Evaluating sampling strategies and logistic regression methods for modelling complex land cover changes

The role of land cover change as a significant component of global change has become increasingly recognized in recent decades. Large databases measuring land cover change, and the data which can potentially be used to explain the observed changes, are also becoming more commonly available. When developing statistical models to investigate observed changes, it is important to be aware that the chosen sampling strategy and modelling techniques can influence results. We present a comparison of three sampling strategies and two forms of grouped logistic regression models (multinomial and ordinal) in the investigation of patterns of successional change after agricultural land abandonment in Switzerland. Results indicated that both ordinal and nominal transitional change occurs in the landscape and that the use of different sampling regimes and modelling techniques as investigative tools yield different results. Synthesis and applications. Our multimodel inference identified successfully a set of consistently selected indicators of land cover change, which can be used to predict further change, including annual average temperature, the number of already overgrown neighbouring areas of land and distance to historically destructive avalanche sites. This allows for more reliable decision making and planning with respect to landscape management. Although both model approaches gave similar results, ordinal regression yielded more parsimonious models that identified the important predictors of land cover change more efficiently. Thus, this approach is favourable where land cover change pattern can be interpreted as an ordinal process. Otherwise, multinomial logistic regression is a viable alternative.

Validation of Logistic Regression Models for Landslide Susceptibility Maps

A wide range of numerical models and tools have been developed over the last decades to support the decision making process in environmental applications, ranging from physical models to a variety of statistically-based methods. In this study, a landslide susceptibility map of a part of Three Gorges Reservoir region of China was produced, employing binary logistic regression analyses. The available information includes the digital elevation model of the region, geological map and different GIS layers including land cover data obtained from satellite imagery. The landslides were observed and documented during the field studies. The validation analysis is exploited to investigate the quality of mapping.

Suicidal expression among school-attending adolescents in a middle-income sub-Saharan country.

We investigated correlates for suicidal expression among adolescents in the Seychelles. Data on 1,432 students (52% females) were derived from the Global School-based Health Survey. Participants were divided into three groups: those with no suicidal behavior (N = 1,199); those with suicide ideation/SI (N = 89); and those reporting SI with a plan to carry out a suicide attempt/SISP (N = 139), each within a 12-month recall period. Using multinomial logistic regression, we examined the strength of associations with social, behavioral and economic indicators while adjusting for covariates. Sixteen percent of school-attending adolescents reported a suicidal expression (10% with a plan/6.2% without). Those reporting SI were younger (relative risk ratio RRR = 0.81; CI = 0.68-0.96), indicated signs of depression (RRR = 1.69; CI = 1.05-2.72) and loneliness (RRR=3.36; CI =1.93-5.84). Tobacco use (RRR = 2.34; CI = 1.32-4.12) and not having close friends (RRR = 3.32; CI = 1.54-7.15) were significantly associated with SI. Those with SISP were more likely to be female (RRR = 0.47; 0.30-0.74), anxious (RRR = 3.04; CI = 1.89-4.88) and lonely (RRR = 1.74; CI = 1.07-2.84). Having no close friends (RRR = 2.98; 1.56-5.69) and using tobacco (RRR = 2.41; 1.48-3.91) were also strongly associated. Having parents who were understanding was protective (RRR = 0.50; CI = 0.31-0.82). Our results suggest that school health promotion programs may benefit from targeting multiple factors associated with suicidal expression. More research, particularly multilevel designs are needed to identify peer and family influences which may modify associations with suicidality.

Prospective evaluation of risk of vertebral fractures using quantitative ultrasound measurements and bone mineral density in a population-based sample of postmenopausal women: results of the Basel Osteoporosis Study.

