A brief conceptual tutorial on multilevel analysis in social epidemiology: investigating contextual phenomena in different groups of people

Variations in hip fracture inpatient care in Japan, Korea, and Taiwan: an analysis of health administrative data

BACKGROUND

Little is known about hip fracture inpatient care in East Asia. This study examined the characteristics of patients, hospitals, and regions associated with delivery of hip fracture surgeries across Japan, Korea, and Taiwan. We also analyzed and compared how the resource use and a short-term outcome of the care in index hospitals varied according to factors in the respective health systems.

METHODS

We developed comparable, nationwide, individual-level health insurance claims datasets linked with hospital- and regional-level statistics across the health systems using common protocols. Generalized linear multi-level analyses were conducted on length of stay (LOS) and total cost of index hospitalization as well as inpatient death.

RESULTS

The majority of patients were female and aged 75 or older. The standardized LOS of the hospitalization for hip fracture surgery was 32.5 (S.D. = 18.7) days in Japan, 24.7 (S.D. = 12.4) days in Korea, and 7.1 (S.D. = 2.9) days in Taiwan. The total cost per admission also widely varied across the systems. Hospitals with a high volume of hip fracture surgeries had a lower LOS across all three systems, while other factors associated with LOS and total cost varied across countries.

CONCLUSION

There were wide variations in resource use for hip fracture surgery in the index hospital within and across the three health systems with similar social health insurance schemes in East Asia. Further investigations into the large variations are necessary, along with efforts to overcome the methodological challenges of international comparisons of health system performance.

Association between access to pediatric trauma care and motor vehicle crash death in children: An ecologic analysis of United States counties

BACKGROUND

Access to pediatric trauma care is highly variable across the United States. The purpose of this study was to measure the association between pediatric trauma center care and motor vehicle crash (MVC) mortality in children (<15 years) at the US county level for 5 years (2014-2018).

METHODS

The exposure was defined as the highest level of pediatric trauma care present within each county: (1) pediatric trauma center, (2) adult level 1/2, (3) adult level 3, or (4) no trauma center. Pediatric deaths due to passenger vehicle crashes on public roads were identified from the NHTSA Fatality Analysis Reporting System. Hierarchical negative binomial modeling measured the relationship between highest level of pediatric trauma care and pediatric MVC mortality within counties. Adjusted analyses accounted for population age and sex, emergency medical service response times, helicopter ambulance availability, state traffic safety laws, and measures of rurality.

RESULTS

During the study period 3,067 children died in fatal crashes. We identified 188 pediatric trauma centers in 141 counties. Significant disparities in access to pediatric trauma care were observed. Specifically, 99% of pediatric trauma centers were situated in population-dense urban counties, while 28% of children lived in counties no trauma center. After risk adjustment, counties with pediatric trauma centers had significantly lower rates of pediatric MVC death than those with no trauma center: 0.7 versus 3.2 deaths/100,000 child-years; mortality rate ratio, 0.58; and 95% confidence interval, 0.39 to 0.86. In counties where pediatric trauma centers were absent, adult level 1/2 trauma centers were associated with comparable risk reduction.

CONCLUSION

The presence of pediatric trauma centers was associated with lower rates of MVC death in children. Adult level 1/2 trauma centers appear to offer comparable risk reduction. Where population differences in pediatric trauma mortality are observed, addressing disparities in county-level access to pediatric trauma care may serve as a viable target for system-level improvement.

LEVEL OF EVIDENCE

Epidemiological, level III; Care management, level III.

Physical Activity among Adults Residing in 11 Countries during the COVID-19 Pandemic Lockdown

During the novel coronavirus (COVID-19) pandemic, physical activity (PA) behaviors were altered worldwide due to public health measures such as “lockdown.” This study described PA among adults residing in 11 countries during COVID-19 lockdown and examined factors associated with PA engagement. We conducted a cross-sectional anonymous survey among adults (≥18 years old) in 11 countries (Brazil, Bulgaria, China, India, Ireland, Malaysia, North Macedonia, Singapore, Spain, Turkey, United States). Of 11,775 participants, 63.7% were female and 52.8% were 18–34 years old. More than 40% of participants were insufficiently active (43.9%) and reported a decrease in their PA during lockdown (44.8%). Statistically significant differences were observed in (1) proportions of participants being insufficiently active, (2) level of PA, and (3) decrease in PA across the 11 countries. More stringent governmental policy responses were associated with greater likelihood of being insufficiently active during lockdown (adjusted odds ratio = 1.22, 95% confidence interval = 1.03, 1.45). Higher depression or anxiety scores were associated with greater likelihood of decreased level of PA during lockdown.We found substantial reductions in PA levels during COVID-19 lockdown across countries. Country-specific PA promotion interventions are needed during this and similar global emergencies.

LGB prevalence in schools is associated with unhealthy weight-control behaviors in lesbian, gay, and bisexual youth: a multilevel analysis

Background

Previous studies have found that a school climate of more heteronormativity is associated with adverse effects on the mental health of LGB students. Accordingly, our aim was to assess the association between lower LGB prevalence in schools and unhealthy weight-control behaviors among LGB youth.

Methods

A cross-sectional, multilevel study based in public high schools in the city of Olinda, Northeast Brazil. A multilevel logistic regression was performed, including 2500 adolescents enrolled in 27 schools. The contextual variable was the prevalence of LGB youth in each school (as a proxy for heteronormativity in schools), while the outcome was unhealthy weight-control behaviors (fasting, purging, and taking diet pills). We controlled for socioeconomic characteristics (age, sex, receiving a family allowance), obesity, and self-reported happiness.

Results

Lower LGB prevalence in schools was associated with higher odds of engaging in unhealthy weight-control behaviors (OR: 1.5, 95%CI: 1.0, 2.2) among all youth, regardless of sexual orientation. No cross-level interactions between school context and individual characteristics were statistically significant.

Conclusion

Lower LGB prevalence in schools was associated with a higher risk of unhealthy weight-control behaviors in youth regardless of sexual orientation, which may reflect either the contextual influence of school climate, or may be due to residual confounding.

Variability in time to surgery for patients with acute thoracolumbar spinal cord injuries

There are limited data pertaining to current practices in timing of surgical decompression for acute thoracolumbar spinal cord injury (SCI). We conducted a retrospective cohort study to evaluate variability in timing between- and within-trauma centers in North America; and to identify patient- and hospital-level factors associated with treatment delay. Adults with acute thoracolumbar SCI who underwent decompressive surgery within five days of injury at participating trauma centers in the American College of Surgeons Trauma Quality Improvement Program were included. Mixed-effects regression with a random intercept for trauma center was used to model the outcome of time to surgical decompression and assess risk-adjusted variability in surgery timeliness across centers. 3,948 patients admitted to 214 TQIP centers were eligible. 28 centers were outliers, with a significantly shorter or longer time to surgery than average. Case-mix and hospital characteristics explained < 1% of between-hospital variability in surgical timing. Moreover, only 7% of surgical timing variability within-centers was explained by case-mix characteristics. The adjusted intraclass correlation coefficient of 12% suggested poor correlation of surgical timing for patients with similar characteristics treated at the same center. These findings support the need for further research into the optimal timing of surgical intervention for thoracolumbar SCI.

Tracking progress in anthropometric failure among under-five children in Ethiopia: a geospatial and multilevel analysis

Background

Undernutrition is a major public health concern among under-five children in many developing countries. This work evaluated the overall prevalence of under-nutrition by using a composite index of anthropometric failure (CIAF), which helps in the detection of children with multiple anthropometric failures. This research also includes the Spatio-temporal distribution of childhood anthropometric failures across time.

Methods

Secondary data was obtained from the Ethiopian Demographic and Health Survey for the survey 2005, 2011, and 2016 years. Data included 23,864 samples of children between the ages of 0–59 months, which is a nationally representative sample in Ethiopia. Other than descriptive statistics, the multivariate multilevel logistic regression was used to identify associated factors, and Getis-Ord spatial statistical tools were employed to identify high and low hotspots areas of anthropometric failures.

Result

The prevalence obtained with CIAF in 2005, 2011, and 2016 was, 53.5 %, 51 %, and 46.2 % of children were suffering from under-nutrition respectively. The spatial analysis revealed areas that are at a higher risk of anthropometric failures consistently were found in northern parts of the country, largely in the Amhara, Tigray, and Afar regions. Multilevel logistic regression analysis showed that the risk of anthropometric failure was higher among older children, had low birth weight, had a mother with low BMI, was in a rural area, had mothers and fathers without formal education.

Conclusions

In addition to identifying wasted, stunted, and underweight children, CIAF also identified children with multiple conditions, which are often overlooked in nutritional surveys. As revealed by this composite index, the prevalence of anthropometric failure remains considerably high and its spatial distribution also significantly varied across the regions in the country. The established socio-demographic characteristics and districts with a higher risk of anthropometric failure can be used to develop localized intervention and prevention strategies to improve Ethiopian children’s nutritional status and healthcare.

Determinants of neonatal near miss in Australia: A multilevel analysis

BACKGROUND

Neonatal Near Miss (NNM) is a situation where a newborn narrowly survived the neonatal period. It has been hypothesized that identifying factors that contribute to the occurrence of NNM and taking timely interventions could enhance the quality of newborn care. However, there is limited evidence in Australia. This study aimed to identify the determinants of NNM in Australia.

