Statistical and Substantive Inferences in Public Health: Issues in the Application of Multilevel Models

Used economy market insight: Sailboat industry pricing mechanism and regional effects

With the popularity of circular economy around the world, transactions in the second-hand sailboat market are extremely active. Determining pricing strategies and exploring their regional effects is a blank area of existing research and has important practical and statistical significance. Therefore, this article uses the random forest model and XGBoost algorithm to identify core price indicators, and uses an innovative rolling NAR dynamic neural network model to simulate and predict second-hand sailboat price data. On this basis, we also constructed a regional effect multi-level model (RE-MLM) from three levels: geography, economy and country to clarify the impact of geographical areas on sailboat prices. The research results show that, first of all, the price of second-hand sailboats fluctuates greatly, and the predicted value better reflects the overall average price level. Secondly, there are significant regional differences in price levels across regions, economies and ethnic groups. Therefore, the price of second-hand sailboats is affected by many factors and has obvious regional effects. In addition, the model evaluation results show that the model constructed in this study has good accuracy, validity, portability and versatility, and can be extended to price simulation and regional analysis of different markets in different regions.

Ebbing Strength, Fading Power: Unveiling the Impact of Persistent Fatigue on Muscle Performance in COVID-19 Survivors

The total number of confirmed cases of COVID-19 caused by SARS-CoV-2 virus infection is over 621 million. Post-COVID-19 syndrome, also known as long COVID or long-haul COVID, refers to a persistent condition where individuals experience symptoms and health issues after the acute phase of COVID-19. The aim of this study was to assess the strength and fatigue of skeletal muscles in people recovered from COVID-19. A total of 94 individuals took part in this cross-sectional study, with 45 participants (referred to as the Post-COVID Cohort, PCC) and 49 healthy age-matched volunteers (Healthy Control Cohort, HCC). This research article uses the direct dynamometry method to provide a detailed analysis of post-COVID survivors' strength and power characteristics. The Biodex System 4 Pro was utilized to evaluate muscle strength characteristics during the fatigue test. The fatigue work in extensors and flexors was significantly higher in the PCC. The PCC also showed significantly less power in both extensors and flexors compared to the HCC. In conclusion, this study provides compelling evidence of the impact of post-COVID-19 fatigue on muscle performance, highlighting the importance of considering these effects in the rehabilitation and care of individuals recovering from the virus. PCC achieved lower muscle strength values than HCC.

Differences in Covid-19 mortality among persons 70 years and older in an integrated care setting in region Stockholm: a multi-level analysis between March 2020-February 2021

BACKGROUND

In Norrtälje municipality, within Region Stockholm, there is a joint integrated care organisation providing health and social care, which may have facilitated a more coordinated response to the covid-19 pandemic compared to the otherwise decentralised Swedish system. This study compares the risk of covid-19 mortality among persons 70 years and older, in the municipalities of Stockholm, Södertälje, and Norrtälje, while considering area and individual risk factors.

METHODS

A population-based study using linked register data to examine covid-19 mortality among those 70 + years (N = 127,575) within the municipalities of interest between the periods March-August 2020 and September 2020-February 2021. The effect of individual and area level variables on covid-19 mortality among inhabitants in 68 catchment areas were examined using multi-level logistic models.

RESULTS

Individual factors associated with covid-19 mortality were sex, older age, primary education, country of birth and poorer health as indicated by the Charlson Co-morbidity Index. The area-level variables associated were high deprivation (OR: 1.56, CI: 1.18-2.08), population density (OR: 1.14, CI: 1.08-1.21), and usual care. Together, this explained 85.7% of the variation between catchment areas in period 1 and most variation was due to individual risk factors in period 2. Little of the residual variation was attributed to differences between catchment areas.

CONCLUSION

Integrated care in Norrtälje may have facilitated a more coordinated response during period 1, compared to municipalities with usual care. In the future, integrated care should be considered as an approach to better protect and meet the care needs of older people during emergency situations.

Association between cardiopulmonary resuscitation audit results with in-situ simulation and in-hospital cardiac arrest outcomes and key performance indicators

INTRODUCTION

In-situ simulation (ISS) is a method to evaluate the performance of hospital units in performing cardiopulmonary resuscitation (CPR). It is conducted by placing a high-fidelity mannequin at hospital units with simulated scenarios and having each unit's performance evaluated. However, little is known about its impact on actual patient outcomes. Therefore, we aimed to evaluate the association between the ISS results and actual outcomes of patients with in-hospital cardiac arrest (IHCA).

METHODS

This retrospective study was conducted by reviewing Siriraj Hospital's CPR ISS results in association with the data of IHCA patients between January 2012 and January 2019. Actual outcomes were determined by patients' outcomes (sustained return of spontaneous circulation (ROSC) and survival to hospital discharge) and arrest performance indicators (time-to-first-epinephrine and time-to-defibrillation). These outcomes were investigated for association with the ISS scores in multilevel regression models with hospital units as clusters.

RESULTS

There were 2146 cardiac arrests included with sustained ROSC rate of 65.3% and survival to hospital discharge rate of 12.9%. Higher ISS scores were significantly associated with improved sustained ROSC rate (adjusted odds ratio 1.32 (95%CI 1.04, 1.67); p = 0.01) and a decrease in time-to-defibrillation (-0.42 (95%CI -0.73, -0.11); p = 0.009). Although higher scores were also associated with better survival to hospital discharge and a decrease in time-to-first-epinephrine, most models for these outcomes failed to reach statistical significance.

CONCLUSION

CPR ISS results were associated with some important patient outcomes and arrest performance indicators. Therefore, it may be an appropriate performance evaluation method that can guide the direction of improvement.

The Level of Health Culture Related to Heart Disease Among Students of the Faculty of Artificial Intelligence at Al-Balqa Applied University

Increasing awareness of any disease, especially heart disease, is crucial to improve health culture in general. The lack of communication between the different departments of social and health institutions may hinder such increasing awareness due to the lack of enough research that highlights this problem. As health culture education related to heart diseases raises young people's awareness, it improves their lives by developing their knowledge and changing their attitudes, habits, and behaviours towards the risk factors related to such diseases. Therefore, this research aimed to determine the level of health culture related to heart disease among students at Al-Balqa Applied University. The descriptive approach was used in its analytical and survey styles to achieve the research objective, and the research sample consisted of 221 male and female students. The results showed that the total score for the level of health culture related to heart disease among the students was average. In light of the results, the researcher presented several recommendations. The most important are holding health education and awareness seminars and workshops for university students in the field of heart disease and its prevention and activating the role of Al-Balqa Applied University in the continuous guidance and counselling of students in all disciplines and levels to increase the health culture of heart disease.

Trust in government regarding COVID-19 and its associations with preventive health behaviour and prosocial behaviour during the pandemic: a cross-sectional and longitudinal study

BACKGROUND

The effective implementation of government policies and measures for controlling the coronavirus disease 2019 (COVID-19) pandemic requires compliance from the public. This study aimed to examine cross-sectional and longitudinal associations of trust in government regarding COVID-19 control with the adoption of recommended health behaviours and prosocial behaviours, and potential determinants of trust in government during the pandemic.

METHODS

This study analysed data from the PsyCorona Survey, an international project on COVID-19 that included 23 733 participants from 23 countries (representative in age and gender distributions by country) at baseline survey and 7785 participants who also completed follow-up surveys. Specification curve analysis was used to examine concurrent associations between trust in government and self-reported behaviours. We further used structural equation model to explore potential determinants of trust in government. Multilevel linear regressions were used to examine associations between baseline trust and longitudinal behavioural changes.

