 5-HTT binding as a predictor of one-year remission from a MDE, examining binding in brain regions implicated in the pathophysiology of major depressive disorder (MDD). Methods 5-HTT binding was quantified using positron emission tomography (PET) with [ 11C]McN5652 in 19 currently depressed subjects with MDD and 41 healthy controls. Depressed subjects received open, naturalistic antidepressant treatment. Remission status was determined one year after PET scan and treatment initiation. Results Significant differences in 5-HTT binding among the three groups (healthy controls, remitters, and non-remitters) were observed in a linear mixed-effects model. Post hoc, non-remitters had lower 5-HTT binding than controls in midbrain, amygdala, and anterior cingulate. Remitters did not differ significantly from controls or non-remitters in 5-HTT binding. Remitters did not differ from non-remitters in clinical characteristics apart from greater family history of depression among non-remitters. A logistic regression model fit to determine the capacity of baseline 5-HTT binding to predict remission status at one year yielded a coefficient that was suggestive but not significant ( p = 0.057). Limitations The small sample size and heterogeneous treatments received reduced statistical power to detect differences in binding based on clinical outcome. Conclusions Lower pretreatment 5-HTT binding may be predictive of non-remission from major depression following one year of naturalistic antidepressant treatment. Future studies using standardized treatment are warranted.

Are Biomarker Levels a Good Follow-Up Tool for Evaluating Obstructive Sleep Apnea Syndrome Treatments?

Background: Polysomnography is the standard method for evaluating results of treatments for obstructive sleep apnea syndrome (OSAS), but it is time-consuming and often unavailable. Objectives: This study aimed to analyze the correlations between sleepiness, anthropometric parameters, exhaled breath condensate (EBC) and serum cytokine levels with changes in the apnea-hypopnea index (AHI), in order to identify potential measurements that could be an alternative to polysomnography. Methods: Based on AHI results, 22 non-OSAS and 68 OSAS cases were followed up. Among the 68 patients, 5 underwent surgery, 2 were treated with oral appliances, 33 were treated using continuous positive airway pressure and 28 were untreated. AHI, anthropometric parameters, serum and EBC levels of interleukin-6 (IL-6), interleukin-10 (IL-10), tumor necrosis factor-α (TNF-α) and 8-isoprostane were measured at baseline and after a 2-month follow-up. Correlations between the AHI and the other parameters were examined. Akaike’s information criterion (AIC) was used to evaluate the fit of logistic regression models, predicting improvements in the severity of the condition from the parameters. Results: IL-6 and TNF-α in EBC and serum gave the highest correlation coefficients (r = 0.62 and r = 0.71 in EBC; r = 0.58 and r = 0.66 in serum, respectively) as well as the lowest AIC values (63.87, 68.97; 62.65, 70.64, respectively). Reductions in waist circumference and weight also correlated with changes in AHI (r = 0.69, AIC = 80.26 and 88.76, respectively). Conclusions: IL-6 and TNF-α measurements may be used in OSAS treatment follow-ups, when polysomnography is not available. Waist circumference and weight could also be used when cytokine laboratory facilities are unavailable.

Understanding the Role of Reactions to Race-Based Treatment in Breast and Cervical Cancer Screening

Racial and ethnic disparities in breast and cervical cancer mortality persist despite effective screening methods. We examined associations between race/ethnicity and Pap testing within three years or mammography within two years, controlling for a composite reactions-to-race-based-treatment variable created using data from the 2002 and 2004 Behavioral Risk Factor Surveillance System Reactions to Race module, which assessed respondents experiences based on one's race. We calculated prevalence of Pap testing (for women aged > or = 18) and mammography (for women aged > or = 40) by race, and fit logistic regression models to estimate the strength of association of reactions to race-based treatment with screening and race--before and after controlling for demographics, socioeconomic status, health status, smoking and healthcare access. In the reduced model, black women were more likely (2.03: 95% CI: 1.55-2.65) to be screened for cervical cancer than whites. Reactions to race-based treatment did not impact the odds of black women receiving Pap tests or mammograms. Given current racial and ethnic disparities in breast and cervical cancer mortality, we suggest that more attention needs to be focused on follow-up of abnormal results and state-of-the art treatment for black and Hispanic women.

