User login
Thiazide Diuretic Utilization Within the VA
Hypertension is one of the most common cardiovascular disease (CVD) states, affecting nearly half of all adults in the United States.1 Numerous classes of antihypertensives are available for blood pressure (BP) management, including thiazide diuretics, which contain both thiazide and thiazide-like agents. Thiazide diuretics available in the US include hydrochlorothiazide (HCTZ), chlorthalidone, metolazone, and indapamide. These agents are commonly used and recommended as first-line treatment in the current 2017 American College of Cardiology/American Heart Association (ACC/AHA) guideline for the prevention, detection, evaluation, and management of high BP in adults.2
The ACC/AHA guideline recommends chlorthalidone as the preferred thiazide diuretic.2 This recommendation is based on its prolonged half-life compared with other thiazide agents, as well as the reduction of CVD seen with chlorthalidone in previous trials. The main evidence supporting chlorthalidone use comes from the ALLHAT trial, which compared chlorthalidone, amlodipine, and lisinopril in patients with hypertension. The primary composite outcome of fatal coronary artery disease or nonfatal myocardial infarction was not significantly different between groups. However, when looking at the incidence of heart failure, chlorthalidone was superior to both amlodipine and lisinopril.3 In the TOMHS trial, chlorthalidone was more effective in reducing left ventricular hypertrophy than amlodipine, enalapril, doxazosin, or acebutolol.4 Furthermore, both a systematic review and a retrospective cohort analysis suggested that chlorthalidone may be associated with improved CVD outcomes compared with HCTZ.5,6 However, prospective randomized trial data is needed to confirm the superiority of chlorthalidone over other thiazide diuretics.
HCTZ has historically been the most common thiazide diuretic.7 However, with the available evidence and 2017 ACC/AHA BP guideline recommendations, it is unclear whether this trend continues and what impact it may have on CVD outcomes. It is unclear which thiazide diuretic is most commonly used in the US Department of Veterans Affairs (VA) health care system. The purpose of this project was to evaluate current thiazide diuretic utilization within the VA.
Methods
This retrospective, observational study evaluated the prescribing pattern of thiazide diuretics from all VA health care systems from January 1, 2016, to January 21, 2022. Thiazide diuretic agents included in this study were HCTZ, chlorthalidone, indapamide, and any combination antihypertensive products that included these 3 thiazide diuretics. Metolazone was excluded as it is commonly used in the setting of diuretic resistance with heart failure. Data was obtained from the VA Corporate Data Warehouse (CDW) and divided into 2 cohorts: the active and historic cohorts. The active cohort was of primary interest and included any active VA thiazide diuretic prescriptions on January 21, 2022. The historic cohort included thiazide prescriptions assessed at yearly intervals from January 1, 2016, to December 31, 2021. This date range was selected to assess what impact the 2017 ACC/AHA BP guideline had on clinician preferences and thiazide diuretic prescribing rates.
Within the active cohort, demographic data, vital information, and concomitant potassium or magnesium supplementation were collected. Baseline characteristics included were age, sex, race and ethnicity, and BP. Patients with > 1 race or ethnicity reported were categorized as other. The first BP reading documented after the active thiazide diuretic initiation date was included for analysis to capture on-therapy BPs while limiting confounding factors due to other potential antihypertensive changes. This project was ruled exempt from institutional review board review by the West Palm Beach VA Healthcare System Research and Development Committee.
The primary outcome was the evaluation of utilization rates of each thiazide in the active cohort, reported as a proportion of overall thiazide class utilization within the VA. Secondary outcomes in the active thiazide cohort included concomitant potassium or magnesium supplement utilization rates in each of the thiazide groups, BP values, and BP control rates. BP control was defined as a systolic BP < 130 mm Hg and a diastolic BP < 80 mm Hg. Finally, the change in thiazide diuretic utilization patterns from January 1, 2016, to December 31, 2021, was evaluated in the historic cohort.
Statistical Analysis
Data collection and analysis were completed using the CDW analyzed with Microsoft SQL Server Management Studio 18 and Microsoft Excel. All exported data to Microsoft Excel was kept in a secure network drive that was only accessible to the authors. Protected health information remained confidential per VA policy and the Health Insurance Portability and Accountability Act.
