Last week, two committees in the U.S. House of Representatives passed out of the committee two health bills, jointly referred to as the American Health Care Act (AHCA), to modify the Affordable Care Act (ACA or Obamacare).  Congressional leaders plan that a third committee is going to consider the AHCA this week and that the U.S. House of Representatives is going to consider the bill on the House floor next week.

With the ACA, federal funds covered the entire cost of coverage of newly-eligible Medicaid enrollees who had income less than or equal to 138% of the federal poverty level.  Under this Medicaid expansion, 32 states and the District of Columbia have expanded Medicaid pursuant to the Act and 14 million people have gained access to Medicaid coverage due to this expansion.

The AHCA amends Medicaid in ways that will likely affect the health care of homeless persons.  First, the bill would not allow further enrollment under the ACA Medicaid expansion of otherwise able-bodied, non-pregnant adults who qualifies for Medicaid solely based on their income being less than or equal to 138% of the federal poverty level after January 1, 2020.  After that date, only those persons already enrolled in Medicaid pursuant to the expansion eligibility will be allowed to remain covered if they do not become ineligible for more than a month.  Second, the bill ends the “enhanced federal medical assistance percentage” (FMAP), which will reduce the percent of Medicaid costs covered by the federal government from 100% to a lower amount equal to the rate for other enrollees in the state (ranging from 50% to 75% with an average of about 57%).  Third, the bill restructures the program’s federal funding whereby it currently covers all eligible persons to a funding structure with a limited per-capita cap starting in 2020 that will likely reduce the amount of federal funds that states will receive.  In addition, the bill replaces the ACA’s subsidies based on income and cost of coverage with age-based tax credits that do not take into account income or the cost of coverage as the subsidies currently do and that may not be sufficient to cover health insurance, especially for older Americans.

Of the more than 14 million people who have acquired health care coverage through the Medicaid expansion, hundreds of thousands of them likely are homeless individuals.  As reported in an earlier blog, according to data from the Health Resources and Services Administration (HRSA), the number of homeless adults visiting a Health Care for the Homeless (HCH) Health Center who reported being uninsured decreased by 149,000 and the number who reported having Medicaid coverage concomitantly increased by 154,000 from 2013 to 2015.  While the overall percent of homeless patients at HCH Health Centers who were uninsured decreased 21% from 61.5% to 40.0% and the overall percent who reported having Medicaid coverage conversely increased 19% from 26.9% to 45.6% from 2013 to 2015, most of these changes were likely in the states that expanded Medicaid.

The Congressional Budget Office estimates that 14 million people will lose Medicaid coverage by 2026 due to the AHCA.  With the number of people estimated to lose coverage roughly equal to the number that have acquired coverage through the Medicaid expansion, it is likely that hundreds of thousands of homeless people will lose their Medicaid coverage and will again be relegated to the ranks of the uninsured.

To estimate the effect of the AHCA and its Medicaid modifications on the health insurance and health and welfare of homeless people, the health insurance status and preventive care and treatment of homeless patients at HCH Health Centers where all of the patients were homeless in the states that had expanded and in the states that did not expand Medicaid were statistically evaluated using Independent t-tests.  Data were available for about one-fourth (24.9% or 209,103 of 840,130) of the homeless patients seen at HCH Health Centers in 12 of the 30 Medicaid-expansion states (including the District of Columbia) and 5 of the 21 no-Medicaid-expansion states as of January 1, 2015.

The data in the table below strongly support that the ACA’s Medicaid expansion helped greatly decreased the percent of homeless persons without health insurance and greatly increased the percent of homeless persons with Medicaid.  The percent of homeless patients who were uninsured was statistically significantly lower in Medicaid-expansion states (29.5%) than in states that had not expanded Medicaid (84.5%).  Likewise, the percent of homeless patients who had Medicaid coverage or coverage through the Children’s Health Insurance Program (CHIP) was statistically significantly higher in Medicaid-expansion states (58.1%) than in states that had not expanded Medicaid (8.2%).

The other health insurance data appeared to highlight synchrony with the Medicaid changes and the dearth of other health insurance options available for homeless persons.  The percent of homeless patients who had Medicare was statistically significantly higher in Medicaid-expansion states (9.3%) than in states that had not expanded Medicaid (2.7%).  This result could be because some of the homeless persons who were newly eligible for Medicaid were also eligible for Medicare and state Medicaid agencies provide information on Medicaid-eligible persons who qualify for Medicare Parts A and B to the Social Security Administration to increase enrollment in Medicare.  The percent of homeless patients who had Other Third Party Insurance did not statistically differ between the Medicaid-expansion states (3.1%) and the states that had not expanded Medicaid (5.3%).