METHODS

Data from the 1973-78 cohort of the Australian Longitudinal Study on Women's Health (ALSWH) were linked with state-based Perinatal Data Collections (PDC) for 3655 mothers and 5526 newborns who were born between 01 January 2007 and 31 December 2015. A newborn was considered as a near miss case if presented with any of the pragmatic criteria (gestational age <32 weeks, birth weight <1500 g, five-minute Apgar score <7) and survived the neonatal period. A multilevel multivariable logistic regression model was used to identify the determinants of NNM.

RESULTS

Of the total 5526 live births included in this study, 95 live births met the criteria for NNM, corresponding to an incidence of 17.2 per 1000 live births. After controlling for potential confounders, maternal age 31-34 years (AOR = 2.57; 95% CI: 1.05, 6.30) and 35 years and above (AOR = 4.03; 95% CI: 1.58, 10.31), caesarean section (AOR = 2.24; 95% CI: 1.09, 4.57), and gestational hypertension (AOR = 2.63; 95% CI: 1.21, 5.71) increased the odds of NNM.

CONCLUSION

Inclusion of NNM evaluations into newborn care and early screening and interventions for women who become pregnant at older age and those with pregnancy complications could improve the quality of newborn care and reduce neonatal morbidity.

Demand for health services or professionals among adolescents: a multilevel study

This study aimed to analyze the demand for health services or professionals by adolescents with individual determinants and contextual variables (school teaching modality and Social Vulnerability Index). Data were collected from March to June 2018, in state public schools in Olinda, Pernambuco, Brazil, through a structured questionnaire with questions from the Youth Risk Behavior Survey (YRBS) and the National School Health Survey (PeNSE). Among the 2,454 adolescents, the proportion of those looking for a health service was higher among women; in adolescents whose mothers had a minimum of nine schooling years; in the physically active; and whose parents did not receive a family aid (Bolsa Família). We observed that the contextual variables did not influence the demand for health services among adolescents. In the third model of the multilevel analysis, which analyzed the individual variables, we observed that only females (OR=1,80), having a mother with schooling greater than or equal to 9 years of study (OR=1,30), not receiving Bolsa Família (OR=1,23), and physically active (OR=1,32) were associated with an increased demand for health services or professionals.

A Multilevel Analysis of Factors Associated with Childhood Diarrhea in Ethiopia

BACKGROUND

Childhood diarrhea is the major contributor to the deaths of children under the age of 5 years in Ethiopia, but evidence at the national level to identify the contributing factors associated with diarrhea by considering the clustering effects is limited. Hence, this study aimed to identify factors associated with childhood diarrhea at the individual and community levels.

METHODS

A secondary data analysis was conducted using the 2011 and 2016 Ethiopian Demographic and Health Survey (EDHS) data. A total of 23 321 children with their mothers were included in this study, and multilevel logistic regression models were applied for the data analysis.

RESULTS

The odds of diarrhea among female children were 13% lower (AOR = 0.87; 95% CI: 0.79-0.94) compared with male children. The odds of diarrhea among children aged between 13 and 24 months were 31% higher than (AOR = 1.31; 95% CI: 1.17-1.47) their younger counter parts. Children aged ⩾25 months (AOR = 0.50; 95% CI: 0.45-0.56), those whose mothers were unemployed (AOR = 0.79; 95% CI: 0.73-0.87), and children live in households between 2 and 3 under-5 children (AOR = 0.87; 95% CI: 0.79-0.96) were associated with lower odds of experiencing diarrhea. The odds of diarrhea among children whose mother had no formal education were 49% higher than (AOR = 1.49; 95% CI: 1.08-2.07) their counterparts. Besides, children residing in city administrations (AOR = 0.69; 95% CI: 0.58-0.82) had lower odds of experiencing diarrhea than children living in agrarian regions.

CONCLUSIONS

At the individual level (sex and age of the child, mother's employment status, and educational level, and the number of under-5 children) and the community-level (contextual region) were found to be significant factors associated with childhood diarrhea in Ethiopia.

Determinants of the use of nonpharmacological analgesia for labor pain management: a national population-based study

Besides neuraxial analgesia, nonpharmacological methods are also proposed to help women coping with pain during labor. We aimed to identify the individual and organizational factors associated with the use of nonpharmacological analgesia for labor pain management. Women who attempted vaginal delivery with labor analgesia were selected among participants included in the 2016 National Perinatal Survey, a population-based cross-sectional study. Labor analgesia was studied as neuraxial analgesia alone, nonpharmacological analgesia alone, and neuraxial and nonpharmacological analgesia combined. The associations were studied using multilevel multinomial logistic regression. Among the 9231 women included, 62.4% had neuraxial analgesia alone, 6.4% had nonpharmacological analgesia alone, and 31.2% had both. Nonpharmacological analgesia alone or combined with neuraxial analgesia were both associated with high educational level (adjusted odds ratio 1.55; 95% confidence interval [CI], 1.08-2.23 and 1.39; 95% CI, 1.18-1.63), antenatal preference to deliver without neuraxial analgesia, and public maternity unit status. Nonpharmacological analgesia alone was more frequent among multiparous women, and in maternity units with an anesthesiologist not dedicated to delivery unit (1.57; 95% CI, 1.16-2.12) and with the lowest midwife workload (2.15; 95% CI, 1.43-3.22). Neuraxial and nonpharmacological analgesia combined was negatively associated with inadequate prenatal care (0.70; 95% CI, 0.53-0.94). In France, most women who had nonpharmacological analgesia during labor used it as a complementary method to neuraxial analgesia. The use of nonpharmacological analgesia combined with neuraxial analgesia mainly depends on the woman's preference, but also on socioeconomic factors, quality of prenatal care, and care organization.

Households access to improved drinking water sources and toilet facilities in Ethiopia: a multilevel analysis based on 2016 Ethiopian Demographic and Health Survey

OBJECTIVE

This study aimed to assess households access to improved drinking water sources and sanitation facilities and their associated factors in Ethiopia.

DESIGN

Cross-sectional study.

SETTING

Ethiopia.

PARTICIPANTS

Household heads.

PRIMARY OUTCOMES

Access to improved drinking water sources and toilet facilities.

METHODS

We conducted an in-depth secondary data analysis of 2016 Ethiopian Demographic and Health Survey. Data from a total of 16 650 households and 645 clusters were included in the analysis. The households were selected using a stratified two-stage cluster sampling technique. Multilevel binary logistic regression analyses were performed to identify factors associated with access to an improved drinking water source and toilet facilities. Adjusted OR with a 95% CI was reported with p value <0.05 was used to declare a significant association between the covariates and the outcome variables.

RESULTS

The proportions of households' access to improved sources of drinking water and toilet facilities were 69.94% (95% CI: 69.23% to 70.63%) and 25.36% (95% CI: 24.69% to 26.03%), respectively. Households headed by women and households with a better wealth index were positively associated with access to improved drinking water sources. Whereas rural households, ≥30 min round trip to obtain drinking water and region were factors negatively associated with households access to improved drinking water sources. A higher probability of having access to improved toilet facilities: households with heads who had attained higher education, households having better access to improved sources of drinking water and households with better wealth index. While the following households were less likely to have access to improved toilet facilities: households with heads were widowed, households with four to six members, rural households and region.

CONCLUSION

The study found that the proportions of households' access to improved drinking water sources and toilet facilities in Ethiopia were relatively low, which demands the need to tailor strategies to increase the coverage of access to improved drinking water sources and toilet facilities.

Variation in the use of percutaneous coronary interventions among older patients with acute coronary syndromes: a multilevel study in Fukuoka, Japan

BACKGROUND

Variation in health care delivery among regions and hospitals has been observed worldwide and reported to have resulted in health inequalities. Regional variation of percutaneous coronary intervention (PCI) was previously reported in Japan. This study aimed to assess the small-area and hospital-level variations and to examine the influence of patient and hospital characteristics on the use of PCI.

METHODS

Data provided by the Fukuoka Prefecture Latter-stage Elderly Insurance Association was used. There were 11,821 patients aged ≥65 years with acute coronary syndromes who were identified from 2015 to 2017. Three-level multilevel logistic regression analyses were performed to quantify the small-area and hospital variations, as well as, to identify the determinants of PCI use.

RESULTS

The results showed significant variation (δ² = 0.744) and increased PCI use (MOR = 2.425) at the hospital level. After controlling patient- and hospital-level characteristics, a large proportional change in cluster variance was found at the hospital level (PCV 14.7%). Fixed-effect estimation results showed that females, patients aged ≥80 years old, hypertension and dyslipidemia had significant association with the use of PCI. Hospitals with high physician density had a significantly positive relationship with PCI use.

CONCLUSIONS

Patients receiving care in hospitals located in small areas have equitable access to PCI. Hospital-level variation might be originated from the oversupply of physicians. A balanced number of physicians and beds should be taken into consideration during healthcare allocation. A treatment process guideline on PCI targeting older patients is also needed to ensure a more equitable access for healthcare resources.

Early Marriage and Its Determinants among Married Reproductive Age Group Women in Amhara Regional State, Ethiopia: A Multilevel Analysis

Introduction

Amhara region has one of the highest rates of female child early marriage in Ethiopia, with eighty percent of girls in the region being married at the age of eighteen. Therefore, this study was intended to assess the prevalence and determinants of early marriage among women, in Amhara regional state.

