Rationale, design and respondent characteristics of the 2013-2014 New York City Health and Nutrition Examination Survey (NYC HANES 2013-2014)

Patient-reported outcomes for people with diabetes: what and how to measure? A narrative review

Patient-reported outcomes (PROs) are valuable for shared decision making and research. Patient-reported outcome measures (PROMs) are questionnaires used to measure PROs, such as health-related quality of life (HRQL). Although core outcome sets for trials and clinical practice have been developed separately, they, as well as other initiatives, recommend different PROs and PROMs. In research and clinical practice, different PROMs are used (some generic, some disease-specific), which measure many different things. This is a threat to the validity of research and clinical findings in the field of diabetes. In this narrative review, we aim to provide recommendations for the selection of relevant PROs and psychometrically sound PROMs for people with diabetes for use in clinical practice and research. Based on a general conceptual framework of PROs, we suggest that relevant PROs to measure in people with diabetes are: disease-specific symptoms (e.g. worries about hypoglycaemia and diabetes distress), general symptoms (e.g. fatigue and depression), functional status, general health perceptions and overall quality of life. Generic PROMs such as the 36-Item Short Form Health Survey (SF-36), WHO Disability Assessment Schedule (WHODAS 2.0), or Patient-Reported Outcomes Measurement Information System (PROMIS) measures could be considered to measure commonly relevant PROs, supplemented with disease-specific PROMs where needed. However, none of the existing diabetes-specific PROM scales has been sufficiently validated, although the Diabetes Symptom Self-Care Inventory (DSSCI) for measuring diabetes-specific symptoms and the Diabetes Distress Scale (DDS) and Problem Areas in Diabetes (PAID) for measuring distress showed sufficient content validity. Standardisation and use of relevant PROs and psychometrically sound PROMs can help inform people with diabetes about the expected course of disease and treatment, for shared decision making, to monitor outcomes and to improve healthcare. We recommend further validation studies of diabetes-specific PROMs that have sufficient content validity for measuring disease-specific symptoms and consider generic item banks developed based on item response theory for measuring commonly relevant PROs.

Disparities in Asthma Rates Amongst Black Residents of New York City

Asthma is a chronic respiratory condition affecting around 300 million people worldwide. In the United States, Black individuals have a higher burden of asthma than White individuals. The goal of this study was to differentiate the burden of asthma between US-born and foreign-born Black residents of New York City (NYC). We use a multivariable Cox proportional hazard model with a robust variance estimate. The results indicate that foreign-born Black NYC residents have a significantly lower asthma prevalence than US-born (PR = 0.40, 95% CI = 0.21-0.76). Additionally, those 65 years and older have a lower prevalence of asthma compared to those 18-34 years old. This study shows that asthma prevalence is higher amongst US-born Black NYC residents than foreign-born, which may indicate that the healthcare needs of the foreign-born may be different from that of the native-born. Further studies are needed to elucidate this result fully.

How severe would prioritization-induced bottlenecks need to be offset the benefits from prioritizing COVID-19 vaccination to those most at risk in New York City?

BACKGROUND

Prioritization of higher-risk people for COVID-19 vaccination could prevent more deaths, but could slow vaccination speed. We used mathematical modeling to examine the trade-off between vaccination speed and prioritization for individuals age 65+ and essential workers.

METHODS

We used a stochastic, discrete-time susceptible-exposed-infected-recovered (SEIR) model with age- and comorbidity-adjusted COVID-19 outcomes (infections, hospitalizations, and deaths). The model was calibrated to COVID-19 hospitalizations, ICU census, and deaths in NYC. We assumed 10,000 vaccinations per day, initially restricted to healthcare workers and nursing home populations, and subsequently expanded to other populations at alternative times (4, 5, or 6 weeks after vaccine launch) and speeds (20,000, 50,000, 100,000, or 150,000 vaccinations per day), as well as prioritization options (+/- prioritization of people age 65+ and essential workers). In sensitivity analyses, we examined the effect of a SARS-COV-2 variant with greater transmissibility.

RESULTS

To be beneficial, prioritization must not create a bottleneck that decreases vaccination speed by > 50% without a more transmissible variant, or by > 33% with the emergence of the more transmissible variant. More specifically, prioritizing people age 65+ and essential workers increased the number of lives saved per vaccine dose delivered: 3000 deaths could be averted by delivering 83,000 vaccinations per day without prioritization or 50,000 vaccinations per day with prioritization. Other tradeoffs involve vaccination speed and timing. Compared to the slowest-examined vaccination speed of 20,000 vaccinations per day, achieving the fastest-examined vaccination speed of 150,000 vaccinations per day would avert additional 313,700 (28.6%) infections and 1693 (24.1%) deaths. Emergence of a more transmissible variant would double COVID-19 infections, hospitalizations, and deaths over the first 6 months of vaccination. The fastest-examined vaccination speed could only offset the harm of the more transmissible variant if achieved within 5 weeks of vaccine launch.

CONCLUSIONS

Faster vaccination speed with sooner vaccination expansion would save more lives. Prioritization of COVID-19 vaccines to higher-risk populations would be more beneficial only if it does not create an excessive vaccine delivery bottleneck.

Addressing Financial Barriers to Health Care Among People Who are Low-Income and Insured in New York City, 2014–2017

While health care-associated financial burdens among uninsured individuals are well described, few studies have systematically characterized the array of financial and logistical complications faced by insured individuals with low household incomes. In this mixed methods paper, we conducted 6 focus groups with a total of 55 residents and analyzed programmatic administrative records to characterize the specific financial and logistic barriers faced by residents living in public housing in East and Central Harlem, New York City (NYC). Participants included individuals who enrolled in a municipal community health worker (CHW) program designed to close equity gaps in health and social outcomes. Dedicated health advocates (HAs) were explicitly paired with CHWs to provide health insurance and health care navigational assistance. We describe the needs of 150 residents with reported financial barriers to care, as well as the navigational and advocacy strategies taken by HAs to address them. Finally, we outline state-level policy recommendations to help ameliorate the problems experienced by participants. The model of paired CHW–HAs may be helpful in addressing financial barriers for insured populations with low household income and reducing health disparities in other communities.

Black New Yorkers with Type 2 Diabetes: Afro-Caribbean Immigrants Have Lower BMI and Lower Waist Circumference than African Americans

OBJECTIVE

Using the 2013/2014 New York City (NYC) Health and Nutrition Examination Survey (NYCHANES) data, this exploratory study examined whether (a) type 2 diabetes (diabetes) prevalence differed between NYC Afro-Caribbeans and African Americans; (b) anthropometric, biochemical, and sociodemographic diabetes profiles differed between and within groups; and (c) diabetes odds differed between and within groups.

METHODS

Diabetes was defined as prior diagnosis, HbA1c ≥ 6.5% (7.8 mmol/L), or fasting glucose ≥ 126 mg/dL. Weighted logistic regression estimated diabetes odds by nativity and either waist circumference (WC) (cm) or BMI (kg/m²). All regression models controlled for age, hypertension, gender, education, income, marital status, physical activity, and smoking.

RESULTS

Among Afro-Caribbeans (n = 81, 65% female, age (mean ± SE) 49 ± 2 years, BMI 29.2 ± 0.7 kg/m²) and African Americans (n = 118, 50% female, age 47 ± 2 years, BMI 30.3 ± 0.9 kg/m²), Afro-Caribbeans with diabetes had lower BMI (29.9 ± 0.8 kg/m² vs. 34.6 ± 1.7 kg/m², P = 0.01) and lower WC (102 ± 2 cm vs. 114 ± 3 cm, P = 0.002) than African Americans with diabetes. Afro-Caribbeans with diabetes had lower prevalence of obesity (33.2% vs. 74.7%) and higher prevalence of overweight (57.2% vs. 13.5%) (P = 0.02) than African Americans with diabetes. Odds of diabetes did not differ between Afro-Caribbeans and African Americans. In models predicting the effect of WC, diabetes odds increased with WC (OR = 1.07 (95% CI 1.02, 1.11), P = 0.003) and age (OR = 1.09 (95% CI 1.03-1.15), P = 0.003) for African Americans only. In models predicting the effect of BMI, diabetes odds increased for Afro-Caribbeans with age (OR = 1.06 (1.01, 1.11)*, P = 0.04) and hypertension (OR = 5.62 (95% CI 1.04, 30.42), P = 0.045), whereas for African Americans, only age predicted higher diabetes odds (OR = 1.08 (95% CI 1.03, 1.14), P = 0.003).

CONCLUSIONS

In NYC, Afro-Caribbeans with diabetes have lower BMI and lower WC than African Americans with diabetes, but odds of diabetes do not differ. Combining African-descent populations into one group obscures clinical differences and generalizes diabetes risk.

