Hierarchical Logistic Models Research Articles

Factors associated with the use of the prescription monitoring program by prescribers and pharmacists in Texas.

To describe Texas Prescription Monitoring Program (PMP) use and identify predictors of PMP query for opioid and benzodiazepine prescriptions by prescribers and pharmacists. Dispensation and query records from the Texas PMP for opioid and benzodiazepine medications dispensed between October 1, 2016 and December 31, 2018 were linked using common patient identifiers. Autoregressive linear regression was used to assess trends in utilization. Hierarchical logistic models were specified to identify factors associated with provider and pharmacist query of opioid and benzodiazepine prescriptions. Despite a significant increase in the total number of pharmacists (β = 169.85, p < 0.0001) and prescribers (β = 301.59, p < 0.0001) who used the PMP every month, the ratio of active to registered pharmacists (β = -0.0001, p = 0.75) and prescribers (β = -0.0015, p = 0.10) did not change. Pharmacists and prescribers were significantly more likely to query opioid and benzodiazepine prescriptions of 14 days or more, and those issued to patients new to their practice. Pharmacists were most likely to query opioid prescriptions for oxycodone (aOR = 4.51, 95%CI = 4.42-4.60) and prescribers were most likely to query prescriptions for buprenorphine (aOR = 2.24, 95%CI = 2.15-2.35) compared to codeine. Changes in PMP utilization between October 2016 and December 2018 were driven by increasing registration, not increasing frequency of use among registered users. Use of the PMP is inconsistent and dependent upon patient characteristics thus limiting the utility of the PMP as a decision support tool. These results support the need for policy mandating PMP use in Texas and provide a useful baseline and framework to evaluate the effectiveness of mandate implementation.

Variation in use and dosing escalation of renin angiotensin system, mineralocorticoid receptor antagonist, angiotensin receptor neprilysin inhibitor and beta-blocker therapies in heart failure and reduced ejection fraction: Association of comorbidities

In patients with heart failure and reduced ejection fraction (HFrEF), angiotensin converting enzyme inhibitors (ACEi), angiotensin II receptor blockers (ARB), or angiotensin receptor neprilysin inhibitor (ARNI), mineralocorticoid receptor antagonists (MRA), and beta-blockers (βB) are underutilized. It is unknown if patients with and without comorbidities have similar ACEi/ARB/ARNI, MRA, and βB prescription patterns. Baseline data from the CHAMP-HF (Change the Management of Patients with Heart Failure) registry were categorized by history of atrial fibrillation, asthma/chronic lung disease, obstructive sleep apnea, and depression. Using multivariate hierarchical logistic models, associations of ACEi/ARB/ARNI, MRA and βB medication use and dose by comorbidities were assessed after adjusting for patient characteristics. Of 4,815 HFrEF patients from 152 CHAMP-HF sites, ACEi/ARB/ARNI use was lower in patients with more comorbidities, and generally, MRA use was low and βB use was high. In adjusted analyses, patients with HFrEF and comorbid obstructive sleep apnea, vs. without, were more likely to be prescribed ARNI (OR [95% CI]: 1.25 [1.00, 1.55]); P = .047 and MRA (1.31 [1.11, 1.55]); P = .002 and less likely to be prescribed ACEi (0.74 [0.63, 0.88]); P < .001. Patients with atrial fibrillation, vs. without, were less likely to receive ACEi/ARB (0.82 [0.71, 0.95]); P = .006 and any study medication (0.81 [0.67, 0.97]); P = .020. Comorbid lung disease and history of depression were not associated with HFrEF prescriptions. Renin-angiotensin-aldosterone blockade therapy prescription and dose varied by comorbidity status, but βB therapy did not. In quality efforts, leaders need to consider use and dosing of prescriptions in light of prevalent comorbidities.