OBJECTIVE: Prospective studies have shown that quantitative ultrasound (QUS) techniques predict the risk of fracture of the proximal femur with similar standardised risk ratios to dual-energy x-ray absorptiometry (DXA). Few studies have investigated these devices for the prediction of vertebral fractures. The Basel Osteoporosis Study (BOS) is a population-based prospective study to assess the performance of QUS devices and DXA in predicting incident vertebral fractures. METHODS: 432 women aged 60-80 years were followed-up for 3 years. Incident vertebral fractures were assessed radiologically. Bone measurements using DXA (spine and hip) and QUS measurements (calcaneus and proximal phalanges) were performed. Measurements were assessed for their value in predicting incident vertebral fractures using logistic regression. RESULTS: QUS measurements at the calcaneus and DXA measurements discriminated between women with and without incident vertebral fracture, (20% height reduction). The relative risks (RRs) for vertebral fracture, adjusted for age, were 2.3 for the Stiffness Index (SI) and 2.8 for the Quantitative Ultrasound Index (QUI) at the calcaneus and 2.0 for bone mineral density at the lumbar spine. The predictive value (AUC (95% CI)) of QUS measurements at the calcaneus remained highly significant (0.70 for SI, 0.72 for the QUI, and 0.67 for DXA at the lumbar spine) even after adjustment for other confounding variables. CONCLUSIONS: QUS of the calcaneus and bone mineral density measurements were shown to be significant predictors of incident vertebral fracture. The RRs for QUS measurements at the calcaneus are of similar magnitude as for DXA measurements.

Comparison of three bone ultrasounds for the discrimination of subjects with and without osteoporotic fractures among 7562 elderly women.

Bone ultrasound measures (QUSs) can assess fracture risk in the elderly. We compared three QUSs and their association with nonvertebral fracture history in 7562 Swiss women 70-80 years of age. The association between nonvertebral fracture was higher for heel than phalangeal QUS. INTRODUCTION: Because of the high morbidity and mortality associated with osteoporotic fractures, it is essential to detect subjects at risk for such fractures with screening methods. Because quantitative bone ultrasound (QUS) discriminated subjects with osteoporotic fractures from controls in several cross-sectional studies and predicted fractures in prospective studies, QUS could be more practical than DXA for screening. MATERIAL AND METHODS: This cross-sectional and retrospective multicenter (10 centers) study was performed to compare three QUSs (two heel ultrasounds: Achilles+ [GE-Lunar] and Sahara [Hologic]; the phalanges: ultrasound DBM sonic 1200 [IGEA]) for determining by logistic regression nonvertebral fracture odds ratio (OR) in a sample of 7562 Swiss women, 75.3 +/- 3.1 years of age. The two heel QUSs measured the broadband ultrasound attenuation (BUA) and the speed of sound (SOS). In addition, Achilles+ calculated the stiffness index (SI) and the Sahara calculated the quantitative ultrasound index (QUI) from BUA and SOS. The DBM sonic 1200 measured the amplitude-dependent SOS (AD-SOS). RESULTS: Eighty-six women had a history of a traumatic hip fracture after the age of 50, 1594 had a history of forearm fracture, and 2016 had other nonvertebral fractures. No fracture history was reported by 3866 women. Discrimination for hip fracture was higher than for the other nonvertebral fractures. The two heel QUSs had a significantly higher discrimination power than the QUSs of the phalanges, with standardized ORs, adjusted for age and body mass index, ranging from 2.1 to 2.7 (95% CI = 1.6, 3.5) compared with 1.4 (95% CI = 1.1, 1.7) for the AD-SOS of DBM sonic 1200. CONCLUSION: This study showed a high association between heel QUS and hip fracture history in elderly Swiss women. This could justify integration of QUS among screening strategies for identifying elderly women at risk for osteoporotic fractures.

Obesity and overweight associated with lower rates of colorectal cancer screening in Switzerland.

