The relationship between voting restrictions and COVID-19 case and mortality rates between US counties.

BACKGROUND: Since the 2010 election, the number of laws in the U.S. that create barriers to voting has increased dramatically. These laws may have spillover effects on population health by creating a disconnect between voter preferences and political representation, thereby limiting protective public health policies and funding. We examine whether voting restrictions are associated with county-level COVID-19 case and mortality rates. METHODS: To obtain information on restricted access to voting, we used the Cost of Voting Index (COVI), a state-level measure of barriers to voting during a U.S. election from 1996 to 2016. COVID-19 case and mortality rates were obtained from the New York Times' GitHub database (a compilation from multiple academic sources). Multilevel modeling was used to determine whether restrictive voting laws were associated with county-level COVID-19 case and mortality rates after controlling for county-level characteristics from the County Health Rankings. We tested whether associations were heterogeneous across racial and socioeconomic groups. RESULTS: A significant association was observed between increasing voting restrictions and COVID-19 case (ß = 580.5, 95% CI = 3.9, 1157.2) and mortality rates (ß = 16.5, 95% CI = 0.33,32.6) when confounders were included. CONCLUSIONS: Restrictive voting laws were associated with higher COVID-19 case and mortality rates.

Neighborhood Income Inequality and Alcohol Use among Adolescents in Boston, Massachusetts.

OBJECTIVES: Previous research has indicated that area-level income inequality is associated with increased risk in alcohol consumption. However, few studies have been conducted among adolescents living within smaller area units, such as neighborhoods. We investigated whether neighborhood income inequality is associated with alcohol consumption among adolescents. METHODS: We analyzed cross-sectional data from a sample of 1878 adolescents living in 38 neighborhoods participating in the 2008 Boston Youth Survey. Multilevel logistic regression modeling was used to determine the role of neighborhood income inequality and the odds for alcohol consumption and to determine if social cohesion and depressive symptoms were mediators. RESULTS: In comparison to the first tertile of income inequality, or the most equal neighborhood, adolescent participants living in the second tertile (AOR = 1.20, 95% CI: 0.89, 1.61) and third tertile (AOR = 1.44, 95% CI: 1.06, 1.96) were more likely to have consumed alcohol in the last 30 days. Social cohesion and depressive symptoms were not observed to mediate this relationship. CONCLUSIONS: Findings indicate that the distribution of incomes within urban areas may be related to alcohol consumption among adolescents. To prevent alcohol consumption, public health practitioners should prioritize prevention efforts for adolescents living in neighborhoods with large gaps between rich and poor.

Barriers to Voting and Access to Health Insurance Among US Adults: A Cross-Sectional Study.

Background: Many states in the United States (US) have introduced barriers to impede voting among individuals from socio-economically disadvantaged groups. This may reduce representation thereby decreasing access to lifesaving goods, such as health insurance. Methods: We used cross-sectional data from 242,727 adults in the 50 states and District of Columbia participating in the US 2017 Behavioral Risk Factor Surveillance System (BRFSS). To quantify access to voting, the Cost of Voting Index (COVI), a global measure of barriers to voting within a state during a US election was used. Multilevel modeling was used to determine whether barriers to voting were associated with health insurance status after adjusting for individual- and state-level covariates. Analyses were stratified by racial/ethnic identity, household income, and age group. Findings: A one standard deviation (SD) increase in COVI score was associated with an overall increased odds of being uninsured (OR=1.25; 95% CI=1.22, 1.28). This association was also present for Non-Hispanic Black (OR=1.18; 95% CI=1.13,1.22), Hispanic (1.18; 95% CI=1.15,1.21), and Asian (OR=1.45;95%CI=1.27,1.66), and other Non-Hispanic (OR=1.12, 95% CI=1.06, 1.18) US adults, but not for White Non-Hispanic and Native US adults. Likewise, a one SD increase in COVI among adults from low-income households was associated with an increased odds of being uninsured (OR=1.32; 95% CI=1.26,1.38) but there was no association among individuals with incomes greater than $75,000. This association was similar for younger US adults (OR=1.22; 95%CI=1.20,1.24) but not among those aged 45 to 64. Interpretation: Groups commonly targeted by voting restriction laws-those with low incomes, who are racial minorities, and who are young-are also less likely to be insured in states with more voting restrictions. However, those who are wealthier, white or older are no more likely to be uninsured irrespective of the level of voting restrictions. Funding: Pabayo is a Tier II Canada Research Chair.