Driver injury severity analysis of crashes in a western China's rural mountainous county: Taking crash compatibility difference into consideration

China's traffic safety attracts increasing research interest. Official data show that crashes in the western region of China are more severe than those in the eastern region. However, research on crash severity in western China is scarce. This study applied a hierarchical Bayesian logistic model to examine the significant factors related to crash and vehicle/driver levels and their heterogeneous impacts on the severity of drivers' injury. Crash data were collected from Lintao, a rural mountainous county in western China. A variable was proposed to measure the relative difference between the crashworthiness of one vehicle and the aggressivity of the other vehicle in the mixed traffic flow. Results indicated that the majority of the total variance was induced by between-crash variance, showing the suitability of the utilized hierarchical modeling approach. One crash-level variable and six vehicle/driver-level variables, namely, road type, compatibility difference, age, vehicle type, drunk driving, driving unregistered vehicle, and driving years, significantly affected modeling drivers' injury severities. Among these variables, road type (national and provincial), age (young and senior drivers), driving unregistered vehicle, and drunk driving tended to increase the odds of crash-related mortality. Driving years (new drivers with less than six years of driving experience) and vehicle type (heavy vehicle) were likely to decrease the probability of fatal outcomes. Compatibility difference was relatively significant, and the possibilities of mortality in single vehicle crashes were higher than those in multivehicle and pedestrian-involved crashes. The developed methodology and estimation results provided insights into the internal mechanism of rural crashes and effective countermeasures to prevent rural crashes. • A hierarchical Bayesian logistic model was established. • Data were collected from a rural county in western China with highly mixed traffic. • A variable measuring crash compatibility difference within a crash was employed. • The effects of seven types of violation on driver injury outcome were examined. • Seven variables were significant, while gender was not.

Predictive accuracy of a hierarchical logistic model of cumulative SARS-CoV-2 case growth until May 2020

BackgroundInfectious disease predictions models, including virtually all epidemiological models describing the spread of the SARS-CoV-2 pandemic, are rarely evaluated empirically. The aim of the present study was to investigate the predictive accuracy of a prognostic model for forecasting the development of the cumulative number of reported SARS-CoV-2 cases in countries and administrative regions worldwide until the end of May 2020.MethodsThe cumulative number of reported SARS-CoV-2 cases was forecasted in 251 regions with a horizon of two weeks, one month, and two months using a hierarchical logistic model at the end of March 2020. Forecasts were compared to actual observations by using a series of evaluation metrics.ResultsOn average, predictive accuracy was very high in nearly all regions at the two weeks forecast, high in most regions at the one month forecast, and notable in the majority of the regions at the two months forecast. Higher accuracy was associated with the availability of more data for estimation and with a more pronounced cumulative case growth from the first case to the date of estimation. In some strongly affected regions, cumulative case counts were considerably underestimated.ConclusionsWith keeping its limitations in mind, the investigated model may be used for the preparation and distribution of resources during the initial phase of epidemics. Future research should primarily address the model’s assumptions and its scope of applicability. In addition, establishing a relationship with known mechanisms and traditional epidemiological models of disease transmission would be desirable.

S0282 Adenoma Detection Rates for Colonoscopy and Associated Factors: Results From a U.S. Sample Using the GIQuIC Registry

INTRODUCTION: Adenoma Detection rate (ADR) is an established quality indicator for screening colonoscopy. However, reported ADRs are highly variable across practices, and national or population-based estimates are not available. The aim of our study was to study the ADR, variability of rates over time and factors associated with detection rates of ADR in a national sample of patients undergoing colonoscopy using the GIQuIC registry. METHODS: We used colonoscopies submitted to the GIQuIC registry from 2014 to 2018 on adults ages 50–89. Only the first colonoscopy record per patient was included. Indications for colonoscopy were categorized as screening, diagnostic and surveillance. We used hierarchical logistic models to study factors associated with ADR. RESULTS: A total of 2,646,833 colonoscopies were performed by 1,169 endoscopists during the study period. The average ADR for screening colonoscopies per endoscopist was 36.80% (SD 10.21), 44.08 (SD 10.98) in males and 31.20 (SD 9.65) in females. Adjusted to the US population, the ADR was 39.08%. There was a significant increase in ADR from screening colonoscopies over the study period from 33.93% in 2014 to 38.12% in 2018. Clinically significant factors associated with higher ADR were age (OR 1.28; 95% CI 1.27–1.29 for 60–69; 1.57, 95% CI 1.55–1.58 for 70–79 compared to 50–59 years), male gender (OR 1.57; 95% CI 1.56–1.58), surveillance indication (vs. screening; OR 1.24; 95% CI 1.22, 1.26) and longer withdrawal times (>11 minutes vs. ≤6 minutes) (OR 10.07; 95% CI 9.51–10.66). CONCLUSION: Population based estimates of detection rates of ADR are 36.60% and have increased over time. Factors associated with ADR were patient age, gender and longer withdrawal time.Figure 1.: ADR for men and women over 20-14-2018.Figure 2.: Trends of ADR by indication over time.