Methods

The data were extracted from the 2016 Ethiopian Demographic and Health Survey. The study included a sample of 2887 (weighted) married women from 645 clusters in Amhara regional state. The data were collected using a two-stage cluster design that includes the selection of enumeration areas as a first stage and selection of households as a second stage. A multilevel logistic regression model was fitted to determine the individual and community-level factors associated with early marriage.

Result

The study revealed that 73% [95% CI 71.38, 74.62] of women aged 15–49 years were married before 18 years old. In the multilevel multivariable model; living as a rural dweller (AOR = 4.33; 95% CI: 2.17, 8.64), no education (AOR = 2.52; 95% CI: 2.23, 9.51), attending only primary education (AOR = 2.31; 95% CI: 1.68, 8.53), parental decision-maker when to get marriage (AOR = 3.44; 95% CI: 2.20, 5.39), being poorer (AOR = 1.38; 95% CI: 1.16, 4.83), and poorest wealth status (AOR = 2.37; 95% CI: 2.19, 7.83) were the independent predictors of early marriage.

Conclusion

The prevalence of early marriage was high in Amhara region compared to other regions of the country. Therefore, the regional government should give due attention to access to education and encourage women's decision-making power upon the time of marriage especially those residing in rural parts of the region.

Mental Health among Adults during the COVID-19 Pandemic Lockdown: A Cross-Sectional Multi-Country Comparison

Despite the global impact of COVID-19, studies comparing the effects of COVID-19 on population mental health across countries are sparse. This study aimed to compare anxiety and depression symptoms during the COVID-19 lockdown among adults from 11 countries and to examine their associations with country-level COVID-19 factors and personal COVID-19 exposure. A cross-sectional survey was conducted among adults (≥18 years) in 11 countries (Brazil, Bulgaria, China, India, Ireland, North Macedonia, Malaysia, Singapore, Spain, Turkey, United States). Mental health (anxiety, depression, resilient coping, hope) and other study data were collected between June–August 2020. Of the 13,263 participants, 62.8% were female and 51.7% were 18–34 years old. Participants living in Brazil had the highest anxiety and depression symptoms while participants living in Singapore had the lowest. Greater personal COVID-19 exposure was associated with increased anxiety and depression symptoms, but country-level COVID-19 factors were not. Higher levels of hope were associated with reduced anxiety and depression; higher levels of resilient coping were associated with reduced anxiety but not depression. Substantial variations exist in anxiety and depression symptoms across countries during the COVID-19 lockdown, with personal COVID-19 exposure being a significant risk factor. Strategies that mitigate COVID-19 exposure and enhance hope and resilience may reduce anxiety and depression during global emergencies.

Assessing the Impact of Individual Characteristics and Neighborhood Socioeconomic Status During the COVID-19 Pandemic in the Provinces of Milan and Lodi

Social inequalities in health are known to be influenced by the socioeconomic status of the territory in which people live. In the context of the ongoing coronavirus disease 2019 (COVID-19) pandemic, this study is aimed at assessing the role of 5 area-level indicators in shaping the risk of contagion in the provinces of Milan and Lodi (Lombardy, Italy), namely: educational disadvantage, unemployment, housing crowding, mobility, and population density. The study area includes the municipalities at the origin of the first Italian epidemic outbreak. Data on COVID-19 patients from the Integrated Datawarehouse for COVID Analysis in Milan were used and matched with aggregate-level data from the National Institute of Statistics Italy (Istat). Multilevel logistic regression models were used to estimate the association between the census block-level predictors and COVID-19 infection, independently of age, sex, country of birth, and preexisting health conditions. All the variables were significantly associated with the outcome, with different effects before and after the lockdown and according to the province of residence. This suggests a pattern of socioeconomic inequalities in the outbreak, which should be taken into account in the eventuality of future epidemics to contain their spread and its related disparities.

Guideline-Concordant Insulin Infusion Initiation Among Critically Ill Patients With Sepsis

OBJECTIVE

Our objective was to benchmark rates of guideline-concordant insulin infusion initiation, identify factors associated with guideline-concordant insulin practices, and examine the association between hospital-level guideline concordance and mortality among critically ill patients with sepsis.

METHODS

We performed a multicenter retrospective cohort study of intensive care patients with sepsis who were eligible for insulin infusion initiation according to American Diabetes Association and Surviving Sepsis guidelines (persistent blood sugar ≥180 mg/dL). We then identified patients who were initiated on insulin infusions within 24 hours of eligibility. We examined patient- and hospital-level factors associated with guideline-concordant insulin infusion initiation and explored the association between the hospital-level proportion of patients who received guideline-concordant insulin infusions and hospital mortality.

RESULTS

Among 5453 guideline-eligible patients with sepsis, 13.4% were initiated on insulin infusions. Factors most strongly associated with guideline-concordant insulin infusion initiation were mechanical ventilation and hospital of admission. The hospital-level proportion of patients who received guideline-concordant insulin infusions were not associated with mortality. Among 1501 intensive care unit patients with sepsis who were started on insulin infusions, 37.0% were initiated at a blood glucose level below 180 mg/dL, the guideline-recommended starting threshold.

CONCLUSION

Guideline-concordant insulin infusion initiation was uncommon among patients with sepsis admitted to U.S. intensive care units and was determined in large part by hospital of admission. The degree to which hospitals were guideline-concordant were not associated with mortality.

Individual and Area-level Factors Contributing to the Geographic Variation in Ambulatory Care Sensitive Conditions in Finland: A Register-based Study

BACKGROUND

Measuring primary health care (PHC) performance through hospitalizations for ambulatory care sensitive conditions (ACSCs) remains controversial-recent cross-sectional research claims that its geographic variation associates more with individual socioeconomic position (SEP) and health status than PHC supply.

OBJECTIVES

To clarify the usage of ACSCs as a PHC performance indicator by quantifying how disease burden, both PHC and hospital supply and spatial access contribute over time to geographic variation in Finland when individual SEP and comorbidities were adjusted for.

METHODS

The Finnish Care Register for Health Care provided hospitalizations for ACSCs (divided further into subgroups of acute, chronic, and vaccine-preventable causes) in 2011-2017. With 3-level nested multilevel Poisson models-individuals, PHC authorities, and hospital authorities-we estimated the proportion of the variance in ACSCs explained by selected factors at 3 time periods.

RESULTS

In age-adjusted and sex-adjusted analysis of total ACSCs the variances between hospital authorities was nearly twice that between PHC authorities. Individual SEP and comorbidities explained 19%-30% of the variance between PHC authorities and 25%-36% between hospital authorities; and area-level disease burden and arrangement and usage of hospital care a further 14%-16% and 32%-33%-evening out the unexplained variances between PHC and hospital authorities.

CONCLUSIONS

Alongside individual factors, areas' disease burden and factors related to hospital care explained the excess variances in ACSCs captured by hospital authorities. Our consistent findings over time suggest that the local strain on health care and the regional arrangement of hospital services affect ACSCs-necessitating caution when comparing areas' PHC performance through ACSCs.

Spatiotemporal variations in ischemic heart disease mortality and related risk factors in China between 2010 and 2015: a multilevel analysis

Background

To explore the relationship between geographical differences of mortality and related risk factors in ischemic heart disease (IHD) in China.

Methods

Data were collected from the nationally representative China Mortality Surveillance System to calculate annual IHD mortality counts (2010–2015). Descriptive analysis was used to analyze the IHD mortality among Chinese population from 2010 to 2015. Negative binomial regression was used to investigate potential spatiotemporal variation and correlations with age, gender, urbanization, and region.

Results

The overall IHD mortality was 221.17/100,000, accounting for 1.51 million deaths in 2015. The standardized IHD mortality rate increased by 5.51% from 2010 to 2015 among people aged 40 years and older. Multilevel analysis indicated significant differences in gender, regions, and age. High urbanization rate (risk ratio [RR] = 0.728, 95% confidence interval [CI] = (0.631, 0.840)) and average high-density lipoprotein (HDL) (RR = 0.741, 95%CI: 0.616,0.891) were negatively associated with IHD mortality. IHD mortality was significantly higher in populations with a low rate of medical insurance coverage (RR = 1.218, 95%CI: 1.007, 1.473), as well as the average body mass index (BMI) (RR = 1.436, 95%CI: 1.135, 1.817) and systolic blood pressure (SBP) (RR = 1.310, 95%CI: 1.019, 1.684). While the relationship with current smoking rate, excessive intake of red meat, insufficient vegetable or fruits intake didn’t show the statistical significance. The negative correlation between the average sedentary time and IHD mortality was not conclusive due to the possible deviation of the data.

Conclusions

The mortality of IHD showed an upward trend for people aged 40 years and older in China during 2010–2015, which should be paid attention to. Therefore, some risk factors should be controlled, such as SBP, overweight/obesity. HDL is a protective factor, as well as higher urbanization rate, family income level, and medical insurance coverage.