Baseline demographics were evaluated across thiazide arms using descriptive statistics. The primary outcome was assessed and a χ2 test with a single comparison α level of 0.05 with Bonferroni correction to adjust for multiple comparisons when appropriate. For the secondary outcomes, analysis of continuous data was assessed using analysis of variance (ANOVA), and nominal data were assessed with a χ2 test with a single comparison α level of 0.05 and Bonferroni correction to adjust for multiple comparisons where appropriate. When comparing all 3 thiazide groups, after the Bonferroni correction, P < .01667 was considered statistically significant to avoid a type 1 error in a family of statistical tests.
Results
As of January 21, 2022, the active thiazide cohort yielded 628,994 thiazide prescriptions within the VA nationwide. Most patients were male, with female patients representing 8.4%, 6.6%, and 5.6% of the HCTZ, chlorthalidone, and indapamide arms, respectively (Table 1). Utilization rates were significantly different between thiazide groups (P < .001). HCTZ was the most prescribed thiazide diuretic (84.6%) followed by chlorthalidone (14.9%) and indapamide (0.5%) (Table 2).
BP values documented after prescription initiation date were available for few individuals in the HCTZ, chlorthalidone, and indapamide groups (0.3%, 0.2%, and 0.5%, respectively). Overall, the mean BP values were similar among thiazide groups: 135/79 mm Hg for HCTZ, 137/78 mm Hg for chlorthalidone, and 133/79 mm Hg for indapamide (P = .32). BP control was also similar with control rates of 26.0%, 27.1%, and 33.3% for those on HCTZ, chlorthalidone, and indapamide, respectively (P = .75). The use of concomitant potassium or magnesium supplementation was significantly different between thiazide groups with rates of 12.4%, 22.6%, and 27.1% for HCTZ, chlorthalidone, and indapamide, respectively (P < .001). When comparing chlorthalidone to HCTZ, there was a significantly higher rate of concomitant supplementation with chlorthalidone (P < .001) (Table 3).
In the historic cohort, HCTZ utilization decreased from 90.2% to 83.5% (P < .001) and chlorthalidone utilization increased significantly from 9.3% to 16.0% (P < .001) (Figure). There was no significant change in the use of indapamide during this period (P = .73). Yearly trends from 2016 to 2021 are listed in Table 4.
Discussion
The findings of our evaluation demonstrate that despite the 2017 ACC/AHA BP guideline recommendations for using chlorthalidone, HCTZ predominates as the most prescribed thiazide diuretic within the VA. However, since the publication of this guideline, there has been an increase in chlorthalidone prescribing and a decrease in HCTZ prescribing within the VA.
A 2010 study by Ernst and colleagues revealed a similar trend to what was seen in our study. At that time, HCTZ was the most prescribed thiazide encompassing 95% of total thiazide utilization; however, chlorthalidone utilization increased from 1.1% in 2003 to 2.4% in 2008.8 In comparing our chlorthalidone utilization rates with these results, 9.3% in 2016 and 16.0% in 2021, the change in chlorthalidone prescribing from 2003 to 2016 represents a more than linear increase. This trend continued in our study from 2016 to 2021; the expected chlorthalidone utilization would be 21.2% in 2021 if it followed the 2003 to 2016 rate of change. Thus the trend in increasing chlorthalidone use predated the 2017 guideline recommendation. Nonetheless, this change in the thiazide prescribing pattern represents a positive shift in practice.
