Studies show women do not receive aggressive cardiovascular interventions and may not be given guideline-based treatment to reduce cardiac events. We describe cholesterol treatment in an academic practice of family and internal medicine physicians to understand factors associated with achievement of guideline-based treatment goals in women compared with men. Primary care patients aged 40 to 75 years were included if they were prescribed a statin, had a Framingham risk score of ≥10%, had diabetes, or had atherosclerotic cardiovascular disease. Patients were classified into Adult Treatment Panel III categories and assessed to whether they were in compliance with Adult Treatment Panel III guidelines. Odds ratios of goal adherence between women and men were calculated, and a multivariate model for goal achievement was created. In 2,747 patients, women were less likely to achieve cholesterol goals (odds ratio [OR] 0.82; 95% confidence interval [CI] 0.70 to 0.95) despite having more prescriptions for statins (48% vs 39%, p <0.001). More women than men failed to reach low-density lipoprotein goals because they were not prescribed a statin (OR 0.69; 95% CI 0.56 to 0.85) and women on high-intensity statins were less likely than men to achieve goals (OR 0.51; 95% CI 0.27 to 0.96). In all patients, diabetes was associated with nonattainment of cholesterol goals, but in high-risk women, the presence of diabetes improved goal achievement. In conclusion, women achieved guideline-based cholesterol recommendations at a lower rate than men, even when individual goals are considered.
Gender differences between the rates of intensive treatments for coronary artery disease were observed in the 1990s and continue today. To explain this disparity, the “Yentl Syndrome” was described, stating that women must present “just like a man” to receive aggressive cardiovascular treatment. Although women are at decreased risk of cardiovascular disease compared with men of similar age and risk factors, they are still undertreated in many practice settings even if they are determined to be at risk. Although previous literature argues that women may derive less benefit from lipid treatment as men, recent data show that statins have similar efficacy for either gender in both primary and secondary prevention of cardiovascular events. The reasons for the gender gap in treatment are likely multifactorial and could include a gender-based difference in the biology of coronary artery disease and response to medicines. Knowing that large data sets can help understand differences in cardiac risk management, we queried clinical data to assess gender differences in lipid treatment in an academic primary care practice.
Methods
A Primary Care Patient Data Registry was created from records of 27,225 patients enrolled in SLUCare, the academic clinical practice of the School of Medicine at Saint Louis University (family medicine = 10,994 and internal medicine = 16,231). The Saint Louis University Institutional Review Board approved the development of the registry and subsequent analysis. All patients aged 40 to 75 years with at least 1 visit from January 1, 2012, to July 31, 2013 were eligible for the study (n = 9,808). Patients were included if gender, race, blood pressure, and smoking status in addition to cholesterol testing and statin dose were available (n = 4,536). Clinical atherosclerotic cardiovascular disease (ASCVD) was determined based on the International Classification of Diseases, Ninth Revision , codes for coronary artery disease, peripheral vascular disease, and cerebrovascular disease. Presence of diabetes and hypertension was identified by the International Classification of Diseases, Ninth Revision, Clinical Modification, codes, and family history of cardiovascular disease was determined from the electronic medical record (EPIC Systems, Verona, Wisconsin). Cholesterol medicine use was gathered from prescription data and self-reported patient medication history. Statin dose intensity was classified as low, moderate, or high based on the 2013 American College of Cardiology/American Heart Association guidelines.
We calculated the Framingham Risk Score (FRS) for all patients. We selected a cohort at risk for cardiovascular disease, defined as patients with a Framingham risk of ≥10%, currently taking a statin, or with a diabetes or ASCVD (n = 2,747). All patients were classified to 2004 Adult Treatment Panel III (ATPIII) risk groups using FRS and the number of risk factors (smoking, hypertension, low high-density lipoprotein, family history of cardiovascular disease and age >44 for men, >54 for women). The primary outcome was achievement of ATPIII treatment goals, using the most recent low-density lipoprotein (LDL) value. Statin prescriptions and statin dose intensity (no treatment, low intensity, medium intensity, or high intensity) were also compared between men and women using chi-square tests. The number and percentage of men and women in each treatment group who achieved cholesterol goals were compared by odds ratios.
Bivariate analyses were performed using logistic regression, with LDL goal attainment as the outcome variable. All factors found to be significantly associated with meeting goals in bivariate analyses were included as exposure variables in multivariate logistic regression. To conduct a stratified analysis, patients were subdivided into a high-risk group (defined as having a Framingham risk of ≥20%, diabetes, or ASCVD) and a low-risk group (defined as having a FRS <20% and without diabetes or ASCVD). As in the combined sample, bivariate and multivariate logistic regressions were performed to determine the associations between the predictor variables and goal attainment in each stratum of risk. Data were analyzed with SPSS version 20 (IBM, Armonk, New York), and statistical significance was determined by 2-tailed p values ≤0.05.
Results
Of the 2,747 patients at risk for cardiovascular disease, 1,341 (49%) were women ( Table 1 ) and 1,657 were classified as high risk. Women were significantly older and more likely to have diabetes, hypertension, a positive family history of heart disease, and a higher BMI. Women had significantly lower mean Framingham and ASCVD risk scores. More women were identified as nonwhite and less likely to go to an internal medicine clinic. No significant differences were found in rates of smoking and ASCVD by gender.
| Patient Characteristics | Men n=1406 | Women n=1341 | p Value ∗ |
|---|---|---|---|
| Age (years) | 59±8.3 | 61±8.0 | p<0.001 |
| Race | |||
| White | 828 (59%) | 575 (43%) | p<0.001 |
| Black | 575 (38%) | 746 (56%) | |
| Other | 38 (2.7%) | 20 (1.5%) | |
| ASCVD | 245 (17%) | 204 (15%) | p=0.12 |
| Smoker | 251 (18%) | 228 (17%) | p=0.56 |
| Diabetes mellitus | 409 (29%) | 589 (44%) | p<0.001 |
| Hypertension | 967 (69%) | 1086 (81%) | p<0.001 |
| Family history of CVD | 915 (65%) | 976 (73%) | p<0.001 |
| Body Mass Index (kg/m 2 ) | 31±7 | 34±8 | p<0.001 |
| Internal Medicine Clinic | 1044 (74%) | 950 (71%) | p=0.045 |
| Framingham Risk Score | 24±13 | 16±10 | p<0.001 |
| ASCVD Risk Score | 16±10 | 13±10 | p<0.001 |
| Statin treatment | 554 (39%) | 647 (48%) | p<0.001 |
∗ p Values calculated by Pearson chi-square or ANOVA as appropriate.