Changes in General Health and Mental Health Outcomes in an Urban Population Over a Decade: A Population-Representative Analysis Stratified by Sexual Orientation

Purpose: We explored population-level changes in general and mental health outcomes among sexual minority and heterosexual New York City (NYC) adults. Methods: Using the NYC Health and Nutrition Examination Surveys, we analyzed data from 2931 adults surveyed between 2004 and 2014. Sexual minority (LGB+) participants included those who identified as lesbian, gay, bisexual, something else, or not sure. Prevalence estimates of general health, mental health services use, prescription use for a mental/emotional condition, and mental/emotional disability were calculated. Changes in these estimates were compared across survey iterations with two-sided t-tests. Multivariate log binomial regression modeling was also employed. Results: Across the study period, LGB+ adults reported a decrease in fair/poor general health (24%-18%). Compared to 2004, LGB+ adults in 2014 were more likely to use mental health services (15%-27%), take prescription medication for a mental/emotional condition (11%-20%), and have a mental/emotional disability limiting work (5%-10%). Point estimates showed similar changes over time among both LGB+ and heterosexual adults, but some changes were not statistically significant. We also found that bisexual adults utilized mental health services (prevalence ratio [PR] = 2.15; 95% confidence interval [CI]: 1.34-3.44) and medications (PR = 2.92; 95% CI: 1.72-4.96) more than heterosexual adults. Conclusion: Although reporting fair/poor general health decreased, the prevalence of using mental health services, using prescription medication, and having a mental/emotional disability increased for both LGB+ and heterosexual adults in NYC. These findings may be related to greater mental health literacy and awareness or to other population-wide trends.

Social distancing and mask-wearing could avoid recurrent stay-at-home restrictions during COVID-19 respiratory pandemic in New York City

Stay-at-home restrictions such as closure of non-essential businesses were effective at reducing SARS-CoV-2 transmission in New York City (NYC) in the spring of 2020. Relaxation of these restrictions was desirable for resuming economic and social activities, but could only occur in conjunction with measures to mitigate the expected resurgence of new infections, in particular social distancing and mask-wearing. We projected the impact of individuals’ adherence to social distancing and mask-wearing on the duration, frequency, and recurrence of stay-at-home restrictions in NYC. We applied a stochastic discrete time-series model to simulate community transmission and household secondary transmission in NYC. The model was calibrated to hospitalizations, ICU admissions, and COVID-attributable deaths over March–July 2020 after accounting for the distribution of age and chronic health conditions in NYC. We projected daily new infections and hospitalizations up to May 31, 2021 under the different levels of adherence to social distancing and mask-wearing after relaxation of stay-at-home restrictions. We assumed that the relaxation of stay-at-home policies would occur in the context of adaptive reopening, where a new hospitalization rate of ≥ 2 per 100,000 residents would trigger reinstatement of stay-at-home restrictions while a new hospitalization rate of ≤ 0.8 per 100,000 residents would trigger relaxation of stay-at-home restrictions. Without social distancing and mask-wearing, simulated relaxation of stay-at-home restrictions led to epidemic resurgence and necessary reinstatement of stay-at-home restrictions within 42 days. NYC would have stayed fully open for 26% of the time until May 31, 2021, alternating reinstatement and relaxation of stay-at-home restrictions in four cycles. At a low (50%) level of adherence to mask-wearing, NYC would have needed to implement stay-at-home restrictions between 8% and 32% of the time depending on individual adherence to social distancing. At moderate to high levels of adherence to mask-wearing without social distancing, NYC would have needed to implement stay-at-home restrictions. In threshold analyses, avoiding reinstatement of stay-at-home restrictions required a minimum of 60% adherence to mask-wearing at 50% adherence to social distancing. With low adherence to mask-wearing and social distancing, reinstatement of stay-at-home restrictions in NYC was inevitable. High levels of adherence to social distancing and mask-wearing could have attributed to avoiding recurrent surges without reinstatement of stay-at-home restrictions.

Assessing the Health and Economic Impact of a Potential Menthol Cigarette Ban in New York City: a Modeling Study

Menthol in cigarettes increases nicotine dependence and decreases the chances of successful smoking cessation. In New York City (NYC), nearly half of current smokers usually smoke menthol cigarettes. Female and non-Latino Black individuals were more likely to smoke menthol-flavored cigarettes compared to males and other races and ethnicities. Although the US Food and Drug Administration recently announced that it will ban menthol cigarettes, it is unclear how the policy would affect population health and health disparities in NYC. To inform potential policymaking, we used a microsimulation model of cardiovascular disease (CVD) to project the long-term health and economic impact of a potential menthol ban in NYC. Our model projected that there could be 57,232 (95% CI: 51,967-62,497) myocardial infarction (MI) cases and 52,195 (95% CI: 47,446-56,945) stroke cases per 1 million adult smokers in NYC over a 20-year period without the menthol ban policy. With the menthol ban policy, 2,862 MI cases and 1,983 stroke cases per 1 million adults could be averted over a 20-year period. The model also projected that an average of $1,836 in healthcare costs per person, or $1.62 billion among all adult smokers, could be saved over a 20-year period due to the implementation of a menthol ban policy. Results from subgroup analyses showed that women, particularly Black women, would have more reductions in adverse CVD outcomes from the potential implementation of the menthol ban policy compared to males and other racial and ethnic subgroups, which implies that the policy could reduce sex and racial and ethnic CVD disparities. Findings from our study provide policymakers with evidence to support policies that limit access to menthol cigarettes and potentially address racial and ethnic disparities in smoking-related disease burden.

Data Architecture to Support Real-Time Data Analytics for the Population-Based HIV Impact Assessments

BACKGROUND AND SETTING

Electronic data capture facilitates timely use of data. Population-based HIV impact assessments (PHIAs) were led by host governments, with funding from the President's Emergency Plan for AIDS Relief, technical assistance from the Centers for Disease Control, and implementation support from ICAP at Columbia University. We described data architectures, code-based processes, and resulting data volume and quality for 14 national PHIA surveys with concurrent timelines and varied country-level data governance (2015-2020).

METHODS

PHIA project data were collected through tablets, point-of-care and laboratory testing instruments, and inventory management systems, using open-source software, vendor solutions, and custom-built software. Data were securely uploaded to the PHIA data warehouse daily or weekly and then used to populate survey-monitoring dashboards and return timely laboratory-based test results on an ongoing basis. Automated data processing allowed timely reporting of survey results.

RESULTS

Fourteen data architectures were successfully established, and data from more than 450,000 participants in 30,000 files across 13 countries with completed PHIAs, and blood draws producing approximately 6000 aliquots each week per country, were securely collected, transmitted, and processed by 17 full-time equivalent staff. More than 25,600 viral load results were returned to clinics of participants' choice. Data cleaning was not needed for 98.5% of household and 99.2% of individual questionnaires.

CONCLUSION

The PHIA data architecture permitted secure, simultaneous collection and transmission of high-quality interview and biomarker data across multiple countries, quick turnaround time of laboratory-based biomarker results, and rapid dissemination of survey outcomes to guide President's Emergency Plan for AIDS Relief epidemic control.

Increasing obesity odds among foreign-born New Yorkers are not explained by eating out, age at arrival, or duration of residence: results from NYC HANES 2004 and 2013/2014

BACKGROUND

Among the foreign-born in the United States (US) dietary acculturation and eating out may increase obesity risk. Using the 2004 (N = 1952) and 2013/14 (N = 1481) New York City (NYC) Health and Nutrition Examination Surveys, we compared for the foreign-born and US-born by survey year: 1) odds of obesity; 2) association between eating out and obesity and 3) effect of age at arrival and duration of residence among the foreign-born. Weighted logistic regression estimated odds of obesity.

RESULTS

Compared to the US-born, the foreign-born had lower odds of obesity in 2004, (aOR = 0.51 (95%CI 0.37-0.70), P = <.0001). Odds were no different in 2013/14. In 2013/14 the foreign-born who ate out had lower obesity odds (aOR = 0.49 (95%CI 0.31-0.77), P = 0.0022). The foreign-born living in the US≥10 years had greater odds of obesity in 2004 (aOR = 1.73 (95%CI 1.08-2.79), P = 0.0233) but not in 2013/14.

CONCLUSIONS

Eating out does not explain increasing obesity odds among the foreign-born.

Associations of Health Conditions and Health-Related Determinants with Disability among New York City Adult Residents

Introduction

Population-based disability prevalence estimates are limited for New York City (NYC). We examined the association of several health and health-related measures with disability among NYC residents aged 20-64 years.

Methods

We used information from 1,314 adults who participated in the 2013-2014 NYC Health and Nutrition Examination Survey (HANES). We categorized survey participants as having a disability if they reported a physical, mental, and/or emotional problem preventing work or if they reported difficulty walking without special equipment because of a health problem. We used log-binomial regression to quantify the association of each exposure with disability before and after adjustment for select covariates.

