By Lisa Zamosky
You can’t talk about the Affordable Care Act without citing the number of people in the country today who currently have no health insurance. After all, one of the law’s main goals is to cover millions of uninsured Americans.
A reader of this blog, Don, asked for an explanation of how the numbers of uninsured are determined. Here’s Don’s question:
The Federal government and the news media have been throwing around numbers, claiming there are between 20 million and 40 million Americans that do not currently have health insurance. Perhaps someone can provide me with a valid, provable answer for the following questions:
How the heck can anyone know for sure exactly how many people in the USA are currently without health insurance?
What is the valid source for this information and how was it obtained?
Show Me the Numbers
In short, the Census Bureau is the main source of how many people in the U.S. have health insurance (or don’t).
“The Census is the only survey in the land that allows you to trend health insurance coverage over time because it’s been done annually for so long,” says Catherine Hoffman, Deputy Director of the California Medicaid Research Institute. “With all its warts, it really is the gold standard.”
The Census Bureau uses three different surveys to compile data on the number of uninsured Americans.
Here’s a rundown of the surveys and what each captures:
1. The Current Population Survey (CPS). The CPS is the core of most estimates you see, according to Hoffman. This is a monthly survey of about 50,000 households that the Census Bureau conducts mostly to estimate the rate of unemployment.
But each year in March, a supplement to CPS called the Annual Social and Economic Supplement (ASCE) is produced. It surveys nearly 80,000 households and contains detailed questions about health insurance, including whether those surveyed had coverage during the previous year.
It’s important to mention that the results are self-reported, leaving some room for error. “When you ask people about their health insurance coverage sometimes they don’t really know,” Hoffman says.
Here’s an example of how people’s confusion about their coverage can play out and skew the numbers: People with public health insurance, like Medicaid, are asked if they’ve had private insurance coverage over the past year. But often private insurance companies administer public programs, a detail frequently unrecognized by the insurance holder.
“They might say ‘I have Blue Cross or United or Kaiser,’ so they get counted as having private insurance,” Hoffman says of people covered by government programs.
For that reason, it’s likely the CPS underreports the number of people who have Medicaid and over reports the number of people in the country with private insurance. That means there’s a good chance that the number of uninsured people mentioned in the news is close to the real number, but probably a little low.
2. The American Community Survey (ACS): This survey drills down to a greater level of detail, allowing the Census Bureau to capture the number of people with insurance at the state level. There is an annual ACS survey that has the ability to generate estimates of insurance coverage for communities as small as 65,000 people, and it asks whether people currently have insurance (as opposed to the CPS, which asks about coverage during the previous year).
3. The Survey of Income and Program Participation (SIPP): SIPP is conducted three times each year over the course of three to four years. In addition to collecting information about income, it obtains detailed health insurance data, identifying how long people stay uninsured, how many people during that time period gain coverage, and any changes to a person’s insurance coverage in a given year.
You’ll notice that each survey asks about a person’s insurance coverage during different periods of time (such as: Do you currently have insurance? vs. Have you had coverage during the past year?). Hoffman points out that what time frame a survey asks about will produce different estimates of health coverage.
The number of Americans currently without health insurance is about 50 million. For an interesting take on just how big that number is, take a look at this graphic created by the Washington Post.
And then share your thoughts and questions in the comments section below. What other statements do you commonly read in the press that leave you questioning where they came from and how accurate they are?