Wireless substitution: state-level estimates from the National Health Interview Survey, January-December 2007.

OBJECTIVES: This report presents state-level estimates of the percentage of households that do not have a landline telephone, but do have at least one wireless telephone. These wireless-only households made up 14.7% of U.S. households in 2007. The report also presents state-level estimates of the percentage of adults living in wireless-only households. These wireless-only adults made up 13.6% of U.S. adults in 2007. METHODS: A two-sample modeling strategy was used to estimate the prevalence of wireless-only households and adults by state. This modeling was based on data from the 2007 National Health Interview Survey and the 2008 Current Population Survey's Annual and Social Economic Supplement. RESULTS: The results show that the prevalence of wireless-only households and adults in 2007 varied substantially across states. State-level estimates ranged from 5.1% (Vermont) to 26.2% (Oklahoma) of households and from 4.0% (Delaware) to 25.1% (Oklahoma) of adults. In addition, approximately one out of four adults (25.4%) living in the District of Columbia were wireless-only.

Disparities in pediatric asthma hospitalizations.

OBJECTIVE: The purpose of this project was to determine to what extent rural children are hospitalized for asthma, an ambulatory care sensitive condition defined by the Agency for Healthcare Research and Quality pediatric quality indicators; to analyze differences in hospitalization rates for asthma by state and by rurality; and to examine the relationships between asthma hospitalization rates and poverty, health insurance, and physician supply. METHODS: The project used 2001 through 2004 hospital inpatient discharge data for children aged 2 to 17 years from six geographically diverse states in the Healthcare Cost and Utilization Project. County-level poverty, uninsurance estimates, and physician data came from the 2004 Area Resource File. Pediatric Quality Indicator software was used to calculate county-level admission rates for asthma. Multivariate regression models were specified to assess how sensitive hospitalization rates were to characteristics of the children's counties of residence. RESULTS: Pediatric asthma hospitalization rates per 100,000 children aged 2 to 17 years varied by state ranging from 51.1 to 185.9. When comparing all six states, rural children were the most likely to be hospitalized for asthma. However, after controlling for rurality, poverty, uninsurance, and physician supply, uninsurance was the only variable to significantly impact hospitalization rates. CONCLUSIONS: These findings indicate that there are significant differences in pediatric asthma hospitalizations rates by and within states, which may best be addressed by targeting public health and healthcare interventions. In addition, the findings support efforts to increase health insurance coverage for children, especially rural children who are less likely to be insured.

Measuring adequacy of coverage for the privately insured: new state estimates to monitor trends in health insurance coverage.

The privately insured are assuming a greater share of the costs of their health care, yet little is known about changes in out-of-pocket spending at the state level. The central problem is that national surveys with the relevant data are not designed to generate state-level estimates. The study addresses this shortcoming by using a two-sample modeling approach to estimate state-level measures of out-of-pocket spending relative to income for privately insured adults and children. National data from the Medical Expenditure Panel Survey-Household Component and state representative data from the Current Population Survey are used. Variation in out-of-pocket spending over time and across states is shown, highlighting concern about the adequacy of coverage for 2.9% of privately insured children and 7.8% of privately insured adults. Out-of-pocket spending relative to income is an important indicator of access to care and should be monitored at the state level.

Medicaid undercount and bias to estimates of uninsurance: new estimates and existing evidence.

OBJECTIVE: To examine whether known Medicaid enrollees misreport their health insurance coverage in surveys and the extent to which misreports of lack of coverage bias estimates of uninsurance. DATA SOURCE: Primary survey data from the Medicaid Undercount Experiment. STUDY DESIGN: Analyze new data from surveys of Medicaid enrollees in California, Florida, and Pennsylvania and summarize existing research examining bias in coverage estimates due to misreports among Medicaid enrollees. DATA COLLECTION METHOD: Subjects were randomly drawn from Medicaid administrative records and were surveyed by telephone. PRINCIPAL FINDINGS AND CONCLUSIONS: Cumulative evidence shows that a small percentage of Medicaid enrollees mistakenly report being uninsured, resulting in modest upward bias in estimates of uninsurance. A somewhat larger percentage of enrollees report having some other type of coverage than no coverage, biasing Medicaid enrollment estimates downward but not biasing estimates of uninsurance significantly upward. Implications for policy makers' confidence in survey estimates of coverage are discussed.