Results

Overall, 12.4% of the study's NYC residents aged 20-64 years had a disability. After adjustment, disability prevalence was significantly greater among those who reported having unmet health care needs (prevalence ratio [PR] = 1.75, 95% CI: 1.18-2.57) and those who reported fair/poor general health (PR = 2.33, 95% CI: 1.68-3.24). The probability of disability was greater among NYC residents with arthritis (PR = 2.66, 95% CI: 1.85-3.98) and hypertension (PR = 1.48, 95% CI: 1.04-2.11) when compared with those without these conditions. Disability was also associated with depression (PR = 2.96, 95% CI: 2.06-4.25), anxiety (PR = 2.89, 95% CI: 2.15-3.88), and post-traumatic stress disorder (PR = 2.55, 95% CI: 1.66-3.91). Disability, however, was not associated with diabetes.

Conclusion

Disability is more prevalent among those with unmet health care needs, fair/poor general health, arthritis, hypertension, depression, anxiety, and PTSD in these NYC residents, aged 20-64 years. These findings have implications for NYC's strategic planning initiatives, which can be better targeted to groups disproportionately affected by disability.

Cardiovascular Disease Risk Factors Among Immigrant and US-Born Adults in New York City

OBJECTIVES

Immigrant adults tend to have better health than native-born adults despite lower incomes, but the health advantage decreases with length of residence. To determine whether immigrant adults have a health advantage over US-born adults in New York City, we compared cardiovascular disease (CVD) risk factors among both groups.

METHODS

Using data from the New York City Health and Nutrition Examination Survey 2013-2014, we assessed health insurance coverage, health behaviors, and health conditions, comparing adults ages ≥20 born in the 50 states or the District of Columbia (US-born) with adults born in a US territory or outside the United States (immigrants, following the National Health and Nutrition Examination Survey) and comparing US-born adults with (1) adults who immigrated recently (≤10 years) and (2) adults who immigrated earlier (>10 years).

RESULTS

For immigrant adults, the mean time since arrival in the United States was 21.8 years. Immigrant adults were significantly more likely than US-born adults to lack health insurance (22% vs 12%), report fair or poor health (26% vs 17%), have hypertension (30% vs 23%), and have diabetes (20% vs 11%) but significantly less likely to smoke (18% vs 27%) (all P < .05). Comparable proportions of immigrant adults and US-born adults were overweight or obese (67% vs 63%) and reported CVD (both 7%). Immigrant adults who arrived recently were less likely than immigrant adults who arrived earlier to have diabetes or high cholesterol but did not differ overall from US-born adults.

CONCLUSIONS

Our findings may help guide prevention programs and policy efforts to ensure that immigrant adults remain healthy.

Metabolic syndrome among New York City (NYC) adults: change in prevalence from 2004 to 2013-2014 using New York City Health and Nutrition Examination Survey

PURPOSE

In this study we aim to estimate the change in metabolic syndrome (MetS) prevalence among New York City (NYC) adults between 2004 and 2013-2014 and identify key subgroups at risk.

METHODS

We analyzed data from NYC Health and Nutrition Examination Survey. MetS was defined as having at least three of the following: abdominal obesity, low HDL, elevated triglycerides, glucose dysregulation, and elevated blood pressure. We calculated age-standardized MetS prevalence, change in prevalence over time, and prevalence ratios by gender and race/ethnicity groups. We also tested for additive interaction.

RESULTS

In 2013-2014 MetS prevalence among NYC adults was 24.4% (95% CI, 21.4-27.6). Adults 65+ years and Asian adults had the highest prevalence (45.6% and 33.8%, respectively). Abdominal obesity was the most prevalent MetS component in 2004 and 2013-2014 (50.7% each time). Between 2004 and 2013-2014, MetS decreased by 18.2% (P = .04) among women. The decrease paralleled similar declines in elevated triglycerides and glucose dysregulation. In 2013-14, non-Latino Black women had higher risk of MetS than non-Latino Black men and non-Latino White adults.

CONCLUSION

Age and racial/ethnic disparities in MetS prevalence in NYC were persistent from 2004 to 2013-2014, with Asian adults and non-Latino Black women at particularly high risk.

Risk Factors Associated With Early-Onset Colorectal Cancer

BACKGROUND & AIMS

The incidence of colorectal cancer (CRC) is increasing in individuals younger than 50 years, who do not usually undergo screening if they are of average risk. We sought to identify risk factors for CRC in this population.

METHODS

We compared sociodemographic and medical characteristics of patients who received a diagnosis of CRC at an age of 18-49 years (early-onset) with patients who received a diagnosis of CRC at an age of 50 years or older (late-onset) and with age-matched, cancer-free individuals (controls) at a tertiary academic hospital. We collected data from all adult patients with a diagnosis of CRC from January 1, 2011 through April 3, 2017 from electronic health records. Associations with risk factors were assessed using univariable and multivariable logistic regression models.

RESULTS

We identified 269 patients with early-onset CRC, 2802 with late-onset CRC, and 1122 controls. Compared with controls, patients with early-onset CRC were more likely to be male (odds ratio [OR], 1.87; 95% CI, 1.39-2.51), have inflammatory bowel disease (IBD) (3% vs 0.4% for controls; univariable P < .01), and have a family history of CRC (OR, 8.61; CI, 4.83-15.75). Prevalence values of well-established modifiable CRC risk factors, including obesity, smoking, and diabetes, were similar. Compared to patients with late-onset CRC, patients with early-onset CRC were more likely to be male (OR, 1.44; 95% CI, 1.11-1.87), black (OR, 1.73; 95% CI, 1.08-2.65) or Asian (OR, 2.60; 95% CI, 1.57-4.15), and have IBD (OR, 2.97; 95% CI, 1.16-6.63) or a family history of CRC (OR, 2.87; 95% CI, 1.89-4.25). Sensitivity analyses excluding IBD and family history of CRC showed comparable results. Early-onset CRC was more likely than late-onset disease to be detected in the left colon or rectum (75% vs 59%, P = .02) and at a late stage of tumor development (77% vs 62%, P = .01).

CONCLUSIONS

In a retrospective study of patients with early-onset CRC vs late-onset CRC or no cancer, we identified non-modifiable risk factors, including sex, race, IBD, and family history of CRC, to be associated with early-onset CRC.

Using Urine Drug Testing to Estimate the Prevalence of Drug Use : Lessons Learned From the New York City Health and Nutrition Examination Survey, 2013-2014

OBJECTIVES

Estimating the prevalence of drug use in the general population is important given its potential health consequences but is challenging. Self-reported surveys on drug use have inherent limitations that underestimate drug use. We evaluated the performance of linking urine drug testing with a local, representative health examination survey in estimating the prevalence of drug use in New York City (NYC).

METHODS

We used urine drug testing from the NYC Health and Nutrition Examination Survey (NYC HANES) to estimate the prevalence of drug use (benzodiazepines, cocaine, heroin, and opioid analgesics) among the study sample and compare the findings with self-reported responses to questions about past-12-month drug use from the same survey.

RESULTS

Of 1527 respondents to NYC HANES, urine drug testing was performed on 1297 (84.9%) participants who provided urine and consented to future studies. Self-reported responses gave past-12-month weighted estimates for heroin, cocaine, or any prescription drug misuse of 13.8% (95% CI, 11.6%-16.3%), for prescription drug misuse of 9.9% (95% CI, 8.1%-12.1%), and for heroin or cocaine use of 6.1% (95% CI, 4.7%-7.9%). Urine drug testing gave past-12-month weighted estimates for any drug use of 4.3% (95% CI, 3.0%-6.0%), for use of any prescription drug of 2.8% (95% CI, 1.9%-4.1%), and for heroin or cocaine use of 2.0% (95% CI, 1.2%-3.6%).

CONCLUSION

Urine drug testing provided underestimates for the prevalence of drug use at a population level compared with self-report. Researchers should use other methods to estimate the prevalence of drug use on a population level.

Prevalence estimation by joint use of big data and health survey: a demonstration study using electronic health records in New York city

Background

Electronic Health Records (EHR) has been increasingly used as a tool to monitor population health. However, subject-level errors in the records can yield biased estimates of health indicators. There is an urgent need for methods to estimate the prevalence of health indicators using large and real-time EHR while correcting the potential bias.

Methods

We demonstrate joint analyses of EHR and a smaller gold-standard health survey. We first adopted Mosteller’s method that pools two estimators, among which one is potentially biased. It only requires knowing the prevalence estimates from two data sources and their standard errors. Then, we adopted the method of Schenker et al., which uses multiple imputations of subject-level health outcomes that are missing for the subjects in EHR. This procedure requires information to link some subjects between two sources and modeling the mechanism of misclassification in EHR as well as modeling inclusion probabilities to both sources.