The data in the second table indicate that the AHCA’s changes to Medicaid will likely negatively impact the health of homeless people and could lead to increased deaths of homeless people due to cardiovascular disease and cancer and other diseases when they are unable to actually obtain screening, treatment, and medication.

The percent of homeless adults age 18 to 85 diagnosed with hypertension with controlled blood pressure (meaning their last blood pressure reading was less than 140/90) was statistically significantly lower in states that had not expanded Medicaid (46.9%) than in Medicaid-expansion states (59.1%).  This significant difference in blood pressure control may be because homeless patients without insurance, including Medicaid, are unable to afford the two or more types of blood pressure medications necessary for most hypertensive patients to achieve blood pressure control.  As high blood pressure is fatal and is also a risk factor for heart disease and stroke and cardiovascular disease is the leading cause of death of Americans, the lack of expanding Medicaid may cause preventable deaths to thousands of homeless adults in the states that have not expanded Medicaid.

The other category that involved actual treatment was the percent of patients with uncontrolled diabetes (meaning having an Hba1c level > 9%) and the percent of patients with uncontrolled diabetes was slightly higher in states that had not expanded Medicaid than in states that had expanded Medicaid.  There was not a statistically significant difference in levels in the two groups of states, which could be because lifestyle changes such as losing weight and exercising could be more successful at lowering Hba1c levels than blood pressure levels or the target level reported of an Hba1c level of 9%, which is used instead of 7% to determine the people most at risk of diabetes complications, is easier to reach for diabetes than the target level of 140/90 is to reach for blood pressure.

Four of the categories — Asthma Treatment, Cholesterol Treatment, Heart Attack/Stroke Treatment, and HIV Linkage to Care — are not likely to be different in the two groups of states for patients with access to a health center because the measured variables did not involve medication, treatment, or access to specialists and there was no statistical difference in the levels for three of the four categories.  The levels for asthma treatment were not statistically different, which may be because Asthma Treatment measured the patients age 5 to 40 diagnosed with asthma who have an acceptable pharmacological treatment plan, including merely receiving a prescription for an inhaled corticosteroid or an accepted alternative medication.  The levels for heart attack/stroke treatment were not statistically different, which may be because Heart Attack/Stroke Treatment measured documentation such as merely being prescribed or dispensed aspirin or another antithrombotic therapy and even use of aspirin is relatively inexpensive and available without a prescription and therefore more readily available for people without insurance.  The levels for HIV Linkage to Care were not statistically different, which may be because HIV Linkage to Care measured whether or not patients newly-diagnosed with HIV by a health center staff were merely seen for follow-up treatment within 90 days of the first ever diagnosis.  The levels of the fourth category (Cholesterol Treatment) were statistically different, but without real meaning because Cholesterol Treatment merely measured whether adult patients diagnosed with coronary artery disease received a prescription for or were provided or were taking lipid-lowering medications.

The HRSA reports two preventive care measures — Cervical Cancer Screening and Colorectal Cancer Screening — and as with the treatment measures, statistical differences in the outcomes varied for patients in states that expanded Medicaid and that had not yet expanded Medicaid based on whether there was the provision of the preventive care at health centers and clinics without regard to insurance status.  The levels for cervical cancer screening were not statistically different and this is likely because this screening is available at no- or low-cost at local county health departments, federally qualified health centers, and women’s clinics such as Planned Parenthood.  The percent of homeless adults age 51 to 74 who were appropriately screened for colorectal cancer was statistically higher in Medicaid-expansion states (27.8%) than in states that had not expanded Medicaid (16.9%).  It is likely significantly fewer homeless patients received colorectal cancer screening in states that had not expanded Medicaid because in contrast to the widely available no- or low-cost cervical cancer screening, the colorectal cancer screening procedure likely involves anesthesia, the procedure and its recovery take several hours, the procedure is performed by a specialist (gastroenterologist) whereas a Pap Smear can be performed by family physicians or internists that are more available at health centers, and no- or low-cost colorectal cancer screening is not available in states that had not expanded Medicaid.

The HRSA data show not only that homeless patients are statistically less likely to be uninsured and statistically more likely to have Medicaid in states that expanded Medicaid, but that the vast difference in insurance and Medicaid coverage of homeless patients also affected the actual cancer preventive care and cardiovascular health of homeless patients with better preventive care and treatment in states that expanded Medicaid than in states that had not expanded Medicaid.  Without cancer detection and blood pressure and cardiovascular treatment, the lives of homeless people are at increased risk and their lives may be depend on having health care insurance coverage.  Therefore, Congress should amend the AHCA to eliminate the modifications to the Medicaid program that the bill makes and should work to expand health insurance coverage for homeless adults and children in every state.