The impact of gaps in health insurance coverage on immunization status for young children.

OBJECTIVE: To examine the impact of full-year versus intermittent public and private health insurance coverage on the immunization status of children aged 19-35 months. DATA SOURCE: 2001 State and Local Area Integrated Telephone Survey's National Survey of Children with Special Health Care Needs (NS-CSHCN) and the 2000-2002 National Immunization Survey (NIS). STUDY DESIGN: Linked health insurance data from 2001 NS-CSHCN with verified immunization status from the 2000-2002 NIS for a nationally representative sample of 8,861 nonspecial health care needs children. Estimated adjusted rates of up-to-date (UTD) immunization status using multivariate logistic regressions for seven recommended immunizations and three series. PRINCIPAL FINDINGS: Children with public full-year coverage were significantly more likely to be UTD for two series of recommended vaccines, (4:3:1:3) and (4:3:1:3:3), compared with children with private full-year coverage. For three out of 10 immunizations and series tested, children with private part-year coverage were significantly less likely to be UTD than children with private full-year coverage. CONCLUSIONS: Our findings raise concerns about access to needed immunizations for children with gaps in private health insurance coverage and challenge the prevailing belief that private health insurance represents the gold standard with regard to UTD status for young children.

Variation in quality of diabetes care at the levels of patient, physician, and clinic.

INTRODUCTION: We studied variance in glycated hemoglobin (HbA1c) values among adults with diabetes to identify variation in quality of diabetes care at the levels of patient, physician, and clinic, and to identify which levels contribute the most to variation and which variables at each level are related to quality of diabetes care. METHODS: Study subjects were 120 primary care physicians and their 2589 eligible adult patients with diabetes seen at 18 clinics. The dependent variable was HbA1c values recorded in clinical databases. Multivariate hierarchical models were used to partition variation in HbA1c values across the levels of patient, physician, or clinic and to identify significant predictors of HbA1c at each level. RESULTS: More than 95% of variance in HbA1c values was attributable to the patient level. Much less variance was seen at the physician and clinic level. Inclusion of patient and physician covariates did not substantially change this pattern of results. Intensification of pharmacotherapy (t = -7.40, P < .01) and patient age (t = 2.10, P < .05) were related to favorable change in HbA1c. Physician age, physician specialty, number of diabetes patients per physician, patient comorbidity, and clinic assignment did not predict change in HbA1c value. The overall model with covariates explained 11.8% of change in HbA1c value over time. CONCLUSION: These data suggest that most variance in HbA1c values is attributable to patient factors, although physicians play a major role in some patient factors (e.g., intensification of medication). These findings may lead to more effective care-improvement strategies and accountability measures.

Accuracy in self-reported health insurance coverage among Medicaid enrollees.

The largest portion of the Medicaid undercount is caused by survey reporting error--that is, Medicaid recipients misreport their enrollment in health insurance coverage surveys. In this study, we sampled known Medicaid enrollees to learn how they respond to health insurance questions and to document correlates of accurate and inaccurate reports. We found that Medicaid enrollees are fairly accurate reporters of insurance status and type of coverage, but some do report being uninsured. Multivariate analyses point to the prominent role of program-related factors in the accuracy of reports. Our findings suggest that the Medicaid undercount should not undermine confidence in survey-based estimates of uninsurance.

Estimating regression standard errors with data from the Current Population Survey's public use file.

This study examines whether reasonable standard errors for multivariate models can be calculated using the public use file of the Current Population Survey's Annual Social and Economic Supplement (CPS ASEC). We restrict our analysis to the 2003 CPS ASEC and model three dependent variables at the individual level. income, poverty, and health insurance coverage. We compare standard error estimates performed on the CPS ASEC public use file with those obtained from the Census Bureau's restricted internal data that include all the relevant sampling information needed to compute standard errors adjusted for the complex survey sample design. Our analysis shows that the multivariate standard error estimates derived from the public use CPS ASEC following our specification perform relatively well compared to the estimates derived from the internal Census Bureau file. However, it is essential that users of CPS ASEC data do not simply choose any available method since three of the methods commonly used for adjusting for the complex sample design produce substantially different estimates.