Results

In a simulation study, both estimators yielded negligible bias even when EHR was biased. They performed as well as health survey estimator when EHR bias was large and better than health survey estimator when EHR bias was moderate. It may be challenging to model the misclassification mechanism in real data for the subject-level imputation estimator. We illustrated the methods analyzing six health indicators from 2013 to 14 NYC HANES and the 2013 NYC Macroscope, and a study that linked some subjects in both data sources.

Conclusions

When a small gold-standard health survey exists, it can serve as a safeguard against potential bias in EHR through the joint analysis of the two sources.

Calibrating Local Population-Based Blood Pressure Data from NYC HANES 2013-2014

New York City Health and Nutrition Examination Survey (NYC HANES) was a population-based cross-sectional survey of NYC adults conducted twice, in 2004 and again in 2013-2014, to monitor the health of NYC adults 20 years or older. While blood pressure was measured in both surveys, an auscultatory mercury sphygmomanometer was used to measure blood pressure in clinics in 2004, and an oscillometric LifeSource UA-789AC monitor was used in homes in 2013-2014. To assess comparability of blood pressure results across both surveys, we undertook a randomized study comparing blood pressure (BP) readings by the two devices. Blood pressure measuring protocols followed the 2013 Association for the Advancement in Medical instrumentation guidelines for non-invasive blood pressure device. Data from 167 volunteers were analyzed for this purpose.Paired t tests were used to test for significant difference in mean systolic and diastolic blood pressure between devices for overall and by mid-arm circumference categories. To test for systematic differences between the two devices, we generated Bland-Altman graphs. Sensitivity, specificity, and Kappa statistics were calculated to assess between-device agreement for high (≥ 130/80 mmHg) and not high (< 130/80 mmHg) blood pressure, with mercury set as the reference.Systolic and diastolic blood pressure measured by LifeSource UA-789AC were on average 2.0 and 1.1 mmHg higher, respectively, than those of the mercury sphygmomanometer systolic and diastolic blood pressure readings (P < 0.05). Sensitivity was 81%, specificity was 96%, and the Kappa coefficient was 75%. The Bland-Altman graphs showed that the between-device difference did not vary as a function of the average of the two devices for systolic blood pressure and was larger in the lower and upper ends for diastolic blood pressure. Given the observed differences in systolic and diastolic blood pressure readings between the two blood pressure measurement approaches, we calibrated NYC HANES 2013-2014 blood pressure data by predicting mercury blood pressure values from LifeSource blood pressure values. The mean systolic and diastolic blood pressure in NYC HANES 2013-2014 were lower when data were calibrated.

Sociodemographic variation in the oral microbiome

PURPOSE

Variations in the oral microbiome are potentially implicated in social inequalities in oral disease, cancers, and metabolic disease. We describe sociodemographic variation of oral microbiomes in a diverse sample.

METHODS

We performed 16S rRNA sequencing on mouthwash specimens in a subsample (n = 282) of the 2013-2014 population-based New York City Health and Nutrition Examination Study. We examined differential abundance of 216 operational taxonomic units, and alpha and beta diversity by age, sex, income, education, nativity, and race/ethnicity. For comparison, we examined differential abundance by diet, smoking status, and oral health behaviors.

RESULTS

Sixty-nine operational taxonomic units were differentially abundant by any sociodemographic variable (false discovery rate < 0.01), including 27 by race/ethnicity, 21 by family income, 19 by education, 3 by sex. We found 49 differentially abundant by smoking status, 23 by diet, 12 by oral health behaviors. Genera differing for multiple sociodemographic characteristics included Lactobacillus, Prevotella, Porphyromonas, Fusobacterium.

CONCLUSIONS

We identified oral microbiome variation consistent with health inequalities, more taxa differing by race/ethnicity than diet, and more by SES variables than oral health behaviors. Investigation is warranted into possible mediating effects of the oral microbiome in social disparities in oral and metabolic diseases and cancers.

Tobacco exposure associated with oral microbiota oxygen utilization in the New York City Health and Nutrition Examination Study

PURPOSE

The effect of tobacco exposure on the oral microbiome has not been established.

METHODS

We performed amplicon sequencing of the 16S ribosomal RNA gene V4 variable region to estimate bacterial community characteristics in 259 oral rinse samples, selected based on self-reported smoking and serum cotinine levels, from the 2013-2014 New York City Health and Nutrition Examination Study. We identified differentially abundant operational taxonomic units (OTUs) by primary and secondhand tobacco exposure, and used "microbe set enrichment analysis" to assess shifts in microbial oxygen utilization.

RESULTS

Cigarette smoking was associated with depletion of aerobic OTUs (Enrichment Score test statistic ES = -0.75, P = .002) with a minority (29%) of aerobic OTUs enriched in current smokers compared with never smokers. Consistent shifts in the microbiota were observed for current cigarette smokers as for nonsmokers with secondhand exposure as measured by serum cotinine levels. Differential abundance findings were similar in crude and adjusted analyses.

CONCLUSIONS

Results support a plausible link between tobacco exposure and shifts in the oral microbiome at the population level through three lines of evidence: (1) a shift in microbiota oxygen utilization associated with primary tobacco smoke exposure; (2) consistency of abundance fold changes associated with current smoking and shifts along the gradient of secondhand smoke exposure among nonsmokers; and (3) consistency after adjusting for a priori hypothesized confounders.

Impact of a Municipal Policy Restricting Trans Fatty Acid Use in New York City Restaurants on Serum Trans Fatty Acid Levels in Adults

OBJECTIVES

To estimate the impact of the 2006 policy restricting use of trans fatty acids (TFAs) in New York City restaurants on change in serum TFA concentrations in New York City adults.

METHODS

Two cross-sectional population-based New York City Health and Nutrition Examination Surveys conducted in 2004 (n = 212) and 2013-2014 (n = 247) provided estimates of serum TFA exposure and average frequency of weekly restaurant meals. We estimated the geometric mean of the sum of serum TFAs by year and restaurant meal frequency by using linear regression.

RESULTS

Among those who ate less than 1 restaurant meal per week, geometric mean of the sum of serum TFAs declined 51.1% (95% confidence interval [CI] = 42.7, 58.3)-from 44.6 (95% CI = 39.7, 50.1) to 21.8 (95% CI = 19.3, 24.5) micromoles per liter. The decline in the geometric mean was greater (P for interaction = .04) among those who ate 4 or more restaurant meals per week: 61.6% (95% CI = 55.8, 66.7) or from 54.6 (95% CI = 49.3, 60.5) to 21.0 (95% CI = 18.9, 23.3) micromoles per liter.

CONCLUSIONS

New York City adult serum TFA concentrations declined between 2004 and 2014. The indication of greater decline in serum TFAs among those eating restaurant meals more frequently suggests that the municipal restriction on TFA use was effective in reducing TFA exposure. Public Health Implications. Local policies focused on restaurants can promote nutritional improvements.

Using Calibration to Reduce Measurement Error in Prevalence Estimates Based on Electronic Health Records

INTRODUCTION

Increasing adoption of electronic health record (EHR) systems by health care providers presents an opportunity for EHR-based population health surveillance. EHR data, however, may be subject to measurement error because of factors such as data entry errors and lack of documentation by physicians. We investigated the use of a calibration model to reduce bias of prevalence estimates from the New York City (NYC) Macroscope, an EHR-based surveillance system.

METHODS

We calibrated 6 health indicators to the 2013-2014 NYC Health and Nutrition Examination Survey (NYC HANES) data: hypertension, diabetes, smoking, obesity, influenza vaccination, and depression. We classified indicators into having low measurement error or high measurement error on the basis of whether the proportion of misclassification (ie, false-negative or false-positive cases) was greater than 15% in 190 reviewed charts. We compared bias (ie, absolute difference between NYC Macroscope estimates and NYC HANES estimates) before and after calibration.

RESULTS

The health indicators with low measurement error had the same bias after calibration as before calibration (diabetes, 2.5 percentage points; smoking, 2.5 percentage points; obesity, 3.5 percentage points; hypertension, 1.1 percentage points). For indicators with high measurement error, bias decreased from 10.8 to 2.5 percentage points for depression, and from 26.7 to 8.4 percentage points for influenza vaccination.

CONCLUSION

The calibration model has the potential to reduce bias of prevalence estimates from EHR-based surveillance systems for indicators with high measurement errors. Further research is warranted to assess the utility of the current calibration model for other EHR data and additional indicators.