Screening for colorectal cancer (CRC) is associated with reduced CRC mortality, but low screening rates have been reported in several settings. The aim of the study was to assess predictors of low CRC screening in Switzerland. A retrospective cohort of a random sample of 940 patients aged 50-80 years followed for 2 years from four Swiss University primary care settings was used. Patients with illegal residency status and a history of CRC or colorectal polyps were excluded. We abstracted sociodemographic data of patients and physicians, patient health status, and indicators derived from RAND's Quality Assessment Tools from medical charts. We defined CRC screening as colonoscopy in the last 10 years, flexible sigmoidoscopy in the last 5 years, or fecal occult blood testing in the last 2 years. We used bivariate and multivariate logistic regression analyses. Of 940 patients (mean age 63.9 years, 42.7% women), 316 (33.6%) had undergone CRC screening. In multivariate analysis, birthplace in a country outside of Western Europe and North America [odds ratio (OR) 0.65, 95% confidence interval (CI) 0.45-0.97], male sex of the physician in charge (OR 0.67, 95% CI 0.50-0.91), BMI 25.0-29.9 kg/m (OR 0.66, CI 0.46-0.96) and at least 30.0 kg/m (OR 0.61, CI 0.40-0.90) were associated with lower CRC screening rates. Obesity, overweight, birthplace outside of Western Europe and North America, and male sex of the physician in charge were associated with lower CRC screening rates in Swiss University primary care settings. Physician perception of obesity and its impact on their recommendation for CRC screening might be a target for further research.

Individual patient data systematic review and meta-analysis of optic nerve sheath diameter ultrasonography for detecting raised intracranial pressure: protocol of the ONSD research group.

BACKGROUND: The purpose of the optic nerve sheath diameter (ONSD) research group project is to establish an individual patient-level database from high quality studies of ONSD ultrasonography for the detection of raised intracranial pressure (ICP), and to perform a systematic review and an individual patient data meta-analysis (IPDMA), which will provide a cutoff value to help physicians making decisions and encourage further research. Previous meta-analyses were able to assess the diagnostic accuracy of ONSD ultrasonography in detecting raised ICP but failed to determine a precise cutoff value. Thus, the ONSD research group was founded to synthesize data from several recent studies on the subject and to provide evidence on the diagnostic accuracy of ONSD ultrasonography in detecting raised ICP. METHODS: This IPDMA will be conducted in different phases. First, we will systematically search for eligible studies. To be eligible, studies must have compared ONSD ultrasonography to invasive intracranial devices, the current reference standard for diagnosing raised ICP. Subsequently, we will assess the quality of studies included based on the QUADAS-2 tool, and then collect and validate individual patient data. The objectives of the primary analyses will be to assess the diagnostic accuracy of ONSD ultrasonography and to determine a precise cutoff value for detecting raised ICP. Secondly, we will construct a logistic regression model to assess whether patient and study characteristics influence diagnostic accuracy. DISCUSSION: We believe that this IPD MA will provide the most reliable basis for the assessment of diagnostic accuracy of ONSD ultrasonography for detecting raised ICP and to provide a cutoff value. We also hope that the creation of the ONSD research group will encourage further study. TRIAL REGISTRATION: PROSPERO registration number: CRD42012003072.

Epidemiology and outcome of fungemia in a cancer Cohort of the Infectious Diseases Group (IDG) of the European Organization for Research and Treatment of Cancer (EORTC 65031).

BACKGROUND: Anti-cancer treatment and the cancer population have evolved since the last European Organisation for Research and Treatment of Cancer (EORTC) fungemia survey, and there are few recent large epidemiological studies. METHODS: This was a prospective cohort study including 145 030 admissions of patients with cancer from 13 EORTC centers. Incidence, clinical characteristics, and outcome of fungemia were analyzed. RESULTS: Fungemia occurred in 333 (0.23%; 95% confidence interval [CI], .21-.26) patients, ranging from 0.15% in patients with solid tumors to 1.55% in hematopoietic stem cell transplantation recipients. In 297 evaluable patients age ranged from 17 to 88 years (median 56 years), 144 (48%) patients were female, 165 (56%) had solid tumors, and 140 (47%) had hematological malignancies. Fungemia including polymicrobial infection was due to: Candida spp. in 267 (90%), C. albicans in 128 (48%), and other Candida spp. in 145 (54%) patients. Favorable overall response was achieved in 113 (46.5%) patients by week 2. After 4 weeks, the survival rate was 64% (95% CI, 59%-70%) and was not significantly different between Candida spp. Multivariable logistic regression identified baseline septic shock (odds ratio [OR] 3.04, 95% CI, 1.22-7.58) and tachypnoea as poor prognostic factors (OR 2.95, 95% CI, 1.66-5.24), while antifungal prophylaxis prior to fungemia (OR 0.20, 95% CI, .06-.62) and remission of underlying cancer (OR, 0.18; 95% CI, .06-.50) were protective. CONCLUSIONS: Fungemia, mostly due to Candida spp., was rare in cancer patients from EORTC centers but was associated with substantial mortality. Antifungal prophylaxis and remission of cancer predicted better survival.