RESULTS

Higher trust in government regarding COVID-19 control was significantly associated with higher adoption of health behaviours (handwashing, avoiding crowded space, self-quarantine) and prosocial behaviours in specification curve analyses (median standardised β = 0.173 and 0.229, p < 0.001). Government perceived as well organised, disseminating clear messages and knowledge on COVID-19, and perceived fairness were positively associated with trust in government (standardised β = 0.358, 0.230, 0.056, and 0.249, p < 0.01). Higher trust at baseline survey was significantly associated with lower rate of decline in health behaviours over time (p for interaction = 0.001).

CONCLUSIONS

These results highlighted the importance of trust in government in the control of COVID-19.

Trust in government regarding COVID-19 and its associations with preventive health behaviour and prosocial behaviour during the pandemic: a cross-sectional and longitudinal study

BACKGROUND

The effective implementation of government policies and measures for controlling the coronavirus disease 2019 (COVID-19) pandemic requires compliance from the public. This study aimed to examine cross-sectional and longitudinal associations of trust in government regarding COVID-19 control with the adoption of recommended health behaviours and prosocial behaviours, and potential determinants of trust in government during the pandemic.

METHODS

This study analysed data from the PsyCorona Survey, an international project on COVID-19 that included 23 733 participants from 23 countries (representative in age and gender distributions by country) at baseline survey and 7785 participants who also completed follow-up surveys. Specification curve analysis was used to examine concurrent associations between trust in government and self-reported behaviours. We further used structural equation model to explore potential determinants of trust in government. Multilevel linear regressions were used to examine associations between baseline trust and longitudinal behavioural changes.

RESULTS

Higher trust in government regarding COVID-19 control was significantly associated with higher adoption of health behaviours (handwashing, avoiding crowded space, self-quarantine) and prosocial behaviours in specification curve analyses (median standardised β = 0.173 and 0.229, p < 0.001). Government perceived as well organised, disseminating clear messages and knowledge on COVID-19, and perceived fairness were positively associated with trust in government (standardised β = 0.358, 0.230, 0.056, and 0.249, p < 0.01). Higher trust at baseline survey was significantly associated with lower rate of decline in health behaviours over time (p for interaction = 0.001).

CONCLUSIONS

These results highlighted the importance of trust in government in the control of COVID-19.

Physician influence on medication adherence, evidence from a population-based cohort

BACKGROUND

The overall impact of physician prescribers on population-level adherence rates are unknown. We aimed to quantify the influence of general practitioner (GP) physician prescribers on the outcome of optimal statin medication adherence.

METHODS

We conducted a retrospective cohort study using health administrative databases from Saskatchewan, Canada. Participants included physician prescribers and their patients beginning a new statin medication between January 1, 2012 and December 31, 2017. We grouped prescribers based on the prevalence of optimal adherence (i.e., proportion of days covered ≥ 80%) within their patient group. Also, we constructed multivariable logistic regression analyses on optimal statin adherence using two-level non-linear mixed-effects models containing patient and prescriber-level characteristics. An intraclass correlation coefficient was used to estimate the physician effect.

RESULTS

We identified 1,562 GPs prescribing to 51,874 new statin users. The median percentage of optimal statin adherence across GPs was 52.4% (inter-quartile range: 35.7% to 65.5%). GP prescribers with the highest patient adherence (versus the lowest) had patients who were older (median age 61.0 vs 55.0, p<0.0001) and sicker (prior hospitalization 39.4% vs 16.4%, p<0.001). After accounting for patient-level factors, only 6.4% of the observed variance in optimal adherence between patients could be attributed to GP prescribers (p<0.001). The majority of GP prescriber influence (5.2% out of 6.4%) was attributed to the variance unexplained by patient and prescriber variables.

INTERPRETATION

The overall impact of GP prescribers on statin adherence appears to be very limited. Even "high-performing" physicians face significant levels of sub-optimal adherence among their patients.

Multilevel analysis of factors associated with perinatal intimate partner violence among postpartum population in Southern Ethiopia

Violence around pregnancy is critical in nature and major public health problem worldwide. Thus, the present study aims to determine the extent of perinatal partner violence and to identify its individual and community-level factors among postpartum women in Southern Ethiopia. A total of 1342 postpartum women nested in 38 'Kebles' (clusters) were enumerated using multistage-clustered sampling techniques for multilevel analysis. Different parameters were computed for model comparison and model fitness. The overall prevalence of intimate partner violence before, during, and/or after pregnancy was estimated to be 39.9% [95% CI 36.9-44.5]. About 18% of women reported continuous abuse over the perinatal period. Postpartum women who live in rural areas [adjusted odds ratio (AOR) = 2.46; 95% CI 1.21-5.01], or in neighborhoods with high IPV favoring norms [AOR = 1.49; 95%CI 1.01-2.20], high female literacy [AOR = 2.84; 95%CI 1.62-5.01], high female autonomy [AOR = 2.06; 95%CI 1.36-3.12], or in neighborhoods with lower wealth status [AOR = 1.74; 95%CI 1.14-2.66] were more likely to encounter PIPV. The complex patterns of interplaying factors operating at different levels could put pregnant or postpartum women at higher risk of IPV victimization. Therefore, policies that prioritize the improvement of contextual factors, particularly norms toward IPV and women's empowerment are likely to be the most effective interventions.

Impact of National Pandemic Lockdowns on Perceived Threat of Immigrants: A Natural Quasi-Experiment Across 23 Countries

Xenophobia and anti-immigrant attacks rose during the COVID-19 pandemic, yet this may not be solely due to the disease threat. According to theories of frustration and scapegoating, situational obstructions and deprivation can motivate prejudice against outgroups. Using a global natural quasi-experimental design, this study tests whether the restrictiveness of national lockdowns can explain higher individual-level perceptions of immigrant threat. Data of 45,894 participants from 23 countries were analyzed. Both lockdown duration and lockdown severity were positively associated with individuals’ perceived threat of immigrants. The lockdown effects were independent of objective and subjective measures of disease threat, and there was no evidence that disease threat drives people’s prejudice toward immigrants. Subgroup analysis suggested the lockdown effects were reliable in Europe and the Americas, but not in Asia. These findings suggest a need to mitigate frustration and scapegoating when implementing lockdowns, and to distinguish the influence of societal restrictions from disease threat.

Associations Among Select State Policies and the Nutritional Quality of Household Packaged Food Purchases in the United States from 2008 Through 2017

BACKGROUND

Policy interventions are important public health tools because they can reach large numbers of people. State context has been associated with health outcomes, yet few studies have examined the extent to which state-level policies are associated with dietary quality.

OBJECTIVES

The objective of this study was to evaluate whether state policies are associated with the nutritional quality of household packaged food purchases.

DESIGN

This observational study used data from Nielsen Homescan, an open-cohort household panel where participants track purchases, and a combination of state-level food and social safety net policy variables from 2008 through 2017.

PARTICIPANTS AND SETTING

This study included 615,634 household-year observations in the United States from 2008 through 2017. Household-year observations were excluded in the case that a household did not make a minimum number of purchases and in the case that they had incorrect geographic information. The final analytic sample was 611,719 household-years.

MAIN OUTCOME MEASURES

Study outcomes included a set of nutrition-related measures of public health interest, including nutrients of concern (eg, sugar, saturated fat, and sodium) and calories from specific food groups (eg, fruits, nonstarchy vegetables, processed meats, mixed dishes, sugar-sweetened beverages, and desserts and snacks).

STATISTICAL ANALYSIS

This study used multilevel generalized linear models with state fixed effects on three samples: all households, only households with low income, and only households with low educational attainment.

RESULTS

Few significant associations were found between healthy food retail policies and the nutritional quality of purchases, and mixed associations were found between social safety net policies and lower or higher quality packaged food purchases.

CONCLUSIONS

Little evidence was found that state policy context in 2008 through 2017 was associated with the quality of packaged food purchases. However, variation in state policies is increasing over time, warranting future research into the relationship between these policies, the quality of packaged food purchases, and the rest of the diet.