Augmenting community-level social determinants of health data with individual-level survey data

Social determinants of health (SDH) such as education and socioeconomic status are strongly associated with health and health outcomes. Incorporating SDH variables into clinical data sets could therefore improve the accuracy of predictive analytics, but individual-level SDH are rarely available and must be inferred from community-level data. We propose a method for doing so leveraging the joint probability distribution of the basic demographics available from the patient's clinical record and known community-level SDH. We demonstrate the method using two data sets, the New York City (NYC) subset of the US census data and the NYC Health and Nutrition Estimation Survey (NYCHANES) and provide sample results for 2 census tracts in NYC. The advantage of this approach is that it does not simplistically assume that all residents within a census tract share the same average/median socioeconomic status, but instead recognizes and leverages the strong known associations between demographics and SDH within localities. Results could explain some of the discrepancies appearing in the SDH-big data literature. Future studies are needed for using the augmented SHD to improve clinically relevant use cases, such as predictive analytics.

Tracking Declines in Mercury Exposure in the New York City Adult Population, 2004-2014

Mercury is a toxic metal that can be measured in human blood and urine. Population-based biomonitoring from 2004 guided New York City (NYC) Department of Health and Mental Hygiene (DOHMH) efforts to reduce exposures by educating the public about risks and benefits of fish consumption-a predominant source of exposure in the general population-and removing mercury-containing skin-lightening creams and other consumer products from the marketplace. We describe changes in exposures over the past decade in relation to these local public health actions and in the context of national changes by comparing mercury concentrations measured in blood (1201 specimens) and urine (1408 specimens) from the NYC Health and Nutrition Examination Survey (NYC HANES) 2013-2014 with measurements from NYC HANES 2004 and National Health and Nutrition Examination Surveys (NHANES) 2003-2004 and 2013-2014. We found that NYC adult blood and urine geometric mean mercury concentrations decreased 46% and 45%, respectively. Adult New Yorkers with blood mercury concentration ≥ 5 μg/L (the New York State reportable level) declined from 24.8% (95% CL = 22.2%, 27.7%) to 12.0% (95% CL = 10.1%, 14.3%). The decline in blood mercury in NYC was greater than the national decline, while the decline in urine mercury was similar. As in 2004, Asian New Yorkers had higher blood mercury concentrations than other racial/ethnic groups. Foreign-born adults of East or Southeast Asian origin had the highest prevalence of reportable levels (29.7%; 95% CL = 21.0%, 40.1%) across sociodemographic groups, and Asians generally were the most frequent fish consumers, eating on average 11 fish meals in the past month compared with 7 among other groups (p < 0.001). Fish consumption patterns were similar over time, and fish continues to be consumed more frequently in NYC than nationwide (24.7% of NYC adults ate fish ten or more times in the past 30 days vs. 14.7% nationally, p < 0.001). The findings are consistent with the hypothesis that blood mercury levels have declined in part because of local and national efforts to promote consumption of lower mercury fish. Local NYC efforts may have accelerated the reduction in exposure. Having "silver-colored fillings" on five or more teeth was associated with the highest 95th percentile for urine mercury (4.06 μg/L; 95% CL = 3.1, 5.9). An estimated 5.5% of the adult population (95% CL = 4.3%, 7.0%) reported using a skin-lightening cream in the past 30 days, but there was little evidence that use was associated with elevated urine mercury in 2013-14.

The Burden of Depression in New York City Adults: Results from the 2013-14 NYC Health and Nutrition Examination Survey

Depression is responsible for a large burden of disability in the USA. We estimated the prevalence of depression in the New York City (NYC) adult population in 2013-14 and examined associations with demographics, health behaviors, and employment status. Data from the 2013-14 New York City Health and Nutrition Examination Survey, a population-based examination study, were analyzed, and 1459 participants met the inclusion criteria for this analysis. We defined current symptomatic depression by a Patient Health Questionnaire (PHQ-9) score ≥ 10. Overall, 8.3% of NYC adults had current symptomatic depression. New Yorkers with current symptomatic depression were significantly more likely to be female, Latino, and unemployed yet not looking for work; they were also significantly more likely to have less than a high school education and to live in a high-poverty neighborhood. Socioeconomic inequalities in mental health persist in NYC and highlight the need for better diagnosis and treatment.

Change in Obesity Prevalence among New York City Adults: the NYC Health and Nutrition Examination Survey, 2004 and 2013-2014

The objective of this study was to measure change in obesity prevalence among New York City (NYC) adults from 2004 to 2013-2014 and assess variation across sociodemographic subgroups. We used objectively measured height and weight data from the NYC Health and Nutrition Examination Survey to calculate relative percent change in obesity (≥ 30 kg/m2) between 2004 (n = 1987) and 2013-2014 (n = 1489) among all NYC adults and sociodemographic subgroups. We also examined changes in self-reported proxies for energy imbalance. Estimates were age-standardized and statistical significance was evaluated using two-tailed T tests and multivariable regression (p < 0.05). Between 2004 and 2013-2014, obesity increased from 27.5 to 32.4% (p = 0.01). Prevalence remained stable and high among women (31.2 to 32.8%, p = 0.53), but increased among men (23.4 to 32.0%, p = 0.002), especially among non-Latino White men and men age ≥ 65 years. Black adults had the highest prevalence in 2013-2014 (37.1%) and Asian adults experienced the largest increase (20.1 to 29.2%, p = 0.06), especially Asian women. Foreign-born participants and participants lacking health insurance also had large increases in obesity. We observed increases in eating out and screen time over time and no improvements in physical activity. Our findings show increases in obesity in NYC in the past decade, with important sociodemographic differences.

Change in Diabetes Prevalence and Control among New York City Adults: NYC Health and Nutrition Examination Surveys 2004-2014

National examination surveys provide trend information on diabetes prevalence, diagnoses, and control. Few localities have access to such information. Using a similar design as the National Health and Nutrition Examination Survey (NHANES), two NYC Health and Nutrition Examination Surveys (NYC HANES) were conducted over a decade, recruiting adults ≥ 20 years using household probability samples (n = 1808 in 2004; n = 1246 in 2013-2014) and physical exam survey methods benchmarked against NHANES. Participants had diagnosed diabetes if told by a health provider they had diabetes, and undiagnosed diabetes if they had no diagnosis but a fasting plasma glucose ≥ 126 mg/dl or A1C ≥ 6.5%. We found that between 2004 and 2014, total diabetes prevalence (diagnosed and undiagnosed) in NYC increased from 13.4 to 16.0% (P = 0.089). In 2013-2014, racial/ethnic disparities in diabetes burden had widened; diabetes was highest among Asians (24.6%), and prevalence was significantly lower among non-Hispanic white adults (7.7%) compared to that among other racial/ethnic groups (P < 0.001). Among adults with diabetes, the proportion of cases diagnosed increased from 68.3 to 77.3% (P = 0.234), and diagnosed cases with very poor control (A1C > 9%), decreased from 26.9 to 18.0% (P = 0.269), though both were non-significant. While local racial/ethnic disparities in diabetes prevalence persist, findings suggest modest improvements in diabetes diagnosis and management.

Disparities in Sleep Problems by Sexual Orientation among New York City Adults: an Analysis of the New York City Health and Nutrition Examination Survey, 2013-2014

We examined disparities in sleep problems by sexual orientation among a population-based sample of adults, using data from the New York City (NYC) Health and Nutrition Examination Survey (NYC HANES), a population-based, cross-sectional survey conducted in 2013-2014 (n = 1220). Two log binomial regression models were created to assess the relative prevalence of sleep problems by sexual orientation. In model 1, heterosexual adults served as the reference category, controlling for gender, age, race/ethnicity, education, marital status, and family income. And in model 2, heterosexual men served as the reference category, controlling for age, race/ethnicity, education, marital status, and family income. We found that almost 42% of NYC adults reported sleep problems in the past 2 weeks. Bisexual adults had 1.4 times the relative risk of sleep problems compared to heterosexual adults (p = 0.037). Compared to heterosexual men, heterosexual and bisexual women had 1.3 and 1.6 times the risk of sleep problems, respectively (p < 0.05). Overall, adults who self-identified as bisexual had a significantly greater risk of sleep problems than adults who self-identified as heterosexual.

Gender and Race Disparities in Cardiovascular Disease Risk Factors among New York City Adults: New York City Health and Nutrition Examination Survey (NYC HANES) 2013-2014

While gender and racial/ethnic disparities in cardiovascular disease (CVD) risk factors have each been well characterized, few studies have comprehensively examined how patterns of major CVD risk factors vary and intersect across gender and major racial/ethnic groups, considered together. Using data from New York City Health and Nutrition Examination Survey 2013-2014-a population-based, cross-sectional survey of NYC residents ages 20 years and older-we measured prevalence of obesity, hypertension, hypercholesterolemia, smoking, and diabetes across gender and race/ethnicity groups for 1527 individuals. We used logistic regression with predicted marginal to estimate age-adjusted prevalence ratio by gender and race/ethnicity groups and assess for potential additive and multiplicative interaction. Overall, women had lower prevalence of CVD risk factors than men, with less hypertension (p = 0.040), lower triglycerides (p < 0.001), higher HDL (p < 0.001), and a greater likelihood of a heart healthy lifestyle, more likely not to smoke and to follow a healthy diet (p < 0.05). When further stratified by race/ethnicity, however, the female advantage was largely restricted to non-Latino white women. Non-Latino black women had significantly higher risk of being overweight or obese, having hypertension, and having diabetes than non-Latino white men or women, or than non-Latino black men (p < 0.05). Non-Latino black women also had higher total cholesterol compared to non-Latino black men (184.4 vs 170.5 mg/dL, p = 0.010). Despite efforts to improve cardiovascular health and narrow disparities, non-Latino black women continue to have a higher burden of CVD risk factors than other gender and racial/ethnic groups. This study highlights the importance of assessing for intersectionality between gender and race/ethnicity groups when examining CVD risk factors.

