Abstract
This study uses national survey data to assess the prevalence and health-related correlates of housing problems among community health center patients.
In 2016, the Health Resources and Services Administration (HRSA) Health Center Program provided primary care to more than 25 million medically underserved patients through a nationwide network of community health center (CHC), health care for the homeless, migrant health center, and public housing primary care clinics. Although the latter 3 clinic types serve individuals with housing problems by definition, little is known about the scope of housing problems among CHC patients, who constitute 91% of Health Center Program patients nationally. We used data from a national survey to assess the prevalence and health-related correlates of housing problems among CHC patients.
Methods
The Partners Human Research Committee exempted this study. We analyzed the 2014 Health Center Patient Survey, a nationally representative, cross-sectional, in-person survey of Health Center Program patients conducted by RTI International from September 2014 through April 2015 using a 3-stage sampling design. First-stage sampling units were Health Center Program grantees, stratified by funding stream, substratified by other characteristics, and sampled with probability proportional to size. Second-stage sampling units were clinic sites within grantees. Third-stage sampling units were patients sampled consecutively at clinics if they had made 1 prior visit or more within the past year; 91.4% of those eligible completed interviews. We confined our analysis to CHC patients aged 18 years or older.
We used responses to items assessing living circumstances to create 5 mutually exclusive housing categories: (1) homeless—usually slept during the past week in an emergency shelter, transitional shelter, or car; anywhere outside; or any other place not meant for habitation; (2) doubled-up—past-week residence in a house, apartment, or room that they did not rent or own (doubled-up individuals are considered homeless by HRSA but not by the US Department of Housing and Urban Development); (3) unstably housed—past-week residence in their own place but moved 2 or more times in the past year or was unable to pay the rent or mortgage at any time; (4) stably housed, previously homeless—past-week residence in their own place without the above difficulties but previously homeless, reflecting potential housing risk; and (5) stably housed, never homeless—no current or prior housing problems.
Other variables included self-reported demographic characteristics, health status indicators, and measures of health care use and access, each defined in the Table.
Table. Characteristics of Patients With vs Without Housing Problems Among Patients at US Community Health Centers.
| Any Current or Prior Housing Problem (n = 1329) | No Current or Prior Housing Problem (n = 1819) | P Valuea | |||
|---|---|---|---|---|---|
| Unweighted No. of Patients | Weighted % (95% CI) | Unweighted No. of Patients | Weighted % (95% CI) | ||
| Demographic | |||||
| Age, y | .001 | ||||
| 18-44 | 554 | 54.2 (48.2-60.1) | 676 | 52.0 (45.9-58.2) | |
| 45-64 | 663 | 38.4 (32.1-44.6) | 835 | 33.4 (26.7-40.1) | |
| ≥65 | 112 | 7.5 (3.6-11.3) | 308 | 14.6 (5.9-23.2) | |
| Women | 919 | 67.3 (63.1-71.5) | 1277 | 64.0 (58.0-69.9) | .30 |
| Race/ethnicity | <.001 | ||||
| Non-Hispanic white | 432 | 57.6 (50.1-65.1) | 448 | 47.2 (39.0-55.4) | |
| Non-Hispanic black | 284 | 16.1 (11.7-20.5) | 378 | 20.4 (14.4-26.4) | |
| Non-Hispanic Asian | 43 | 0.8 (0.2-1.3) | 222 | 3.6 (0.7-6.5) | |
| Non-Hispanic other | 163 | 5.4 (2.9-7.9) | 216 | 4.6 (2.6-6.7) | |
| Hispanic | 406 | 20.1 (14.2-26.0) | 553 | 24.2 (17.4-30.9) | |
| High school diploma | 854 | 68.2 (62.4-73.9) | 1101 | 65.2 (60.1-70.2) | .28 |
| Currently employed | 432 | 35.4 (30.3-40.5) | 713 | 41.9 (35.6-48.2) | .07 |
| Federal poverty level, % | <.001 | ||||
| ≤100 | 859 | 60.1 (54.6-65.5) | 919 | 50.9 (44.7-57.0) | |
| 101-199 | 359 | 31.5 (26.2-36.7) | 580 | 28.4 (24.2-32.5) | |
| ≥200 | 102 | 8.4 (5.3-11.6) | 307 | 20.8 (15.7-25.8) | |
| Health insurance | <.001 | ||||
| None | 265 | 29.7 (21.8-37.5) | 337 | 24.2 (18.9-29.4) | |
| Public | 779 | 50.6 (43.4-57.8) | 922 | 44.8 (39.7-49.9) | |
| Private | 262 | 19.4 (15.0-23.7) | 559 | 30.1 (24.9-35.4) | |
| Health Indicators | |||||
| Fair or poor general health | 697 | 44.8 (39.1-50.5) | 786 | 37.6 (33.7-41.4) | .04 |
| Multiple chronic conditionsb | 708 | 52.7 (46.9-58.5) | 876 | 45.0 (40.1-50.0) | .02 |
| Functional impairmentc | 236 | 17.6 (13.0-22.2) | 185 | 9.1 (6.1-12.2) | <.001 |
| Sensory impairmentd | 360 | 26.6 (21.3-31.8) | 328 | 19.0 (14.4-23.5) | .007 |
| Fair or poor oral health | 679 | 46.8 (42.1-51.6) | 749 | 35.5 (31.3-39.8) | .001 |
| Serious mental illnesse | 236 | 19.0 (14.1-23.8) | 112 | 6.2 (3.7-8.7) | <.001 |
| High psychological distressf | 265 | 22.2 (18.1-26.3) | 153 | 7.5 (5.0-9.9) | <.001 |
| Lifetime drug useg | 729 | 58.7 (51.8-65.5) | 541 | 36.8 (29.9-43.6) | <.001 |