On the job illness and injury resulting in lost work time among a national cohort of emergency medical services professionals

The objective of this study was to estimate the prevalence and incidence of job-related illness or injury resulting in lost work time among a national cohort of Emergency Medical Services (EMS) professionals. Also, it was hypothesized that individual and work life characteristics were associated with the occurrence of illnesses or injury. Data for this analysis were obtained from the Longitudinal Emergency Medical Technician Attributes and Demographics Study (LEADS), a prospective study of EMS professionals. The outcome variable of interest was self-reported absence from their EMS job due to an EMS work related illness or injury. The prevalence and incidence of injury with lost work time was estimated using cross-sectional and follow-up data. Multivariable logistic regression analyses were performed to determine if individual and work life characteristics were associated with occupational injury. The prevalence of job-related illness or injury with time away from work was estimated at 9.4%, while the 1-year incidence was estimated at 8.1 per 100 EMS providers. The results from the logistic regression model fit to follow-up data indicate that increasing call volume (OR=3.12 for very high vs. moderate, 95% CI 1.40-6.97), an urban work environment (OR=2.79, 95% CI 1.65-4.72) and a history of back problems (OR=1.72, 95% CI 1.06-2.78) were associated with reporting job-related illness or injury. Results from this analysis provide estimates of the prevalence and incidence of on the job illness and injury resulting in lost work time among a national cohort of EMS professionals.

Recurrence of Ischemic Placental Disease

To test the hypothesis that the presence of preeclampsia, small for gestational age (SGA)-birth, and placental abruption in the first pregnancy confers increased risk in the second pregnancy. A retrospective cohort study entailing a case-crossover analysis was performed based on women who had two consecutive singleton live births (n=154,810) between 1989 and 1997 in Missouri. Small for gestational age was defined as infants with birth weight below the 10th centile for gestational age. Risk and recurrence of ischemic placental disease was assessed from fitting logistic regression models after adjusting for several confounders. Preeclampsia in the first pregnancy was associated with significantly increased risk of preeclampsia (odds ratio 7.03, 95% confidence interval 6.51, 7.59), SGA (odds ratio 1.16, 95% confidence interval 1.06, 1.27), and placental abruption (odds ratio 1.90, 95% confidence interval 1.51, 2.38) in the second pregnancy. Similarly, women with SGA and abruption in the first pregnancy were associated with increased risks of all other conditions in the second pregnancy. Women with preeclampsia, SGA, and placental abruption in their first pregnancy--conditions that constitute ischemic placental disease--are at substantially increased risk of recurrence of any or all these conditions in their second pregnancy. Although causes of these conditions remain largely speculative, these entities may manifest through a common pathway of ischemic placental disease with significant risk of recurrence.

Climate factors affecting conception rate of high producing dairy cows in northeastern Spain

Summer heat stress is a main factor related to low conception rate in high producing dairy herds in warm areas worldwide. We assessed the impact of several climate variables on conception rate in high producing dairy cows in northeastern Spain by examining 10,964 inseminations. The temperature–humidity index (THI) was compared with maximum temperature in terms of its efficiency at predicting conception rate. The following data were recorded for each animal: herd, lactation number, insemination number, insemination date, inseminating bull, and AI technician along with climate variables such as mean and maximum temperatures, rainfall, mean and maximum THI for individual time points Days 7 to 1 before insemination, the day of insemination and 1, 2 and 3 days after insemination. Averages were also established for the following periods: from 7 days before insemination to the insemination day, from 3 days before insemination to the insemination day and from the insemination day to 3 days postinsemination. Based on the odds ratios, the likelihood of conception rate increased significantly by factors of 1.48, 1.47, 1.5, and 1.1 for the respective maximum THI classes <70, 71–75, 76–80, and 81–85 only on Day 3 before AI, while on the insemination day, it increased by factors of 1.73, 1.53, 1.11, and 1.3 for the respective maximum THI classes <70, 71–75, 76–80, and 81–85. In a subsequent logistic regression excluding mean and maximum THI, the effectiveness of temperature at predicting conception rate was evaluated. Although high, the fit of the second logistic regression model was slightly lower than that of the full model ( P = 0.88 versus P = 0.98, respectively) and the information provided by the THI model. The likelihood of conception rate increased significantly by factors of 1.5, 1.2, 1.0, 1.0 for the respective maximum temperature classes <20, 21–25, 26–30, 31–35 °C on Day 1 after AI. The choice of the THI or temperature to monitor the farm environment would have to depend on the particular farm and situation. In our study conditions, the use of maximum temperature alone gives a new point of view regarding the information provided by the THI variables.