State- and county-level income inequality and infant mortality in the USA in 2010: a cohort study.

OBJECTIVES: We examined the relationship between income inequality and the risk for infant/neonatal mortality at the state and county level and tested possible mediators of this relationship. METHODS: We first linked state and county Gini coefficients to US Vital Statistics 2010 Cohort Linked Birth and Infant Death records (n = 3,954,325). We then fit multilevel models to test whether income inequality was associated with infant/neonatal mortality. County-level factors were tested as potential mediators. RESULTS: Adjusted analyses indicated that income inequality at the county level-but not at the state level-was associated with increased odds of infant mortality (OR 1.14, 95% CI 1.10, 1.18) and neonatal death (OR 1.17, 95% CI 1.12, 1.23). Our mediators explained most of this variation. Bivariate analyses revealed associations between 3 county-level measures-patient-to-physician ratio, the violent crime rate, and sexually transmitted infection rate-and infant and neonatal mortality. Proportion of college-educated adults was associated with decreased odds for neonatal mortality. CONCLUSIONS: Local variations in access to care, the rate of sexually transmitted disease, and crime are associated with infant mortality, while variations in college education in addition to these mediators explain neonatal mortality. To reduce infant and neonatal mortality, experiments are needed to examine the effectiveness of policies targeted at reducing income inequality and improving healthcare access, policing, and educational opportunities.

Laws Restricting Access to Abortion Services and Infant Mortality Risk in the United States.

OBJECTIVES: Since the US Supreme Court's 1973 Roe v. Wade decision legalizing abortion, states have enacted laws restricting access to abortion services. Previous studies suggest that restricting access to abortion is a risk factor for adverse maternal and infant health. The objective of this investigation is to study the relationship between the type and the number of state-level restrictive abortion laws and infant mortality risk. METHODS: We used data on 11,972,629 infants and mothers from the US Cohort Linked Birth/Infant Death Data Files 2008-2010. State-level abortion laws included Medicaid funding restrictions, mandatory parental involvement, mandatory counseling, mandatory waiting period, and two-visit laws. Multilevel logistic regression was used to determine whether type or number of state-level restrictive abortion laws during year of birth were associated with odds of infant mortality. RESULTS: Compared to infants living in states with no restrictive laws, infants living in states with one or two restrictive laws (adjusted odds ratio (AOR) = 1.08; 95% confidence interval [CI] = 0.99-1.18) and those living in states with 3 to 5 restrictive laws (AOR = 1.10; 95% CI = 1.01-1.20) were more likely to die. Separate analyses examining the relationship between parental involvement laws and infant mortality risk, stratified by maternal age, indicated that significant associations were observed among mothers aged ≤19 years (AOR = 1.09, 95% CI = 1.00-1.19), and 20 to 25 years (AOR = 1.10, 95% CI = 1.03-1.17). No significant association was observed among infants born to older mothers. CONCLUSION: Restricting access to abortion services may increase the risk for infant mortality.

Differential relationship between state-level minimum wage and infant mortality risk among US infants born to white and black mothers.

BACKGROUND: Compared to other Organisation for Economic Co-operation and Development (OECD) nations, US infant mortality rates (IMRs) are particularly high. These differences are partially driven by racial disparities, with non-Hispanic black having IMRs that are twice those of non-Hispanic white. Income inequality (the gap between rich and poor) is associated with infant mortality. One proposed way to decrease income inequality (and possibly to improve birth outcomes) is to increase the minimum wage. We aimed to elucidate the relationship between state-level minimum wage and infant mortality risk using individual-level and state-level data. We also determined whether observed associations were heterogeneous across racial groups. METHODS: Data were from US Vital Statistics 2010 Cohort Linked Birth and Infant Death records and the 2010 US Bureau of Labor Statistics. We fit multilevel logistic models to test whether state minimum wage was associated with infant mortality. Minimum wage was standardised using the z-transformation and was dichotomised (high vs low) at the 75th percentile. Analyses were stratified by mother's race (non-Hispanic black vs non-Hispanic white). RESULTS: High minimum wage (adjusted OR (AOR)=0.93, 95% CI 0.83 to 1.03) was associated with decreased odds of infant mortality but was not statistically significant. High minimum wage was significantly associated with reduced infant mortality among non-Hispanic black infants (AOR=0.80, 95% CI 0.68 to 0.94) but not among non-Hispanic white infants (AOR=1.04, 95% CI 0.92 to 1.17). CONCLUSIONS: Increasing the minimum wage might be beneficial to infant health, especially among non-Hispanic black infants, and thus might decrease the racial disparity in infant mortality.