Characterizing the vulnerability of frequent emergency department users by applying a conceptual framework: a controlled, cross-sectional study.

BACKGROUND: Frequent emergency department (ED) users meet several of the criteria of vulnerability, but this needs to be further examined taking into consideration all vulnerability's different dimensions. This study aimed to characterize frequent ED users and to define risk factors of frequent ED use within a universal health care coverage system, applying a conceptual framework of vulnerability. METHODS: A controlled, cross-sectional study comparing frequent ED users to a control group of non-frequent users was conducted at the Lausanne University Hospital, Switzerland. Frequent users were defined as patients with five or more visits to the ED in the previous 12 months. The two groups were compared using validated scales for each one of the five dimensions of an innovative conceptual framework: socio-demographic characteristics; somatic, mental, and risk-behavior indicators; and use of health care services. Independent t-tests, Wilcoxon rank-sum tests, Pearson's Chi-squared test and Fisher's exact test were used for the comparison. To examine the -related to vulnerability- risk factors for being a frequent ED user, univariate and multivariate logistic regression models were used. RESULTS: We compared 226 frequent users and 173 controls. Frequent users had more vulnerabilities in all five dimensions of the conceptual framework. They were younger, and more often immigrants from low/middle-income countries or unemployed, had more somatic and psychiatric comorbidities, were more often tobacco users, and had more primary care physician (PCP) visits. The most significant frequent ED use risk factors were a history of more than three hospital admissions in the previous 12 months (adj OR:23.2, 95%CI = 9.1-59.2), the absence of a PCP (adj OR:8.4, 95%CI = 2.1-32.7), living less than 5 km from an ED (adj OR:4.4, 95%CI = 2.1-9.0), and household income lower than USD 2,800/month (adj OR:4.3, 95%CI = 2.0-9.2). CONCLUSIONS: Frequent ED users within a universal health coverage system form a highly vulnerable population, when taking into account all five dimensions of a conceptual framework of vulnerability. The predictive factors identified could be useful in the early detection of future frequent users, in order to address their specific needs and decrease vulnerability, a key priority for health care policy makers. Application of the conceptual framework in future research is warranted.

Outcome and Experience from the Swiss Mammography Screening Pilot Programmes

BACKGROUND:The Swiss breast cancer screening pilot programme was conducted in 3 districts of theFrench-speaking canton of Vaud (ca. 300,000 resident women) between October 1993 and January 1999.Women aged 50 to 69 were invited by mail every 2 years for a free of charge screening mammography (doubleview, multiple reading). This first ever-organised cancer screening programme in Switzerland showed thefeasibility and acceptability of this kind of public health intervention in the liberal Swiss healthcare system, whichwas the main objective of the pilot programme. This mammographic screening programme was extended to thewhole canton in 1999, and contributed to the implementation of similar programmes in 2 neighbouring cantons. OBJECTIVE:To appraise the use, the quality and the effectiveness of the Swiss screening pilot programme. METHODS:About 15,000 women (aged 50-69) were enrolled. Logistic regression analyses were performedseparately to identify determinants of initial and subsequent attendance. Standard indicators of quality,effectiveness and impact of the programme were assessed and compared with European recommendations. Tothis intent, linkage with data from the Vaud Cancer Registry was performed. RESULTS:About half the target population was screened at least once during the pilot trial. Participation washigher among Swiss than foreigners, among widowed or married women than among single, divorced or separatedones. Attendance also increased with age and decreasing distance between residence and the dedicatedscreening centre. Apart from Swiss citizenship, socio-demographic factors were not associated with reattendance.Intensity of prior recruitment, outcome of previous screening test (positive vs. negative) and indicators of women'shealth behaviour (time of last mammography prior to initial screen, smoking status) were the main determinants ofreattendance. Programme performance and quality indicators were, overall, in line with European Guidelines. Theywere overall more favourable among 60-69 than 50-59 year-olds and improved over time. CONCLUSION:The objectives of the pilot programme were met. Even if participation should increase in order toreach European standards, performance indicators overall met quality requirements. Ways to improve screeninguse, quality and effectiveness were devised and taken into account for the generalisation of the programme.