Androgen receptor amplification is associated with increased cell proliferation in prostate cancer

Mechanisms of prostate cancer progression during hormonal therapy and the pathobiologic consequences of androgen receptor (AR) gene amplification are inadequately known. To further investigate the hypothesis that AR gene amplification is associated with increased cell proliferation, we analyzed 123 paraffin-embedded prostate cancer specimens from men who experienced tumor relapse during androgen withdrawal therapy. We used fluorescence in situ hybridization to quantify AR gene copy number and Ki-67 immunohistochemistry to determine cell proliferation. One third of the tumors showed AR gene amplification. Among tumors with AR amplification, the mean cell proliferation rate was 19.8 (SD, 12.3; 95% confidence interval [CI], 15.4-24.1), whereas it was 13.0 (SD, 15.9; 95% CI, 9.1-16.8) in tumors without amplification (P = .032). In the best fitting logistic regression model, only proliferation remained significant (P = .040). When the median Ki-67 labeling index (6.7%) of all tumors was used as a cutoff point, the tumors with AR amplification were more frequently highly proliferating than tumors with no amplification (P = .010; odds ratio, 3.4; 95% CI, 1.4-8.3). Our results imply that progression of prostate cancer during androgen withdrawal therapy is associated with AR gene amplification and increased cell proliferation rate in one third of tumors. We suggest that AR gene amplification is an important molecular mechanism underlying the increase in proliferation rate of a substantial fraction of recurrent prostate carcinomas. However, efforts should be targeted to develop prostate cancer cell lines to study causal relationships between AR gene amplification and various biologic variables.

Prospective Breast Cancer Risk Prediction Model for Women Undergoing Screening Mammography

Risk prediction models for breast cancer can be improved by the addition of recently identified risk factors, including breast density and use of hormone therapy. We used prospective risk information to predict a diagnosis of breast cancer in a cohort of 1 million women undergoing screening mammography. There were 2,392,998 eligible screening mammograms from women without previously diagnosed breast cancer who had had a prior mammogram in the preceding 5 years. Within 1 year of the screening mammogram, 11,638 women were diagnosed with breast cancer. Separate logistic regression risk models were constructed for premenopausal and postmenopausal examinations by use of a stringent (P<.0001) criterion for the inclusion of risk factors. Risk models were constructed with 75% of the data and validated with the remaining 25%. Concordance of the predicted with the observed outcomes was assessed by a concordance (c) statistic after logistic regression model fit. All statistical tests were two-sided. Statistically significant risk factors for breast cancer diagnosis among premenopausal women included age, breast density, family history of breast cancer, and a prior breast procedure. For postmenopausal women, the statistically significant factors included age, breast density, race, ethnicity, family history of breast cancer, a prior breast procedure, body mass index, natural menopause, hormone therapy, and a prior false-positive mammogram. The model may identify high-risk women better than the Gail model, although predictive accuracy was only moderate. The c statistics were 0.631 (95% confidence interval [CI] = 0.618 to 0.644) for premenopausal women and 0.624 (95% CI = 0.619 to 0.630) for postmenopausal women. Breast density is a strong additional risk factor for breast cancer, although it is unknown whether reduction in breast density would reduce risk. Our risk model may be able to identify women at high risk for breast cancer for preventive interventions or more intensive surveillance.

Recurrence of spontaneous versus medically indicated preterm birth

Despite the increased tendency of preterm birth to recur, little is known with regard to recurrence risks for spontaneous and medically indicated preterm birth as well as recurrence risks in relation to severity of preterm birth. We examined the recurrence of spontaneous and medically indicated preterm birth. A population-based, retrospective cohort study of births in Missouri (1989 to 1997) was carried out with analyses restricted to women who delivered their first 2 consecutive singleton live births (n = 154,809). Women who experienced spontaneous onset of labor and subsequently delivered preterm (less than 35 weeks) were classified as spontaneous preterm birth. Medically indicated preterm birth included women who delivered preterm through a labor induction or a prelabor cesarean delivery. Risk and odds ratio of preterm birth recurrence were derived from fitting multivariate conditional logistic regression models after adjusting for potential confounders. If the first pregnancy resulted in a spontaneous preterm birth, then affected women were more likely to deliver preterm spontaneously (adjusted odds ratio 3.6, 95% confidence interval 3.2, 4.0) and also as a medically indicated preterm birth (odds ratio 2.5, 95% confidence interval 2.1, 3.0) in the second birth. Similarly, if the first pregnancy resulted in a medically indicated preterm birth, affected women were 10.6-fold (95% confidence interval 10.1, 12.4) more likely to deliver preterm because of medical indications in the second pregnancy as well as preterm spontaneously (odds ratio 1.6, 95% confidence interval 1.3, 2.1). The greatest risk of recurrence of preterm birth in the second pregnancy tended to occur around the same gestational age as preterm birth in the first pregnancy, regardless of the clinical subtype. The observation that spontaneous preterm birth is not only associated with increased recurrence of spontaneous but also medically indicated preterm birth and vice versa, suggests that the 2 clinical subtypes may share common etiologies.