Secularism, family ties and loneliness: A multilevel longitudinal study of ten European societies

OBJECTIVES

Mass media suggest rising political and religious concern about secularism-induced decline of the family. Implications for loneliness remain unexamined. The current study filled this gap.

METHODS

Data were from 10 national probability samples in the Survey of Health, Ageing and Retirement in Europe. Multilevel longitudinal models tested linkages of societal secularism with loneliness, their mediation by specific family relationships, and the role of this cultural dimension in weakening associations of family ties with loneliness. Both weighted Maximum Likelihood and unweighted Bayesian analyses were conducted, separately for each gender.

RESULTS

Societal secularism was not positively linked to either gender's loneliness. Associations with family ties were inconsistent, with only men's average partnered status lower in more secular settings. Nor did any positive indirect effects emerge. Moderation results were also inconsistent, with secularism only weakening linkages of some family dimensions with loneliness. Bayesian estimates were generally nonsignificant.

DISCUSSION

Societal secularism may not be a risk factor for loneliness or for weak family ties. Results stand at odds with religious and political rhetoric on secularism-induced decline of the family, and its individual and societal consequences.

Interrogating structural inequalities in COVID-19 mortality in England and Wales

BACKGROUND

Numerous observational studies have highlighted structural inequalities in COVID-19 mortality in the UK. Such studies often fail to consider the hierarchical, spatial nature of such inequalities in their analysis, leading to the potential for bias and an inability to reach conclusions about the most appropriate structural levels for policy intervention.

METHODS

We use publicly available population data on COVID-19-related mortality and all-cause mortality between March and July 2020 in England and Wales to investigate the spatial scale of such inequalities. We propose a multiscale approach to simultaneously consider three spatial scales at which processes driving inequality may act and apportion inequality between these.

RESULTS

Adjusting for population age structure and number of local care homes we find highest regional inequality in March and June/July. We find finer grained within region inequality increased steadily from March until July. The importance of spatial context increases over the study period. No analogous pattern is visible for non-COVID-19 mortality. Higher relative deprivation is associated with increased COVID-19 mortality at all stages of the pandemic but does not explain structural inequalities.

CONCLUSIONS

Results support initial stochastic viral introduction in the South, with initially high inequality decreasing before the establishment of regional trends by June and July, prior to reported regionality of the 'second-wave'. We outline how this framework can help identify structural factors driving such processes, and offer suggestions for a long-term, locally targeted model of pandemic relief in tandem with regional support to buffer the social context of the area.

Interrogating structural inequalities in COVID-19 mortality in England and Wales

BACKGROUND

Numerous observational studies have highlighted structural inequalities in COVID-19 mortality in the UK. Such studies often fail to consider the hierarchical, spatial nature of such inequalities in their analysis, leading to the potential for bias and an inability to reach conclusions about the most appropriate structural levels for policy intervention.

METHODS

We use publicly available population data on COVID-19-related mortality and all-cause mortality between March and July 2020 in England and Wales to investigate the spatial scale of such inequalities. We propose a multiscale approach to simultaneously consider three spatial scales at which processes driving inequality may act and apportion inequality between these.

RESULTS

Adjusting for population age structure and number of local care homes we find highest regional inequality in March and June/July. We find finer grained within region inequality increased steadily from March until July. The importance of spatial context increases over the study period. No analogous pattern is visible for non-COVID-19 mortality. Higher relative deprivation is associated with increased COVID-19 mortality at all stages of the pandemic but does not explain structural inequalities.

CONCLUSIONS

Results support initial stochastic viral introduction in the South, with initially high inequality decreasing before the establishment of regional trends by June and July, prior to reported regionality of the 'second-wave'. We outline how this framework can help identify structural factors driving such processes, and offer suggestions for a long-term, locally targeted model of pandemic relief in tandem with regional support to buffer the social context of the area.

Nutritional status and concomitant factors of stunting among pre-school children in Malda, India: A micro-level study using a multilevel approach

Background

Malnutrition was the main cause of death among children below 5 years in every state of India in 2017. Despite several flagship programmes and schemes implemented by the Government of India, the latest edition of the Global Nutrition Report 2018 addressed that India tops in the number of stunted children, which is a matter of concern. Thus, a micro-level study was designed to know the level of nutritional status and to study this by various disaggregate levels, as well as to examine the risk factors of stunting among pre-school children aged 36–59 months in Malda.

Method

A primary cross-sectional quantitative survey was conducted using structured questionnaires following a multi-stage, stratified simple random sampling procedure in 2018. A sum of 731 mothers with at least one eligible child aged 36–59 months were the study participants. Anthropometric measures of children were collected following the WHO child growth standard. Children were classified as stunted, wasted, and underweight if their HAZ, WHZ, and WAZ scores, respectively, were less than −2SD. The random intercept multilevel logistic regression model has been employed to estimate the effects of possible risk factors on childhood stunting.

Results

The prevalence of stunting in the study area is 40% among children aged 36–59 months, which is a very high prevalence as per the WHO’s cut-off values (≥40%) for public health significance. Results of the multilevel analysis revealed that preceding birth interval, low birth weight, duration of breastfeeding, mother’s age at birth, mother’s education, and occupation are the associated risk factors of stunting. Among them, low birth weight (OR 2.22, 95% CI: 1.44–3.41) and bidi worker as mothers’ occupation (OR 1.92, 95% CI: 1.18–3.12) are the most influencing factors of stunting. Further, about 14 and 86% variation in stunting lie at community and child/household level, respectively.

Conclusion

Special attention needs to be placed on the modifiable risk factors of childhood stunting. Policy interventions should direct community health workers to encourage women as well as their male partners to increase birth interval using various family planning practices, provide extra care for low birth weight baby, that can help to reduce childhood stunting.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-11704-w.

Public satisfaction with health care system in 30 countries: The effects of individual characteristics and social contexts

In this article, the associations among individual socio-economic characteristics, the institutional set-up of health care systems, and satisfaction with the health care system are investigated. Data from the 2011 International Social Survey Program (30 countries, 34,212 respondents) is used. Multilevel analyses across countries have shown how the state financing context affects satisfaction at the individual level. Consistent with previous research, at individual level, personal experiences with medical providers, age, gender as well as income are significant predictors of satisfaction with the health care system. At the country level, real input indicators such as density of physicians and density of hospital beds are negative predictors of satisfaction with the health care system whereas the percentage of total health expenditures comprised by public sources is a positive predictor of satisfaction with the health care system. However, findings from the cross-level interactions indicate that the negative effect of lower income is more prominent in predominantly publicly-funded health care systems. Specifically, in primarily publicly-funded health care systems, the model-predicted probability of satisfaction with the health care system is higher, but the gap in the probability of satisfaction with the health care system between individuals with lower income and those with higher income is greater than that in mostly privately-financed health care systems. The findings in this study suggest that the future direction of health care system reform should be focused on balancing the distribution of resources between private and public sectors.

The Role of Education in the Transition towards Sustainable Agriculture: A Family Farm Learning Perspective

This paper deals with the analysis of decision-making processes at the family-farm level with reference to the transition towards sustainable agriculture. Despite literature that has underlined the relevance of education in strategic decision making, less attention has been devoted to the (family) collective decision-making process by taking into account the maximum level of education of the family members regardless of the position and whether they are the manager of the family farm or not. Therefore, this paper tries to fill a gap in literature by emphasizing the family farm’s collective decision-making process. In order to empirically measure this relevance, an econometric model was carried out that allowed us to evidence clear differences in the transition paths among various typologies of family farms on the basis of the level of education. Our results confirm the impact of education at the collective family level on transition towards more sustainable agricultural practices. This is particularly true in remote rural areas, where the transition is realized with higher intensity with respect to other territorial contexts. This brings about policy implications on enskilling farmers and upgrading their level of human capital.

