Survey Data on Health Insurance Coverage for 2013 and 2014. Mid-2014: Gallup-Healthways Well-Being Index (WBI)

10/31/2014

Analysis of data from the Gallup-Healthways WBI by Sommers et al. shows a decline of 26 percent in the uninsured rate for nonelderly adults. The study estimated a 5.2 percentage point reduction in the uninsured rate for 18 to 64 year olds in the second quarter of 2014, compared to the base period from the first quarter of 2012 through the third quarter of 2013.5 This decrease in the uninsured rate translates to 10.3 million additional adults becoming insured since the start of the open enrollment period on October 1, 2013 (see Table 1). These estimates are adjusted for the prior trend and for changes in income and employment, and so provide a reasonable estimate of the impact of the Affordable Care Act on insurance coverage.

Percent Decline in the Uninsured Rate: Second Quarter of 2014 compared to First Quarter 2012 through Third Quarter 2013 (Gallup-Healthways Well-Being Index)


Percent Decline in the Uninsured Rate: Second Quarter of 2014 compared to First Quarter 2012 through Third Quarter 2013

Total PopulationBaseline Uninsured Rate (2012 Q1-2013 Q3)Absolute Change in Uninsured Rate (percentage points)Relative Change in Uninsured Rate (percent)Number Gaining Coverage
Total198,462,00020.3%5.226%10,320,000
White123,437,00014.3%4.028%4,937,000
African American25,211,00022.4%6.830%1,714,000
Latino34,017,00041.8%7.718%2,619,000
Other/Unknown15,797,00017.5%6.638%1,049,000

 

Note: Other/Unknown includes Asian American, American Indian and Alaska Native, and Native Hawaiian and Other Pacific Islander, who are not broken out separately due to sample sizes.

Source: ASPE analysis of Gallup-Healthways Well-Being Index (WBI) data, adapting the basic approach used in Benjamin D. Sommers, Thomas Musco, Kenneth Finegold, Munira Z. Gunja, Amy Burke, and Audrey McDowell, “Health Reform and Changes in Health Insurance Coverage in 2014,” New England Journal of Medicine, July 23, 2014 (http://www.nejm.org/doi/full/10.1056/NEJMsr1406753, accessed October 27, 2014).


The January 2012-September 2013 uninsured rate for Whites was 14.3 percent, with a 4.0 percentage point reduction in the second quarter of 2014, translating to a 28 percent relative decline (Table 1). For African Americans there was a 30 percent relative decline (22.4 percent uninsured baseline rate and a 6.8 percentage point reduction in the second quarter of 2014) and for Hispanics there was an 18 percent relative decline (a 41.8 percent uninsured baseline rate and a 7.7 percentage point reduction in the second quarter of 2014).6 The Commonwealth Fund found that among Latinos living in states that did not expand their Medicaid programs, the uninsured rate remained statistically unchanged (33 percent).7 Some large states with substantial Latino populations, including Florida and Texas, did not take part in the Medicaid Expansion.

Skopec, Musco, and Sommers found that unadjusted Gallup-Healthways WBI annual estimates of the share of 18-64 year olds without health insurance tracked those from multiple federal surveys for 2008-2011.8 Similarly, Figure 1 shows that the unadjusted Gallup-Healthways WBI quarterly estimates closely track NHIS quarterly data for the period covered by both datasets. Annual data on the percentage of 18 to 64 year olds who were uninsured are also similar, with Gallup-Healthways WBI estimates of 20.2 percent uninsured in 2012 and 20.7 percent uninsured in 2013, compared with NHIS estimates of 20.9 percent uninsured in 2012 and 20.4 percent uninsured in 2013.9

Figure 1. Percentage of Adults ages 18-64 who were Uninsured by Year and Quarter, 2010-2014. NHIS and Gallup-Healthways Well-Being Index

Sources: Robin A. Cohen and Michael E. Martinez, Health insurance coverage: Early release of estimates from the National Health Interview Survey, January–March 2014. National Center for Health Statistics. September 2014 (http://www.cdc.gov/nchs/data/nhis/earlyrelease/insur201409.pdf, accessed October 27, 2014); ASPE analysis of Gallup-Healthways Well-Being Index data.