Factors associated with anaemia among women of reproductive age in Ethiopia: Multilevel ordinal logistic regression analysis

Anaemia has prevailed as a mild to severe public health problem in Ethiopian women of reproductive age. Many studies carried out on anaemia have been limited to subnational assessments and subgroups of women. The effects of potential factors thought to affect anaemia and severity levels of anaemia have not been well considered. Therefore, this study identifies individual, household and community level factors associated with anaemia among women of reproductive age in Ethiopia applying multilevel ordinal logistic regression models. Proportional odds assumption was tested by likelihood ratio test. About 35.6% of the variation on anaemia was due to between household and community level differences. Pregnancy (adjusted odds ratio [AOR] = 2.30, 95% confidence interval [CI]: 1.82, 2.91), HIV (AOR = 2.40, 95% CI: 1.76, 3.25), giving birth once (AOR = 1.2, 95% CI: 1.05, 1.40), giving birth more than once (AOR = 1.4, 95% CI: 1.19, 1.71), living with five or more family members (AOR = 1.24, 95% CI: 1.05, 1.47), living in poorest households (AOR = 1.34, 95% CI: 1.2, 1.61) and rural area (AOR = 1.57, 95% CI: 1.28, 1.92) were associated with greater odds of more severe anaemia compared with their respective counter parts. Secondary and above education (AOR = 0.83, 95% CI: 0.70, 0.97) and use of pills, implants or injectable (AOR = 0.67, 95% CI: 0.59, 0.77) were associated with lower odds of more severe anaemia. Anaemia prevention and control programmes need to be strengthened for women living with HIV/AIDS and during pregnancy. Household poverty reduction and social protection services need to be strengthened and integrated in anaemia prevention and management activities in women.

Individual- and country-level correlates of female permanent contraception use in sub-Saharan Africa

Background

Female permanent contraception is a cost-effective contraceptive method that can help clients with the desire to limit childbearing achieve their reproductive intention. However, despite its benefits, the use of FPC remains low in sub-Saharan Africa (SSA), and limited studies have examined the correlates of its uptake. In this study, we assessed the individual- and country-level factors associated with the use of FPC among married or in-union women using modern contraceptive methods to limit childbearing in SSA.

Methods

This study was a secondary data analysis of individual- and country-level data obtained from the Demographic and Health Surveys (DHS) Program and three open data repositories. The study included 29,777 married or in-union women aged 15–49 years using modern contraceptive methods to limit childbearing from DHS conducted in 33 sub-Sahara African countries between 2010 and 2018. We performed descriptive statistics and fitted multilevel logistic regression models to determine the predisposing, enabling, and need factors associated with the use of FPC.

Results

Approximately 13% of the women used FPC. About 20% of the variance in the odds of using FPC was attributable to between-country differences. In the full model, the significant individual-level factors associated with the use of FPC compared with other modern contraceptive methods were: age (odds ratio [OR] = 1.10; 95%CI = 1.08–1.12), living children (OR = 1.11, 95%CI = 1.04–1.16), high household wealth (OR = 1.39, 95%CI = 1.18–1.64), rural residence (OR = 0.83, 95% CI = 0.71–0.97), joint contraceptive decision with partner (OR = 1.68, 95% = 1.43–1.99), contraceptive decision by partner and others (OR = 2.46, 95% = 1.97–3.07), and the number of living children less than the ideal number of children (OR = 1.40, 95%CI = 1.21–1.62). The significantly associated country-level factors were births attended by skilled health providers (OR = 1.03, 95%CI = 1.00–1.05) and density of medical doctors (OR = 1.37, 95%CI = 1.01–1.85).

Conclusions

Our results suggest that both individual- and country-level factors affect uptake of FPC in SSA. Increasing geographic, economic, and psychosocial access to FPC may improve its uptake in SSA.

Investigating the association of health system characteristics and health care utilization: a multilevel model in China's ageing population

Background

To achieve universal health coverage in China, it is necessary to identify access barriers to health care. This study examined the association between health system characteristics and health care utilization in China and identified factors associated with accessing health care among the mid-aged and elderly.

Methods

Data were obtained from the 2015 China Health and Retirement Longitudinal Study, and 17 370 respondents aged 45 and above were included in the analysis. The dependent variables were the use of outpatient and inpatient care among respondents. Health system characteristics at the provincial level were measured using the density of doctors and ward beds, health expenditure per visit/admission and health financing. A two-level logistic regression model was constructed to examine association between health care utilization and health system characteristics, controlling for predisposing, enabling and need variables.

Results

Of the 17 370 respondents, 18.3% had utilized outpatient care and 13.7% had utilized inpatient care in 2015. Increases in the share of out-of-pocket (OOP) payments as total health spending at the provincial-level was less likely to be associated with outpatient care utilization (odds ratio (OR) = 0.96, 95% confidence interval (CI) = 0.93-0.98) among the mid-age and elderly population. Increases in the share of OOP payments (OR = 0.98, 95% CI = 0.97-1.00) and health expenditure per admission (OR = 0.20, 95% CI = 0.04-0.88) were less likely to be associated with inpatient care utilization, while increases in the density of beds (OR = 1.26, 95% CI = 1.10-1.43) was more likely to be associated with inpatient care utilization. gross domestic product (GDP) per capita at the provincial level and types of health insurance owned by respondents were significantly related to both inpatient and outpatient care utilization.

Conclusions

Low affordability of the mid-aged and elderly population is the main barrier to utilizing health care in China. In order to improve access to health care, the government should make more efforts, such as improving health insurance reimbursement rates and implementing prospective provider payment methods, to decrease OOP payment for the ageing population.

Prevalence and predictors of alcohol use among adult males in Ethiopia: multilevel analysis of Ethiopian Demographic and Health Survey 2016

BACKGROUND

Alcohol is a psychoactive substance that is widely consumed in the world. Alcohol use is one of the world's leading risk factors for disease and disability. It affects individuals' physical, mental, economic, and social issues. To our knowledge, there is limited study on alcohol consumption and associated factors. Therefore, this study aimed to determine the prevalence and predictors of alcohol use in Ethiopia by using the 2016 Ethiopian Demographic and Health Survey.

METHODS

This study was based on the most recent Ethiopian Demographic and Health Survey 2016. A total of 12,594 men at the age of 15 to 59 were included in this study. Considering the hierarchical nature of EDHS data, a multilevel logistic regression model was applied. The ICC, MOR, and the LR test were done to assess the presence of a significant clustering effect. Besides, deviance was used for model comparison since the models were nested models. Variables with a p value ≤ 0.2 in the bivariable analysis were considered for the multivariable analysis. In the multilevel logistic regression, the adjusted odds ratio (AOR) with 95% confidence interval (CI) was reported to declare the strength and significance of the association between the dependent variable and independent variables.

RESULTS

The prevalence of alcohol drinking in this study was 46.64% with a 95% CI of 45.00 to 47.00%. Age groups 30-44 (AOR = 1.30, 95% CI 1.08, 1.56) and 45-59 (AOR = 1.38, 95% CI 1.10, 1.74), Orthodox religion follower (AOR = 0.36, 95% CI 0.24, 0.55), media exposure (AOR = 1.67, 95% CI 1.41, 2.20), khat chewing (AOR = 3.08, 95% CI 2.54, 3.74), smoking (AOR = 2.18, 95% CI 1.71, 2.79), having no occupation (AOR = 0.34, 95% CI 0.22, 0.51), and region were the predictors of alcohol use in Ethiopia.

CONCLUSIONS

Nearly half of the Ethiopian population reported alcohol use at least once in their lifetime. Old age, Orthodox religion followers, media exposure, khat chewing, smoking, and having no occupation were predictors of alcohol use in Ethiopia. Therefore, health education about the risk of alcohol used is highly recommended. In addition, khat chewing and smoking control mechanisms should be designed and given special attention. Advertising alcohol through media is better to be controlled. Job opportunities should also be created for those who have no occupation to mitigate alcohol use in Ethiopia.

Spatial pattern of perinatal mortality and its determinants in Ethiopia: Data from Ethiopian Demographic and Health Survey 2016

BACKGROUND

The perinatal mortality rate in Ethiopia is among the highest in Sub Saharan Africa. The aim of this study was to identify the spatial patterns and determinants of perinatal mortality in the country using a national representative 2016 Ethiopia Demographic and Health Survey (EDHS) data.

METHODS

The analysis was completed utilizing data from 2016 Ethiopian Demographic and Health Survey. This data captured the information of 5 years preceding the survey period. A total of 7230 women who at delivered at seven or more months gestational age nested within 622 enumeration areas (EAs) were used. Statistical analysis was performed by using STATA version 14.1, by considering the hierarchical nature of the data. Multilevel logistic regression models were fitted to identify community and individual-level factors associated with perinatal mortality. ArcGIS version 10.1 was used for spatial analysis. Moran's, I statistics fitted to identify global autocorrelation and local autocorrelation was identified using SatSCan version 9.6.

RESULTS

The spatial distribution of perinatal mortality in Ethiopia revealed a clustering pattern. The global Moran's I value was 0.047 with p-value <0.001. Perinatal mortality was positively associated with the maternal age, being from rural residence, history of terminating a pregnancy, and place of delivery, while negatively associated with partners' educational level, higher wealth index, longer birth interval, female being head of household and the number of antenatal care (ANC) follow up.

CONCLUSIONS

In Ethiopia, the perinatal mortality is high and had spatial variations across the country. Strengthening partner's education, family planning for longer birth interval, ANC, and delivery services are essential to reduce perinatal mortality and achieve sustainable development goals in Ethiopia. Disparities in perinatal mortality rates should be addressed alongside efforts to address inequities in maternal and neonatal healthcare services all over the country.

Cross-classified Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) to evaluate hospital performance: the case of hospital differences in patient survival after acute myocardial infarction

OBJECTIVE

To describe a novel strategy, Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) to evaluate hospital performance, by analysing differences in 30-day mortality after a first-ever acute myocardial infarction (AMI) in Sweden.