Goodness-of-fit tests for logistic regression models when data are collected using a complex sampling design

Logistic regression models are frequently used in epidemiological studies for estimating associations that demographic, behavioral, and risk factor variables have on a dichotomous outcome, such as disease being present versus absent. After the coefficients in a logistic regression model have been estimated, goodness-of-fit of the resulting model should be examined, particularly if the purpose of the model is to estimate probabilities of event occurrences. While various goodness-of-fit tests have been proposed, the properties of these tests have been studied under the assumption that observations selected were independent and identically distributed. Increasingly, epidemiologists are using large-scale sample survey data when fitting logistic regression models, such as the National Health Interview Survey or the National Health and Nutrition Examination Survey. Unfortunately, for such situations no goodness-of-fit testing procedures have been developed or implemented in available software. To address this problem, goodness-of-fit tests for logistic regression models when data are collected using complex sampling designs are proposed. Properties of the proposed tests were examined using extensive simulation studies and results were compared to traditional goodness-of-fit tests. A Stata ado function svylogitgof for estimating the F -adjusted mean residual test after svylogit fit is available at the author's website http://www.people.vcu.edu/~kjarcher/Research/Data.htm.

Nonparametric smoothing in modeling logistic regression

This paper is emphasized on the logistic regression model fit with continuous and categorical covariates. A test statistic based on non-parametric local linear regression technique with optimal bandwidth which is chosen by cross validation method is proposed. This proposed test does not require a space partition of covariates or groups of the fitted values to compensate a small expected cell size. The expectation and variance of the proposed test statistic are computed and the sampling distributions of test statistic for various of logistic regression models are evaluated. We use simulations to compare the power of the new test with that of the current assessing methods for different logistic models. The proposed method is illustrated by using data from Caplehorn (1991) for the heroin addicts.

Evaluating satellite and climate data-derived indices as fire risk indicators in savanna ecosystems

The repeated occurrence of severe wildfires has highlighted the need for development of effective vegetation monitoring tools. We compared the performance of indices derived from satellite and climate data as a first step toward an operational tool for fire risk assessment in savanna ecosystems. Field collected fire activity data were used to evaluate the potential of the normalized difference vegetation index (NDVI), normalized difference water index (NDWI), and the meteorological Keetch-Byram drought idex (KBDI) to assess fire risk. Performance measures extracted from the binary logistic regression model fit were used to quantitatively rank indices in terms of their effectiveness as fire risk indicators. NDWI performed better when compared to NDVI and KBDI based on the results from the ranking method. The c-index, a measure of predictive ability, indicated that the NDWI can be used to predict seasonal fire activity (c=0.78). The time lag at the start of the fire season between time-series of fire activity data and the selected indices also was studied to evaluate the ability to predict the start of the fire season. The results showed that NDVI, NDWI, and KBDI can be used to predict the start of the fire season. NDWI consequently had the highest capacity to monitor fire activity and was able to detect the start of the fire season in savanna ecosystems. It is shown that the evaluation of satellite- and meteorological fire risk indices is essential before the indices are used for operational purposes to obtain more accurate maps of fire risk for the temporal and spatial allocation of fire prevention or fire management.