State-level income inequality and mortality among infants born in the United States 2007-2010: A Cohort Study.

BACKGROUND: United States state-level income inequality is positively associated with infant mortality in ecological studies. We exploit spatiotemporal variations in a large dataset containing individual-level data to conduct a cohort study and to investigate whether current income inequality and increases in income inequality are associated with infant and neonatal mortality risk over the period of the 2007-2010 Great Recession in the United States. METHODS: We used data on 16,145,716 infants and their mothers from the 2007-2010 United States Statistics Linked Infant Birth and Death Records. Multilevel logistic regression was used to determine whether 1) US state-level income inequality, as measured by Z-transformed Gini coefficients in the year of birth and 2) change in Gini coefficient between 1990 and year of birth (2007-2010), predicted infant or neonatal mortality. Our analyses adjusted for both individual and state-level covariates. RESULTS: From 2007 to 2010 there were 98,002 infant deaths: an infant mortality rate of 6.07 infant deaths per 1000 live births. When controlling for state and individual level characteristics, there was no significant relationship between Gini Z-score and infant mortality risk. However, the observed increase in the Gini Z-score was associated with a small but significant increase likelihood of infant mortality (AOR = 1.03 to 1.06 from 2007 to 2010). Similar findings were observed when the neonatal mortality was the outcome (AOR = 1.05 to 1.13 from 2007 to 2010). CONCLUSIONS: Infants born in states with greater changes in income inequality between 1990 and 2007 to 2010 experienced a greater likelihood of infant and neonatal mortality.

Structural Racism and Odds for Infant Mortality Among Infants Born in the United States 2010.

OBJECTIVES: While ecological studies indicate that high levels of structural racism within US states are associated with elevated infant mortality rates, studies using individual-level data are needed. To determine whether indicators of structural racism are associated with the individual odds for infant mortality among white and black infants in the US. METHODS: We used data on 2,163,096 white and 590,081 black infants from the 2010 US Cohort Linked Birth/Infant Death Data Files. Structural racism indicators were ratios of relative proportions of blacks to whites for these domains: electoral (registered to vote and voted; state legislature representation), employment (civilian labor force; employed; in management; with a bachelor's degree), and justice system (sentenced to death; incarcerated). Multilevel logistic regression was used to determine whether structural racism indicators were risk factors of infant mortality. RESULTS: Compared to the lowest tertile ratio of relative proportions of blacks to whites with a bachelor's degree or higher-indicative of low structural racism-black infants, but not whites, in states with moderate (OR = 1.12, 95% CI = 0.94, 1.32) and high tertiles (OR = 1.25, 95% CI = 1.03, 1.51) had higher odds of infant mortality. CONCLUSIONS: Educational and judicial indicators of structural racism were associated with infant mortality among blacks. Decreasing structural racism could prevent black infant deaths.

Racial Discrimination in Health Care and Utilization of Health Care: a Cross-sectional Study of California Adults.

BACKGROUND: Racial and ethnic discrimination in health care have been associated with suboptimal use of health care. However, limited research has examined how facets of health care utilization influence, and are influenced by, discrimination. OBJECTIVE: This study aimed to determine if type of insurance coverage and location of usual source of care used were associated with perceptions of racial or ethnic discrimination in health care. Additionally, this study examined if perceived racial or ethnic discrimination influenced delaying or forgoing prescriptions or medical care. DESIGN: Data from the 2015-2016 California Health Interview Survey were used. Logistic regression models estimated odds of perceiving racial or ethnic discrimination from insurance type and location of usual source of care. Logistic regression models estimated odds of delaying or forgoing medical care or prescriptions. PARTICIPANTS: Responses for 39,171 adults aged 18 and over were used. MAIN MEASURES: Key health care utilization variables were as follows: current insurance coverage, location of usual source of care, delaying or forgoing medical care, and delaying or forgoing prescriptions. We examined if these effects differed by race. Ever experiencing racial or ethnic discrimination in the health care setting functioned as a dependent and independent variable in analyses. KEY RESULTS: When insurance type and location of care were included in the same model, only the former was associated with perceived discrimination. Specifically, those with Medicaid had 66% higher odds of perceiving discrimination, relative to those with employer-sponsored coverage (AOR = 1.66; 95% CI 1.11, 2.47). Race did not moderate the impact of discrimination. Perceived discrimination was associated with higher odds of delaying or forgoing both prescriptions (AOR = 1.97; 95% CI 1.26, 3.09) and medical care (AOR = 1.84; 95% CI 1.31, 2.59). CONCLUSIONS: Health care providers have an opportunity to improve the experiences of their patients, particularly those with publicly sponsored coverage.