Hospital size, uncertainty, and pay-for-performance.

We construct statistical models to assess whether hospital size will impact the ability to identify "true" hospital ranks in pay-for-performance (P4P) programs. We use Bayesian hierarchical models to estimate the uncertainty associated with the ranking of hospitals by their raw composite score values for three medical conditions: acute myocardial infarction (AMI), heart failure (HF), and community acquired pneumonia (PN). The results indicate a dramatic inverse relationship between the size of the hospital and its expected range of ranking positions for its true or stabilized mean rank. The smallest hospitals among the augmented dataset would likely experience five to seven times more uncertainty concerning their true ranks.

Unstable inferences? An examination of complex survey sample design adjustments using the Current Population Survey for health services research.

Statistical analysis of the Current Population Survey's Annual Social and Economic Supplement is used widely in health services research. However, the statistical evidence cited from the Current Population Survey (CPS) is not always consistent because researchers use a variety of methods to produce standard errors that are fundamental to significance tests. This analysis examines the 2002 Annual Social and Economic Supplement's (ASEC) estimates of national and state average income, national and state poverty rates, and national and state health insurance coverage rates. Findings show that the standard error estimates derived from the public use CPS data perform poorly compared with the survey design-based estimates derived from restricted internal data, and that the generalized variance parameters currently used by the U.S. Census Bureau in its ASEC reports and funding formula inputs perform erratically. Because the majority of published research (both by academics and Census Bureau analysts) does not make use of the survey design-based information available only on the internal ASEC data file, we argue that the Census Bureau ought to use alternative methods for its official ASEC reports. We also argue that for public use data the Census Bureau should produce a set of replicate weights for the ASEC or release a set of sample design variables that incorporate statistical "noise" to maintain respondent confidentiality (e.g., pseudo-primary sampling units) as other federal government surveys do. This is essential to make appropriate inferences using the ASEC data regarding statistical significance and estimate variance for health policy analysis.

Pharmacist staffing, technology use, and implementation of medication safety practices in rural hospitals.

CONTEXT: Medication safety is clearly an important quality issue for rural hospitals. However, rural hospitals face special challenges implementing medication safety practices in terms of their staffing and financial and technical resources. PURPOSE: This study assessed the capacity of small rural hospitals to implement medication safety practices, with particular focus on pharmacist staffing and the availability of technology. METHODS: A telephone survey of a national random sample of small rural hospitals was conducted from March to May 2005 (N = 387 hospitals, 94.6% response rate). Survey respondents included pharmacists (89%) and directors of nursing (11%). Multivariate analyses examined the relationships between hospital organizational and financial variables and (1) the amount of pharmacist staffing; (2) use of pharmacy computers for medication safety activities; and (3) implementation of medication safety practices. FINDINGS: Many small rural hospitals have limited hours of on-site pharmacist coverage. Almost one quarter of hospitals either do not have a pharmacy computer or are not using it for clinical purposes. Half of the hospitals have implemented 4 key medication safety practices. Level of pharmacist staffing, use of technology, and implementation of medication safety practices are significantly related to hospital financial status and accreditation. CONCLUSIONS: Implementation of protocols related to medication use and key medication safety practices are areas where small rural hospitals could improve. The study results support a continuation of Medicare cost-based reimbursement policies to help ensure financial stability and support quality and patient safety activities in small rural hospitals.

Does a health plan effort to increase smokers' awareness of cessation medication coverage increase utilization and cessation?