DESIGN

Cross-classified study.

SETTING

68 Swedish hospitals.

PARTICIPANTS

43 247 patients admitted between 2007 and 2009, with a first-ever AMI.

PRIMARY AND SECONDARY OUTCOME MEASURES

We evaluate hospital performance by analysing differences in 30-day mortality after a first-ever AMI using a cross-classified multilevel analysis. We classified the patients into 10 categories according to a risk score (RS) for 30-day mortality and created 680 strata defined by combining hospital and RS categories.

RESULTS

In the cross-classified multilevel analysis the overall RS adjusted hospital 30-day mortality in Sweden was 4.78% and the between-hospital variation was very small (variance partition coefficient (VPC)=0.70%, area under the curve (AUC)=0.54). The benchmark value was therefore achieved by all hospitals. However, as expected, there were large differences between the RS categories (VPC=34.13%, AUC=0.77) CONCLUSIONS: MAIHDA is a useful tool to evaluate hospital performance. The benefit of this novel approach to adjusting for patient RS is that it allowed one to estimate separate VPCs and AUC statistics to simultaneously evaluate the influence of RS categories and hospital differences on mortality. At the time of our analysis, all hospitals in Sweden were performing homogeneously well. That is, the benchmark target for 30-day mortality was fully achieved and there were not relevant hospital differences. Therefore, possible quality interventions should be universal and oriented to maintain the high hospital quality of care.

Digital support principles for sustained mathematics learning in disadvantaged students

This study addresses the pressing issue of how to raise the performance of disadvantaged students in mathematics. We combined established findings on effective instruction with emerging research addressing the specific needs of disadvantaged students. A sample of N = 260 disadvantaged 6th-graders received 4 weeks (15 lessons) of fraction instruction either as usual or evidence-based instruction, with and without digital learning support (i.e., interactivity, adaptivity, and immediate explanatory feedback). To examine the sustainability of effects, we assessed students' fraction knowledge immediately after the 4 weeks and once again after a period of additional 8 weeks. Generalized linear mixed models revealed that students only benefitted from evidence-based instruction if digital support was available in addition. Digital support principles implemented in evidence-based instruction helped disadvantaged students to acquire mathematics knowledge-and to maintain this knowledge.

Multilevel influences of women's empowerment and economic resources on risky sexual behaviour among young women in Zomba district, Malawi

Gender disparities are pronounced in Zomba district, Malawi. Among women aged 15-49 years, HIV prevalence is 16.8%, compared with 9.3% among men of the same age. Complex structural factors are associated with risky sexual behaviour leading to HIV infection. This study's objective was to explore associations between multilevel measures of economic resources and women's empowerment with risky sexual behaviour among young women in Zomba. Four measures of risky sexual behaviour were examined: ever had sex, condom use and two indices measuring age during sexual activity and partner history. Multilevel regression models and regression models with cluster-robust standard errors were used to estimate associations, stratified by school enrolment status. Among the schoolgirl stratum, the percentage of girls enrolled in school at the community level had protective associations with ever having sex (OR = 0.76; 95% CI: 0.60, 0.96) and condom use (OR = 1.06; 95% CI: 1.01, 1.11). Belief in the right to refuse sex was protective against ever having sex (OR = 0.76; 95% CI: 0.60, 0.96). Participants from households with no secondary school education had higher odds of ever having sex (OR = 1.59; 95% CI: 1.14, 2.22). Among the dropout stratum, participants who had not achieved a secondary school level of education had riskier Age Factor and Partner History Factor scores (β = 0.51; 95% CI: 0.23, 0.79, and β = 0.24; 95% CI: 0.07, 0.41, respectively). Participants from households without a secondary school level of education had riskier Age Factor scores (β = 0.26; 95% CI: 0.03, 0.48). Across strata, the most consistent variables associated with risky sexual behaviour were those related to education, including girl's level of education, highest level of education of her household of origin and the community percentage of girls enrolled in school. These results suggest that programmes seeking to reduce risky sexual behaviour among young women in Malawi should consider the role of improving access to education at multiple levels.

Spatial variations and associated factors of modern contraceptive use in Ethiopia: a spatial and multilevel analysis

OBJECTIVE

To assess spatial variations in modern contraceptive use and to identify factors associated with it among married women in Ethiopia.

DESIGN

Cross-sectional analysis of population-based and health facility data.

SETTING

Ethiopia Demographic and Health Survey data linked to Service Provision Assessment data.

POPULATION

8473 married women and 1020 facilities that reported providing family planning services.

METHODS

A linked secondary data analysis of population and health facility data was carried out. Both multilevel and spatial analyses were conducted to identify key determinants of women's use of modern contraceptive and spatial clustering of modern contraceptive use.

MAIN OUTCOME MEASURE

Modern contraceptive use.

RESULTS

About 24% of the variation in the use of modern contraception was accounted for by location. A one-unit increase in the mean score of health facilities' readiness to provide short-term modern contraceptives in a typical region was associated with a 20-fold increase in the odds of modern contraceptive use (adjusted OR (AOR) 20.49, 95% CI 1.44 to 29.54). In the spatial analysis, it was found that Addis Ababa and the Amhara region had high clusters of modern contraceptive use rates. On the other hand, low rates of contraceptive use were clustered in the Afar and Somali regions.

CONCLUSION

There were significant variations in the use of modern contraceptives across the different regions of Ethiopia. Therefore, regions with low contraceptive rates and high fertility rates should be targeted for scaling up and tailoring of services to the culture and lifestyles of the population of those regions.

Association of remote imaging photoplethysmography and cutaneous perfusion in volunteers

Remote imaging photoplethysmography (iPPG) senses the cardiac pulse in outer skin layers and is responsive to mean arterial pressure and pulse pressure in critically ill patients. Whether iPPG is sufficiently sensitive to monitor cutaneous perfusion is not known. This study aimed at determining the response of iPPG to changes in cutaneous perfusion measured by  Laser speckle imaging (LSI). Thirty-seven volunteers were engaged in a cognitive test known to evoke autonomic nervous activity and a Heat test. Simultaneous measurements of iPPG and LSI were taken at baseline and during cutaneous perfusion challenges. A perfusion index (PI) was calculated to assess iPPG signal strength. The response of iPPG to the challenges and its relation to LSI were determined. PI of iPPG significantly increased in response to autonomic nervous stimuli and to the Heat test by 5.8% (p = 0.005) and 11.1% (p < 0.001), respectively. PI was associated with LSI measures of cutaneous perfusion throughout experiments (p < 0.001). iPPG responses to study task correlated with those of LSI (r = 0.62, p < 0.001) and were comparable among subjects. iPPG is sensitive to autonomic nervous activity in volunteers and is closely associated with cutaneous perfusion.

Examining Individual- and Community-Level Factors Affecting Skilled Delivery Care among Women Who Received Adequate Antenatal Care in Ethiopia: Using Multilevel Analysis

Introduction

Maternal mortality continues to be a major public health and development challenge in Africa even after the permissible commitment of the international community. Although the use of skilled delivery care is the key intervention and is effective to lower maternal mortality rates, it is still at a lower proportion. The study is aimed at investigating the individual- and community level factors affecting the use of skilled delivery care among those women who had received adequate antenatal care.

Materials and Methods

Data were extracted from the 2016 Ethiopian Demographic and Health Survey on women aged 15-49 years and gave birth within five years prior to the survey (N = 957). Multilevel logistic regression model with two levels were fitted to assess the influence of the individual- and community-level factors on the use of skilled delivery care.

Results

Women who were exposed to media were more likely to use skilled delivery care (OR = 1.81; 95% CI: 1.20-2.74). Having six or more birth order (OR = 0.33; 95% CI: 0.16-0.69) and residing in rural areas (OR = 0.40; 95% CI: 0.21-0.79) were associated with less likelihood use of skilled delivery care. Attaining primary and secondary educational level, being older women, being from the richest household, and having a urine test during antenatal visits were significantly associated with the use of skilled delivery care. The value of intraclass correlation coefficient supported a significant community-level effect on the likelihood of using skilled delivery care.

Conclusions

Factors operating both at the individual level and community level were found significantly associated with the use of skilled delivery care in Ethiopia. A considerable variation at community level accounts for the difference in the use of skilled delivery level.

Education and income-based inequality in tooth loss among Brazilian adults: does the place you live make a difference?

BACKGROUND

Socioeconomic inequalities in tooth loss might be minimized or potentialized by the characteristics of the context where people live. We examined whether there is contextual variation in socioeconomic inequalities in tooth loss across Brazilian municipalities.

METHODS

Data from the 2010 National Oral Health Survey of 9633 adults living in 157 Brazilian municipalities were used. The individual socioeconomic indicators were education and household income. At the municipal level, we used the Municipal Human Development Index (HDI) as our contextual indicator of socioeconomic status (low:< 0.699 versus high: > 0.70). The Relative (RII) and Slope (SII) Indexes of Inequality, Relative (RCI), and Absolute (ACI) Concentration Indexes were calculated to compare the magnitude of education and income-based inequalities among municipalities with low versus high HDI. Multilevel Poisson regression models with random intercepts and slopes were developed.