Elderly hospitalization and the New-type Rural Cooperative Medical Scheme (NCMS) in China: multi-stage cross-sectional surveys of Jiangxi province.

BACKGROUND: Studies assessing the impacts of China's New-type Rural Cooperative Medical Scheme (NCMS) reform of 2003 among rural elderly have been limited. METHOD: Multistage stratified cluster sampling household surveys of 1838, 1924, 1879, 1888, 1890 and 1896 households from 27 villages in Jiangxi province were conducted in 2003/2004, 2006, 2008, 2010, 2012 and 2014. Data from older adults age 65 and above were analyzed. Weighted logistic regression was applied to find factors of elderly hospitalization services. RESULTS: Since 2003, hospitalization rates for elderly increased, while rates of patients leaving against medical advice and patients avoiding the hospital decreased (P < 0.05). Factors associated with a higher likelihood of reporting hospitalization in the past year for elderly were the per-capita financial level V in 2012 for NCMS (Adjusted Odds Ratios [aOR]: 2.295), the level VI in 2014 (aOR: 3.045) versus the level I in 2003 and chronic disease (aOR: 2.089) versus not having a chronic disease. Lower rate of elderly left against medical advice was associated with the financial level V in 2012 (aOR: 0.099) versus the level I. The higher rate of hospital avoidance was associated with chronic disease status (aOR: 5.759) versus not having a chronic disease, while the lower rate was associated with the financial level VI in 2014 (aOR: 0.143) versus the level I. Among reporting reasons for elderly hospital avoidance, the cost-related reasons just dropped slightly over the years. CONCLUSIONS: NCMS improved access to health services for older adults. The utilization of hospitalization services for rural elderly increased gradually, but cost-related barriers remained the primary reporting barrier to accessing hospitalization services.

New cooperative medical financing policy and hospitalization in rural China: multi-stage cross-sectional surveys.

BACKGROUND: In 2003 China began to implement the New-type rural Cooperative Medical System (NCMS). This provided enhanced funding for hospital-based medical services among farmers. We examined self-reported utilization data for evidence of changes following the new policy. METHODS: We conducted a multistage stratified random cluster sampling method for Jiangxi Province, China. Data were collected via five surveys in 2003-4, 2006, 2008, 2010, and 2012. The study compared the rates of hospitalization, early discharge, and hospital avoidance as descriptive indices after weighting the data. Weighted multiple logistic regression analysis was used. Multi-stage cross-sectional analysis was used to explore the reasons for early discharge and for avoiding the hospital during illness. RESULTS: We found that the rates of hospitalization, early discharge and hospital avoidance showed upward, downward and downward changes respectively. The logistic regression analysis showed that, controlling for other factors, the financing level significantly affected the changes of the three indexes (p<0.05). The proportion of finance-related early discharge and hospital avoidance dropped significantly (p<0.05). CONCLUSIONS: NCMS improved the utilization of in-hospital services step by step as time went on, and greatly alleviated cost-related barriers to accessing health services. Even so, because costs continue to restrict access to services we should continue the NCMS policy and improve its guarantee levels.

A Comparison of HIV Knowledge, Attitudes, and Sources of STI Information Between Female and Male College Students in Taiwan.

Taiwan withdrew from the United Nations in 1971, which led to missed opportunities for participating in global HIV/AIDS programs and made Taiwan more vulnerable to HIV. Employing a questionnaire of 996 college students in Taiwan, the authors assessed and compared female and male HIV/AIDS-related knowledge, attitudes, and sources of HIV/sexually transmitted infections (STI) information. Students reported moderate knowledge and attitudes. Females had more positive attitudes toward people with HIV/AIDS than males. Most participants reported learning about HIV and STIs from traditional media, school teachers, and the Internet. We suggest evidence-based educational interventions for students should include targeted electronic and cultural awareness strategies.

The association of state law to breastfeeding practices in the US.