Accuracy of the need for dental prosthesis and associated factors between elderly in Recife, Brazil

Abstract Background The latest national oral health survey showed a high prevalence of the need for dental prostheses between the Brazilian elderly. To classify this need, normative (clinical) and subjective (self-reported) criteria must be considered since patients' self-perception takes into account social and functional issues that arise with oral health problems. Few studies investigate the agreement between these criteria, as well as its determinants. Thus, this study aimed to investigate the accuracy of the need for the use of total dental prosthesis and factors associated with the agreement between criteria. Methods Cross-sectional study, carried out in three municipalities in the Metropolitan Region of Recife, with a random sample of 816 elderly people from 65 to 74 years old. The dependent variable was the accuracy, calculated by the agreement between the self-reported and the normative need for a total dental prosthesis, and the independents were assembled in three blocks (socioeconomic/demographic, access to oral health services and self-perceived oral health). Hierarchical logistic models were conducted for total upper prosthesis (TUP) and total lower prosthesis (TLP). Results The self-perception of the need for prosthetic use presented an accuracy of 75.9% (95% CI = 72.8-78.7%) for TUP and 78.6% (95% CI = 75.6-81.3%) for TLP. In the multiple analysis, the accuracy for TUP and TLP needs holds an association with the variables: family income, age and time since the last dental appointment. Conclusions In conclusion, the self-perception of need for dental prosthesis demonstrates potential applicability for the elderly, presenting notable accuracy values. It suggests that studies based on patients' self-reports should be stimulated, aiming for the evaluation and validation of self-reported criteria in different contexts and cultures. Furthermore, the identification of accuracy associated factors can help to build more meaningful questions to be used in future surveys. Key messages The use of the self-reported need for total dental prosthesis may be feasible when considering lower cost, reduced time of execution and ease of use in population epidemiological surveys. Application of self-reporting as an epidemiological tool for planning and monitoring oral health services, incorporating it in the form of indicators for oral health surveillance.

Factors affecting mammography screening behaviour among rural Vietnamese women.

To identify factors associated with mammography screening behaviour and its predictors among rural Vietnamese women. A predictive correlational study involving 120 women aged≥40years was conducted in the suburbs of Hanoi, Vietnam, in July 2018 by using Breast Cancer Awareness Measurement and the Champion Health Beliefs Model Scale. Mammography screening behaviour was assessed by asking participants about their previous mammography experience. Only 16.7% of participants had undergone mammography screening. High education levels, high monthly family income, having family members or friends with breast cancer, and receiving physicians' recommendations increased the likelihood of participants screening for breast cancer. Mean scores on perceived susceptibility and perceived barriers differed significantly between participants who had and those who had not undergone screening (t=4.31; p<.001; t=-5.05; p<.001, respectively). Perceived susceptibility and perceived barriers significantly increased the predictive power of the hierarchical logistic model (critical value=6.16; [df=2]; p=.046). Perceived barriers were the most significant predictors of screening behaviour (odds ratio 0.84; 95% CI, 0.71-0.99; p=.039). Efforts are necessary to increase mammography awareness in the community and promote screening rates in Vietnam.

Variation Over Time and Factors Associated With Detection Rates of Sessile Serrated Lesion Across the United States: Results Form a National Sample Using the GIQuIC Registry.

Sessile serrated lesions (SSLs) are important precursor lesions for the CpG island-methylated pathway to colorectal cancer. The reported detection rates of SSL are highly variable, and national or population-based estimates are not available. Patient-, provider-, and procedure-level factors associated with the detection rates of SSL have not been well described. The aim of our study was to study the detection rates of SSL, variability of rates over time, and factors associated with detection rates of SSL in a national sample of patients undergoing colonoscopy using the GIQuIC registry. We used colonoscopies submitted to the GIQuIC registry from 2014 to 2017 on adults, aged 18-89 years. Only the first colonoscopy record per patient was included. Indications for colonoscopy were categorized as screening, diagnostic, and surveillance. We used the hierarchical logistic models to study the factors associated with the detection rates of SSL. The Cochrane-Armitage test was used to study the significance of trend over time. There were a total of 5,173,211 colonoscopies performed by 3,934 endoscopists during the study period. Among the 2,101,082 screening colonoscopies over the study period in adults older than or equal to 50 years that were complete to the cecum, the average detection rate per endoscopist for SSL was 6.43% (SD 5.18) and 6.25% standardized for the 2010 US population. There was a significant increase in the detection rates of SSLs from screening colonoscopies over the study period from 4.99% in 2014 to 7.09% in 2017 (P trend <0.001). Clinically significant factors associated with higher detection rates of SSL were longer withdrawal times (>11 minutes vs ≤6 minutes) (odds ratio [OR] 9.61; 9.03-10.24), adequate preparation (OR 1.25; 1.22-1.28), female sex (OR 1.17; 1.16-1.18), and use of a specialized gastrointestinal pathology group (OR 1.12; 95% confidence interval 1.04, 1.19). Population-based estimates of the detection rates of SSL are 6% and have increased over time.