Use of orthopedic shoes in patients with degenerative disorders of the foot

Jannink MJ, IJzerman MJ, Groothuis-Oudshoorn K, Stewart RE, Groothoff JW, Lankhorst GJ. Use of orthopedic shoes in patients with degenerative disorders of the foot. Arch Phys Med Rehabil 2005;86:687–92. Objectives To study the actual use of orthopedic shoes by patients with degenerative foot disorders and to identify factors associated with use and nonuse, based on the parameters of the International Organization for Standardization definition of usability: effectiveness, efficiency, satisfaction, and context of use. Design Multicenter, prospective cohort study. Setting Outpatient clinics of 7 rehabilitation centers in the Netherlands. Participants One hundred consecutive patients with degenerative foot disorders. Interventions Not applicable. Main outcome measures Usability was assessed by means of the Questionnaire for Usability Evaluation of orthopedic shoes. Results Seventy of 93 patients with degenerative foot disorders wore their orthopedic shoes for more than 3 days a week after 3 months of follow-up. Factors significantly associated with the actual use of orthopedic shoes were (1) increase in stance duration (effectiveness odds ratio [OR]=2.14; 95% confidence interval [CI], 1.19–3.85), (2) decrease in skin abnormalities (effectiveness OR=1.35; 95% CI, 1.02–1.8]), (3) problems experienced with putting on and taking off orthopedic shoes (efficiency OR=.46; 95% CI, .26–.82), and (4) cosmetic appearance of orthopedic shoes (satisfaction OR=1.54; 95% CI, 1.1–2.15). The overall fit of the multiple logistic regression model ( R 2) was 56.3%. Conclusions By adding efficiency and satisfaction factors and not focusing only on the effectiveness factors, the amount of explained variance increases, and it becomes possible to evaluate and design products for people with special needs more comprehensively.

Autosomal-dominant polycystic kidney disease as a risk factor for diabetes mellitus following renal transplantation

Posttransplant diabetes mellitus is an important complication of renal transplantation that is associated with a significant impact on quality of life and an increase in long-term morbidity and mortality. Autosomal-dominant polycystic kidney disease (ADPKD) is a hereditary disease that commonly leads to end-stage renal disease (ESRD) in adulthood. The association between ADPKD and posttransplant diabetes mellitus has not been previously studied in a large cohort of patients. To address this question, we studied a cohort of 135 patients with ADPKD who received a first renal-only transplant between January 1985 and December 1999. An age, race, and date of transplant-matched cohort of 135 non-ADPKD subjects were used as the control population. The cohorts were similar at baseline for gender distribution, body mass index (BMI), proportion of obese subjects (BMI greater than 30 kg/m(2)), family history of diabetes mellitus, and type of donor (deceased or living). At 12 months, the incidence of posttransplant diabetes mellitus was significantly higher in patients with ADPKD when compared to the controls (17% vs. 7.4%) (P= 0.016), despite no significant differences in the BMI, percent increase in BMI, number of acute rejections, prednisone dose at 3 and 6 months, use of diuretics or beta blockers, delayed graft function, or serum creatinine levels. The proportion of subjects requiring insulin was significantly higher in the ADPKD group (11.1% vs. 3%) (P= 0.009). Variables significantly associated with posttransplant diabetes mellitus at 1 year by bivariate analyses were the diagnosis of ADPKD (P= 0.02), BMI at transplant (P= 0.04), obesity at 12 months (P= 0.01), and delayed graft function (P= 0.02). Gender of recipient (P= 0.9), family history of diabetes (P= 0.3), prednisone dose at 3 months (P= 0.9) and 6 months (P= 0.7), acute rejection (P= 0.9), use of beta blockers or tacrolimus (P= 0.8), deceased donor transplant (P= 0.2), and serum creatinine at 1 year (P= 0.5) were not associated with posttransplant diabetes mellitus. A trend toward increased incidence of posttransplant diabetes mellitus was found with the use of diuretics post transplant (P= 0.054). By multivariable analyses, in patients with ADPKD, the adjusted (by all the variables listed above) relative risk for development of posttransplant diabetes mellitus was 2.87 (95% CI = 1.24-6.65) (P= 0.014). Only the diagnosis of ADPKD (RR = 2.9) (P= 0.01), obesity at 1 year (RR 2.5) (P= 0.017), and delayed graft function (RR 2.4) (P= 0.03) contributed significantly to the fit of a stepwise logistic regression model. Patient survival was significantly worse in the cohort of patients who developed posttransplant diabetes mellitus (median survival 109.3 vs. 121 months) (P= 0.008). In our study patients with ADPKD were at a threefold increased risk for development of posttransplant diabetes mellitus within the first year following renal transplantation. Development of posttransplant diabetes mellitus was associated with a significant detrimental impact on patient survival. Further studies are needed to provide insight into the mechanisms of the association between ADPKD and posttransplant diabetes mellitus.