Estimating county-level vaccination coverage using small area estimation with the National Immunization Survey-Child.

The National Immunization Survey-Child (NIS-Child) provides annual vaccination coverage estimates in the United States for children aged 19 through 35 months, nationally, for each state, and for select local areas and territories. There is a need for vaccination coverage estimates for smaller geographic areas to support local authority planning and identify counties with potentially low vaccination coverage for possible further intervention. We describe small area estimation methods using 2008-2018 NIS-Child data to generate county-level estimates for children up to two years of age born 2007-2011 and 2012-2016. We applied an empirical best linear unbiased prediction method to combine direct estimates of vaccination coverage with model-based prediction using county-level predictors regarding health and demographic characteristics. We review the predictors commonly selected for the small area models and note multiple predictors related to barriers to vaccination.

Developing a sampling methodology for timely reporting of population-based COVID-19-associated hospitalization surveillance in the United States, COVID-NET 2020-2021.

BACKGROUND: The COVID-19-Associated Hospitalization Surveillance Network (COVID-NET) required a sampling methodology that allowed for production of timely population-based clinical estimates to inform the ongoing US COVID-19 pandemic response. METHODS: We developed a flexible sampling approach that considered reporting delays, differential hospitalized case burden across surveillance sites, and changing geographic and demographic trends over time. We incorporated weighting methods to adjust for the probability of selection and non-response, and to calibrate the sampled case distribution to the population distribution on demographics. We additionally developed procedures for variance estimation. RESULTS: Between March 2020 and June 2021, 19,293 (10.4%) of all adult hospitalized cases were sampled for chart abstraction. Variance estimates for select variables of interest were within desired ranges. CONCLUSIONS: COVID-NET's sampling methodology allowed for reporting of robust and timely, population-based data on the clinical epidemiology of COVID-19-associated hospitalizations and evolving trends over time, while attempting to reduce data collection burden on surveillance sites. Such methods may provide a general framework for other surveillance systems needing to quickly and efficiently collect and disseminate data for public health action.

Estimation tools for reducing the impact of sampling and nonresponse errors in dual-frame RDD telephone surveys.

We discuss alternative estimators of the population total given a dual-frame random-digit-dial (RDD) telephone survey in which samples are selected from landline and cell phone sampling frames. The estimators are subject to sampling and nonsampling errors. To reduce sampling variability when an optimum balance of landline and cell phone samples is not feasible, we develop an application of shrinkage estimation. We demonstrate the implications for survey weighting of a differential nonresponse mechanism by telephone status. We illustrate these ideas using data from the National Immunization Survey-Child, a large dual-frame RDD telephone survey sponsored by the Centers for Disease Control and Prevention and conducted to measure the vaccination status of American children aged 19 to 35 months.

Spending per Medicare Beneficiary Is Higher in Hospital-Owned Small- and Medium-Sized Physician Practices.

OBJECTIVE: To examine the relationship of physician versus hospital ownership of small- and medium-sized practices with spending and utilization of care. DATA SOURCE/STUDY SETTING/DATA COLLECTION: Survey data for 1,045 primary care-based practices of 1-19 physicians linked to Medicare claims data for 2008 for 282,372 beneficiaries attributed to the 3,010 physicians in these practices. STUDY DESIGN: We used generalized linear models to estimate the associations between practice characteristics and outcomes (emergency department visits, index admissions, readmissions, and spending). PRINCIPAL FINDINGS: Beneficiaries linked to hospital-owned practices had 7.3 percent more emergency department visits and 6.4 percent higher total spending compared to beneficiaries linked to physician-owned practices. CONCLUSIONS: Physician practices are increasingly being purchased by hospitals. This may result in higher total spending on care.

Statistical Methodology of the National Immunization Survey, 2005-2014.

The National Immunization Survey (NIS) family of surveys includes NIS-Child, which monitors vaccination coverage for the U.S. population of children aged 19-35 months; NIS- Teen, which monitors vaccination coverage for the U.S. population of adolescents aged 13-17; and NIS-Flu, which monitors influenza vaccination coverage for the U.S. population of children aged 6 months through 17 years. This report describes the methods used in this family of surveys during the 2005-2014 period.

Trends in hospital ownership of physician practices and the effect on processes to improve quality.

OBJECTIVES: Reports suggest a trend for physician practices to change ownership from physicians to hospitals. It remains unclear how this change affects quality of patient care. We report the effect of a change to hospital ownership on the use of care management processes (CMPs) and health information technology (IT) among practices in the United States. STUDY DESIGN: Trend analyses of 3 large national surveys of physician practices. METHODS: We included 2 cohorts of practices: large practices with 20 or more physicians and small/medium practices with fewer than 20 physicians. The main outcomes were the changes in CMP and health IT indices among practices that were acquired by hospitals. We used multivariate logistic regression to assess these changes. RESULTS: Large practices acquired by hospitals had larger increases in their CMP index than those that remained physician-owned (11.0-point increase vs 7.0-point decrease; adjusted P = .03). Small/medium practices acquired by hospitals had smaller but significantly higher increases in their CMP score (3.8 points vs 2.6 points; adjusted P = .04). Among all practices, there were no significant differences in the change of the health IT index. CONCLUSIONS: We found a significant increase in the use of CMPs among practices that were acquired by hospitals and no difference in health IT use. These findings suggest that a trend for hospitals to own physician practices may have a positive effect on chronic disease management and quality of care.