Two-stage Bayesian hierarchical modeling for blinded and unblinded safety monitoring in randomized clinical trials

BackgroundMonitoring and reporting of drug safety during a clinical trial is essential to its success. More recent attention to drug safety has encouraged statistical methods development for monitoring and detecting potential safety signals. This paper investigates the potential impact of the process of the blinded investigator identifying a potential safety signal, which should be further investigated by the Data and Safety Monitoring Board with an unblinded safety data analysis.MethodsIn this paper, two-stage Bayesian hierarchical models are proposed for safety signal detection following a pre-specified set of interim analyses that are applied to efficacy. At stage 1, a hierarchical blinded model uses blinded safety data to detect a potential safety signal and at stage 2, a hierarchical logistic model is applied to confirm the signal with unblinded safety data.ResultsAny interim safety monitoring analysis is usually scheduled via negotiation between the trial sponsor and the Data and Safety Monitoring Board. The proposed safety monitoring process starts once 53 subjects have been enrolled into an eight-arm phase II clinical trial for the first interim analysis. Operating characteristics describing the performance of this proposed workflow are investigated using simulations based on the different scenarios.ConclusionsThe two-stage Bayesian safety procedure in this paper provides a statistical view to monitor safety during the clinical trials. The proposed two-stage monitoring model has an excellent accuracy of detecting and flagging a potential safety signal at stage 1, and with the most important feature that further action at stage 2 could confirm the safety issue.

Insatisfação com a imagem corporal e fatores associados em adolescentes

This study aimed to examine the factors associated with body dissatisfaction in adolescents. This is a cross-sectional study with students from two public and four private schools in the metropolitan region of Rio de Janeiro. Body dissatisfaction was evaluated using the Body Silhouette Scale, by the difference between the image they perceive as their current image and the one they would like to have. The association was assessed by hierarchical logistic regression multinomial model, by estimating odds ratios (OR) and 95% confidence intervals (95%CI). Among the 1,019 adolescents evaluated (13-19 years), 75% showed body dissatisfaction, 41.4% wishing for a smaller silhouette and 33.7% wishing for larger silhouettes. The wish for a smaller silhouette was more significant in girls, overweight adolescents, those with an unsatisfactory meal pattern, and with higher waist circumference. Adolescents that had been exposed to teasing due to their body shape increased the probability of wishing for both smaller and larger silhouettes. Strategies are required for a more positive perception of body image, especially for girls and for overweight adolescents, which provide guidance on adequate meal consumption and prevent exposure to peer teasing, valuing coexistence and well-being in the face of existing bodily differences.

Open the gates wider: affirmative action and dropping out

Affirmative action policies are oftentimes pitted against the need of universities to maintain meritocratic standards in enrollment. The current study tackles this institutional dilemma against the standard of student attrition. It does so by analyzing records of 41,483 undergraduate students who attended the Hebrew University of Jerusalem (2003–2015). Approximately 5% of the students were marked eligible by an affirmative action policy that is unique to Israel. Specifically, a non-governmental organization oversees the Israeli indirect affirmative action policy and grants applicants with the status largely based on class. Descriptive statistics reveal no significant differences in dropout rates between affirmative action students and their normative peers. To verify those patterns, we test hierarchical logistic models alongside advanced decision tree models. The findings show the superiority of first-year grade point average and other academic indicators in predicting dropouts. They also confirm that students who are eligible for affirmative action depart at virtually the same rates as normative students and other risk groups. We conclude by suggesting that under certain conditions, universities do not pay any price by admitting students through the backdoor of affirmative action. Therefore, universities can and should open their gates wider.