Figure 1 presents the 2004 ATPIII risk strata and percentage of LDL goal attainment across each stratum, according to gender. Overall, female gender was significantly associated with failure to meet LDL goals. This relationship was unchanged when using the 2001 ATPIII guidelines that included an LDL goal of <100 mg/dl for patients with known ASCVD (odds ratio [OR] 0.79; 95% confidence interval [CI] 0.68 to 0.92). In several ATPIII risk categories, women failed to meet LDL goals at a higher rate than men, including patients with ASCVD and those with 2 risk factors and an FRS of 10 to <20% without diabetes or ASCVD. Despite lower attainment of LDL goals, more women in the cross-sectional cohort were prescribed statins. In the high-risk group, there were no significant differences in statin prescriptions. More women were prescribed statins in the lower risk groups, including patients with 2 risk factors and an FRS of 10% to 20% and those whose goal was an LDL <190 mg/dl.
Table 2 shows statin dose intensity compared with achievement of ATPIII goals by gender. No significant trend in the intensity of statin prescriptions was found between men and women. For those not prescribed a statin, women were less likely to achieve their goals than men. Although no association was found with groups prescribed low- and medium-intensity statins, a significant association was found in those prescribed a high-intensity statin where women met goals less frequently than men.
| Statin Treatment | Men | Women | Men at Goal | Women at Goal | OR (95% CI) |
|---|---|---|---|---|---|
| No treatment | 852 ( 61%) ∗ | 694 ( 52%) ∗ | 430 ( 51%) | 287 ( 41%) | 0.69 ( 0.56-0.85) |
| Low Intensity | 87 (6%) | 83 (6%) | 41 (47%) | 46 (55%) | 1.4 (0.76-2.6) |
| Medium Intensity | 393 (28)% | 476 (36%) | 214 (55%) | 258 (54%) | 0.99 (0.76-1.3) |
| High Intensity | 74 (5%) | 88 (7%) | 40 ( 54%) | 33 ( 38%) | 0.51 ( 0.27-0.96) |
| Total | 1406 | 1314 | 725 ( 52%) | 624 ( 47%) | 0.82 ( 0.70-0.95) |
∗ p values calculated by Pearson chi-square or ANOVA as appropriate.
The association of demographic variables and failure of achieving LDL targets for the entire cohort and patients at high risk for cardiovascular disease using bivariate and multivariate analyses is listed in Table 3 . There was an association between goal nonadherence and female gender. Clinic enrollment was associated with goal adherence. In bivariate analyses, patients in internal medicine clinics were less likely to adhere to cholesterol goals. In multivariate analyses adjusting for covariates, patients in internal medicine clinics were more likely to achieve LDL goals.
| Variable | Bivariate Analysis n=2747 | Multivariate Analysis n=2747 | Bivariate Analysis for High Risk Group n=1657 | Multivariate Analysis for High Risk Group n=1657 | ||||
|---|---|---|---|---|---|---|---|---|
| Odds Ratio | CI (Odds, 95%) | Odds Ratio | CI (Odds, 95%) | Odds Ratio | CI (Odds, 95%) | Odds Ratio | CI (Odds, 95%) | |
| Sex (Female) | 0.82 | (0.70-0.95) | 0.84 | (0.72 – 0.98) | 0.89 | (0.72-1.09) | 0.61 | (0.48 – 0.76) |
| Race (non-White) | 0.80 | (0.69-0.93) | 0.94 | (0.90-1.11) | 1.10 | (0.89-1.34) | 0.96 | (0.77-1.21) |
| Diabetes mellitus | 0.68 | (0.59-0.80) | 0.75 | (0.63 – 0.88) | 2.88 | (2.30 – 3.62) | 3.13 | (2.44 – 4.03) |
| Hypertension | 0.73 | (0.61-0.86) | 0.92 | (0.76-1.10) | 1.23 | (0.94-1.61) | 1.31 | (0.98-1.74) |
| ASCVD | 0.23 | (0.18-0.30) | 0.24 | (0.19 – 0.31) | 0.44 | (0.34 – 0.57) | 0.51 | (0.39 – 0.66) |
| Smoker | 0.74 | (0.61-0.91) | 0.79 | (0.64 – 0.97) | 0.89 | (0.69-1.14) | 1.13 | (0.86-1.49) |
| Family History of CVD | 0.80 | (0.68-0.94) | 0.77 | (0.65 – 0.92) | 0.87 | (0.70-1.07) | 0.87 | (0.69-1.09) |
| Internal Medicine | 0.69 | (0.58-0.82) | 1.37 | (1.15 – 1.64) | 0.76 | (0.60 – 0.97) | 1.28 | (0.99-1.64) |
| BMI ≥30 (kg/m 2 ) | 0.87 | (0.75-1.02) | —– | 1.16 | (0.94-1.43) | —– | ||
| Statin treatment | 1.28 | (1.10-1.49) | —– | 1.15 | (0.94-1.41) | —– | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