Managing chronic illness: physician practices increased the use of care management and medical home processes.

The effective management of patients with chronic illnesses is critical to bending the curve of health care spending in the United States and is a crucial test for health care reform. In this article we used data from three national surveys of physician practices between 2006 and 2013 to determine the extent to which practices of all sizes have increased their use of evidence-based care management processes associated with patient-centered medical homes for patients with asthma, congestive heart failure, depression, and diabetes. We found relatively large increases over time in the overall use of these processes for small and medium-size practices as well as for large practices. However, the large practices used fewer than half of the recommended processes, on average. We also identified the individual processes whose use increased the most and show that greater use of care management processes is positively associated with public reporting of patient experience and clinical quality and with pay-for-performance.

Small primary care physician practices have low rates of preventable hospital admissions.

Nearly two-thirds of US office-based physicians work in practices of fewer than seven physicians. It is often assumed that larger practices provide better care, although there is little evidence for or against this assumption. What is the relationship between practice size--and other practice characteristics, such as ownership or use of medical home processes--and the quality of care? We conducted a national survey of 1,045 primary care-based practices with nineteen or fewer physicians to determine practice characteristics. We used Medicare data to calculate practices' rate of potentially preventable hospital admissions (ambulatory care-sensitive admissions). Compared to practices with 10-19 physicians, practices with 1-2 physicians had 33 percent fewer preventable admissions, and practices with 3-9 physicians had 27 percent fewer. Physician-owned practices had fewer preventable admissions than hospital-owned practices. In an era when health care reform appears to be driving physicians into larger organizations, it is important to measure the comparative performance of practices of all sizes, to learn more about how small practices provide patient care, and to learn more about the types of organizational structures--such as independent practice associations--that may make it possible for small practices to share resources that are useful for improving the quality of care.

Physician practice participation in accountable care organizations: the emergence of the unicorn.

OBJECTIVE: To provide the first nationally based information on physician practice involvement in ACOs. DATA SOURCES/STUDY SETTING: Primary data from the third National Survey of Physician Organizations (January 2012-May 2013). STUDY DESIGN: We conducted a 40-minute phone survey in a sample of physician practices. A nationally representative sample of practices was surveyed in order to provide estimates of organizational characteristics, care management processes, ACO participation, and related variables for four major chronic illnesses. DATA COLLECTION/EXTRACTION METHODS: We evaluated the associations between ACO participation, organizational characteristics, and a 25-point index of patient-centered medical home processes. PRINCIPAL FINDINGS: We found that 23.7 percent of physician practices (n = 280) reported joining an ACO; 15.7 percent (n = 186) were planning to become involved within the next 12 months and 60.6 percent (n = 717) reported no involvement and no plans to become involved. Larger practices, those receiving patients from an IPA and/or PHO, those that were physician-owned versus hospital/health system-owned, those located in New England, and those with greater patient-centered medical home (PCMH) care management processes were more likely to have joined an ACO. CONCLUSIONS: Physician practices that are currently participating in ACOs appear to be relatively large, or to be members of an IPA or PHO, are less likely to be hospital-owned and are more likely to use more care management processes than nonparticipating practices.

Independent practice associations and physician-hospital organizations can improve care management for smaller practices.

Pay-for-performance, public reporting, and accountable care organization programs place pressures on physicians to use health information technology and organized care management processes to improve the care they provide. But physician practices that are not large may lack the resources and size to implement such processes. We used data from a unique national survey of 1,164 practices with fewer than twenty physicians to provide the first information available on the extent to which independent practice associations (IPAs) and physician-hospital organizations (PHOs) might make it possible for these smaller practices to share resources to improve care. Nearly a quarter of the practices participated in an IPA or a PHO that accounted for a significant proportion of their patients. On average, practices participating in these organizations provided nearly three times as many care management processes for patients with chronic conditions as nonparticipating practices did (10.4 versus 3.8). Half of these processes were provided only by IPAs or PHOs. These organizations may provide a way for small and medium-size practices to systematically improve care and participate in accountable care organizations.

Quantifying bias in a health survey: modeling total survey error in the national immunization survey.

Random-digit-dial telephone surveys are experiencing both declining response rates and increasing under-coverage due to the prevalence of households that substitute a wireless telephone for their residential landline telephone. These changes increase the potential for bias in survey estimates and heighten the need for survey researchers to evaluate the sources and magnitudes of potential bias. We apply a Monte Carlo simulation-based approach to assess bias in the NIS, a land-line telephone survey of 19-35 month-old children used to obtain national vaccination coverage estimates. We develop a model describing the survey stages at which component nonsampling error may be introduced due to nonresponse and under-coverage. We use that model and components of error estimated in special studies to quantify the extent to which noncoverage and nonresponse may bias the vaccination coverage estimates obtained from the NIS and present a distribution of the total survey error. Results indicated that the total error followed a normal distribution with mean of 1.72 per cent(95 per cent CI: 1.71, 1.74 per cent) and final adjusted survey weights corrected for this error. Although small, the largest contributor to error in terms of magnitude was nonresponse of immunization providers. The total error was most sensitive to declines in coverage due to cell phone only households. These results indicate that, while response rates and coverage may be declining, total survey error is quite small. Since response rates have historically been used to proxy for total survey error, the finding that these rates do not accurately reflect bias is important for evaluation of survey data. Published in 2011 by John Wiley & Sons, Ltd.