National Analysis of Coronary Artery Bypass Grafting in Autoimmune Connective Tissue Disease

Autoimmune connective tissue diseases (CTDs) are associated with accelerated atherosclerosis and inflammation, while often requiring immunosuppression. Large-scale outcomes of coronary artery bypass graft (CABG) surgery in this population have not been reported thus far. This study characterized trends in use of CABG in patients with CTDs and the impact of the disease on mortality, in-hospital complications, length of stay, and costs. The 2005 to 2015 National Inpatient Sample was used to identify all adult patients undergoing isolated CABG. The CTDs cohort included rheumatoid arthritis, lupus erythematosus, and antiphospholipid syndrome (APLS), among others. Hierarchical multivariable logistic models were used to calculate the independent impact of CTDs on clinical outcomes and costs. Of an estimated 2,101,591 patients, 41,567 (1.8%) were diagnosed with CTDs (rheumatoid arthritis, 58%; systemic lupus erythematosus, 12%; APLS, 11%) Although the overall annual use of CABG decreased, the proportion of patients with CTDs receiving the operation significantly increased. After adjusting for patient and hospital characteristics, CTDs were not associated with increased mortality (adjusted odds ratio [AOR], 0.91; P= .34) but were protective against cardiovascular (AOR, 0.92; P < .003), neurologic (AOR, 0.81; P= .01), and infectious (AOR, 0.80; P= .01) complications. The diagnosis of CTDs was also predictive of reduced length of hospital stay (β-coefficient= -0.40; P < .001) and costs (β-coefficient, -$1200; P= .01). On subgroup analysis patients with APLS had significantly increased odds of mortality (AOR, 1.5) and increased renal (AOR, 1.3), infectious (AOR, 1.7), and thromboembolic (AOR, 4.3) complications (all P < .05). CABG in patients with CTDs provides acceptable outcomes and paradoxically improved resource use. However CABG in patients with APLSwarrants careful consideration given inferior outcomes.

Quality of supervisor behaviour, workplace social capital and psychological well-being.

Employees with a higher level of workplace social capital (i.e. relationships, trust and reciprocity at work) have a lower risk of mental health problems. Supervisor behaviour may be a predictor of workplace social capital. To examine the associations between supervisor behaviour, workplace social capital and psychological well-being. We conducted a secondary analysis of the sixth European Working Conditions Survey. The study sample included 28 900 employees in 35 European countries with an immediate line manager. Depression was assessed with the WHO-5 Well-Being Index. Supervisor behaviour quality and workplace social capital were each measured with a 6-item question. Association between quality of supervisor behaviour and workplace social capital was analysed using a hierarchical linear modelling. A mediation analysis was conducted by using hierarchical logistic models of depression with and without workplace social capital index. A higher supervisor behaviour quality index was associated with an increased workplace social capital index (β, 0.55; 95% confidence interval [CI], 0.51-0.59). A higher workplace social capital index was associated with lower odds of depression (odds ratio [OR], 0.89; 95% CI, 0.87-0.90). A higher supervisor behaviour quality index was associated with lower odds of depression (OR, 0.90; 95% CI, 0.89-0.92); 58% of the effect was mediated by workplace social capital. Our findings support the hypothesis that better supervisor behaviour quality increases workplace social capital and contributes to psychological well-being of employees. This may be useful for planning organizational interventions to enhance mental health of workers.

Effects of stress on eating behaviours in adolescents: a daily diary investigation

Objective The current study aimed to use ecologically valid assessments to examine within- and between-person associations between adolescents’ daily stress and unhealthy eating behaviours. Design: Daily diary assessments were used to investigate whether daily perceived stress and negative life events were associated with naturally occurring eating behaviours across the span of one week in a community-based sample of 88 adolescents. Results: Logistic hierarchical models demonstrated between-person effects of stress on eating, such that adolescents who endorsed greater perceived stress and more daily negative life events, on average, across the one-week period, also reported elevated rates of craving tasty foods and trouble stopping the consumption of tasty foods. Within-person findings showed that some eating behaviours were associated with day-to-day, within-person fluctuations in stressful experiences. On days marked by higher than usual perceived stress (relative to one’s own average levels during the week), adolescents reported an increased likelihood of eating to cope with difficult emotions. Conclusions: Findings indicate that daily fluctuations in stress may contribute to unhealthy eating behaviours during adolescence, and that associations between stress and eating behaviour may differ depending on whether they are assessed within-person or between-persons.