A limitation of nongovernmental surveys such as the Gallup-Healthways WBI is that their response rates are low compared with federal surveys. The Gallup-Healthways WBI response rate, measured according to the American Association for Public Opinion Research (AAPOR)’s criteria for complex surveys, stood at 7 percent for the Gallup-Healthways Well-Being portion of the survey, which includes the health insurance question. The Urban Institute reports a lower response rate of 5 percent for its Health Reform Monitoring Survey.10

Production estimates for federal surveys, which have more resources to convert non-respondents to respondents through incentives and follow-up interviews—and in which participation may be mandatory, or thought to be so—are based on higher, though not perfect, response rates. Skopec, Musco, and Sommers report a 91-93 percent response rate for the CPS-ASEC, 93-98 percent for the ACS, 80-90 percent for NHIS, 63-71 percent for the Medical Expenditure Panel Survey (MEPS), and 50-58 percent for the Behavioral Risk Factor Surveillance System (BRFSS).11 Even among federal surveys, however, response rates may be lower for content tests than for production estimates: Brault reports an unweighted response rate of 43.1 percent for the 2013 CPS-ASEC Content Test.12

Survey methodologists are, however, increasingly arguing that survey response rates may be a poor indicator of the level of bias in survey estimates13 Substantial efforts by federal and academic researchers over the last decade or so have concluded that low response rates do not necessarily indicate bias, and that scarce survey resources are better devoted to analysis of potential biases and correction of them through reweighting than to pushing response rates slightly higher, which does not appear to reduce survey response bias as much as had previously been thought. As Brault suggests, “While little can be done ex post to address the problems of small sample size, adjustment can be done to minimize the effect of non-response bias.”14


5 Benjamin D. Sommers, Thomas Musco, Kenneth Finegold, Munira Z. Gunja, Amy Burke, and Audrey McDowell, “Health Reform and Changes in Health Insurance Coverage in 2014,” New England Journal of Medicine, July 23, 2014 (http://www.nejm.org/doi/full/10.1056/NEJMsr1406753, accessed October 27, 2014).

6 Benjamin D. Sommers, Thomas Musco, Kenneth Finegold, Munira Z. Gunja, Amy Burke, and Audrey McDowell, “Health Reform and Changes in Health Insurance Coverage in 2014,” New England Journal of Medicine, July 23, 2014 (http://www.nejm.org/doi/full/10.1056/NEJMsr1406753, accessed October 27, 2014).
7 Michelle M. Doty, Petra W. Rasmussen, and Sara R. Collins, “Catching Up: Latino Health Coverage Gains and Challenges Under the Affordable Care Act,” The Commonwealth Fund, September 2014. (http://www.commonwealthfund.org/~/media/files/publications/issue-brief/2014/sep/1775_doty_catching_up_latino_hlt_coverage_aca_tb_v3.pdf, accessed October 27, 2014).
8 Laura Skopec, Thomas Musco, and Benjamin D. Sommers. 2014. “A potential new data source for assessing the impacts of health reform: Evaluating the Gallup-Healthways Well-Being Index.” Healthcare: http://dx.doi.org/10.1016/j.hjdsi.2014.03.001 (accessed October 27, 2014).
9 ASPE calculations from Gallup-Healthways WBI data; Robin A. Cohen and Michael E. Martinez, Health insurance coverage: Early release of estimates from the National Health Interview Survey, January–March 2014. National Center for Health Statistics. September 2014. (http://www.cdc.gov/nchs/data/nhis/earlyrelease/insur201406.pdf, accessed October 27, 2014).
10 Urban Institute. N.D. “Health Reform Monitoring Survey, “HRMS Frequently Asked Questions, 8) What is the response rate for the HRMS?” (http://hrms.urban.org/faq.html, accessed October 27, 2014).
11 Laura Skopec, Thomas Musco, and Benjamin D. Sommers. 2014. “A potential new data source for assessing the impacts of health reform: Evaluating the Gallup-Healthways Well-Being Index.” Healthcare: http://dx.doi.org/10.1016/j.hjdsi.2014.03.001 (accessed October 27, 2014).
12 Matthew W. Brault, “Non-response Bias in the 2013 CPS ASEC Content Test.” Proceedings of the 2013 Federal Committee on Statistical Methodology (FCSM) Research Conference, November 6. SEHSD Working Paper Number 2014-17. (https://www.census.gov/hhes/www/hlthins/publications/sehsd_wp_2014-17.pdf, accessed October 27, 2014).
13 Michael Davern, “Nonresponse Rates are a Problematic Indicator of Nonresponse Bias in Survey Research,” Health Services Research 48:3 (2013): 905-912.
14 Matthew W. Brault, “Non-response Bias in the 2013 CPS ASEC Content Test.” Proceedings of the 2013 Federal Committee on Statistical Methodology (FCSM) Research Conference, November 6. SEHSD Working Paper Number 2014-17. (https://www.census.gov/hhes/www/hlthins/publications/sehsd_wp_2014-17.pdf, accessed October 27, 2014).

 

View full report

Preview
Download

"ib_InsuranceEstimates.pdf" (pdf, 216.18Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®