| Current alcohol use disorderg | 122 | 10.8 (7.5-14.0) | 84 | 3.6 (1.5-5.8) | <.001 |
| Current cigarette smoking | 462 | 36.1 (30.1-42.1) | 358 | 21.4 (15.4-27.5) | <.001 |
| Health Care Use and Access | |||||
| Health service use | |||||
| Emergency department use, past year | 738 | 53.0 (47.2-58.9) | 706 | 44.1 (39.5-48.7) | .02 |
| Hospitalization, past year | 246 | 17.5 (13.1-22.0) | 228 | 13.6 (11.0-16.3) | .12 |
| Medical care accessh | |||||
| Delayed, past year | 232 | 22.7 (16.8-28.6) | 151 | 11.0 (7.6-14.4) | <.001 |
| Unmet need, past year | 173 | 15.7 (10.1-21.4) | 121 | 11.3 (7.3-15.4) | .10 |
| Prescription medication accessh | |||||
| Delayed, past year | 353 | 29.6 (22.2-36.9) | 234 | 17.0 (12.7-21.3) | <.001 |
| Unmet need, past year | 280 | 27.6 (21.5-33.7) | 172 | 14.0 (10.1-18.0) | <.001 |
Two-tailed P values were obtained using the Rao-Scott χ2 test with strata, cluster, and weight variables to account for the sampling design. P values for multicategory variables reflect general tests of between-group differences across all categories.
≥2 of asthma, cancer, kidney disease, chronic obstructive pulmonary disease, diabetes, congestive heart failure, hepatitis B or C infection, HIV infection, high cholesterol, hypertension, ischemic heart disease, or stroke.
≥1 of difficulty dressing or bathing, needing help with eating, getting in or out of bed or chairs, or toileting.
Serious difficulty hearing seeing (even with glasses).
Diagnosis of schizophrenia or bipolar disorder.
Score of ≥13 on the Kessler 6-item scale of psychological distress in the past 30 d.
Assessed with the WHO Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST).
Participants needing past-year medical care (n = 2262); participants needing past-year prescription medications (n = 2667); “delayed” indicates delay in obtaining and “unmet need” indicates unable to obtain.
We used Rao-Scott χ2 tests with a 2-sided P value of less than .05 for significance to compare respondents with (categories 1-4) vs without (category 5) current or prior housing problems. We examined whether those with housing problems had ever received CHC assistance in finding a place to live. We conducted analyses in SAS (SAS Institute), version 9.4, using strata, cluster, and weight variables to account for the sampling design. Reported percentages are weighted.
Results
Of 3172 adult CHC patients, 3148 provided sufficient information to characterize their housing status. Of these, 1.2% (95% CI, 0.6%-1.8%) reported current homelessness, 9.0% (95% CI, 6.8%-11.2%) reported doubling-up, 26.8% (95% CI, 23.1%-30.6%) reported unstable housing, and 6.5% (95% CI, 4.6%-8.5%) reported stable housing but previous homelessness, totaling 43.6% (95% CI, 39.0%-48.1%) with any history of housing problems. Compared with those without housing problems, participants with housing problems were more likely to report health problems, emergency department use, and delays in care (Table). Twenty-nine percent (95% CI, 4.4%-52.9%) of homeless, 1.1% (95% CI, 0%-2.2%) of doubled-up, and 2.5% (95% CI, 0.8%-4.2%) of unstably housed patients reported CHC assistance in finding a place to live.
Discussion
In this cross-sectional study, 43.6% of adult CHC patients reported housing problems, including 1.2% who reported current homelessness. By comparison, the point prevalence of homelessness in the US population has been estimated at 0.18%. Limitations of this study include reliance on cross-sectional self-report, the lack of a validated measure of housing instability, and the potential lack of generalizability to non-CHC clinic settings. Additionally, we did not examine the correlates of specific housing problems. Nonetheless, the high prevalence of housing problems and their association with adverse health metrics suggests that CHCs should consider universal screening of housing status.
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Health Resources and Services Administration 2016 Health center data. https://bphc.hrsa.gov/uds/datacenter.aspx. Accessed November 1, 2017.
- 2.Health Resources and Services Administration Health Center Patient Survey. https://bphc.hrsa.gov/datareporting/research/hcpsurvey/index.html. Accessed November 2, 2017.
- 3.National Health Care for the Homeless Council What is the official definition of homelessness? https://www.nhchc.org/faq/official-definition-homelessness/. Accessed August 22, 2017.
- 4.Montgomery AE, Fargo JD, Kane V, Culhane DP. Development and validation of an instrument to assess imminent risk of homelessness among veterans. Public Health Rep. 2014;129(5):428-436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kushel MB, Gupta R, Gee L, Haas JS. Housing instability and food insecurity as barriers to health care among low-income Americans. J Gen Intern Med. 2006;21(1):71-77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.National Alliance to End Homelessness The state of homelessness in America. https://endhomelessness.org/homelessness-in-america/homelessness-statistics/state-of-homelessness-report/. Accessed November 15, 2017.