Consequences of ignoring clustering in linear regression

Background

Clustering of observations is a common phenomenon in epidemiological and clinical research. Previous studies have highlighted the importance of using multilevel analysis to account for such clustering, but in practice, methods ignoring clustering are often employed. We used simulated data to explore the circumstances in which failure to account for clustering in linear regression could lead to importantly erroneous conclusions.

Methods

We simulated data following the random-intercept model specification under different scenarios of clustering of a continuous outcome and a single continuous or binary explanatory variable. We fitted random-intercept (RI) and ordinary least squares (OLS) models and compared effect estimates with the “true” value that had been used in simulation. We also assessed the relative precision of effect estimates, and explored the extent to which coverage by 95% confidence intervals and Type I error rates were appropriate.

Results

We found that effect estimates from both types of regression model were on average unbiased. However, deviations from the “true” value were greater when the outcome variable was more clustered. For a continuous explanatory variable, they tended also to be greater for the OLS than the RI model, and when the explanatory variable was less clustered. The precision of effect estimates from the OLS model was overestimated when the explanatory variable varied more between than within clusters, and was somewhat underestimated when the explanatory variable was less clustered. The cluster-unadjusted model gave poor coverage rates by 95% confidence intervals and high Type I error rates when the explanatory variable was continuous. With a binary explanatory variable, coverage rates by 95% confidence intervals and Type I error rates deviated from nominal values when the outcome variable was more clustered, but the direction of the deviation varied according to the overall prevalence of the explanatory variable, and the extent to which it was clustered.

Conclusions

In this study we identified circumstances in which application of an OLS regression model to clustered data is more likely to mislead statistical inference. The potential for error is greatest when the explanatory variable is continuous, and the outcome variable more clustered (intraclass correlation coefficient is ≥ 0.01).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-021-01333-7.

The Urine-culturing Cascade: Variation in Nursing Home Urine Culturing and Association With Antibiotic Use and Clostridiodes difficile Infection

BACKGROUND

Rates of antibiotic use vary widely across nursing homes and cannot be explained by resident characteristics. Antibiotic prescribing for a presumed urinary tract infection is often preceded by inappropriate urine culturing. We examined nursing home urine-culturing practices and their association with antibiotic use.

METHODS

We conducted a longitudinal, multilevel, retrospective cohort study based on quarterly nursing home assessments between April 2014 and January 2017 in 591 nursing homes and covering >90% of nursing home residents in Ontario, Canada. Nursing home urine culturing was measured as the proportion of residents with a urine culture in the prior 14 days. Outcomes included receipt of any systemic antibiotic and any urinary antibiotic (eg, nitrofurantoin, trimethoprim/sulfonamides, ciprofloxacin) in the 30 days after the assessment and Clostridiodes difficile infection in the 90 days after the assessment. Adjusted Poisson regression models accounted for 14 resident covariates.

RESULTS

A total of 131 218 residents in 591 nursing homes were included; 7.9% of resident assessments had a urine culture in the prior 14 days; this proportion was highly variable across the 591 nursing homes (10th percentile = 3.4%, 90th percentile = 14.3%). Before and after adjusting for 14 resident characteristics, nursing home urine culturing predicted total antibiotic use (adjusted risk ratio [RR] per doubling of urine culturing, 1.21; 95% confidence interval [CI], 1.18-1.23), urinary antibiotic use (RR, 1.33; 95% CI, 1.28-1.38), and C. difficile infection (incidence rate ratio, 1.18; 95% CI, 1.07-1.31).

CONCLUSIONS

Nursing homes have highly divergent urine culturing rates; this variability is associated with higher antibiotic use and rates of C. difficile infection.

Influence of program staff on quality of relationships in a community-based youth mentoring program

In many mentoring programs, mentor-youth pairs have the latitude to engage in a wide range of activities together across varying community settings. Within this context, program staff are tasked with supporting development of high-quality relationships between mentors and youth. To date, however, this role of program staff has been largely overlooked in research. The current study investigates potential contributions of program staff to mentoring relationships in the Big Brothers Big Sisters community-based mentoring program over their first 15 months of relationship development with a sample of 450 mentor-youth pairs that were supported by 76 program staff across 10 agencies. Two-level analyses (mentoring relationships nested within program staff) examined characteristics and approaches of program staff as prospective predictors of several facets of mentoring relationship quality as reported by youth: closeness, help with coping, youth-centeredness, growth orientation, and attachment. Staff-reported work engagement and emphasis on adherence to program guidelines as well as supervisor-rated staff competence predicted more favorable mentoring relationship quality. By contrast, a nondirective approach to supporting mentors, as reported by staff, predicted lower relationship quality. These findings suggest that further investigation of program staff influences on mentoring relationship development could be fruitful and ultimately provide a basis for enhancing program effectiveness.

The Use of Geographic Information Systems for Spatial Ecological Risk Assessments: An Example from the Athabasca Oil Sands Area in Canada

There is an acknowledged need in ecotoxicology for methods that integrate spatial analyses in risk assessment. This has resulted in the emergence of landscape ecotoxicology, a subdiscipline of ecotoxicology. However, landscape ecotoxicology has yet to become common practice in risk assessment due to the underdevelopment of techniques and a lack of standardized methods. In the present study, we demonstrate how geographic information systems (GISs) can serve as a standardized platform to integrate data, assess spatial patterns of ecotoxicological data for multiple species, and assess relationships between chemical mixture exposures and effects on biota for landscape ecotoxicological risks assessment. We use data collected under the Joint Oil Sands Monitoring Program in the Athabasca Oil Sands Region in Alberta, Canada. This dataset is composed of concentrations of contaminants including metals and polycyclic aromatic compounds, and health endpoints measured in 1100 biological samples, including tree swallows, amphibians, gull and tern eggs, plants, and mammals. We present 3 examples using a GIS as a platform and geospatial analysis to: 1) integrate data and assess spatial patterns of contaminant exposure in the region, 2) assess spatial patterns of exposures to complex mixtures, and 3) examine patterns of exposures and responses across the landscape. We summarize the methods used in the present study into a workflow for ease of use. The GIS methods allow researchers to identify hot spots of contamination, use georeferenced monitoring data to derive quantitative exposure-response relationships, and assess complex exposures with more realism. Environ Toxicol Chem 2019;38:2797-2810. © 2019 SETAC.

Network Alterations in Comorbid Chronic Pain and Opioid Addiction: An Exploratory Approach

The comorbidity of chronic pain and opioid addiction is a serious problem that has been growing with the practice of prescribing opioids for chronic pain. Neuroimaging research has shown that chronic pain and opioid dependence both affect brain structure and function, but this is the first study to evaluate the neurophysiological alterations in patients with comorbid chronic pain and addiction. Eighteen participants with chronic low back pain and opioid addiction were compared with eighteen age- and sex-matched healthy individuals in a pain-induction fMRI task. Unified structural equation modeling (SEM) with Lagrange multiplier (LM) testing yielded a network model of pain processing for patient and control groups based on 19 a priori defined regions. Tests of differences between groups on specific regression parameters were determined on a path-by-path basis using z-tests corrected for the number of comparisons. Patients with the chronic pain and addiction comorbidity had increased connection strengths; many of these connections were interhemispheric and spanned regions involved in sensory, affective, and cognitive processes. The affected regions included those that are commonly altered in chronic pain or addiction alone, indicating that this comorbidity manifests with neurological symptoms of both disorders. Understanding the neural mechanisms involved in the comorbidity is crucial to finding a comprehensive treatment, rather than treating the symptoms individually.