RESULTS

At the individual level, adults with lower education & income reported more tooth loss. The mean number of missing teeth was 9.52 (95% CI: 7.93-11.13) and 6.95 (95% CI: 6.43-7.49) in municipalities with low and high HDI, respectively. Municipalities with high HDI showed higher relative and absolute education-based inequality. For income-based inequalities, higher SII and RCI was observed in municipalities with lower HDI. A significant cross-level interaction indicated that high-education adults reported fewer missing teeth when they lived in municipalities with high HDI compared to adults with the same education level living in low HDI municipalities. For individuals with the lowest education level, there was no difference in the number of teeth between those from municipalities with high and low HDI.

CONCLUSIONS

There was a social gradient in tooth loss by education and income. Living in disadvantaged municipalities cannot overcome the risk associated with low schooling. The protective effect of higher education can be reduced when people live in disadvantaged areas.

Factors explaining regional variation in under-five mortality in India: An evidence from NFHS-4

Although child mortality has declined in India, pronounced regional and socioeconomic inequality exists. The study examines the effects of individual- and community-level characteristics on under-five mortality and investigates the extent to which they affect regional variation. The study is based on Indian National Family and Health Survey 4, 2015-16 data. A two-level logistic regression model was performed to examine the effects of the socio-economic characteristics, and multivariate decomposition analysis was done to assess the contribution of factors in the inter-regional under-five mortality differentials. Regional variation in the selected variables is observed. For instance, the percentage of children with small childbirth size varied from 9.7% in the southern to 21.6% in the northeastern region. The percentage of poor households, low educated mothers, and childbirths delivered at home facility were higher in the central and eastern region compared to the southern region. The multilevel analysis shows that the region of residence explained 15.8% variance, and community-level characteristics alone could explain 25.3% variation in the risk of under-five deaths. The decomposition analysis indicates that the average number of excess deaths in the central and eastern regions is higher compared to the other regions. The compositional differences account for 50.9% of the under-five mortality gaps between the south and north region, 80.9% of the gap between the south and east, and 42.9% of the gap between the south and central region of India. Special attention and targeted action are needed to address the underlying causes of low birth weight of children in all the regions of India. Region-specific interventions might be priorities; for example, north, and central regions, need an economic and educational uplift, while infrastructural and economic policies should be prioritized for the northeastern region. Also, region-specific community-level interventions are needed to improve child survival in India.

Geographical and sociodemographic differences in discontinuation of medication for Chronic Obstructive Pulmonary Disease - A Cross-Classified Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA)

BACKGROUND

While discontinuation of COPD maintenance medication is a known problem, the proportion of patients with discontinuation and its geographical and sociodemographic distribution are so far unknown in Sweden. Therefore, we analyse this question by applying an innovative approach called multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA).

PATIENTS AND METHODS

We analysed 49,019 patients categorized into 18 sociodemographic contexts and 21 counties of residence. All patients had a hospital COPD diagnosis and had been on inhaled maintenance medication during the 5 years before the study baseline in 2010. We defined "discontinuation" as the absolute lack of retrieval from a pharmacy of any inhaled maintenance medication during 2011. We performed a cross-classified MAIHDA and obtained the average proportion of discontinuation, as well as county and sociodemographic absolute risks, and compared them with a proposed benchmark value of 10%. We calculated the variance partition coefficient (VPC) and the area under the receiver operating characteristics curve (AUC) to quantify county and sociodemographic differences. To summarize the results, we used a framework with 15 scenarios defined by the size of the differences and the level of achievement in relation to the benchmark value.

RESULTS

Around 18% of COPD patients in Sweden discontinued maintenance medication, so the benchmark value was not achieved. There were very small county differences (VPC=0.35%, AUC=0.54). The sociodemographic differences were small (VPC=4.98%, AUC=0.57).

CONCLUSION

Continuity of maintenance medication among COPD patients in Sweden could be improved by reducing the unjustifiably high prevalence of discontinuation. The very small county and small sociodemographic differences should motivate universal interventions across all counties and sociodemographic groups. Geographical analyses should be combined with sociodemographic analyses, and the cross-classified MAIHDA is an appropriate tool to assess health-care quality.

Measuring the completeness of death registration in 2844 Chinese counties in 2018

BACKGROUND

Death registration completeness has never been assessed at the county level in China. Such analyses would provide critical intelligence to monitor the performance of the vital registration system and yield adjustment factors to correct death registration data, thereby increasing their policy utility.

METHODS

We estimated the completeness of death registration for 31 provinces and 2844 counties of China in 2018 based on death data from the China Cause of Death Reporting System (CDRS) by using the empirical completeness method. We computed the root mean square difference (RMSD) of county-level completeness compared with provincial-level completeness to study intra-provincial variations. A two-level (province and county) logistic regression model was fitted to explore the association between county-level registration completeness and a set of covariates reflecting socioeconomic status, healthcare quality, and specific strategies and regulations designed to improve registration.

RESULTS

In 2018, the overall death registration completeness for the CDRS in China was 74.2% (95% uncertainty interval [UI] 66.2-80.4), with very little difference for males and females. Geographical differences in completeness were higher across counties than across provinces. The county-level completeness ranged from 2.4% (95% UI 1.0-5.0%) in Burang County, Tibet, to 100.0% (95% UI 99.9-100.0%) in Guandu District, Yunnan. The coastal provinces of Jiangsu, Guangdong, and Fujian, with higher overall completeness, contained counties with low completeness; conversely, the underdeveloped provinces of Guangxi and Guizhou, with lower overall completeness, included some counties with high completeness. GDP, education, population density, minority population, healthcare access, and registration strategies were important drivers of the geographical differences in registration completeness.

CONCLUSIONS

There are marked inequalities in registration completeness at the county level and within provinces in China. The socioeconomic condition, the implementation of specific registration-enhancing initiatives, and the availability and quality of medical care were the primary drivers of the observed geographical variation. A more strategic approach, with more research, is required to identify the main reasons for death under-reporting, especially in the poorer performing counties, to guide remedial action.

Hospital-level variation in the development of persistent critical illness

PURPOSE

Patients with persistent critical illness may account for up to half of all intensive care unit (ICU) bed-days. It is unknown if there is hospital variation in the development of persistent critical illness and if hospital performance affects the incidence of persistent critical illness.

METHODS

This is a retrospective analysis of Veterans admitted to the Veterans Administration (VA) ICUs from 2015 to 2017. Hospital performance was defined by the risk- and reliability-adjusted 30-day mortality. Persistent critical illness was defined as an ICU length of stay of at least 11 days. We used 2-level multilevel logistic regression models to assess variation in risk- and reliability-adjusted probabilities in the development of persistent critical illness.

RESULTS

In the analysis of 100 hospitals which encompassed 153,512 hospitalizations, 4.9% (N = 7640/153,512) developed persistent critical illness. There was variation in the development of persistent critical illness despite controlling for patient characteristics (intraclass correlation: 0.067, 95% CI 0.049-0.091). Hospitals with higher risk- and reliability-adjusted 30-day mortality had higher probabilities of developing persistent critical illness (predicted probability: 0.057, 95% CI 0.051-0.063, p < 0.01) compared to those with lower risk- and reliability-adjusted 30-day mortality (predicted probability: 0.046, 95% CI 0.041-0.051, p < 0.01). The median odds ratio was 1.4 (95% CI 1.33-1.49) implying that, for two patients with the same physiology on admission at two different VA hospitals, the patient admitted to the hospital with higher adjusted mortality would have 40% greater odds of developing persistent critical illness.

CONCLUSION

Hospitals with higher risk- and reliability-adjusted 30-day mortality have a higher probability of developing persistent critical illness. Understanding the drivers of this variation may identify modifiable factors contributing to the development of persistent critical illness.

A Cross-sectional Study of Hospital Performance on ICU Utilization Practices for Patients with Chronic Obstructive Pulmonary Disease

PURPOSE

Intensive care unit (ICU) resources are a costly but effective commodity used in the management of critically ill patients with chronic obstructive pulmonary disease (COPD). ICU admission decisions are determined by patient diagnosis and severity of illness, but also may be affected by hospital differences in quality and performance. We investigate the variability in ICU utilization for patients with COPD and its association with hospital characteristics.

METHODS

Using a 3M administrative dataset spanning 2008-2013, we conducted a retrospective cohort study of adult patients discharged with COPD at hospitals in three state to determine variability in ICU utilization. Quality metrics were calculated for each hospital using observed-to-expected (O/E) ratios for overall mortality and length of stay. Logistic and multilevel multivariate regression models were constructed, estimating the association between hospital quality metrics on ICU utilization, after adjustment for available clinical factors and hospital characteristics.

RESULTS

In 434 hospitals with 570,517 COPD patient visits, overall ICU admission rate was 33.1% [range 0-89%; median (IQR) 24% (8, 54)]. The addition of patient, hospital, and quality characteristics decreased the overall variability attributable to individual hospital differences seen within our cohort from 40.9 to 33%. Odds of ICU utilization were increased for larger hospitals and those seeing lower pulmonary case volume. Hospitals with better overall O/E ratios for length of stay or mortality had lower ICU utilization.

CONCLUSIONS

Hospital characteristics, including quality metrics, are associated with variability in ICU utilization for COPD patients, with higher ICU utilization seen for lower performing hospitals.

Modelling and prediction of global non-communicable diseases

Background

Non-communicable diseases (NCDs) are the main health and development challenge facing humankind all over the world. They are inextricably linked to socio-economic development. Deaths caused by NCDs should be different in different socio-economic development stages. The stratified heterogeneity of NCD deaths is currently not fully explored.