PURPOSE: To test whether a mailing describing new coverage for smoking cessation medications increases benefit knowledge, utilization, and quitting. METHODS: This randomized controlled trial assigned participants to benefit communication via (1) standard contract changes or (2) enhanced communication with direct-to-member postcards. A sample of 1930 self-identified smokers from two Minnesota health plans took surveys before and 1 year after the benefit's introduction. The follow-up response rate was 80%. A multilevel logistic estimator tested for differences in benefit knowledge and smoking behavior from baseline. RESULTS: More enhanced than standard communication respondents knew about the benefit (39.0% vs. 22.2%, p < .0001) at follow-up. Groups did not differ on bupropion utilization (24.6% vs. 23.1%, p = .92); nicotine replacement therapy utilization (26.9% vs. 25.9%, p = .26), or cessation (12.8% vs. 15.6%, p = .32). CONCLUSION: Although limited by the low intervention intensity and potential social desirability bias, information about new coverage alone does not appear to increase quitting behaviors.

Public program crowd-out of private coverage.

Coverage expansions by Medicaid, SCHIP and other state programs significantly increased the number of people covered by public insurance. Crowd-out occurs when people drop private coverage for public coverage, when those enrolled in public insurance turn down private coverage when eligible, or when employers opt not to offer private insurance because of the existence of a public program. This synthesis examines the extent of crowd-out and whether it can be reduced. Key findings include: Estimates of crowd-out are imprecise and vary depending on the type of coverage expansion; the assumptions, methods and data used; and the time period covered. Crowd-out is more likely to occur in programs that enroll families, and among families with incomes greater than 200 percent FPL. Programs have used waiting periods and cost-sharing to limit crowd-out, but these techniques can be difficult and costly to implement, and may reduce program participation by the uninsured.

Hospital provision of uncompensated care and public program enrollment.

Hospital provision of uncompensated care is partly a function of insurance coverage of state populations. As states expand insurance coverage options and reduce the number of uninsured, hospital provision of uncompensated care should also decrease. Controlling for hospital characteristics and market factors, the authors estimate that increases in MinnesotaCare (a state-subsidized health insurance program for the working poor) enrollment resulted in a 5-year cumulative savings of $58.6 million in hospital uncompensated care costs. Efforts to evaluate access expansions should take into account the costs of the program and the savings associated with reductions in hospital uncompensated care.

Does insurance coverage for drug therapy affect smoking cessation?

Whether insurance coverage for smoking-cessation medicines increases quitting rates is uncertain. In this paper we evaluate the overall effect of a new health plan pharmacy benefit on the use of pharmacotherapy, attempts to quit, and quitting rates. The presence of a smoking-cessation pharmacy benefit as implemented by these health plans produced no change in the use of bupropion, nicotine patches, or nicotine gum, nor did it result in higher rates of quitting smoking. Further studies are needed to test whether greater efforts to make smokers aware of insurance benefits or adding other types of cessation support might lead to any beneficial effects.

Physician smoking-cessation actions: are they dependent on insurance coverage or on patients?

BACKGROUND: Despite good evidence that their smoking-cessation actions can be very effective, physicians have not consistently used the 5A actions (being asked, advised, assessed, assisted, and arranged) recommended in the U.S. Public Health Service tobacco guidelines. We tested the hypothesis that the introduction of coverage for smoking-cessation pharmacotherapy by the health plans covering most of the population in one region would increase physician use of 5A's. METHODS: A cohort of smoking members of two health plans was surveyed before and after the introduction of coverage for smoking cessation. A total of 1560 current smokers with a physician visit in the last year responded to both surveys. The key outcome measures were smoker reports of the guideline 5As for smoking-cessation support during the last physician visit. RESULTS: There were small significant absolute percentage increases only for reports of being assessed (+4.9%, p=0.01) and assisted (set quit date +6.5%, p=0.0004); encouraged to use medications (+8.8%, p=0.03); and given a prescription (+8.6%, p=0.0005). However, these increases were limited to smokers reporting awareness of the coverage, asking for quitting help, or both. CONCLUSION: Coverage for pharmacotherapy alone appears to have had no effect on physician behavior beyond that stimulated by smokers who were aware of the coverage, perhaps because they raised the issue. More research is needed on this suggestion that patients create physician behavior change.