Characteristics of neighborhood environment (social cohesion and safety) and common mental disorders in ELSA-Brasil study: a multilevel analysis

The purpose of this study was to determine if self-reported characteristics of social cohesion and local neighborhood safety positively affect the mental health of their residents, regardless of individual characteristics. A sample of participants in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) baseline was used. The Clinical Interview Schedule-Revised (CIS-R) instrument was used for tracking common mental disorders (CMD). Social cohesion and safety were measured by validated scales of neighborhood environment self-reported characteristics. The multilevel logistic regression model was used to estimate the effect in neighborhoods (level 2) and individuals (level 1), as well as the odds ratios for each neighborhood explanatory variable and social characteristics in the CMD. The results showed that part of the variance (2.3%), in the common mental disorders prevalence is attributed to local neighborhoods. The characteristics of social cohesion and safety remained significant, even after the adjustment of individual explanatory variables. This study confirmed the hypothesis that individuals living in neighborhoods where they perceive low social cohesion and safety present a higher chance of developing CMD.

Focusing neighborhood context and self-rated health in the Pró-Saúde Study

The influence of neighborhood characteristics on self-rated health has been little studied. A multilevel approach using hierarchical models was applied to analyze the relationship between the socioeconomic characteristics in 621 neighborhoods (level 2) in the city of Rio de Janeiro, Brazil, and the self-rated health of 3,054 university employees (level 1) from the baseline of the Pró-Saúde Study. Neighborhoods were created using the SKATER algorithm (Spatial ‘K’luster Analysis by Tree Edge Removal) to cluster census tracts according to four indicators and a minimum population of 5,000 people. After adjustment for individual factors (per capita income, schooling, age, sex, ethnicity, health-related behavior and chronic diseases), low level of neighborhood income and higher numbers of members per household were significantly associated with poor self-rated health. Participants living in medium income-level neighborhoods were 34% more likely to self-rate their health as being poor. Those living in areas with a higher density of members per household were 50% more likely to present poor self-rated health. Neighborhood context influences self-rated health, beyond the effect of individual factors. Worsening neighborhood socioeconomic conditions affect health adversely, which in turn increasing the chance of poor self-rated health.

Estimating the Learning Curve of a Novel Medical Device: Bipolar Sealer Use in Unilateral Total Knee Arthroplasties

BACKGROUND

The use of cost-effectiveness analysis for medical devices has proven to be challenging because of the existence of the learning effects in the device-operator interactions. The need for the relevant analytical framework for assessing the economic value of such technologies has been recognized.

OBJECTIVES

To present a modified difference-in-differences (DID) cost-effectiveness methodology that facilitates visualization of a new health technology's learning curve.

METHODS

Using the Premier Perspective database (Premier Inc., Charlotte, NC), we examined the impact of physicians adopting a bipolar sealer (BPS) to control blood loss in primary unilateral total knee arthroplasties on hospital lengths of stay and total hospitalization costs when compared with two control groups. In our DID approach, we substituted month-from-adoption for the calendar-month-of-adoption in both graphical representations and ordinary least-squares regression results to estimate the effect of the BPS.

RESULTS

The results clearly demonstrated a learning curve associated with the adoption of the BPS technology. Although the reductions in length of stay were immediate, the first postadoption year costs increased by $1335 (extrahospital controls) to $1565 (within-hospital controls). Importantly, and also consistent with a learning curve hypothesis, these initial higher costs were offset by subsequent cost savings in the second and third years postadoption.

CONCLUSIONS

The presented modified DID approach is a suitable and versatile analytical tool for economic evaluation of a slowly diffusing medical device or health technology. It provides a better understanding of the potential learning effects associated with relevant interventions.

Neighborhood Social and Environmental Factors and Asthma Among Children Living in Low-Income Neighborhoods: The Importance of Informal Social Control

Despite the knowledge that children in low-income neighborhoods are particularly vulnerable to asthma, few studies of child asthma focus on variation among low-income neighborhoods. We examined the relationship between child asthma and features associated with neighborhood poverty including safety, social cohesion, informal social control, collective efficacy, and disorder, across a sample of children from low-income neighborhoods (N = 3010; 2005-2007). Results show that the relationship between asthma and poverty is accounted for by family-level characteristics, but informal social control remains significantly and positively related to asthma after accounting for family-level characteristics. We discuss the importance of neighborhood environmental features for children's asthma.

Clinic variation in glycaemic control for children with Type 1 diabetes in England and Wales: a population-based, multilevel analysis

AIM

To understand the scope for improving children's glycaemic outcomes by reducing variation between clinics and examine the role of insulin regimen and clinic characteristics.

METHODS

Cross-sectional analysis of 2012-2013 National Paediatric Diabetes Audit data from 21 773 children aged < 19 years with Type 1 diabetes cared for at 176 clinics organized into 11 regional diabetes networks in England and Wales. Variation in HbA_1c was explored by multilevel models with a random effect for clinic. The impact of clinic context was quantified by computing the per cent of total variation in HbA_1c which occurs between clinics (intraclass correlation coefficient; ICC).

RESULTS

Overall, 69 of the 176 diabetes clinics (39%) had a glycaemic performance that differed significantly from the national average after adjusting for patient case-mix with respect to age, gender, diabetes duration, deprivation and ethnicity. However, differences between clinics accounted for 4.7% of the total variation in HbA_1c . Inclusion of within-clinic HbA_1c standard deviation led to a substantial reduction in ICC to 2.4%. Insulin regimen, clinic volume and diabetes networks had a small or moderate impact on ICC.

CONCLUSIONS

Differences between diabetes clinics accounted for only a small portion of the total variation in glycaemic control because most of the variation was within clinics. This implies that national glycaemic improvements might best be achieved not only by targeting poor centres but also by shifting the whole distribution of clinics to higher levels of quality.

Effects of Acute-Postacute Continuity on Community Discharge and 30-Day Rehospitalization Following Inpatient Rehabilitation

OBJECTIVE

To examine the effects of facility-level acute-postacute continuity on probability of community discharge and 30-day rehospitalization following inpatient rehabilitation.

DATA SOURCES

We used national Medicare enrollment, claims, and assessment data to study 541,097 patients discharged from 1,156 inpatient rehabilitation facilities (IRFs) in 2010-2011.

STUDY DESIGN

We calculated facility-level continuity as the percentages of an IRF's patients admitted from each contributing acute care hospital. Patients were categorized into three groups: low continuity (<26 percent from same hospital that discharged the patient), medium continuity (26-75 percent from same hospital), or high continuity (>75 percent from same hospital). The multivariable models included an interaction term to examine the potential moderating effects of facility type (freestanding facility vs. hospital-based rehabilitation unit) on the relationships between facility-level continuity and our two outcomes: community discharge and 30-day rehospitalization.

PRINCIPAL FINDINGS

Medicare beneficiaries in hospital-based rehabilitation units were more likely to be referred from a high-contributing hospital compared to those in freestanding facilities. However, the association between higher acute-postacute continuity and desirable outcomes is significantly better in freestanding rehabilitation facilities than in hospital-based units.

CONCLUSIONS

Improving continuity is a key premise of health care reform. We found that both observed referral patterns and continuity-related benefits differed markedly by facility type. These findings provide a starting point for health systems establishing or strengthening acute-postacute relationships to improve patient outcomes in this new era of shared accountability and public quality reporting programs.

Prescription trends and the selection of initial oral antidiabetic agents for patients with newly diagnosed type 2 diabetes: a nationwide study

OBJECTIVES

The aim of this study was to examine the characteristics of patients, physicians, and medical facilities, and their association with prescriptions that do not include metformin as the initial oral antidiabetic agent.

STUDY DESIGN

Observational, cross-sectional study.

