Table 5.
Odds Ratios Between Types of Barriers to Follow-up Care and Participant Characteristics
| |
Model 1: Intrapersonal-level barriersa |
Model 2: Interpersonal-level barriersa |
Model 3: Institutional-level barriersa |
|||
|---|---|---|---|---|---|---|
| Characteristics | Reported barrier, Nb N=153 | OR (95% CI) | Reported barrier, Nb N=24 | OR (95% CI) | Reported barrier, Nb N=39 | OR (95% CI) |
| Age (years)c | 153 | 1.03 (1.00, 1.07) | 24 | 1.00 (0.97, 1.03) | 34 | 1.00 (0.96, 1.04) |
| Race | ||||||
| African American | 20 | ref | 3 | ref | 5 | ref |
| Latina | 133 | 17.39 (7.62, 39.68)*** | 21 | 1.67 (0.72, 3.88) | 29 | 2.06 (0.61, 7.01) |
| Education | ||||||
| ≤ High school | 113 | ref | 13 | ref | 23 | ref |
| > High school | 34 | 1.25 (0.52, 3.00) | 10 | 1.11 (0.57, 2.18) | 10 | 1.44 (0.58, 3.59) |
| Employed | ||||||
| No | 90 | ref | 6 | d | 13 | ref |
| Yes | 72 | 0.68 (0.30, 1.53) | 18 | 21 | 2.41 (1.05, 5.54)* | |
| Income | ||||||
| <$10,000 | 61 | ref | 10 | ref | 10 | e |
| $10,000–$19,999 | 50 | 1.99 (0.78, 5.03) | 10 | 1.41 (0.71, 2.82) | 15 | |
| >=$20,000 | 35 | 0.95 (0.36, 2.51) | 3 | 0.89 (0.40, 1.95) | 8 | |
| Marital status | ||||||
| Single/divorced/widow | 84 | ref | 16 | ref | 17 | ref |
| Married | 66 | 1.32 (0.53, 3.31) | 7 | 0.84 (0.43, 1.64) | 17 | 1.62 (0.7, 3.77) |
| Health insurance | ||||||
| Uninsured | 156 | d | 19 | ref | 29 | ref |
| Public | 13 | 4 | 0.76 (0.31, 1.83) | 5 | 0.68 (0.19, 2.34) | |
| Private | 3 | 1 | 1.30 (0.34, 4.95) | 0 | – | |
| Primary care Provider | ||||||
| No | 51 | ref | 4 | ref | 13 | ref |
| Yes | 99 | 0.76 (0.32, 1.77) | 19 | 21 | 0.74 (0.32,1.70) | |
| Lost to follow-up | ||||||
| No | 148 | ref | 21 | ref | 33 | ref |
| Yes | 5 | 0.13 (0.03, 0.69) | 3 | 1 | 0.40 (0.05,3.44) | |
p<.05; **p<.01; ***p<.001.
Each column represents a multivariate logistic regression model that includes independent variables listed in the first column.
Column totals within each characteristic do not add up to the total number of women reporting the barrier due to missing values.
Age included in model as a continuous variable.
Independent variable not included in regression due to its endogenous relationship with the outcome variable.
An initial model (not shown) included the income variable, resulting in no statistically significant predictors; therefore, we excluded this variable as it was correlated with employment. New estimates are presented in Model 3.
OR: Adjusted Odds Ratio, CI: Confidence Intervals.