We assessed the relationship between breastfeeding initiation and duration with laws supportive of breastfeeding enacted at the state level. We analyzed breastfeeding practices using the 2003-2010 National Health and Nutrition Examination Survey. We evaluated three measures of breastfeeding practices: Mother's reported breastfeeding initiation, a proxy report of infants ever being breastfeed, and a proxy report of infants being breastfeed for at least 6 months. Survey data were linked to eight laws supportive of breastfeeding enacted at the state level. The most robust laws associated with increased infant breastfeeding at 6 months were an enforcement provision for workplace pumping laws [OR (95 % CI) 2.0 (1.6, 2.6)] and a jury duty exemption for breastfeeding mothers [OR (95 % CI) 1.7 (1.3, 2.1)]. Having a private area in the workplace to express breast milk [OR (95 % CI) 1.3 (1.1, 1.7)] and having break time to breastfeed or pump [OR (95 % CI) 1.2 (1.0, 1.5)] were also important for infant breastfeeding at 6 months. This research responds to breastfeeding advocates' calls for evidence-based data to generate the necessary political action to enact legislation and laws to protect, promote, and support breastfeeding. We identify the laws with the greatest potential to reach the Healthy People 2020 targets for breastfeeding initiation and duration.

Breastfeeding laws and breastfeeding practices by race and ethnicity.

OBJECTIVES: We sought to examine breastfeeding practices by race and ethnicity in areas with and without eight specific breastfeeding laws. METHODS: The 2003 through 2010 National Health and Nutrition Examination Survey provides national breastfeeding practice information. We assessed eight breastfeeding laws before and after legislation was enacted and linked to population-based estimates of breastfeeding initiation and duration for children between birth and age one. FINDINGS: Relative to Whites, Mexican-American infants were 30% more likely to breastfeed for at least 6 months in areas with laws protecting break-time from work to pump, and 20% more likely to breastfeed for at least 6 months in areas with pumping law enforcement provisions. Unexpectedly, five laws with the intention of supporting breastfeeding duration were significantly less helpful for African-American women relative to White women. African-American women were nearly half as likely to breastfeed for at least 6 months, relative to Whites in areas with provisions to provide break-time from work (adjusted odds ratio [AOR], 0.6; 95% confidence interval [CI], 0.5-0.8), private areas to pump at work (AOR, 0.6; 95% CI, 0.4-0.8), exemption from jury duty (AOR, 0.6; 95% CI, 0.4-0.9), awareness education campaigns (AOR, 0.5; 95% CI, 0.3-0.8), and pumping law enforcement provisions (AOR, 0.6; 95% CI, 0.5-0.8). CONCLUSIONS: Breastfeeding laws influence African Americans and Mexican Americans differently than Whites. Examination of specific laws in conjunction with the interaction of known specific barriers for African-American mothers could help to achieve the Healthy People 2020 goals for breastfeeding.

HIV knowledge and behavior among Chinese college students in China and the United States.

Disease prevention can be improved with a better understanding of student-environment interactions. The purposes of the study were (a) to compare HIV/AIDS-related knowledge, attitudes, sexual behaviors, and sources of HIV/STI information and (b) to investigate the association between sources of HIV/STI information and HIV/AIDS-related knowledge, attitudes, and sexual behaviors among Chinese college students in China and the United States. Comparative, correlational analysis of survey data from 608 students in two countries was conducted. Chinese students in the United States scored higher on knowledge questions. More students in the United States received HIV/STI information from the Internet and family members than did students in China. Traditional media and schoolteachers had a stronger association with participants' HIV-related knowledge, age at first intercourse, and number of sexual partners in both samples than did other sources. The survey revealed incomplete knowledge within both groups. Prevention programs should focus on risky misconceptions and should teach about strategic utilization of media.

HIV/AIDS knowledge and attitudes among Chinese college students in the US.

This study assessed knowledge and attitudes about HIV/AIDS and sources of HIV/STI information among Chinese college students living in the USA and explored specific factors associated with knowledge levels and types of sources of information. We surveyed 133 Chinese students enrolled in three US universities. About 41.4 % believed that HIV could be contracted through mosquito bites, and 22.6 % were unaware that condoms could prevent HIV. Sources of HIV/STI information were the mass media. Males were more likely to demonstrate a higher HIV/AIDS knowledge level than females. Graduate students were more likely to cite television as a source of information, and less likely to mention school teachers, than were undergraduate students. These ethnic minority immigrant students held misconceptions about HIV transmission and prevention, and possibly utilized information of varying quality. Accordingly this study identifies specific objectives for education, including basic biology and diversity issues from evidence-based sources.