METHODS

Patients with incident type 2 diabetes between January 1, 2006, and December 31, 2010, were identified from the Taiwan National Insurance Research Database. We describe trends in the initial prescription of antidiabetic medications that do not contain metformin during the study period. A multivariable logistic model and a multilevel linear model were used in the analysis of factors at a range of levels (patient, physician, and medical facility), which may be associated with the selection of oral antidiabetic drugs.

RESULTS

During the study period, the proportion of prescriptions that did not include metformin declined from 43.8% to 26.2%. Male patients were more likely to obtain non-metformin prescriptions (adjusted odds ratio [OR]: 1.15; 95% confidence interval [CI]: 1.08-1.23), and the likelihood that a patient would be prescribed a non-metformin prescription increased with age. Physicians aged ≥35 years and those with specialties other than endocrinology tended to prescribe non-metformin prescriptions. Metformin was less commonly prescribed in for-profit hospitals (adjusted OR: 1.34, 95% CI: 1.11-1.61) and hospitals in smaller cities (adjusted OR: 1.28, 95% CI: 1.05-1.57) and rural areas (adjusted OR: 1.83, 95% CI: 1.32-2.54).

CONCLUSIONS

Disparities continue to exist in clinical practice with regard to the treatment of diabetes. These inequalities appear to be linked to a variety of factors related to patients, physicians, and medical facilities. Further study will be required to understand the effects of continuing medical education in enhancing adherence to clinical guidelines.

Contextual generalized trust and immunization against the 2009 A(H1N1) pandemic in the American states: A multilevel approach

The aim of the study was to investigate the association between contextual generalized trust and individual-level 2009 A(H1N1) pandemic immunization acceptance. A second aim was to investigate whether knowledge about the A(H1N1) pandemic mediated the association between contextual generalized trust and A(H1N1) immunization acceptance. Data from the National 2009 H1N1 Flu Survey was used. To capture contextual generalized trust, data comes from an aggregation of surveys measuring generalized trust in the American states. To investigate the association between contextual generalized trust and immunization acceptance, while taking potential individual-level confounders into account, multilevel logistic regression was used. The investigation showed contextual generalized trust to be significantly associated with immunization acceptance. However, controlling for knowledge about the A(H1N1) pandemic did not substantially affect the association between contextual generalized trust and immunization acceptance. In conclusion, contextual state-level generalized trust was associated with A(H1N1) immunization, but knowledge about A(H1N1) was not mediating this association.

Determinants of Healthcare Utilisation and Out-of-Pocket Payments in the Context of Free Public Primary Healthcare in Zambia

BACKGROUND

Access to appropriate and affordable healthcare is needed to achieve better health outcomes in Africa. However, access to healthcare remains low, especially among the poor. In Zambia, poor access exists despite the policy by the government to remove user fees in all primary healthcare facilities in the public sector. The paper has two main objectives: (i) to examine the factors associated with healthcare choices among sick people, and (ii) to assess the determinants of the magnitude of out-of-pocket (OOP) payments related to a visit to a health provider.

METHODS

This paper employs a multilevel multinomial logistic regression to model the determinants of an individual's choice of healthcare options following an illness. Further, the study analyses the drivers of the magnitude of OOP expenditure related to a visit to a health provider using a two-part generalised linear model. The analysis is based on a nationally representative healthcare utilisation and expenditure survey that was conducted in 2014.

RESULTS

Household per capita consumption expenditure is significantly associated with increased odds of seeking formal care (odds ratio [OR] = 1.12, P = .000). Living in a household in which the head has a higher level of education is associated with increased odds of seeking formal healthcare (OR = 1.54, P = .000) and (OR = 1.55, P = .01), for secondary and tertiary education, respectively. Rural residence is associated with reduced odds of seeking formal care (OR = 0.706, P = .002). The magnitude of OOP expenditure during a visit is significantly dependent on household economic well-being, distance from a health facility, among other factors. A 10% increase in per capita consumption expenditure was associated with a 0.2% increase in OOP health expenditure while every kilometre travelled was associated with a K0.51 increase in OOP health expenditure.

CONCLUSION

Despite the removal of user fees on public primary healthcare in Zambia, access to healthcare is highly dependent on an individual's socio-economic status, illness type and region of residence. These findings also suggest that the benefits of free public healthcare may not reach the poorest proportionately, which raise implications for increasing access in Zambia and other countries in sub-Saharan Africa.

Meeting the Challenges of Longitudinal Cluster-Based Trials in Schools: Lessons From the Chicago Trial of Positive Action

We describe challenges in the 6-year longitudinal cluster randomized controlled trial (CRCT) of Positive Action (PA), a social-emotional and character development (SECD) program, conducted in 14 low-income, urban Chicago Public Schools. Challenges pertained to logistics of study planning (school recruitment, retention of schools during the trial, consent rates, assessment of student outcomes, and confidentiality), study design (randomization of a small number of schools), fidelity (implementation of PA and control condition activities), and evaluation (restricted range of outcomes, measurement invariance, statistical power, student mobility, and moderators of program effects). Strategies used to address the challenges within each of these areas are discussed. Incorporation of lessons learned from this study may help to improve future evaluations of longitudinal CRCTs, especially those that involve evaluation of school-based interventions for minority populations and urban areas.

Use of the "Exposome" in the Practice of Epidemiology: A Primer on -Omic Technologies

The exposome has been defined as the totality of exposures individuals experience over the course of their lives and how those exposures affect health. Three domains of the exposome have been identified: internal, specific external, and general external. Internal factors are those that are unique to the individual, and specific external factors include occupational exposures and lifestyle factors. The general external domain includes sociodemographic factors such as educational level and financial status. Eliciting information on the exposome is daunting and not feasible at present; the undertaking may never be fully realized. A variety of tools have been identified to measure the exposome. Biomarker measurements will be one of the major tools in exposomic studies. However, exposure data can also be obtained from other sources such as sensors, geographic information systems, and conventional tools such as survey instruments. Proof-of-concept studies are being conducted that show the promise of exposomic investigation and the integration of different kinds of data. The inherent value of exposomic data in epidemiologic studies is that they can provide greater understanding of the relationships among a broad range of chemical and other risk factors and health conditions and ultimately lead to more effective and efficient disease prevention and control.

Spatializing Area-Based Measures of Neighborhood Characteristics for Multilevel Regression Analyses: An Areal Median Filtering Approach

Area-based measures of neighborhood characteristics simply derived from enumeration units (e.g., census tracts or block groups) ignore the potential of spatial spillover effects, and thus incorporating such measures into multilevel regression models may underestimate the neighborhood effects on health. To overcome this limitation, we describe the concept and method of areal median filtering to spatialize area-based measures of neighborhood characteristics for multilevel regression analyses. The areal median filtering approach provides a means to specify or formulate "neighborhoods" as meaningful geographic entities by removing enumeration unit boundaries as the absolute barriers and by pooling information from the neighboring enumeration units. This spatializing process takes into account for the potential of spatial spillover effects and also converts aspatial measures of neighborhood characteristics into spatial measures. From a conceptual and methodological standpoint, incorporating the derived spatial measures into multilevel regression analyses allows us to more accurately examine the relationships between neighborhood characteristics and health. To promote and set the stage for informative research in the future, we provide a few important conceptual and methodological remarks, and discuss possible applications, inherent limitations, and practical solutions for using the areal median filtering approach in the study of neighborhood effects on health.