State-level estimation of diabetes and prediabetes prevalence: Combining national and local survey data and clinical data

Many statisticians and policy researchers are interested in using data generated through the normal delivery of health care services, rather than carefully designed and implemented population-representative surveys, to estimate disease prevalence. These larger databases allow for the estimation of smaller geographies, for example, states, at potentially lower expense. However, these health care records frequently do not cover all of the population of interest and may not collect some covariates that are important for accurate estimation. In a recent paper, the authors have described how to adjust for the incomplete coverage of administrative claims data and electronic health records at the state or local level. This article illustrates how to adjust and combine multiple data sets, namely, national surveys, state-level surveys, claims data, and electronic health record data, to improve estimates of diabetes and prediabetes prevalence, along with the estimates of the method's accuracy. We demonstrate and validate the method using data from three jurisdictions (Alabama, California, and New York City). This method can be applied more generally to other areas and other data sources.

Declines in adult blood lead levels in New York City compared with the United States, 2004-2014

OBJECTIVES

To assess changes in lead exposure in the New York City (NYC) adult population over a 10-year period and to contrast changes with national estimates, overall, and by socio-demographics and smoking status.

METHODS

We used measurements of blood lead levels (BLLs) from NYC resident adults who participated in the NYC Health and Nutrition Examination Surveys (HANES) in 2004 and 2013-2014. We compared estimates of geometric means (GM), 95th percentiles, and prevalence of BLL ≥ 5 µg/dL overall and by subgroups over time, with adults who participated in the National HANES (NHANES) 2001-2004 and 2011-2014.

RESULTS

The GM BLLs among NYC adults declined from 1.79 µg/dL in 2004 to 1.13 µg/dL in 2013-2014 (P < .0001). The declines over this period ranged from 30.1% to 43.2% across socio-demographic groups and smoking status (P < .0001 for all comparisons), and were slightly greater than declines observed nationally. The drop in prevalence of elevated BLLs (≥ 5 µg/dL) was also greater in NYC (4.8-0.5%), compared with NHANES (3.8-2.0%). By 2013-2014, NYC adults with lower annual family income (< $20,000) no longer had higher GM BLLs relative to those with higher incomes (≥ $75,000), a disparity improvement not observed nationally. Likewise, GM BLLs and 95th percentiles for non-Hispanic black adults in NYC were lower than GM BLLs for non-Hispanic white adults. Non-Hispanic Asian adults had the highest GM BLLs compared with other racial/ethnic groups, both in NYC in 2013-14 and nationally in 2011-2014 (1.37 µg/dL, P = .1048 and 1.22 µg/dL, P = .0004, respectively).

CONCLUSION

The lessening of disparity in lead exposure across income groups and decreasing exposure at the high end of the distribution among non-Hispanic black and Asian adults in NYC suggest that regulatory and outreach efforts have effectively targeted these higher exposure risk groups. However, Asian adults still had the highest average BLL, suggesting a need for enhanced outreach to this group. Local surveillance remains an important tool to monitor BLLs of local populations and to inform initiatives to reduce exposures in those at highest risk.

Generalizability of Indicators from the New York City Macroscope Electronic Health Record Surveillance System to Systems Based on Other EHR Platforms

Introduction:

The New York City (NYC) Macroscope is an electronic health record (EHR) surveillance system based on a distributed network of primary care records from the Hub Population Health System. In a previous 3-part series published in eGEMS, we reported the validity of health indicators from the NYC Macroscope; however, questions remained regarding their generalizability to other EHR surveillance systems.

Methods:

We abstracted primary care chart data from more than 20 EHR software systems for 142 participants of the 2013–14 NYC Health and Nutrition Examination Survey who did not contribute data to the NYC Macroscope. We then computed the sensitivity and specificity for indicators, comparing data abstracted from EHRs with survey data.

Results:

Obesity and diabetes indicators had moderate to high sensitivity (0.81–0.96) and high specificity (0.94–0.98). Smoking status and hypertension indicators had moderate sensitivity (0.78–0.90) and moderate to high specificity (0.88–0.98); sensitivity improved when the sample was restricted to records from providers who attested to Stage 1 Meaningful Use. Hyperlipidemia indicators had moderate sensitivity (≥0.72) and low specificity (≤0.59), with minimal changes when restricting to Stage 1 Meaningful Use.

Discussion:

Indicators for obesity and diabetes used in the NYC Macroscope can be adapted to other EHR surveillance systems with minimal validation. However, additional validation of smoking status and hypertension indicators is recommended and further development of hyperlipidemia indicators is needed.

Conclusion:

Our findings suggest that many of the EHR-based surveillance indicators developed and validated for the NYC Macroscope are generalizable for use in other EHR surveillance systems.

Exploring the Link Between Posttraumatic Stress Disorder and inflammation-Related Medical Conditions: An Epidemiological Examination

There have been few epidemiological studies exploring the link between PTSD and inflammation using population-based samples. This study examined the relation between posttraumatic stress disorder (PTSD) and inflammation-related medical conditions using data from the 2013-2014 New York City Health and Nutrition Examination Survey. Using a representative sample of 1,527 residents in New York City, the association between PTSD and 17 inflammation-related medical conditions were examined. Bivariate and multivariable analyses were conducted, adjusting for demographic characteristics and lifetime depression. PTSD was strongly associated with increased odds for hypercholesterolemia, insulin resistance, angina, heart attack, and emphysema with the greatest odds observed for heart attack (OR= 3.94) and emphysema (OR= 4.06). But PTSD was also associated with lower odds for hypertension, type 1 diabetes, asthma, coronary heart disease, stroke, osteoporosis, and a failing kidney with the lowest odds observed for type 1 diabetes (OR= 0.43). These findings suggest a complex link between PTSD and inflammation-related medical conditions.

Comparing Prevalence Estimates From Population-Based Surveys to Inform Surveillance Using Electronic Health Records

INTRODUCTION

Electronic health record (EHR) systems provide an opportunity to use a novel data source for population health surveillance. Validation studies that compare prevalence estimates from EHRs and surveys most often use difference testing, which can, because of large sample sizes, lead to detection of significant differences that are not meaningful. We explored a novel application of the two one-sided t test (TOST) to assess the equivalence of prevalence estimates in 2 population-based surveys to inform margin selection for validating EHR-based surveillance prevalence estimates derived from large samples.

METHODS

We compared prevalence estimates of health indicators in the 2013 Community Health Survey (CHS) and the 2013-2014 New York City Health and Nutrition Examination Survey (NYC HANES) by using TOST, a 2-tailed t test, and other goodness-of-fit measures.

RESULTS

A ±5 percentage-point equivalence margin for a TOST performed well for most health indicators. For health indicators with a prevalence estimate of less than 10% (extreme obesity [CHS, 3.5%; NYC HANES, 5.1%] and serious psychological distress [CHS, 5.2%; NYC HANES, 4.8%]), a ±2.5 percentage-point margin was more consistent with other goodness-of-fit measures than the larger percentage-point margins.

CONCLUSION

A TOST with a ±5 percentage-point margin was useful in establishing equivalence, but a ±2.5 percentage-point margin may be appropriate for health indicators with a prevalence estimate of less than 10%. Equivalence testing can guide future efforts to validate EHR data.

Innovations in Population Health Surveillance: Using Electronic Health Records for Chronic Disease Surveillance

With 87% of providers using electronic health records (EHRs) in the United States, EHRs have the potential to contribute to population health surveillance efforts. However, little is known about using EHR data outside syndromic surveillance and quality improvement. We created an EHR-based population health surveillance system called the New York City (NYC) Macroscope and assessed the validity of diabetes, hyperlipidemia, hypertension, smoking, obesity, depression, and influenza vaccination indicators. The NYC Macroscope uses aggregate data from a network of outpatient practices. We compared 2013 NYC Macroscope prevalence estimates with those from a population-based, in-person examination survey, the 2013-2014 NYC Health and Nutrition Examination Survey. NYC Macroscope diabetes, hypertension, smoking, and obesity prevalence indicators performed well, but depression and influenza vaccination estimates were substantially lower than were survey estimates. Ongoing validation will be important to monitor changes in validity over time as EHR networks mature and to assess new indicators. We discuss NYC's experience and how this project fits into the national context. Sharing lessons learned can help achieve the full potential of EHRs for population health surveillance.