Our evaluation found a significantly higher rate of concomitant potassium or magnesium supplementation with chlorthalidone and indapamide compared with HCTZ in the active cohort. Electrolyte abnormalities are well documented adverse effects associated with thiazide diuretic use.9 A cross-sectional analysis by Ravioli and colleagues revealed thiazide diuretic use was an independent predictor of both hyponatremia (22.1% incidence) and hypokalemia (19% incidence) and that chlorthalidone was associated with the highest risk of electrolyte abnormalities whereas HCTZ was associated with the lowest risk. Their study also found these electrolyte abnormalities to have a dose-dependent relationship with the thiazide diuretic prescribed.10
While Ravioli and colleagues did not address the incidence of hypomagnesemia with thiazide diuretic use, a cross-sectional analysis by Kieboom and colleagues reported a significant increase in hypomagnesemia in patients prescribed thiazide diuretics.11 Although rates of electrolyte abnormalities are reported in the literature, the rates of concomitant supplementation are unclear, especially when compared across thiazide agents. Our study provides insight into the use of concomitant potassium and magnesium supplementation compared between HCTZ, chlorthalidone, and indapamide. In our active cohort, potassium was more commonly prescribed than magnesium. Interestingly, magnesium supplementation accounted for 25.9% of the total supplement use for HCTZ compared with rates of 22.4% and 21.0% for chlorthalidone and indapamide, respectively. It is unclear if this trend highlights a greater incidence of hypomagnesemia with HCTZ or greater clinician awareness to monitor this agent, but this finding may warrant further investigation. In addition, when considering the overall lower rate of supplementation seen with HCTZ in our study, the use of potassium-sparing diuretics should be considered. These agents, including triamterene, amiloride, eplerenone, and spironolactone, can be supplement-sparing and are available in combination products only with HCTZ.
Low chlorthalidone utilization rates are concerning especially given the literature demonstrating CVD benefit with chlorthalidone and the lack of compelling outcomes data to support HCTZ as the preferred agent.3,4 There are several reasons why HCTZ use may be higher in practice. First is clinical inertia, which is defined as a lack of treatment intensification or lack of changing practice patterns, despite evidence-based goals of care.12 HCTZ has been the most widely prescribed thiazide diuretic for years.7 As a result, converting HCTZ to chlorthalidone for a patient with suboptimal BP control may not be considered and instead clinicians may add on another antihypertensive or titrate doses of current antihypertensives.
There is also a consideration for patient adherence. HCTZ has many more combination products available than chlorthalidone and indapamide. If switching a patient from an HCTZ-containing combination product to chlorthalidone, adherence and patient willingness to take another capsule or tablet must be considered. Finally, there may be clinical controversy and questions around switching patients from HCTZ to chlorthalidone. Although the guidelines do not explicitly recommend switching to chlorthalidone, it may be reasonable in most patients unless they have or are at significant risk of electrolyte or metabolic disturbances that may be exacerbated or triggered with conversion.
When converting from HCTZ to chlorthalidone, it is important to consider dosing. Previous studies have demonstrated that chlorthalidone is 1.5 to 2 times more potent than HCTZ.13,14 Therefore, the conversion from HCTZ to chlorthalidone is not 1:1, but instead 50 mg of HCTZ is approximately equal to25 to 37.5 mg of chlorthalidone.14
Limitations
This study was limited by its retrospective design, gaps in data, duplicate active prescription data, and the assessment of concomitant electrolyte supplementation. As with any retrospective study, there is a potential for confounding and a concern for information bias with missing information. This study relied on proper documentation of prescription and demographic information in the Veterans Health Information Systems and Technology Architecture (VistA), as the CDW compiles information from this electronic health record. Strengths of the VistA include ease in clinical functions, documentation, and the ability for records to be updated from any VA facility nationally. However, there is always the possibility of user error and information to be omitted.
In our study, the documentation of BP values and subsequent analysis of overall BP control were limited. For BP values to be included in this study, they had to be recorded after the active thiazide prescription was written and from an in-person encounter documented in VistA. The COVID-19 pandemic shifted the clinical landscape and many primary care appointments during the active cohort evaluation period were conducted virtually. Therefore, patients may not have had formal vitals recorded. There may also be an aspect of selection bias regarding the chlorthalidone group. Although rates of thiazide switching were not assessed, some patients may have been switched from HCTZ or indapamide to chlorthalidone to achieve additional BP control. Thus, patients receiving chlorthalidone may represent a more difficult-to-control hypertensive population, making a finding of similar BP control rates between HCTZ and chlorthalidone an actual positive finding regarding chlorthalidone. Finally, this study did not assess adherence to medications. As the intent of the study was to analyze prescribing patterns, it is impossible to know if the patient was actively taking the medication at the time of assessment. When considering the rates of BP control, there were limited BP values, a potential for selection bias, and neither adherence nor patient self-reported home BP values were assessed. Therefore, the interpretation of overall BP control must be done with caution.