Assessing the Individual, Neighborhood, and Policy Predictors of Disparities in Mental Health Care

This study assesses individual- and area-level predictors of racial/ethnic disparities in mental health care episodes for adults with psychiatric illness. Multilevel regression models are estimated using data from the Medical Expenditure Panel Surveys linked to area-level data sets. Compared with Whites, Blacks and Latinos live in neighborhoods with higher minority density, lower average education, and greater specialist mental health provider density, all of which predict lesser mental health care initiation. Neighborhood-level variables do not have differential effects on mental health care by race/ethnicity. Racial/ethnic disparities arise because minorities are more likely to live in neighborhoods where treatment initiation is low, rather than because of a differential influence of neighborhood disadvantage on treatment initiation for minorities compared with Whites. Low rates of initiation in neighborhoods with a high density of specialists suggest that interventions to increase mental health care specialists, without a focus on treating racial/ethnic minorities, may not reduce access disparities.

Cost Variation in Diabetes Care across Dutch Care Groups?

OBJECTIVE

The introduction of bundled payment for diabetes care in the Netherlands led to the origination of care groups. This study explored to what extent variation in health care costs per patient can be attributed to the performance of care groups. Furthermore, the commonly applied simple mean aggregation was compared with the more advanced generalized linear mixed model (GLMM) to benchmark health care costs per patient between care groups.

DATA SOURCE

Dutch 2009 nationwide insurance claims data of diabetes type 2 patients (104,544 patients, 50 care groups).

STUDY DESIGN

Both a simple mean aggregation and a GLMM approach was applied to rank care groups, using two different health care costs variables: total treatment health care costs and diabetes-specific specialist care costs per diabetes patient.

PRINCIPAL FINDINGS

Care groups varied slightly in the first and mainly in the second indicator. Care group variation was not explained by composition. Although the ranking methods were correlated, some care groups' rank positions differed, with consequences on the top-10 and the low-10 positions.

CONCLUSIONS

Differences between care groups exist when an appropriate indicator and a sophisticated aggregation technique is used. Currently applied benchmarking may have unfair consequences for some care groups.

Sub-Ethnic and Geographic Variations in Out-of-Pocket Private Health Insurance Premiums Among Mid-Life Asians

OBJECTIVE

This study examined out-of-pocket premium burden of mid-life Asian Americans by comparing six sub-groups of Asians after controlling for geographic clustering at the county and state levels.

METHOD

The 2007-2011 National Health Interview Survey was linked to community-level data and analyzed for 4,628 Asians (ages 50-64), including 697 Asian Indians, 1,125 Chinese, 1,393 Filipinos, 434 Japanese, 524 Koreans, and 455 Vietnamese. Non-Hispanic Whites were included as a comparison group ( n = 48,135). Three-level multilevel modeling (state > county > individual) was conducted.

RESULTS

Koreans and Vietnamese were found as vulnerable sub-groups considering their lower private health insurance rates and higher uninsured rates. Among those with private insurance, Asians, specifically Filipinos, paid significantly less than non-Hispanic Whites. Moderate but significant variations in the county- and state-level variance in out-of-pocket premiums were found, especially among mid-life Asians.

DISCUSSION

This study demonstrates the importance of examining within-group heterogeneity and geographic variations in understanding premium burden among mid-life Asians.

Does the Neighborhood Area of Residence Influence Non-Attendance in an Urban Mammography Screening Program? A Multilevel Study in a Swedish City

BACKGROUND AND AIM

The public health impact of population-based mammography screening programs depends on high participation rates. Thus, monitoring participation rates, as well as understanding and considering the factors influencing attendance, is important. With the goal to acquire information on the appropriate level of intervention for increasing screening participation our study aimed to (1) examine whether, over and above individual factors, the neighborhood of residence influences a woman's mammography non-attendance, and (2) evaluate, whether knowing a woman's neighborhood of residence would be sufficient to predict non-attendance.

METHODS

We analyze all women invited to mammography screening in 2005-09, residing in the city of Malmö, Sweden. Information regarding mammography screening attendance was linked to data on area of residence, demographic and socioeconomic characteristics available from Statistics Sweden. The influence of individual and neighborhood factors was assessed by multilevel logistic regression analysis with 29,901 women nested within 212 neighborhoods.

RESULTS

The prevalence of non-attendance among women was 18.3%. After adjusting for individual characteristics, the prevalence in the 212 neighborhoods was 3.6%. Neighborhood of residence had little influence on non-attendance. The multilevel analysis indicates that 8.4% of the total individual differences in the propensity of non-attendance were at the neighborhood level. However, when adjusting for specific individual characteristics this general contextual effect decreased to 1.8%. This minor effect was explained by the sociodemographic characteristic of the neighborhoods. The discriminatory accuracy of classifying women according to their non-attendance was 0.747 when considering only individual level variables, and 0.760 after including neighborhood level as a random effect.

CONCLUSION

Our results suggest that neighborhoods of residence in Malmö, Sweden (as defined by small-area market statistics (SAMS) areas) do not condition women's participation in population based mammography screening. Thus, interventions should be directed to the whole city and target women with a higher risk of non-attendance.

A cross-sectional study of geographic differences in health risk factors among young Australian adults: the role of socioeconomic position

Background

It remains unclear why living outside of an urban environment affects aspects of health, particularly whether these differences can be explained by other factors such as socioeconomic position (SEP). The aim of this study was to compare health risk factors between metropolitan and non-metropolitan young Australian adults and examine whether socioeconomic position (SEP) mediates any differences.

Methods

Cross-sectional data came from an Australia-wide sample of 26–36 year-olds (n = 2567). Information on demographic characteristics, smoking, alcohol consumption, diet, physical activity (PA, mins/week) and mental health were collected by questionnaire, BMI from measured height and weight and daily steps using pedometers. Metropolitan versus non-metropolitan residence was classified from addresses. SEP included individual-level (education, occupation) and area-level measures. Prevalence ratios and ratio of means were calculated using log binomial, log multinomial and linear regression techniques.

Results

Non-metropolitan residents were less likely to meet 2 or more dietary guidelines, reported less leisure-time PA and active commuting but more occupational and domestic PA than metropolitan residents. Non-metropolitan women were more likely to smoke and be obese. No differences in mental health were found. After adjusting for SEP, differences remained significant except for leisure-time PA (men and women) and smoking (women).

Conclusions

Living outside metropolitan areas was associated with more risk factors in these young adults. Individual SEP and area-level disadvantage generally did not explain these differences, suggesting that a focus on geographic location as its own social determinant of health, beyond SEP, is warranted.

Random-effects, fixed-effects and the within-between specification for clustered data in observational health studies: a simulation study

Background

When unaccounted-for group-level characteristics affect an outcome variable, traditional linear regression is inefficient and can be biased. The random- and fixed-effects estimators (RE and FE, respectively) are two competing methods that address these problems. While each estimator controls for otherwise unaccounted-for effects, the two estimators require different assumptions. Health researchers tend to favor RE estimation, while researchers from some other disciplines tend to favor FE estimation. In addition to RE and FE, an alternative method called within-between (WB) was suggested by Mundlak in 1978, although is utilized infrequently.

Methods

We conduct a simulation study to compare RE, FE, and WB estimation across 16,200 scenarios. The scenarios vary in the number of groups, the size of the groups, within-group variation, goodness-of-fit of the model, and the degree to which the model is correctly specified. Estimator preference is determined by lowest mean squared error of the estimated marginal effect and root mean squared error of fitted values.

Results

Although there are scenarios when each estimator is most appropriate, the cases in which traditional RE estimation is preferred are less common. In finite samples, the WB approach outperforms both traditional estimators. The Hausman test guides the practitioner to the estimator with the smallest absolute error only 61% of the time, and in many sample sizes simply applying the WB approach produces smaller absolute errors than following the suggestion of the test.

Conclusions

Specification and estimation should be carefully considered and ultimately guided by the objective of the analysis and characteristics of the data. The WB approach has been underutilized, particularly for inference on marginal effects in small samples. Blindly applying any estimator can lead to bias, inefficiency, and flawed inference.