Feasibility of Using HIV Care-Continuum Outcomes to Identify Geographic Areas for Targeted HIV Testing

BACKGROUND

Improved detection and linkage to care of previously undiagnosed HIV infections require innovative approaches to testing. We sought to determine the feasibility of targeted HIV testing in geographic areas, defined by continuum of care parameters, to identify HIV-infected persons needing linkage or engagement in care.

METHODS

Using HIV surveillance data from Washington, DC, we identified census tracts that had an HIV prevalence >1% and were either above (higher risk areas-HRAs) or below (lower risk areas-LRAs) the median for 3 indicators: monitored viral load, proportion of persons out of care (OOC), and never in care. Community-based HIV rapid testing and participant surveys were conducted in the 20 census tracts meeting the criteria. Areas were mapped using ArcGIS, and descriptive and univariate analyses were conducted comparing the areas and participants.

RESULTS

Among 1471 persons tested, 28 (1.9%) tested HIV positive; 2.1% in HRAs vs. 1.7% in LRAs (P = 0.57). Higher proportions of men (63.7% vs. 56.7%, P = 0.007) and fewer blacks (91.0% vs. 94.6%, P = 0.008) were tested in LRAs vs. HRAs; no differences were observed in risk behaviors between the areas. Among HIV-positive participants, 54% were new diagnoses (n = 9) or OOC (n = 6), all were Black, 64% were men with a median age of 51 years.

CONCLUSIONS

Although significant differences in HIV seropositivity were not observed between testing areas, our approach proved feasible and enabled identification of new diagnoses and OOC HIV-infected persons. This testing paradigm could be adapted in other locales to identify areas for targeted HIV testing and other reengagement efforts.

Racial and Ethnic Subgroup Disparities in Hypertension Prevalence, New York City Health and Nutrition Examination Survey, 2013-2014

Introduction

Racial/ethnic minority adults have higher rates of hypertension than non-Hispanic white adults. We examined the prevalence of hypertension among Hispanic and Asian subgroups in New York City.

Methods

Data from the 2013–2014 New York City Health and Nutrition Examination Survey were used to assess hypertension prevalence among adults (aged ≥20) in New York City (n = 1,476). Hypertension was measured (systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or self-reported hypertension and use of blood pressure medication). Participants self-reported race/ethnicity and country of origin. Multivariable logistic regression models assessed differences in prevalence by race/ethnicity and sociodemographic and health-related characteristics.

Results

Overall hypertension prevalence among adults in New York City was 33.9% (43.5% for non-Hispanic blacks, 38.0% for Asians, 33.0% for Hispanics, and 27.5% for non-Hispanic whites). Among Hispanic adults, prevalence was 39.4% for Dominican, 34.2% for Puerto Rican, and 27.5% for Central/South American adults. Among Asian adults, prevalence was 43.0% for South Asian and 39.9% for East/Southeast Asian adults. Adjusting for age, sex, education, and body mass index, 2 major racial/ethnic minority groups had higher odds of hypertension than non-Hispanic whites: non-Hispanic black (AOR [adjusted odds ratio], 2.6; 95% confidence interval [CI], 1.7–3.9) and Asian (AOR, 2.0; 95% CI, 1.2–3.4) adults. Two subgroups had greater odds of hypertension than the non-Hispanic white group: East/Southeast Asian adults (AOR, 2.8; 95% CI, 1.6–4.9) and Dominican adults (AOR, 1.9; 95% CI, 1.1–3.5).

Conclusion

Racial/ethnic minority subgroups vary in hypertension prevalence, suggesting the need for targeted interventions.

Can Electronic Health Records Be Used for Population Health Surveillance? Validating Population Health Metrics Against Established Survey Data

INTRODUCTION

Electronic health records (EHRs) offer potential for population health surveillance but EHR-based surveillance measures require validation prior to use. We assessed the validity of obesity, smoking, depression, and influenza vaccination indicators from a new EHR surveillance system, the New York City (NYC) Macroscope. This report is the second in a 3-part series describing the development and validation of the NYC Macroscope. The first report describes in detail the infrastructure underlying the NYC Macroscope; design decisions that were made to maximize data quality; characteristics of the population sampled; completeness of data collected; and lessons learned from doing this work. This second report, which addresses concerns related to sampling bias and data quality, describes the methods used to evaluate the validity and robustness of NYC Macroscope prevalence estimates; presents validation results for estimates of obesity, smoking, depression and influenza vaccination; and discusses the implications of our findings for NYC and for other jurisdictions embarking on similar work. The third report applies the same validation methods described in this report to metabolic outcomes, including the prevalence, treatment and control of diabetes, hypertension and hyperlipidemia.

METHODS

NYC Macroscope prevalence estimates, overall and stratified by sex and age group, were compared to reference survey estimates for adult New Yorkers who reported visiting a doctor in the past year. Agreement was evaluated against 5 a priori criteria. Sensitivity and specificity were assessed by examining individual EHR records in a subsample of 48 survey participants.

RESULTS

Among adult New Yorkers in care, the NYC Macroscope prevalence estimate for smoking (15.2%) fell between estimates from NYC HANES (17.7 %) and CHS (14.9%) and met all 5 a priori criteria. The NYC Macroscope obesity prevalence estimate (27.8%) also fell between the NYC HANES (31.3%) and CHS (24.7%) estimates, but met only 3 a priori criteria. Sensitivity and specificity exceeded 0.90 for both the smoking and obesity indicators. The NYC Macroscope estimates of depression and influenza vaccination prevalence were more than 10 percentage points lower than the estimates from either reference survey. While specificity was > 0.90 for both of these indicators, sensitivity was < 0.70.

DISCUSSION

Through this work we have demonstrated that EHR data from a convenience sample of providers can produce acceptable estimates of smoking and obesity prevalence among adult New Yorkers in care; gained a better understanding of the challenges involved in estimating depression prevalence from EHRs; and identified areas for additional research regarding estimation of influenza vaccination prevalence. We have also shared lessons learned about how EHR indicators should be constructed and offer methodologic suggestions for validating them.

CONCLUSIONS

This work adds to a rapidly emerging body of literature about how to define, collect and interpret EHR-based surveillance measures and may help guide other jurisdictions.

Design of the New York City Macroscope: Innovations in Population Health Surveillance Using Electronic Health Records

Introduction:

Electronic health records (EHRs) have the potential to offer real-time, inexpensive standardized health data about chronic health conditions. Despite rapid expansion, EHR data evaluations for chronic disease surveillance have been limited. We present design and methods for the New York City (NYC) Macroscope, an EHR-based chronic disease surveillance system.

This methods report is the first in a three part series describing the development and validation of the NYC Macroscope. This report describes in detail the infrastructure underlying the NYC Macroscope; indicator definitions; design decisions that were made to maximize data quality; characteristics of the population sampled; completeness of data collected; and lessons learned from doing this work. The second report describes the methods used to evaluate the validity and robustness of NYC Macroscope prevalence estimates; presents validation results for estimates of obesity, smoking, depression and influenza vaccination; and discusses the implications of our findings for NYC and for other jurisdictions embarking on similar work. The third report applies the same validation methods to metabolic outcomes, including the prevalence, treatment and control of diabetes, hypertension and hyperlipidemia.

Methods:

We designed the NYC Macroscope for comparison to a local “gold standard,” the 2013–14 NYC Health and Nutrition Examination Survey, and the telephonic 2013 Community Health Survey. NYC Macroscope indicators covered prevalence, treatment, and control of diabetes, hypertension, and hyperlipidemia; and prevalence of influenza vaccination, obesity, depression and smoking. Indicators were stratified by age, sex, and neighborhood poverty, and weighted to the in-care NYC population and limited to primary care patients. Indicator queries were distributed to a virtual network of primary care practices; 392 practices and 716,076 adult patients were retained in the final sample.

Findings:

The NYC Macroscope covered 10% of primary care providers and 15% of all adult patients in NYC in 2013 (8–47% of patients by neighborhood). Data completeness varied by domain from 98% for blood pressure among patients with hypertension to 33% for depression screening.

Discussion:

Design and validation efforts undertaken by NYC are described here to provide one potential blueprint for leveraging EHRs for population health monitoring. To replicate a model like NYC Macroscope, jurisdictions should establish buy-in; build informatics capacity; use standard, simple case defnitions; establish documentation quality thresholds; restrict to primary care providers; and weight the sample to a target population.