Hierarchical analysis of factors associated with hospital readmissions for coronary heart disease: A case-control study.

To analyse, hierarchically, factors associated with hospital readmissions for acute coronary syndrome. Hospital readmissions have risen, especially in patients with multiple comorbidities, which are most often chronic. The leading causes of hospital readmission include acute coronary syndrome, which is costly and often preventable. Determining clinical and nonclinical variables that increase the chances of readmission is important to assess and evaluate patients hospitalised for coronary heart diseases. A case-control study whose dependent variable was hospital readmission for acute coronary syndrome. The study included 277 inpatients, of whom 132 were in their first hospitalisation and 145 had already been hospitalised for acute coronary syndrome. The independent variables for this hierarchical model were sociodemographic conditions, life habits, access to health services and physical health measures. Data were obtained by interviews, anthropometric measurements and patient records. Logistic regression analysis was performed using the stepwise technique, with Microsoft Excel and R version 3.2.3. The research was reported via strengthening the reporting of observational studies in epidemiology (STROBE). In the final hierarchical logistic model, the following risk factors were associated with readmission for acute coronary syndrome: inadequate drug therapy adherence, stress, history of smoking for 30years or more, and the lack of use of primary healthcare services. Clinical and nonclinical variables are related to hospital readmission for acute coronary syndrome and can increase the chance of readmission by up to six times. The predictive model can be used to avoid readmission for acute coronary syndrome, and it represents an advance in the prediction of the occurrence of the outcome. This implies the need for a reorientation of the network for postdischarge care in the first hospitalisation for acute coronary syndrome.

Applying hierarchical logistic models to compare urban and rural roadway modeling of severity of rear-end vehicular crashes

A rear-end crash is a widely studied type of road accident. The road area at the crash scene is a factor that significantly affects the crash severity from rear-end collisions. These road areas may be classified as urban or rural and evince obvious differences such as speed limits, number of intersections, vehicle types, etc. However, no study comparing rear-end crashes occurring in urban and rural areas has yet been conducted. Therefore, the present investigation focused on the comparison of diverse factors affecting the likelihood of rear-end crash severities in the two types of roadways. Additionally, hierarchical logistic models grounded in a spatial basis concept were applied by determining varying parameter estimations with regard to road segments. Additionally, the study compared coefficients with multilevel correlation model and those without multilevel correlation. Four models were established as a result. The data used for the study pertained to rear-end crashes occurring on Thai highways between 2011 and 2015. The results of the data analysis revealed that the model parameters for both urban and rural areas are in the same direction with the larger number of significant parameter values present in the rural rear-end crash model. The significant variables in both the urban and rural road segment models are the seat belt use, and the time of the incident. To conclude, the present study is useful because it provides another perspective of rear-end crashes to encourage policy makers to apply decisions that favor rules that assure safety.

Opting for Exit: Informalization, Social Policy Discontent, and Lack of Good Governance

ABSTRACTThe informal sector challenges economic growth and hinders the abatement of income disparities in developing countries. This study argues that a weak and poorly governed welfare state can cause the informal sector to increase when individuals use it as an exit option from an unsatisfying welfare system. The article explores how the welfare state’s benefit structure and citizens’ trust in institutions to deliver public goods affect the likelihood of informality. A logistic hierarchical model, based on cross-sectional survey data from Latin America and the Caribbean and descriptive panel data from Brazil, is used to test the hypothesis. Findings reveal that social policy discontent, low trust, an elitist distribution of welfare benefits, and dysfunctional institutions increase the likelihood of being informally employed. However, workers with greater agency—the better-educated—seem notably less likely to informalize when social policy benefits are targeted toward their own socioeconomic group.

