Hospital Readmission Risk for Patients with Self-Reported Hearing Loss and Communication Trouble
Communication is essential to good medical care. Patients who have trouble communicating with medical personnel because of hearing loss may be at risk of hospital readmission.
Methods
We studied participants in the Medicare Current Beneficiary Survey (MCBS), a nationally representative rotating panel survey. Individuals who responded between 2010 and 2013, were not institutionalized, were aged 65 and older, and had at least one inpatient claim at an acute care hospital were potentially eligible for the analysis.
Index hospitalizations were identified using criteria from the Centers for Medicare and Medicaid Services (CMS) all-cause readmission program: no transition to a health maintenance organization plan within 30 days of discharge, no death in the hospital or within 30 days of discharge, no same-day transfer to another acute care facility, and no discharge against medical advice. Readmissions were admissions to an acute care hospital within 30 days of index discharge.
Trouble communicating was assessed according to responses to two survey questions: Which describes your hearing (with a hearing aid): no trouble hearing, a little trouble, a lot of trouble, or deaf? and How much trouble do you have communicating with your doctor or other medical personnel because of your difficulty hearing: no trouble, a little trouble, or a lot of trouble? Subjects were classified as having: no trouble (no trouble on the first or second question) or trouble (a little trouble or a lot of trouble on the second question) communicating.
Multivariable models controlled for demographic (age, sex, marital status, education, race, ethnicity, income) and health (Elixhauser comorbidities present during index admission;1 self-rated health) factors. All characteristics were measured between September and December of the calendar year during which the index admission took place, except for Elixhauser comorbidities.
Some subjects had multiple hospitalizations, and following prior investigators,2 we used admissions rather than patients as the unit of analysis in multivariable models, which included clustered standard errors (by patient).
Findings
Of 30,516 potentially eligible respondents 4,426 had at least one index admission. There were with 8,005 index admissions in total; 1,130 index admissions were followed by a 30-day readmission (14.1% readmission rate).
Of those with an index admission, 513 (11.6%) had trouble communicating because of hearing loss. Trouble communicating was associated with older age, sociodemographic disadvantage, more comorbidities, and worse self-rated health (Table 1).
| Characteristic | No Trouble Communicating, n=3,913 | Trouble Communicating, n=513 | P-Value |
|---|---|---|---|
| Age, n (%) | <.001 | ||
| 65–74 | 1,136 (92.7) | 104 (7.3) | |
| 75–84 | 1,760 (89.0) | 217 (11.0) | |
| ≥85 | 837 (81.3) | 192 (18.7) | |
| Sex, n (%) | .009 | ||
| Male | 1,677 (87.0) | 251 (13.0) | |
| Female | 2,136 (89.5) | 262 (10.5) | |
| Race, n (%) | .01 | ||
| White | 3,362 (88.8) | 423 (11.2) | |
| Black | 349 (87.9) | 48 (12.1) | |
| Other | 179 (82.1) | 39 (17.9) | |
| Married or partnered, n (%) | .14 | ||
| Not currently married | 2,024 (87.7) | 283 (12.3) | |
| Currently married | 1,888 (89.1) | 230 (10.9) | |
| Income, n (%) | <.001 | ||
| <$25,000 | 1,660 (85.2) | 288 (14.8) | |
| ≥$25,000 | 2,017 (91.4) | 190 (8.6) | |
| Education, n (%) | <.001 | ||
| <High school | 981 (84.0) | 183 (15.7) | |
| ≥High school | 2,901 (90.0) | 326 (10.1) | |
| Self-rated health, n (%) | <.001 | ||
| Excellent | 365 (93.6) | 25 (6.4) | |
| Very good | 913 (92.7) | 72 (7.3) | |
| Good | 1,331 (89.6) | 154 (10.4) | |
| Fair | 880 (85.3) | 151 (14.6) | |
| Poor | 403 (79.0) | 107 (21.0) | |
| Number of Elixhauser comorbidities, mean±standard deviation | 3.15 ± 1.84 | 3.47 ± 1.84 | <.001 |
- Unweighted statistics. Characteristics were measured when patients first entered the Medicare Current Beneficiary panel, except for Elixhauser comorbidities, which were measured at first index admission. Subjects excluded in comparsion of characteristics if they had missing values.
In unadjusted and adjusted models, trouble communicating was associated with a substantially higher risk of 30-day hospital readmission (unadjusted odds ratio (OR)=1.49, 95% confidence interval (CI)=1.26–1.76; adjusted OR=1.32, 95% CI=1.06–1.64).
Discussion
In this nationally representative sample of older adults hospitalized between 2010 and 2013, 11.6% characterized their hearing difficulties as being sufficiently severe that they had trouble communicating with their doctor or other medical personnel. Those who reported trouble communicating had, on average, 32% greater odds of hospital readmission.
This study relied on self-report of hearing handicap, rather than objective measurement of hearing loss (e.g., pure tone audiometry). Objective measures correlate somewhat loosely with self-report of hearing handicap.3 A strength of this study is reliance on self-report of a plausible mediator between hearing loss and readmission (trouble communicating with medical personnel).
Limitations of the study include possible incomplete control for confounders that might mediate the observed association. Hearing difficulties may have been underreported because of stigma,4 attenuating the association.
Hearing loss is highly prevalent in older adults. Hospitals are often noisy, chaotic settings, where understanding speech is challenging, particularly for people with hearing loss.5-7 Attending to hearing difficulties could improve the quality of hospital care.
Acknowledgments
We acknowledge the support of the New York University Freedman Research Center.
Conflict of Interest: None.
Author Contributions: Conception and design: Blustein, Chang, Weinstein, Chodosh. Acquisition of data and data analysis: Chang. Interpretation: Chang, Blustein, Weinstein, Chodosh. Drafting and final approval: Blustein, Chang, Weinstein, Chodosh.
Sponsor's Role: None.
