Sex-Based Differences in Discharge Disposition and Outcomes for ST-Segment Elevation Myocardial Infarction Patients Within a Regional Network

Introduction

Cardiovascular disease (CVD) is the leading cause of mortality for women, accounting for nearly one of every four deaths, and mortality for women with ST-segment elevation myocardial infarction (STEMI) is higher than for men. ¹ Through regionalization and a systems approach, significant improvements in cardiovascular care have been made throughout the United States during the last few decades, including: (1) availability and timely access to percutaneous coronary intervention (PCI), (2) improved early recognition and involvement by emergency medical services (EMS), (3) enhanced and evolving medications and interventions reflected in guideline recommendations, and (4) targeted consumer awareness campaigns. ² Yet, despite this, sex-related differences in diagnosis, management, and outcomes persist. ³ Potential reasons include biological, pathophysiological, psychosocial, or other inherent multifactorial differences in women in addition to disparities in care. ^3,4

One area that has received relatively little attention is discharge disposition and management following STEMI. Women have more bleeding, heart failure, and other major adverse cardiac events following STEMI. ^{4 –8} Appropriate guideline-based recommendations for women include appropriate adjuvant pharmacology and postdischarge interventions, but there is limited guidance on discharge dispositions that might lead to better long-term outcomes. This is especially important for women, since they most often serve as caregivers in the home, and their absence may have negative psychosocial impact. ⁹ Moreover, women are suggested to have higher prevalence of depression, higher and less family and emotional support than men. ¹⁰ Therefore, longer hospital stays and discharge to facilities rather than to home can create additional psychosocial stressors. Clearly, a better understanding of the variation in discharge disposition patterns between women and men could help provide insight into improving long-term outcomes.

Modern, regionalized cardiovascular systems of care can improve access to reperfusion therapy and outcomes. These systems utilize evidence-based guidelines and large populations to improve care across an entire region, and therefore, there has been significant momentum in growth of the American Heart Association's Mission: Lifeline. ^2,11 Studies from one large Minnesota system suggests that standardized STEMI protocols can mitigate differences between the sexes in both outcomes and management. ¹²

Whether the system approach—using standardized treatment and transport protocols in addition to continuous quality improvement—will negate the sex-based differences in treatment times and outcomes is unclear. Therefore, in this study we seek to first determine if sex differences exist in patient discharge disposition and second to comprehensively assess the impact of sex on reperfusion strategy in hospital outcomes within a large regional cardiovascular system of care.

Materials and Methods

Results

Of the 25,100 myocardial infarctions, 9,674 were STEMI (38.5%), which formed the basis of this study. Average patient age was 60.8 years (IQR: 52–69). There were 7,105 males and 2,569 females (26.6%).

Table 1 presents the main characteristics of the study sample stratified by sex. Women were on average older (by 2.3 years), more likely to present with prior history of diabetes, previous stroke, and CVD, and more likely to present with shock and heart failure before PCI. Women also had overall lower reperfusion rates (93% vs. 95%), primarily due to lower utilization of PCI.

There were significant differences in all outcomes and time metrics, even after multivariate adjustments. Table 2 shows the differences between men and women for quality measures and in-hospital mortality, discharge disposition, and other discharge instructions.

Most notable, disposition upon discharge showed significantly different patterns for women compared with men. Women had nearly 0.45 days longer stay in the hospital and upon discharged were far less likely to be discharged home (by 3.7%) and instead were more likely to be discharged to transitional care or nursing homes. At discharge, they were less likely to receive special dietary modification counseling postdischarge (65.2% vs. 56.9%).

Women experienced a 4-minute longer (7.4%) median D2B, and 28-minute (15.7%) longer TIS, and longer lengths of stay—all representing significant delays in presentation and diagnosis. Unadjusted in-hospital mortality was also significantly higher for women (7.4% vs. 4.6% for men).

We found differences in medications administered at discharge for women versus men. Women received significantly greater antiplatelet and anticoagulant at discharge, but were less likely to receive statins and angiotensin converting enzyme inhibitors (ACE) than men (Table 3).

Based on the significant findings in the univariate analyses from Table 1, regression models were run for D2B, TIS, LOS, and mortality. Gender persisted as statistically significant in all four models. Table 4 shows the regression results for all models. In the adjusted models, women were much more likely to have longer D2B (B = 1.34, p < 0.001), longer TIS (B = 1.67, p < 0.001), and longer lengths of stay (B = 0.321, p < 0.001) and were less likely to live at the time of discharge than men (OR: 0.67; 95% CI: 0.52–0.76).

Discharge patterns remained statistically different for women as well, after applying multivariate controls. We compared discharges to home relative to all other dispositions, and women were significantly less likely to be sent home. Men were nearly 1.277 × more likely to be discharged home than women, after controlling for all confounding factors (OR: 1.277; p < 0.001).