Time to first take-back operation predicts successful primary fascial closure in patients undergoing damage control laparotomy

BACKGROUND

Failure to achieve primary fascial closure (PFC) after damage control laparotomy is costly and carries great morbidity. We hypothesized that time from the initial laparotomy to the first take-back operation would be predictive of successful PFC.

METHODS

Trauma patients managed with open abdominal techniques after damage control laparotomy were prospectively followed at 14 Level 1 trauma centers during a 2-year period. Time to the first take-back was evaluated as a predictor of PFC using hierarchical multivariate logistic regression analysis.

RESULTS

A total of 499 patients underwent damage control laparotomy and were included in this analysis. PFC was achieved in 327 (65.5%) patients. Median time to the first take-back operation was 36 hours (interquartile range 24-48). After we adjusted for patient demographics, resuscitation volumes, and operative characteristics, increasing time to the first take-back was associated with a decreased likelihood of PFC. Specifically, each hour delay in return to the operating room (24 hours after initial laparotomy) was associated with a 1.1% decrease in the odds of PFC (odds ratio 0.989; 95% confidence interval 0.978-0.999; P = .045). In addition, there was a trend towards increased intra-abdominal complications in patients returning after 48 hours (odds ratio 1.80; 95% confidence interval 1.00-3.25; P = .05).

CONCLUSION

Data from this prospective, multicenter study demonstrate that delays in returning to the operating room after damage control laparotomy are associated with reductions in PFC. These findings suggest that emphasis should be placed on returning to the operating room within 24 hours after the initial laparotomy if possible (and no later than 48 hours).

Common risk factor approach to address socioeconomic inequality in the oral health of preschool children--a prospective cohort study

BACKGROUND

Dental caries remains the most prevalent chronic condition in children and a major contributor to poor general health. There is ample evidence of a skewed distribution of oral health, with a small proportion of children in the population bearing the majority of the burden of the disease. This minority group is comprised disproportionately of socioeconomically disadvantaged children. An in-depth longitudinal study is needed to better understand the determinants of child oral health, in order to support effective evidence-based policies and interventions in improving child oral health. The aim of the Study of Mothers' and Infants' Life Events Affecting Oral Health (SMILE) project is to identify and evaluate the relative importance and timing of critical factors that shape the oral health of young children and then to seek to evaluate those factors in their inter-relationship with socioeconomic influences.

METHODS/DESIGN

This investigation will apply an observational prospective study design to a cohort of socioeconomically-diverse South Australian newborns and their mothers, intensively following these dyads as the children grow to toddler age. Mothers of newborn children will be invited to participate in the study in the early post-partum period. At enrolment, data will be collected on parental socioeconomic status, mothers' general and dental health conditions, details of the pregnancy, infant feeding practice and parental health behaviours and practices. Data on diet and feeding practices, oral health behaviours and practices, and dental visiting patterns will be collected at 3, 6, 12 and 24 months of age. When children turn 24-30 months, the children and their mothers/primary care givers will be invited to an oral examination to record oral health status. Anthropometric assessment will also be conducted.

DISCUSSION

This prospective cohort study will examine a wide range of determinants influencing child oral health and related general conditions such as overweight. It will lead to the evaluation of the inter-relationship among main influences and their relative effect on child oral health. The study findings will provide high level evidence of pathways through which socio-environmental factors impact child oral health. It will also provide an opportunity to examine the relationship between oral health and childhood overweight.

Multilevel analysis of the determinants of the global assessment of functioning in an inpatient population

BACKGROUND

The Global Assessment of Functioning (GAF) is a widely used measure of psychiatric symptoms and functioning, yet numerous concerns persist about its reliability and validity. The objective of this study was to determine the extent to which GAF scores reflect physician-related differences in addition to information about patients.

METHODS

This is a secondary analysis of clinical data collected between 2005 and 2010 from inpatients at a psychiatric hospital (N = 1,852). Multilevel modeling was used to estimate the influence of physicians on GAF scores at admission and on the change between admission and discharge, controlling for patient clinical presentation.

RESULTS

Controlling for patient-level predictors, 7% of the residual variance in admission GAF scores and 8% of the residual variance in change scores was at the physician level. The physician-level variance was significantly larger than zero in both models.

CONCLUSIONS

Although statistically significant, estimates of physician-level variance were not overwhelming, suggesting that the GAF was rated in a consistent manner across physicians in this hospital. While results lend support to the utility of the GAF for drawing comparisons between patients seen by different physicians across a large institution, further study is necessary to determine generalizability and to assess differences across multiple institutions.

Neighborhood characteristics and differential risk for depressive and anxiety disorders across racial/ethnic groups in the United States

BACKGROUND

The prevalence of psychiatric disorders varies depending on the person's neighborhood context, their racial/ethnic group, and the specific diagnoses being examined. Less is known about specific neighborhood features that represent differential risk for depressive and anxiety disorders (DAD) across racial/ethnic groups in the United States. This study examines whether neighborhood etiologic factors are associated with DAD, above and beyond individual-level characteristics, and whether these associations are moderated by race/ethnicity.

METHODS

We utilized nationally representative data (N = 13,837) from the Collaborative Psychiatric Epidemiology Studies (CPES-Geocode file). Separate weighted multilevel logistic regression models were fitted for any past-year depressive and/or anxiety disorder, any depressive disorder only, and any anxiety disorder only.

RESULTS

After adjusting for individual-level characteristics, African Americans living in a neighborhood with greater affluence and Afro-Caribbeans residing in more residentially unstable neighborhoods were at increased risk for any past-year depressive disorder as compared to their non-Latino white counterparts. Further, Latinos residing in neighborhoods with greater levels of Latino/immigrant concentration were at increased risk of any past-year anxiety disorder. Lastly, Asians living in neighborhoods with higher levels of economic disadvantage were at decreased risk of any past-year depressive and/or anxiety disorders compared to non-Latino whites, independent of individual-level factors. Differences across subethnic groups are also evident.

CONCLUSIONS

Results suggest neighborhood characteristics operate differently on risk for DAD across racial/ethnic groups. Our findings have important implications for designing and targeting interventions to address DAD risk among racial/ethnic minorities.

Probability interpretations of intraclass reliabilities

Research where many organizations are rated by different samples of individuals such as clients, patients, or employees frequently uses reliabilities computed from intraclass correlations. Consumers of statistical information, such as patients and policy makers, may not have sufficient background for deciding which levels of reliability are acceptable. It is shown that the reliability is related to various probabilities that may be easier to understand, for example, the proportion of organizations that will be classed significantly above (or below) the mean and the probability that an organization is classed correctly given that it is classed significantly above (or below) the mean. One can view these probabilities as the amount of information of the classification and the correctness of the classification. These probabilities have an inverse relationship: given a reliability, one can 'buy' correctness at the cost of informativeness and conversely. This article discusses how this can be used to make judgments about the required level of reliabilities.

Analysis of HIV/AIDS DRG in Portugal: a hierarchical finite mixture model

Inpatient length of stay (LOS) is an important measure of hospital activity, but its empirical distribution is often positively skewed, representing a challenge for statistical analysis. Taking this feature into account, we seek to identify factors that are associated with HIV/AIDS through a hierarchical finite mixture model. A mixture of normal components is applied to adult HIV/AIDS diagnosis-related group data (DRG) from 2008. The model accounts for the demographic and clinical characteristics of the patients, as well the inherent correlation of patients clustered within hospitals. In the present research, a normal mixture distribution was fitted to the logarithm of LOS and it was found that a model with two-components had the best fit, resulting in two subgroups of LOS: a short-stay subgroup and a long-stay subgroup. Associated risk factors for both groups were identified as well as some statistical differences in the hospitals. Our findings provide important information for policy makers in terms of discharge planning and the efficient management of LOS. The presence of "atypical" hospitals also suggests that hospitals should not be viewed or treated as homogenous bodies.