Life partnerships in childhood cancer survivors, their siblings, and the general population.

BACKGROUND: Life partnerships other than marriage are rarely studied in childhood cancer survivors (CCS). We aimed (1) to describe life partnership and marriage in CCS and compare them to life partnerships in siblings and the general population; and (2) to identify socio-demographic and cancer-related factors associated with life partnership and marriage. METHODS: As part of the Swiss Childhood Cancer Survivor Study (SCCSS), a questionnaire was sent to all CCS (aged 20-40 years) registered in the Swiss Childhood Cancer Registry (SCCR), aged <16 years at diagnosis, who had survived ≥ 5 years. The proportion with life partner or married was compared between CSS and siblings and participants in the Swiss Health Survey (SHS). Multivariable logistic regression was used to identify factors associated with life partnership or marriage. RESULTS: We included 1,096 CCS of the SCCSS, 500 siblings and 5,593 participants of the SHS. Fewer CCS (47%) than siblings (61%, P < 0.001) had life partners, and fewer CCS were married (16%) than among the SHS population (26%, P > 0.001). Older (OR = 1.14, P < 0.001) and female CCS (OR = 1.85, <0.001) were more likely to have life partners. CCS who had undergone radiotherapy, bone marrow transplants (global P Treatment = 0.018) or who had a CNS diagnosis (global P Diagnosis < 0.001) were less likely to have life partners. CONCLUSION: CCS are less likely to have life partners than their peers. Most CCS with a life partner were not married. Future research should focus on the effect of these disparities on the quality of life of CCS.

Patient and physician gender concordance in preventive care in university primary care settings.

OBJECTIVE: To assess the quality of preventive care according to physician and patient gender in a country with universal health care coverage. METHODS: We assessed a retrospective cohort study of 1001 randomly selected patients aged 50-80years followed over 2years (2005-2006) in 4 Swiss university primary care settings (Basel, Geneva, Lausanne, Zürich). We used indicators derived from RAND's Quality Assessment Tools and examined percentages of recommended preventive care. Results were adjusted using hierarchical multivariate logistic regression models. RESULTS: 1001 patients (44% women) were followed by 189 physicians (52% women). Female patients received less preventive care than male patients (65.2% vs. 72.1%, p<0.001). Female physicians provided significantly more preventive care than male physicians (p=0.01) to both female (66.7% vs. 63.6%) and male patients (73.4% vs. 70.7%). After multivariate adjustment, differences according to physician (p=0.02) and patient gender (p<0.001) remained statistically significant. Female physicians provided more recommended cancer screening than male physicians (78.4 vs. 71.9%, p=0.01). CONCLUSIONS: In Swiss university primary care settings, female patients receive less preventive care than male patients, with female physicians providing more preventive care than male physicians. Greater attention should be paid to female patients in preventive care and to why female physicians tend to provide better preventive care.

Respiratory symptoms in preterm infants: burden of disease in the first year of life.