Discussion

Interest and research in sex-based differences in the management and outcome of CVD in women are growing, despite that it remains “understudied, underdiagnosed, and undertreated.” ³ We observed in this study that women had differential discharge patterns, including being discharged less frequently to home and with orders for special dietary instructions. We also found that despite the presence of an urban regionalized system of care, disparities exist in all treatment outcomes and metrics. This finding is consistent with previous studies, which found that mortality and ischemic time are significantly greater for females than men. ^{15 –21} It differs from the findings of Wei et al., ¹² which suggest that standardized system protocols can mitigate differences. In the present study, we found that women with STEMI are more likely to be older than men, have greater prevalence of diabetes and other comorbid disease, and have poorer overall outcomes.

Most notable of our findings is that women experiencing acute STEMI are less likely to be discharged home than men. Overall, 88% of women in our sample were discharged home, compared to nearly 92% for men, with the remainder discharged to other facilities, such as rehabilitation or skilled nursing facilities. Discharge planning after STEMI is typically based on cardiologist recommendations of the patient's functional status, ability to perform activities of daily living, ability to comply with medication and other nutritional instructions, and availability of family support and transportation. After PCI, many facilities rely on discharge based on bedside risk prediction tools to estimate a patient's odds of short-term major adverse events and mortality, such as the Zwolle score, Killip class, TIMI score, or any variety of other risk prediction models. ²² These models, although different, incorporate quantitative factors about each patient (e.g., ischemic time, biomarkers, age, and prior history) to make decisions on optimal discharge timing and location. ²³ However, ultimately there will be variations based on individual providers and facilities, due to differences in interpretation and use of these approaches. None of the bedside tools currently incorporate sex as a variable, but here we found that women were significantly less likely to be discharged home than men, even after adjustments for age, insurance, and comorbidity factors. This could be problematic for many reasons.

Recent guidelines specific to women suggest that providers look for atypical symptoms in women, understand risks and benefits of each type of reperfusion strategy, and apply appropriate medical management and behavioral interventions for women. ³ Yet our study shows that this does not seem to be the case. This is especially concerning as gerontological and psychological research on sex show that women are often disadvantaged from longer stays and disposition away from home, due to psychosocial and other factors. ²⁴ Women are often the primary caregiver in the home, so discharging them away from home could influence the development of postdischarge depression and anxiety. Thus, physicians caring for women with STEMI should be aware of the sex effect of discharge dispositions. Of course, there are a myriad of clinical reasons why women received different discharge plans, but the presence of a modern system of care with updated protocols and evidence-based care, as well as adjustments for variations in patient and facility, did not mitigate these differences between women and men.

Guidelines for diagnosis and treatment of STEMI in women should ensure specific focus on the impact of discharge instructions that may negatively impact women more than men. For example, it has been shown that negative emotions have a damaging and lasting effect on health outcomes in patients with coronary artery disease and this in turn may impact patient adherence to postdischarge compliance with medication management and behavioral programs, more so in women than in men. ^{25 –27} Anxiety and depression postdischarge may increase when return to home is delayed and female patients are unable to resume caregiving and other social support roles. ^28,29

It is unclear why women were less likely to receive guideline-based medications at discharge in our study. Perhaps women were less likely to receive antiplatelets because they had fewer PCI procedures with stents and were less likely to receive ACE inhibitors due to better or preserved left ventricular function. However, fewer women than men discharged on statins are more difficult to explain and may relate to ongoing disparities in care.

There are limitations to this study. As a population-based study, we did not have identical baseline characteristics for both men and women, and we did not utilize a control group in the study design. In most cardiovascular studies, women represent a smaller proportion of patients than men. Although we adjusted for multiple confounding variables, future studies could use propensity matching to better control for sample variations between men and women. Furthermore, the lack of angiographic data is a key limitation. Women have higher prevalence of nonobstructive coronary artery disease and diffuse or intermediate lesions that may be difficult to treat or not amenable to PCI. With the absence of angiographic data, it is difficult to conclusively interpret the implications of these differential findings between men and women. In addition, we did not account for other biological and medical conditions that are more common in women (such as coronary artery spasm or Takotsubo syndrome) and which may delay reperfusion in women. It is likely that greater understanding of biological differences and the clinical presentation of women will provide better insight into the design of phenotype-specific guidelines for both women and men. How best to mitigate these outcome and disposition differences both prehospital and at discharge within a developed STEMI system will require additional study.

Conclusions

Based on data from this regional STEMI system in north Texas, we find that sex remains a significant factor influencing cardiovascular outcomes and management. Women had longer treatment times even after age and other multivariate adjustments. Women's medication management was also different, with lower levels of certain medications such as ACE inhibitors, but much higher frequency in antiplatelet and anticoagulant medications at discharge. Most interestingly, women were less likely to be discharged directly home. Such findings could be problematic for women with STEMI, especially given concurrent research suggesting an increased incidence of postdischarge anxiety and depression in women.