Monitoring Prevalence, Treatment, and Control of Metabolic Conditions in New York City Adults Using 2013 Primary Care Electronic Health Records: A Surveillance Validation Study

Introduction:

Electronic health records (EHRs) can potentially extend chronic disease surveillance, but few EHR-based initiatives tracking population-based metrics have been validated for accuracy. We designed a new EHR-based population health surveillance system for New York City (NYC) known as NYC Macroscope. This report is the third in a 3-part series describing the development and validation of that system. The first report describes governance and technical infrastructure underlying the NYC Macroscope. The second report describes validation methods and presents validation results for estimates of obesity, smoking, depression and influenza vaccination. In this third paper we present validation findings for metabolic indicators (hypertension, hyperlipidemia, diabetes).

Methods:

We compared EHR-based estimates to those from a gold standard surveillance source - the 2013–2014 NYC Health and Nutrition Examination Survey (NYC HANES) - overall and stratified by sex and age group, using the two one-sided test of equivalence and other validation criteria.

Results:

EHR-based hypertension prevalence estimates were highly concordant with NYC HANES estimates. Diabetes prevalence estimates were highly concordant when measuring diagnosed diabetes but less so when incorporating laboratory results. Hypercholesterolemia prevalence estimates were less concordant overall. Measures to assess treatment and control of the 3 metabolic conditions performed poorly.

Discussion:

While indicator performance was variable, findings here confirm that a carefully constructed EHR-based surveillance system can generate prevalence estimates comparable to those from gold-standard examination surveys for certain metabolic conditions such as hypertension and diabetes.

Conclusions:

Standardized EHR metrics have potential utility for surveillance at lower annual costs than surveys, especially as representativeness of contributing clinical practices to EHR-based surveillance systems increases.

World Health Organization disability assessment schedule 2.0: An international systematic review

PURPOSE

This systematic review examines research and practical applications of the World Health Organization Disability Assessment Schedule (WHODAS 2.0) as a basis for establishing specific criteria for evaluating relevant international scientific literature. The aims were to establish the extent of international dissemination and use of WHODAS 2.0 and analyze psychometric research on its various translations and adaptations. In particular, we wanted to highlight which psychometric features have been investigated, focusing on the factor structure, reliability, and validity of this instrument.

METHOD

Following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) methodology, we conducted a search for publications focused on "whodas" using the ProQuest, PubMed, and Google Scholar electronic databases.

RESULTS

We identified 810 studies from 94 countries published between 1999 and 2015. WHODAS 2.0 has been translated into 47 languages and dialects and used in 27 areas of research (40% in psychiatry).

CONCLUSIONS

The growing number of studies indicates increasing interest in the WHODAS 2.0 for assessing individual functioning and disability in different settings and individual health conditions. The WHODAS 2.0 shows strong correlations with several other measures of activity limitations; probably due to the fact that it shares the same disability latent variable with them. Implications for Rehabilitation WHODAS 2.0 seems to be a valid, reliable self-report instrument for the assessment of disability. The increasing interest in use of the WHODAS 2.0 extends to rehabilitation and life sciences rather than being limited to psychiatry. WHODAS 2.0 is suitable for assessing health status and disability in a variety of settings and populations. A critical issue for rehabilitation is that a single "minimal clinically important .difference" score for the WHODAS 2.0 has not yet been established.

Health-Specific Information and Communication Technology Use and Its Relationship to Obesity in High-Poverty, Urban Communities: Analysis of a Population-Based Biosocial Survey

Background

More than 35% of American adults are obese. For African American and Hispanic adults, as well as individuals residing in poorer or more racially segregated urban neighborhoods, the likelihood of obesity is even higher. Information and communication technologies (ICTs) may substitute for or complement community-based resources for weight management. However, little is currently known about health-specific ICT use among urban-dwelling people with obesity.

Objective

We describe health-specific ICT use and its relationship to measured obesity among adults in high-poverty urban communities.

Methods

Using data collected between November 2012 and July 2013 from a population-based probability sample of urban-dwelling African American and Hispanic adults residing on the South Side of Chicago, we described patterns of ICT use in relation to measured obesity defined by a body mass index (BMI) of ≥30 kg/m². Among those with BMI≥30 kg/m², we also assessed the association between health-specific ICT use and diagnosed versus undiagnosed obesity as well as differences in health-specific ICT use by self-reported comorbidities, including diabetes and hypertension.

Results

The survey response rate was 44.6% (267 completed surveys/598.4 eligible or likely eligible individuals); 53.2% were African American and 34.6% were Hispanic. More than 35% of the population reported an annual income of less than US $25,000. The population prevalence of measured obesity was 50.2%. People with measured obesity (BMI≥30 kg/m²) were more likely to report both general (81.5% vs 67.0%, P=.04) and health-specific (61.1% vs 41.2%, P=.01) ICT use. In contrast, among those with measured obesity, being told of this diagnosis by a physician was not associated with increased health-specific ICT use. People with measured obesity alone had higher rates of health-specific use than those with comorbid hypertension and/or diabetes diagnoses (77.1% vs 60.7% vs 47.4%, P=.04).

Conclusions

In conclusion, ICT-based health resources may be particularly useful for people in high-poverty urban communities with isolated measured obesity, a population that is at high risk for poor health outcomes.

Exposure to Secondhand Smoke Among Nonsmokers in New York City in the Context of Recent Tobacco Control Policies: Current Status, Changes Over the Past Decade, and National Comparisons

INTRODUCTION

Exposure to secondhand smoke is hazardous and can cause cancer, coronary heart disease, and birth defects. New York City (NYC) and other jurisdictions have established smoke-free air laws in the past 10-15 years.

METHODS

NYC Health and Nutrition Examination Survey (HANES) 2013-2014 was a population-based survey of NYC residents, aged 20 years and older, in which biospecimens were collected and cotinine levels were measured. Secondhand smoke exposure was assessed by demographics and risk factors and compared with that from NYC HANES 2004 and national HANES.

RESULTS

More than a third (37.1%, 95% confidence interval [CI] = 33.3%-41.2%) of nonsmoking adult New Yorkers were exposed to secondhand smoke, defined as a cotinine level of 0.05-10ng/mL. This was significantly lower than in 2004 NYC HANES, when 56.7% (95% CI = 53.6%-59.7%) of nonsmokers were exposed to secondhand smoke, but was greater than the proportion of adults exposed nationwide, as measured by national HANES (24.4%, 95% CI = 22.0%-26.9% in 2011-2012). Men, non-Hispanic blacks, adults aged 20-39, those with less education, and those living in high-poverty neighborhoods were more likely to be exposed.

CONCLUSIONS

There has been a large decrease in secondhand smoke exposure in NYC, although disparities persist. The decrease may be the result of successful policies to limit exposure to secondhand smoke in public places and of smokers smoking fewer cigarettes per day. Yet NYC residents still experience more secondhand smoke exposure than US residents overall. Possible explanations include multiunit housing, greater population density, and pedestrian exposure.

IMPLICATIONS

Measuring exposure to secondhand smoke can be difficult, and few studies have monitored changes over time. This study uses serum cotinine, a nicotine metabolite, from a local population-based examination survey, the NYC HANES 2013-2014, to examine exposure to secondhand smoke in an urban area that has implemented stringent antismoking laws. Comparison with NYC HANES conducted 10 years ago allows for an assessment of changes in the last decade in the context of municipal tobacco control policies. Results may be helpful to jurisdictions considering implementing similar tobacco control policies.

Characterizing Adults Receiving Primary Medical Care in New York City: Implications for Using Electronic Health Records for Chronic Disease Surveillance

Introduction

Electronic health records (EHRs) from primary care providers can be used for chronic disease surveillance; however, EHR-based prevalence estimates may be biased toward people who seek care. This study sought to describe the characteristics of an in-care population and compare them with those of a not-in-care population to inform interpretation of EHR data.

Methods

We used data from the 2013–2014 New York City Health and Nutrition Examination Survey (NYC HANES), considered the gold standard for estimating disease prevalence, and the 2013 Community Health Survey, and classified participants as in care or not in care, on the basis of their report of seeing a health care provider in the previous year. We used χ² tests to compare the distribution of demographic characteristics, health care coverage and access, and chronic conditions between the 2 populations.

Results

According to the Community Health Survey, approximately 4.1 million (71.7%) adults aged 20 or older had seen a health care provider in the previous year; according to NYC HANES, approximately 4.7 million (75.1%) had. In both surveys, the in-care population was more likely to be older, female, non-Hispanic, and insured compared with the not-in-care population. The in-care population from the NYC HANES also had a higher prevalence of diabetes (16.7% vs 6.9%; P < .001), hypercholesterolemia (35.7% vs 22.3%; P < .001), and hypertension (35.5% vs 26.4%; P < .001) than the not-in-care population.

Conclusion

Systematic differences between in-care and not-in-care populations warrant caution in using primary care data to generalize to the population at large. Future efforts to use primary care data for chronic disease surveillance need to consider the intended purpose of data collected in these systems as well as the characteristics of the population using primary care.