Methods

Countries were classified according to their socio-economic types and development stages, which were illustrated as a tree-like structure called Geotree. NCD deaths were linked to the countries and so were attached to the Geotree, which was modelled by a multilevel model (MLM) approach. Accordingly, the levels of NCD death indexes were predicted for 2030.

Results

Through the Geotree structure constructed in the study, it can be seen that the NCD death index has obvious stratified heterogeneity; that is, the NCD death index shows different trends in different country types and socio-economic development stages. In the first-level branches (country type), as national income increases, NCD mortality rate decreases and the proportion of NCD deaths to total deaths increases. In the secondary-level trunks (socio-economic development stage), as a country’s development stage rises, the NCD mortality rate decreases and the proportion of NCD deaths to total deaths increases. In addition, combined with the hierarchical nature of the evolution tree model, the MLM was used to predict the global NCD death index for 2030. The result was that by 2030, the global average age-standardized NCD mortality rate would be 510.54 (per 100,000 population) and the global average mortality for NCD deaths of the total number of deaths would be 75.26%.

Conclusions

This study found that there is a significant association between socio-economic factors and NCD death indicators in the tree-like structure. In the Geotree, countries on the same branch or trunk can learn from countries with higher development stages to formulate more effective NCD response policies and find the right prevention and treatment path.

Individual and community level factors associated with anemia among lactating mothers in Ethiopia using data from Ethiopian demographic and health survey, 2016; a multilevel analysis

BACKGROUND

Maternal anemia is a worldwide public health problem especially in developing countries including Ethiopia. The anemia burden among lactating mothers was higher in Africa particularly in Ethiopia, and scant attention was paid. To date, there is limited evidence on community level determinants of anemia among lactating mothers in Ethiopia. This study, therefore, aimed to assess the prevalence and factors associated with anemia among lactating mothers in Ethiopia.

METHODS

Secondary data analysis was employed using the 2016 Ethiopian Demographic and Health Survey. A total weighted sample of 4658 lactating women was included. A multilevel logistic regression model was used to identify individual and community level determinants of anemia during lactation. Finally, the adjusted odds ratio with a 95% confidence interval was reported.

RESULTS

The overall prevalence of anemia was 28.3% (95% CI; 26.7, 30.0) with the higher regional prevalence in Somali (68.3%) and Afar (47.2%) regions. Current modern contraceptive use [AOR = 0.71; 95% CI: 0.58, 0.87], Poorer [AOR = 0.77; 95% CI: 0.61, 0.98], middle [AOR = 0.74; 95% CI: 0.56, 0.97], rich [AOR = 0.64; 95% CI: 0.46, 0.85], and richest [AOR = 0.66; 95% CI: 0.43, 0.98] wealth index, being working within the 12 months preceding the survey [AOR = 0.77; 95% CI: 0.64, 0.92], and taking iron during pregnancy [AOR = 0.82; 95% CI: 0.68, 0.98] were associated with lower odds of anemia. Whereas, being female household head [AOR = 1.22; 95% CI: 1.01, 1.49], having two births [AOR = 1.27; 95% CI: 1.04, 1.55] and three to four births [AOR = 1.53; 95% CI: 1.14, 2.06] within 5 years, and higher community illiteracy level [AOR = 1.06; 95% CI: 1.06, 1.70] were associated with the increased odds of anemia during lactation.

CONCLUSION

In this study the prevalence of anemia among lactating mothers was high. It was affected by both individual and community level factors. Therefore, focusing on family planning services especially on modern contraceptive methods, iron supplementation during pregnancy, child spacing, and improving community literacy could decrease anemia during lactation.

Spatial Distribution and Factors Associated with Khat Chewing among Adult Males 15-59 Years in Ethiopia Using a Secondary Analysis of Ethiopian Demographic and Health Survey 2016: Spatial and Multilevel Analysis

Background

Khat chewing has become prevalent in the world due to the improvement of road and air transportation. In Ethiopia, khat chewing is more prevalent and widely practiced by men. Khat has a negative effect on social, economic, and mental health. There is variation in khat cultivation, use, and factors that associated with khat chewing in the Ethiopian regions. Therefore, this study is aimed at showing spatial distribution and factors associated with khat chewing among male adults 15-59 years in Ethiopia.

Methods

A total of 12,594 men were included in this study. ArcGIS version 10.7 software was used to show the spatial distribution of chewing khat among adult men in Ethiopia. The Bernoulli model was applied using Kilduff SaTScan version 9.6 software to identify significant purely spatial clusters for chewing khat in Ethiopia. A multilevel logistic regression model was fitted to identify factors associated with khat chewing. A P value < 0.05 was taken to declare statistically significant predictors.

Results

The EDHS 2016 survey showed that the high proportion of chewing khat was found in Dire Dawa, Harari, Southern Oromia, Somali, and Benishangul Gumuz regions. In spatial scan statistics analysis, a total of 126 clusters (LLR = 946.60, P value < 0.001) were identified. Age group 30-44 years old (AOR = 1.60, 95% CI: 1.37, 1.86) and 45-59 years old (AOR = 1.33, 95% CI: 1.09, 1.61), being single (AOR = 1.86, 95% CI: 1.64, 2.12), Muslim religion followers (AOR = 15.03, 95% CI: 11.90, 18.90), media exposed (AOR = 0.77, 95% CI: 0.68, 0.86), had work (AOR = 2.48, 95% CI: 2.08, 2.95), alcohol drinker (AOR = 3.75, 95% CI: 3.10, 4.53), and region (Afar, Amhara, Benishangul Gumuz, Gambela, Harari, Oromia, Somali, Southern Nations, Nationalities, and People's Region (SNNPR), and Tigray) and two cities (Addis Ababa and Dire Dawa) were statistically significant factors affecting chewing khat in Ethiopia.

Conclusions

In Ethiopia, the spatial distribution of khat chewing among adult men was nonrandom. A high proportion of khat chewing was observed in Dire Dawa, Harari, Southern Oromia, Somali, and Benishangul Gumuz regions. Older age group, being single marital status, alcohol drinker, media unexposed, had no work, and Muslim religion follower were factors affecting khat chewing. Policymakers should be given spatial attention in reducing the prevalence of chewing khat by teaching the health impact of khat chewing through media in the identified regions.

A multilevel analysis of overweight and obesity among non-pregnant women of reproductive age in Malawi: evidence from the 2015-16 Malawi Demographic and Health Survey

BACKGROUND

Overweight and obesity are well-known risk factors for non-communicable diseases such as cardiovascular disease, diabetes, some cancers and musculoskeletal disorders. In Malawi, the proportion of women who are overweight/obese has doubled, from 10% in 1992 to 21% in 2015-16. Therefore we aimed to explore the individual- and community-level factors associated with overweight and obesity among non-pregnant women of child-bearing age.

METHODS

Secondary analysis of the 2015-16 Malawi Demographic and Health Survey was conducted. Overweight was defined as a body mass index (BMI) of 25 to <30 kg/m2, while obesity was defined as a BMI ≥30 kg/m2. Two-level multilevel multivariable logistic regression models were constructed using a logit-link function with a binomial distribution on 7326 women living in 850 different communities.

RESULTS

At the individual level, the adjusted multilevel regression results showed that women 15-19 y of age as well as women from the poorest households had reduced odds of being overweight/obese. However, women with white collar jobs and women who were affiliated with the Church of Central Africa Presbyterian and the Roman Catholic church had increased odds of being overweight/obese. At the community level, women from urban areas and women who resided in communities with a low percentage of media exposure had increased odds of being overweight or obese. While women from poor communities had reduced odds of being overweight and obese, the proportion change in variance showed that 56, 77 and 78% of total variations in the odds of overweight, obese and overweight/obese across the communities were explained by both individual- and community-level factors. The median odds ratio showed that the likelihood of maternal overweight, obese and overweight/obese increased by 63, 39 and 84% when the women moved from low- to high-risk neighbourhoods.

CONCLUSIONS

Older women, Christian women, women with white collar jobs and women from the richest households should be targeted during policy formulation. At the community level, media coverage should be spread evenly so that health awareness messages, consequences and means of overweight and obesity prevention are getting to the targeted women. Our study revealed evidence of clustering effects of overweight and obesity at the community level, hence neighbourhood variations with respect to maternal overweight and obesity should be taken into account when designing nutritional policies.

Acknowledging the role of patient heterogeneity in hospital outcome reporting: Mortality after acute myocardial infarction in five European countries

Background

Hospital performance, presented as the comparison of average measurements, dismisses that hospital outcomes may vary across types of patients. We aim at drawing out the relevance of accounting for patient heterogeneity when reporting on hospital performance.

Methods

An observational study on administrative data from virtually all 2009 hospital admissions for Acute Myocardial Infarction (AMI) discharged in Denmark, Portugal, Slovenia, Spain, and Sweden. Hospital performance was proxied using in-hospital risk-adjusted mortality. Multilevel Regression Modelling (MLRM) was used to assess differences in hospital performance, comparing the estimates of random intercept modelling (capturing hospital general contextual effects (GCE)), and random slope modelling (capturing hospital contextual effects for patients with and without congestive heart failure -CHF). The weighted Kappa Index (KI) was used to assess the agreement between performance estimates.