Sexual behaviors, decisional balance, and self-efficacy among a sample of Chinese college students in the United States.

OBJECTIVES: To assess sexual behaviors and condom use including perceived benefits, barriers, and self-efficacy among Chinese college students in the United States. PARTICIPANTS: One hundred thirty-three Chinese undergraduate and graduate students studying at 3 US universities. METHODS: Self-report Internet questionnaire grounded in the Transtheoretical Model (TTM) framework analyzed with multivariate linear regression. RESULTS: More than half (57.9%) have initiated sex. Only 27.3% used condoms every time. Frequent reasons for not using condoms were trust in partner, reduced pleasure, and partner dislike for condoms. Participants in the earlier TTM stages (precontemplation, contemplation, preparation) reported more worry about making partners angry if condoms were used and lower self-efficacy in situations involving alcohol or drug use and perceived low-risk scenarios. CONCLUSIONS: Interventions should increase condom availability to students and should also make condom use more acceptable, more often expected, and easier to discuss.

Self-reported responses to medication therapy management services for older adults: analysis of a 5-year program.

BACKGROUND: Medication therapy management (MTM) services provide essential reviews of drug regimens and are increasingly recognized as beneficial to patient safety, improved health outcomes, and cost savings. OBJECTIVE: To assess patient behavioral outcomes from an MTM service, including actions following receipt of a pharmacist report. METHODS: A retrospective analysis of an MTM program at the Sanford Center for Aging (NV, USA) was conducted. Outcome measures included whether the patient discussed the review with the physician, whether any changes in the client's drug regimen occurred, and whether the client feels more knowledgeable about his or her medications. Predictor variables included basic demographics, prescription insurance status, number of prescriptions taken, self-reported health status, and use of medications considered to be high risk. The analysis plan involved the use of multivariate logistic regression models. RESULTS: The odds of discussing the medication review with physicians, making changes recommended in the report, and both discussing and making a change were 65%, 60%, and 67%, respectively, lower among those below poverty level than among those above poverty level; 95% confidence intervals (CIs): 0.15, 0.80; 0.18, 0.85; and 0.15, 0.73, respectively. The odds of those using high-risk drugs of making changes in drug regimens, and of discussing with physicians and making changes together, were 2 times higher than the odds of those not using these drugs, 95% CIs: 1.02, 4.31 and 1.20, 4.87, respectively. The likelihood of those reporting good or excellent health of doing the combination of discussing the MTM report with physicians and to make a drug regimen change was 2 times greater than for those reporting poor to fair health, 95% CI: 1.08, 3.65. Gender, ethnicity, age group, rural status, prescription drug insurance, and high polypharmacy were not significant factors for acting on the medication review in the adjusted model. CONCLUSION: MTM services are associated with enhanced patient self-advocacy, but like other interventions, they are constrained by social disparities. Greater attention to the resources of target populations to respond to pharmacy services is merited.

Sexual behavior, stages of condom use, and self-efficacy among college students in Taiwan.

The purposes of this study were: (1) to assess sexual behaviors and condom use behaviors; (2) to compare sexual behaviors and condom use behaviors between gender groups; and (3) to explore differences in specific items of self-efficacy to practicing condom use by the transtheoretical model stages of readiness to change among college students in Taiwan. A survey of students at two universities yielded 996 valid responses. The survey questions collected reports of demographic information, sexual history, condom use in general, and likely condom use in specific situations in relation to self-efficacy. Only 27.8% (n=277) reported ever having had sex, of these only 31.4% used condoms every time (those in action and maintenance). Condom use among women was lower than among men with men 5.1 times more likely to use condoms to prevent sexually transmitted infections (OR=5.1, 95% CI: 2.14-12.16, p=0.0002). The stages of change model with reported attitudes (self-efficacy) toward condom use in specific situations. The Tukey-Kramer multiple comparisons showed that participants in the maintenance stage reported significantly higher scores than those in the pre-contemplation, contemplation, and preparation stages for all 10 self-efficacy items (p<0.0001). Circumstances that are the most challenging for condom adherence across the stages are: partner preference to forego use, situations involving alcohol and drug use, and perceived low-risk scenarios. College students in Taiwan would benefit from targeted interventions that link risky sex to alcohol, and that address the interpersonal pressure within relationships that compel women students to practice unsafe sex.