Additionally, duplicate prescriptions were noted in the active cohort. Rates of duplication were 0.2%, 0.08%, and 0.09% for HCTZ, chlorthalidone, and indapamide, respectively. With these small percentages, we felt this would not have a significant impact on the overall thiazide use trends seen in our study. Patients can receive prescriptions from multiple VA facilities and may have > 1 active prescriptions. This has been mitigated in recent years with the introduction of the OneVA program, allowing pharmacists to access any prescription on file from any VA facility and refill if needed (except controlled substance prescriptions). However, there are certain instances in which duplicate prescriptions may be necessary. These include patients enrolled and receiving care at another VA facility (eg, traveling for part of a year) and patients hospitalized at a different facility and given medications on discharge.
With the overall low rate of duplication prescriptions seen in each thiazide group, we determined that this was not large enough to cause substantial variation in the results of this evaluation and was unlikely to alter the results. This study also does not inform on the incidence of switching between thiazide diuretics. If a patient was switched from HCTZ to chlorthalidone in 2017, for example, a prescription for HCTZ and chlorthalidone would have been reported in this study. We felt that the change in chlorthalidone prescribing from January 1, 2016, to December 31, 2021, would reflect overall utilization rates, which may include switching from HCTZ or indapamide to chlorthalidone in addition to new chlorthalidone prescriptions.
Finally, there are confounders and trends in concomitant potassium or magnesium supplementation that were not accounted for in our study. These include concomitant loop diuretics or other medications that may cause electrolyte abnormalities and the dose-dependent relationship between thiazide diuretics and electrolyte abnormalities.10 Actual laboratory values were not included in this analysis and thus we cannot assess whether supplementation or management of electrolyte disturbances was clinically appropriate.
Conclusions
Thiazide utilization patterns have shifted possibly due to the 2017 ACC/AHA BP guideline recommendations. However, HCTZ continues to be the most widely prescribed thiazide diuretic within the VA. There is a need for future projects and clinician education to increase the implementation of guideline-recommended therapy within the VA, particularly regarding chlorthalidone use.
1. Centers for Disease Control and Prevention. Hypertension cascade: hypertension prevalence, treatment and control estimates among U.S. adults aged 18 years and older applying the criteria from the American College of Cardiology and American Heart Association’s 2017 Hypertension Guideline—NHANES 2015–2018. Updated May 12, 2023. Accessed October 12, 2023. https://millionhearts.hhs.gov/data-reports/hypertension-prevalence.html
2. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. doi:10.1161/HYP.0000000000000065
3. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288(23):2981-2997. doi:10.1001/jama.288.23.2981
4. Liebson PR, Grandits GA, Dianzumba S, et al. Comparison of five antihypertensive monotherapies and placebo for change in left ventricular mass in patients receiving nutritional-hygienic therapy in the Treatment of Mild Hypertension Study (TOMHS). Circulation. 1995;91(3):698-706. doi:10.1161/01.cir.91.3.698
5. Roush GC, Holford TR, Guddati AK. Chlorthalidone compared with hydrochlorothiazide in reducing cardiovascular events: systematic review and network meta-analyses. Hypertension. 2012;59(6):1110-1117. doi:10.1161/HYPERTENSIONAHA.112.191106
6. Dorsch MP, Gillespie BW, Erickson SR, Bleske BE, Weder AB. Chlorthalidone reduces cardiovascular events compared with hydrochlorothiazide: a retrospective cohort analysis. Hypertension. 2011;57(4):689-694. doi:10.1161/HYPERTENSIONAHA.110.161505