Results

We analysed 46,875 admissions of AMI, 6,314 with coexistent CHF, discharged from 107 hospitals. The overall in-hospital mortality rate was 5.2%, ranging from 4% in Sweden to 6.9% in Portugal. The MLRM with random slope outperformed the model with only random intercept, highlighting a much higher GCE in CHF patients [VPC = 8.34 (CI95% 4.94 to 13.03) and MOR = 1.69 (CI95% 1.62 to 2.21) vs. VPC = 3.9 (CI95% 2.4 to 5.9), MOR of 1.42 (CI95% 1.31 to 1.54) without CHF]. No agreement was observed between estimates [KI = -0,02 (CI95% -0,08 to 0.04].

Conclusions

The different GCE in AMI patients with and without CHF, along with the lack of agreement in estimates, suggests that accounting for patient heterogeneity is required to adequately characterize and report on hospital performance.

Routinely measuring symptom burden and health-related quality of life in dialysis patients: first results from the Dutch registry of patient-reported outcome measures

Background

The use of patient-reported outcome measures (PROMs) is becoming increasingly important in healthcare. However, incorporation of PROMs into routine nephrological care is challenging. This study describes the first experience with PROMs in Dutch routine dialysis care.

Methods

A pilot study was conducted in dialysis patients in 16 centres. Patients were invited to complete PROMs at baseline and 3 and 6 months. PROMs consisted of the 12-item short-form and Dialysis Symptom Index to assess health-related quality of life (HRQoL) and symptom burden. Response rates, HRQoL and symptom burden scores were analysed. Qualitative research methods were used to gain insight into patients’ views on using PROMs in clinical practice.

Results

In total, 512 patients (36%) completed 908 PROMs (24%) across three time points. Response rates varied from 6 to 70% among centres. Mean scores for physical and mental HRQoL were 35.6 [standard deviation (SD) 10.2] and 47.7 (SD 10.6), respectively. Patients experienced on average 10.8 (SD 6.1) symptoms with a symptom burden score of 30.7 (SD 22.0). Only 1–3% of the variation in PROM scores can be explained by differences between centres. Patients perceived discussing their HRQoL and symptom scores as insightful and valuable. Individual feedback on results was considered crucial.

Conclusions

The first results show low average response rates with high variability among centres. Dialysis patients experienced a high symptom burden and poor HRQoL. Using PROMs at the individual patient level is suitable and may improve patient–professional communication and shared decision making. Further research is needed to investigate how the collection and the use of PROMs can be successfully integrated into routine care to improve healthcare quality and outcomes.

Binge drinking, alcohol outlet density and associated factors: a multilevel analysis among adolescents in Belo Horizonte, Minas Gerais State, Brazil

Our study sought to evaluate the prevalence of binge drinking in adolescents and its association with density of alcohol outlets around schools. This cross-sectional study was conducted in Belo Horizonte, Minas Gerais State, Brazil, with 436 high-school students aged between 17 and 19 and enrolled in 18 public and private schools. The students completed the Alcohol Use Disorder Identification Test (AUDIT-C), consisting of questions about alcohol consumption by parents and siblings, and socioeconomic status (type of school, mother's education level). Data from geographic information systems were used to estimate the density of alcohol outlets around schools participating. The association between exploratory variables and binge drinking was investigated using multilevel logistic regression analysis (p < 0.05) with random intercepts and fixed slopes. A three-step sequential modeling strategy was adopted. The prevalence of binge drinking was 39.9%. The alcohol consumption among adolescents was lower for those studying in areas with low density of alcohol outlets around schools (OR = 0.32; 95%CI: 0.14; 0.73) and the consumption of alcohol by mothers was associated with binge drinking among adolescents (OR = 1.94; 95%CI: 1.14; 3.30). Our study concluded that binge drinking among adolescents was associated with density of alcohol outlets around the schools and mother's alcohol consumption.

Holliday E, Loxton D. Individual and community level determinants of short birth interval in Ethiopia: A multilevel analysis

Background

The World Health Organization recommends a minimum of 33 months between two consecutive live births to reduce the risk of adverse maternal and child health outcomes. However, determinants of short birth interval have not been well understood in Ethiopia.

Objective

The aim of this study was to assess individual- and community-level determinants of short birth interval among women in Ethiopia.

Methods

A detailed analysis of the 2016 Ethiopian Demographic and Health Survey data was performed. A total of 8,448 women were included in the analysis. A two-level multilevel logistic regression analysis was used to identify associated individual- and community-level factors and estimate between-community variance.

Results

At the individual-level, women aged between 20 and 24 years at first marriage (AOR = 1.37; 95% CI: 1.18–1.60), women aged between 25 and 29 years at first marriage (AOR = 1.65; 95% CI: 1.20–2.25), having a husband who attended higher education (AOR = 1.32; 95% CI: 1.01–1.73), being unemployed (AOR = 1.16; 95% CI: 1.03–1.31), having an unemployed husband (AOR = 1.23; 95% CI: 1.04–1.45), being in the poorest wealth quintile (AOR = 1.82; 95% CI: 1.39–2.39), being in the poorer wealth quintile (AOR = 1.58; 95% CI: 1.21–2.06), being in the middle wealth quintile (AOR = 1.61; 95% CI: 1.24–2.10), being in the richer wealth quintile (AOR = 1.54; 95% CI: 1.19–2.00), increased total number of children born before the index child (AOR = 1.07; 95% CI: 1.03–1.10) and death of the preceding child (AOR = 1.97; 95% CI: 1.59–2.45) were associated with increased odds of short birth interval. At the community-level, living in a pastoralist region (AOR = 2.01; 95% CI: 1.68–2.39), being a city dweller (AOR = 1.75; 95% CI: 1.38–2.22), high community-level female illiteracy (AOR = 1.23; 95% CI: 1.05–1.45) and increased distance to health facilities (AOR = 1.32; 95% CI: 1.11–1.56) were associated with higher odds of experiencing short birth interval. Random effects showed significant variation in short birth interval between communities.

Conclusion

Determinants of short birth interval are varied and complex. Multifaceted intervention approaches supported by policy initiatives are required to prevent short birth interval.

The interplay between national and parental unemployment in relation to adolescent life satisfaction in 27 countries: analyses of repeated cross-sectional school surveys

BACKGROUND

Previous research shows that parental unemployment is associated with low life satisfaction in adolescents. It is unclear whether this translates to an association between national unemployment and adolescent life satisfaction, and whether such a contextual association is entirely explained by parental unemployment, or if it changes as a function thereof. For adults, associations have been shown between unemployment and mental health, including that national unemployment can affect mental health and life satisfaction of both the employed and the unemployed, but to different degrees. The aim of this paper is to analyse how national unemployment levels are related to adolescent life satisfaction, across countries as well as over time within a country, and to what extent and in what ways such an association depends on whether the individual's own parents are unemployed or not.

METHODS

Repeated cross-sectional data on adolescents' (aged 11, 13 and 15 years, n = 386,402) life satisfaction and parental unemployment were collected in the Health Behaviour in School-aged Children (HBSC) survey, in 27 countries and 74 country-years, across 2001/02, 2005/06 and 2009/10 survey cycles. We linked this data to national harmonised unemployment rates provided by OECD and tested their associations using multilevel linear regression, including interaction terms between national and parental unemployment.

RESULTS

Higher national unemployment rates were related to lower adolescent life satisfaction, cross-sectionally between countries but not over time within countries. The verified association was significant for adolescents with and without unemployed parents, but stronger so in adolescents with unemployed fathers or both parents unemployed. Having an unemployed father, mother och both parents was in itself related to lower life satisfaction.

CONCLUSION

Living in a country with higher national unemployment seems to be related to lower adolescent life satisfaction, whether parents are unemployed or not, although stronger among adolescents where the father or both parents are unemployed. However, variation in unemployment over the years did not show an association with adolescent life satisfaction.

Clinic and patient variation in intermediate clinical outcomes for type 2 diabetes: a multilevel analysis

Background

Variation at different levels of diabetes care has not yet been quantified for low- and middle-income countries. Understanding this variation and its magnitude is important to guide policy makers in designing effective interventions. This study aims to quantify the variation in the control of glycated haemoglobin (HbA1c), systolic blood pressure (SBP) and low-density lipoprotein cholesterol (LDL-C) for type 2 diabetes (T2D) patients at the clinic and patient level and determine patient and clinic factors associated with control of these outcomes in T2D.

Methods

This is a cross-sectional study within the baseline data from the impact evaluation of the Enhanced Primary Health Care (EnPHC) intervention on 40 public clinics in Malaysia. Patients aged 30 and above, diagnosed with T2D, had a clinic visit for T2D between 01 Nov 2016 and 30 April 2017 and had at least one HbA1c, SBP and LDL-C measurement within 1 year from the date of visit were included for analysis. Multilevel linear regression adjusting for patient and clinic characteristics was used to quantify variation at the clinic and patient levels for each outcome.

Results

Variation in intermediate clinical outcomes in T2D lies predominantly (93% and above) at the patient level. The strongest predictors for poor disease control in T2D were the proxy measures for disease severity including duration of diabetes, presence of microvascular complications, being on insulin therapy and number of antihypertensives. Among the three outcomes, HbA1c and LDL-C results provide greatest opportunity for improvement.

Conclusion

Clinic variation in HbA1c, SBP and LDL-C accounts for a small percentage from total variation. Findings from this study suggest that standardised interventions need to be applied across all clinics, with a focus on customizing therapy based on individual patient characteristics.