OBJECTIVE: While respiratory symptoms in the first year of life are relatively well described for term infants, data for preterm infants are scarce. We aimed to describe the burden of respiratory disease in a group of preterm infants with and without bronchopulmonary dysplasia (BPD) and to assess the association of respiratory symptoms with perinatal, genetic and environmental risk factors. METHODS: Single centre birth cohort study: prospective recording of perinatal risk factors and retrospective assessment of respiratory symptoms during the first year of life by standardised questionnaires. MAIN OUTCOME MEASURES: Cough and wheeze (common symptoms), re-hospitalisation and need for inhalation therapy (severe outcomes). PATIENTS: 126 preterms (median gestational age 28.7 weeks; 78 with, 48 without BPD) hospitalised at the University Children's Hospital of Bern, Switzerland 1999-2006. RESULTS: Cough occurred in 80%, wheeze in 44%, re-hospitalisation in 25% and long term inhalation therapy in wheezers in 13% of the preterm infants. Using logistic regression, the main risk factor for common symptoms was frequent contact with other children. Severe outcomes were associated with maximal peak inspiratory pressure, arterial cord blood pH, APGAR- and CRIB-Score. CONCLUSIONS: Cough in preterm infants is as common as in term infants, whereas wheeze, inhalation therapy and re-hospitalisations occur more often. Severe outcomes are associated with perinatal risk factors. Preterm infants who did not qualify for BPD according to latest guidelines also showed a significant burden of respiratory disease in the first year of life.

Determinants of generic drug substitution in Switzerland.

Background: Since generic drugs have the same therapeutic effect as the original formulation but at generally lower costs, their use should be more heavily promoted. However, a considerable number of barriers to their wider use have been observed in many countries. The present study examines the influence of patients, physicians and certain characteristics of the generics' market on generic substitution in Switzerland.Methods: We used reimbursement claims' data submitted to a large health insurer by insured individuals living in one of Switzerland's three linguistic regions during 2003. All dispensed drugs studied here were substitutable. The outcome (use of a generic or not) was modelled by logistic regression, adjusted for patients' characteristics (gender, age, treatment complexity, substitution groups) and with several variables describing reimbursement incentives (deductible, co-payments) and the generics' market (prices, packaging, co-branded original, number of available generics, etc.).Results: The overall generics' substitution rate for 173,212 dispensed prescriptions was 31%, though this varied considerably across cantons. Poor health status (older patients, complex treatments) was associated with lower generic use. Higher rates were associated with higher out-of-pocket costs, greater price differences between the original and the generic, and with the number of generics on the market, while reformulation and repackaging were associated with lower rates. The substitution rate was 13% lower among hospital physicians. The adoption of the prescribing practices of the canton with the highest substitution rate would increase substitution in other cantons to as much as 26%.Conclusions: Patient health status explained a part of the reluctance to substitute an original formulation by a generic. Economic incentives were efficient, but with a moderate global effect. The huge interregional differences indicated that prescribing behaviours and beliefs are probably the main determinant of generic substitution.

Health-risk behaviors and quality of life among young men.

PURPOSE: To examine the associations between substance use and other health-risk behaviors and quality of life (QOL) among young men. METHODS: The analytical sample consisted of 5,306 young Swiss men who participated in the Cohort Study on Substance Use Risk Factors. Associations between seven distinct self-reported health-risk behaviors (risky single-occasion drinking; volume drinking; cigarette smoking; cannabis use; use of any other illicit drugs; sexual intercourse without a condom; low physical activity) were assessed via chi-square analysis. Logistic regression analyses were conducted to study the associations between each particular health-risk behavior and either physical or mental QOL (assessed with the SF-12v2) while adjusting for socio-demographic variables and the presence of all other health-risk behaviors. RESULTS: Most health-risk behaviors co-occurred. However, low physical activity was not or negatively related to other health-risk behaviors. Almost all health-risk behaviors were associated with a greater likelihood of compromised QOL. However, sexual intercourse without a condom (not associated with both physical and mental QOL) and frequent risky single-occasion drinking (not related to mental QOL after adjusting for the presence of other health-risk behaviors; positively associated with physical QOL) differed from this pattern. CONCLUSIONS: Health-risk behaviors are mostly associated with compromised QOL. However, sexual intercourse without a condom and frequent risky single-occasion drinking differ from this pattern and are therefore possibly particularly difficult to change relative to other health-risk behaviors.