7. Vongpatanasin W. Hydrochlorothiazide is not the most useful nor versatile thiazide diuretic. Curr Opin Cardiol. 2015;30(4):361-365. doi:10.1097/HCO.0000000000000178
8. Ernst ME, Lund BC. Renewed interest in chlorthalidone: evidence from the Veterans Health Administration. J Clin Hypertens (Greenwich). 2010;12(12):927-934. doi:10.1111/j.1751-7176.2010.00373.x
9. Greenberg A. Diuretic complications. Am J Med Sci. 2000;319(1):10-24. doi:10.1016/S0002-9629(15)40676-7
10. Ravioli S, Bahmad S, Funk GC, Schwarz C, Exadaktylos A, Lindner G. Risk of electrolyte disorders, syncope, and falls in patients taking thiazide diuretics: results of a cross-sectional study. Am J Med. 2021;134(9):1148-1154. doi:10.1016/j.amjmed.2021.04.007
11. Kieboom BCT, Zietse R, Ikram MA, Hoorn EJ, Stricker BH. Thiazide but not loop diuretics is associated with hypomagnesaemia in the general population. Pharmacoepidemiol Drug Saf. 2018;27(11):1166-1173. doi:10.1002/pds.4636
12. O’Connor PJ, Sperl-Hillen JAM, Johnson PE, et al. Clinical Inertia and Outpatient Medical Errors. In: Henriksen K, Battles JB, Marks ES, et al, editors. Advances in Patient Safety: From Research to Implementation (Volume 2: Concepts and Methodology). Rockville (MD): Agency for Healthcare Research and Quality (US); 2005. https://www.ncbi.nlm.nih.gov/books/NBK20513/
13. Carter BL, Ernst ME, Cohen JD. Hydrochlorothiazide versus chlorthalidone: evidence supporting their interchangeability. Hypertension. 2004;43(1):4-9. doi:10.1161/01.HYP.0000103632.19915.0E
14. Liang W, Ma H, Cao L, Yan W, Yang J. Comparison of thiazide-like diuretics versus thiazide-type diuretics: a meta-analysis. J Cell Mol Med. 2017;21(11):2634-2642. doi:10.1111/jcmm.13205
Hypertension is one of the most common cardiovascular disease (CVD) states, affecting nearly half of all adults in the United States.1 Numerous classes of antihypertensives are available for blood pressure (BP) management, including thiazide diuretics, which contain both thiazide and thiazide-like agents. Thiazide diuretics available in the US include hydrochlorothiazide (HCTZ), chlorthalidone, metolazone, and indapamide. These agents are commonly used and recommended as first-line treatment in the current 2017 American College of Cardiology/American Heart Association (ACC/AHA) guideline for the prevention, detection, evaluation, and management of high BP in adults.2
The ACC/AHA guideline recommends chlorthalidone as the preferred thiazide diuretic.2 This recommendation is based on its prolonged half-life compared with other thiazide agents, as well as the reduction of CVD seen with chlorthalidone in previous trials. The main evidence supporting chlorthalidone use comes from the ALLHAT trial, which compared chlorthalidone, amlodipine, and lisinopril in patients with hypertension. The primary composite outcome of fatal coronary artery disease or nonfatal myocardial infarction was not significantly different between groups. However, when looking at the incidence of heart failure, chlorthalidone was superior to both amlodipine and lisinopril.3 In the TOMHS trial, chlorthalidone was more effective in reducing left ventricular hypertrophy than amlodipine, enalapril, doxazosin, or acebutolol.4 Furthermore, both a systematic review and a retrospective cohort analysis suggested that chlorthalidone may be associated with improved CVD outcomes compared with HCTZ.5,6 However, prospective randomized trial data is needed to confirm the superiority of chlorthalidone over other thiazide diuretics.
HCTZ has historically been the most common thiazide diuretic.7 However, with the available evidence and 2017 ACC/AHA BP guideline recommendations, it is unclear whether this trend continues and what impact it may have on CVD outcomes. It is unclear which thiazide diuretic is most commonly used in the US Department of Veterans Affairs (VA) health care system. The purpose of this project was to evaluate current thiazide diuretic utilization within the VA.
