| Literature DB >> 36232206 |
Ryan D Talbert1, Emma D Macy2.
Abstract
A large body of research has documented the far-reaching health consequences of mass incarceration in the United States. Yet, less scholarship has examined the relationship between former incarceration and oral health, a key reflection of health and disease occurring within the rest of the body. Using data extracted from the National Survey of American Life (n = 3343), this study examines associations among former incarceration status, duration of detention, and self-reported oral health among African American women and men. Results from gender-stratified ordered logistic models reveal that formerly incarcerated African American men and women experience significantly poorer oral health than their never incarcerated counterparts even after controlling for important social determinants of health. Furthermore, oral health is curvilinearly associated with the length of time that men are incarcerated such that odds of poor health decrease as detention duration increases up to approximately 15 years incarcerated. After 15 years of detainment, the odds of poor health tend to increase as duration increases. Findings extend research identifying gendered spillover health consequences of contact with the criminal legal system. Health professionals and policymakers should be conscious of incarceration as an important deleterious experience for the immediate and long-term condition of people's teeth, mouth, and gums.Entities:
Keywords: African Americans; gender; gums; incarceration; oral health; prison term; teeth
Mesh:
Year: 2022 PMID: 36232206 PMCID: PMC9566785 DOI: 10.3390/ijerph191912906
Source DB: PubMed Journal: Int J Environ Res Public Health ISSN: 1660-4601 Impact factor: 4.614
Descriptive statistics for African American women and men in the National Survey of American Life, 2003.
| African | African | |||
|---|---|---|---|---|
|
| Mean/% | SD | Mean/% | SD |
|
| ||||
| Self-rated oral health * ( | 3.08 | (1.21) | 3.18 | (0.98) |
|
| ||||
| Formerly incarcerated * | 6.33% | — | 21.05% | — |
| Years spent incarcerated * a ( | 1.81 | (3.50) | 2.35 | (4.11) |
|
| ||||
| Age (in years; | 42.64 | (17.52) | 41.58 | (14.12) |
| Education (in years; | 12.48 | (2.59) | 12.50 | (2.21) |
| Employed (yes = 1) * | 63.68% | — | 72.67% | — |
| No insurance (yes = 1) | 20.02% | — | 23.01% | — |
| Federal program insurance * (yes = 1) | 26.20% | — | 15.18% | — |
| Employee sponsored insurance * (yes = 1) | 52.78% | — | 61.81% | — |
| Married/cohabiting (yes = 1) * | 35.88% | — | 50.30% | — |
| Formerly married * (yes = 1) | 32.25 % | — | 19.14% | — |
| Never married (yes = 1) | 31.86% | — | 30.56% | — |
| Sample size | 2144 | 1166 | ||
Note: Analyses are corrected for the sampling design. Means and percentages (%) are presented with standard deviations in parentheses (SD). * Means/proportions different at p < 0.05. a Only among people who have experienced incarceration.
Ordered logistic regression models predicting oral health for African American women and men in the National Survey of American Life, 2003.
| Self-Rated Oral Health ( | ||||||||
|---|---|---|---|---|---|---|---|---|
| African American Women | African American Men | |||||||
| Model 1 | Model 2 | Model 3 | Model 4 | |||||
|
| OR | SE | OR | SE | OR | SE | OR | SE |
|
| ||||||||
| Formerly incarcerated (yes = 1) | 0.65 ** | (0.10) | — | — | 0.74 * | (0.09) | — | — |
| Years spent Incarcerated | — | — | 0.93 | (0.09) | — | — | 1.16 ** | (0.06) |
| Years spent Incarcerated | — | — | 1.00 | (0.00) | — | — | 0.99 * | (0.00) |
|
| ||||||||
| Age (in years) | 0.98 *** | (0.01) | 0.94 * | (0.02) | 0.98 *** | (0.01) | 0.98 | (0.01) |
| Education (in years) | 1.09 *** | (0.02) | 1.12 * | (0.02) | 1.09 ** | (0.04) | 1.21 * | (0.10) |
| Employed (yes = 1) | 1.01 | (0.14) | 0.93 | (0.57) | 1.53 * | (0.25) | 2.28 * | (0.85) |
| Federal program Insurance (ref = no insurance) | 0.79 | (0.10) | 0.83 | (0.51) | 0.97 | (0.18) | 1.22 | (0.45) |
| Employee sponsored | 1.32 * | (0.14) | 1.43 | (0.74) | 1.05 | (0.17) | 1.28 | (0.44) |
| Formerly married (ref = married/cohabiting) | 0.81 * | (0.08) | 1.13 | (0.52) | 0.78 | (0.12) | 0.52 * | (0.17) |
| Never married (ref = married/cohabiting) | 1.13 | (0.12) | 1.59 | (0.57) | 1.13 | (0.16) | 1.46 | (0.42) |
| Sample size | 2144 | 127 | 1166 | 252 | ||||
| BIC | 4396.310 | 318.592 | 3333.870 | 773.058 | ||||
| McFadden Pseudo R2 | 0.031 | 0.080 | 0.031 | 0.060 | ||||
Note: Analyses are corrected for the sampling design. Odds ratios (OR) are presented with rounded standard errors (SE) in parentheses for African American women and men. Reference groups for multicategory measures are indicated using the abbreviation ref. Models 2 and 4 only include adults who have ever been incarcerated. BIC indicates the Bayesian Information Criterion. * p < 0.05; ** p < 0.01; *** p < 0.001 (two-tailed tests).
Figure 1Predictions of Oral Health among African American Women in the National Survey of American Life, 2003. Note: Fully-adjusted estimates generated from Model 1 in Table 2 (n = 2144). Probabilities of poor and fair health are significantly higher for formerly incarcerated women. Probabilities of very good and excellent health are significantly lower for formerly incarcerated women. * Probabilities of an outcome are significantly different for formerly and never incarcerated women at p < 0.05.
Figure 2Predictions of Oral Health among African American Men in the National Survey of American Life, 2003. Note: Fully-adjusted estimates generated from Model 3 in Table 2 (n = 1166). Probabilities of poor, fair, and good health are significantly higher for formerly incarcerated men. Probabilities of very good and excellent health are significantly lower for formerly incarcerated men. * Probabilities of an outcome are significantly different for formerly and never incarcerated men at p < 0.05.
Figure 3Predictions of Oral Health by Number of Years Incarcerated among Formerly Incarcerated African American Men in the National Survey of American Life, 2003. Note: Fully-adjusted estimates generated from Model 4 in Table 2 (n = 252). * Probabilities of an outcome across years incarcerated are significantly different for zero at p < 0.05 indicating curvilinear associations.