The critical role of the institution‐led market in the technological catch‐up of emerging market enterprises: evidence from Chinese enterprises

Innovation scholars have, in the past, applied the concept of ‘windows of opportunity’ to study latecomer firms’ catch‐up. Previous research treats this concept as three separate aspects, i.e., technological, institutional and market. The role of government was seen as being concerned only with institutional windows of opportunity. However, governments in emerging markets exert influence not only through traditional means of institutional support but, also, via market‐driven mechanisms. The former refers to state procurement, resource provision, legislation, and administrative control, whereas the latter is concerned with demand creation, resource allocation, and the regulation of market orders. This multifaceted nature of government in promoting economic growth, guiding technological development, and influencing enterprise behavior remains under‐researched. Yet, it plays a crucial role in the catch‐up of emerging market enterprises. Therefore, based on innovation studies literature and an institution‐based view of international business, the present research proposes a new construct, termed the institution‐led market, with the aim to encapsulate the complex role of government in the catch‐up of emerging market enterprises. The institution‐led market is defined as a unique type of market that is well‐timed and strategically created by the government and supported by institutional policies and resources. A large database of 259 Chinese firms in 37 industries was created and analyzed using a hierarchical logistic model to empirically test the relationship between the institution‐led market and technological catch‐up of emerging market enterprises. We demonstrate that the institution‐led market positively affects the catch‐up of emerging market enterprises; furthermore, it significantly moderates the effect of technological discontinuity on the catch‐up. Finally, the theoretical contributions and managerial implications of the present research are discussed.

“Are They Aware, and Why?” Bayesian Analysis of Predictors of Smallholder Farmers’ Awareness of Climate Change and Its Risks to Agriculture

While climate change threatens global food security, health, and nutrition outcomes, Africa is more vulnerable because its economies largely depend on rain-fed agriculture. Thus, there is need for agricultural producers in Africa to employ robust adaptive measures that withstand the risks of climate change. However, the success of adaptation measures to climate change primarily depends on the communities’ knowledge or awareness of climate change and its risks. Nonetheless, existing empirical research is still limited to illuminate farmers’ awareness of the climate change problem. This study employs a Bayesian hierarchical logistic model, estimated using Hamiltonian Monte Carlo (HMC) methods, to empirically determine drivers of smallholder farmers’ awareness of climate change and its risks to agriculture in Zambia. The results suggest that on average, 77% of farmers in Zambia are aware of climate change and its risks to agriculture. We find socio-demographics, climate change information sources, climate change adaptive factors, and climate change impact-related shocks as predictors of the expression of climate change awareness. We suggest that farmers should be given all the necessary information about climate change and its risks to agriculture. Most importantly, the drivers identified can assist policymakers to provide the effective extension and advisory services that would enhance the understanding of climate change among farmers in synergy with appropriate farm-level climate-smart agricultural practices.

Abstract P158: USING CHA 2 DS 2 - VASC SCORE TO PREDICT CLINICAL OUTCOMES IN PATIENTS WITH INFECTIVE ENDOCARDITIS

Background: The CHA 2 DS 2 - VASC score is a well-validated stroke risk stratification tool in patients with non-valvular atrial fibrillation. Most of the clinical variables used have been associated with poor outcomes in patients with infective endocarditis (IE). We aim to determine its usefulness in predicting outcomes in patients with IE. Methods: We included 35,570 patients with IE from the national inpatient sample (2009 - 2012). CHA 2 DS 2 - VASC score was calculated for each patient. Hierarchical two-level logistic and linear models were used to estimate outcomes for CHA 2 DS 2 - VASC scores 1 to 8, using score 0 as the referent score. Outcomes of interest were overall in-hospital mortality, cardiogenic shock, atrioventricular block (AVB) and valve replacement. All clinical characteristics were defined per the ICD-9-CM codes. Results: The mean age of the sample was 59.81 +/- 18 years. Of the 35570 patients with IE, overall mortality rate was 11.4% (n= 4038/35570), cardiogenic shock 2.6 % (n=912/35570), AVB 4% (n=1426/35570) and 10.3% (n=3652/35570) required valve replacement. Higher CHA2DS2-VASc score were associated with increased mortality from 8.1 % for a score of 0 to 21.7 % for score of 8. Cardiogenic shock increased from 5.2% for a score of 0 to 24 % for a score of 3. AVB increased from 5.5% for score of 0 to 17.5% for score of 4. While valve replacement surgery increased from 1.3% for score of 0 to 3.8% for score of 5. Figure, 1 revealed the adjusted odds ratios for outcomes per CHA 2 DS 2 - VASC score 1 to 8, using 0 as the referent score, a higher score was significantly associated with increased odds for overall mortality, cardiogenic shock, AVB, and valve replacement. Conclusion: In patients with infective endocarditis, CHA2DS2-VASc score can serve as a risk assessment tool to predict outcomes. &lt;!--EndFragment--&gt;