Methods
This retrospective, observational study evaluated the prescribing pattern of thiazide diuretics from all VA health care systems from January 1, 2016, to January 21, 2022. Thiazide diuretic agents included in this study were HCTZ, chlorthalidone, indapamide, and any combination antihypertensive products that included these 3 thiazide diuretics. Metolazone was excluded as it is commonly used in the setting of diuretic resistance with heart failure. Data was obtained from the VA Corporate Data Warehouse (CDW) and divided into 2 cohorts: the active and historic cohorts. The active cohort was of primary interest and included any active VA thiazide diuretic prescriptions on January 21, 2022. The historic cohort included thiazide prescriptions assessed at yearly intervals from January 1, 2016, to December 31, 2021. This date range was selected to assess what impact the 2017 ACC/AHA BP guideline had on clinician preferences and thiazide diuretic prescribing rates.
Within the active cohort, demographic data, vital information, and concomitant potassium or magnesium supplementation were collected. Baseline characteristics included were age, sex, race and ethnicity, and BP. Patients with > 1 race or ethnicity reported were categorized as other. The first BP reading documented after the active thiazide diuretic initiation date was included for analysis to capture on-therapy BPs while limiting confounding factors due to other potential antihypertensive changes. This project was ruled exempt from institutional review board review by the West Palm Beach VA Healthcare System Research and Development Committee.
The primary outcome was the evaluation of utilization rates of each thiazide in the active cohort, reported as a proportion of overall thiazide class utilization within the VA. Secondary outcomes in the active thiazide cohort included concomitant potassium or magnesium supplement utilization rates in each of the thiazide groups, BP values, and BP control rates. BP control was defined as a systolic BP < 130 mm Hg and a diastolic BP < 80 mm Hg. Finally, the change in thiazide diuretic utilization patterns from January 1, 2016, to December 31, 2021, was evaluated in the historic cohort.
Statistical Analysis
Data collection and analysis were completed using the CDW analyzed with Microsoft SQL Server Management Studio 18 and Microsoft Excel. All exported data to Microsoft Excel was kept in a secure network drive that was only accessible to the authors. Protected health information remained confidential per VA policy and the Health Insurance Portability and Accountability Act.
Baseline demographics were evaluated across thiazide arms using descriptive statistics. The primary outcome was assessed and a χ2 test with a single comparison α level of 0.05 with Bonferroni correction to adjust for multiple comparisons when appropriate. For the secondary outcomes, analysis of continuous data was assessed using analysis of variance (ANOVA), and nominal data were assessed with a χ2 test with a single comparison α level of 0.05 and Bonferroni correction to adjust for multiple comparisons where appropriate. When comparing all 3 thiazide groups, after the Bonferroni correction, P < .01667 was considered statistically significant to avoid a type 1 error in a family of statistical tests.
Results
As of January 21, 2022, the active thiazide cohort yielded 628,994 thiazide prescriptions within the VA nationwide. Most patients were male, with female patients representing 8.4%, 6.6%, and 5.6% of the HCTZ, chlorthalidone, and indapamide arms, respectively (Table 1). Utilization rates were significantly different between thiazide groups (P < .001). HCTZ was the most prescribed thiazide diuretic (84.6%) followed by chlorthalidone (14.9%) and indapamide (0.5%) (Table 2).
BP values documented after prescription initiation date were available for few individuals in the HCTZ, chlorthalidone, and indapamide groups (0.3%, 0.2%, and 0.5%, respectively). Overall, the mean BP values were similar among thiazide groups: 135/79 mm Hg for HCTZ, 137/78 mm Hg for chlorthalidone, and 133/79 mm Hg for indapamide (P = .32). BP control was also similar with control rates of 26.0%, 27.1%, and 33.3% for those on HCTZ, chlorthalidone, and indapamide, respectively (P = .75). The use of concomitant potassium or magnesium supplementation was significantly different between thiazide groups with rates of 12.4%, 22.6%, and 27.1% for HCTZ, chlorthalidone, and indapamide, respectively (P < .001). When comparing chlorthalidone to HCTZ, there was a significantly higher rate of concomitant supplementation with chlorthalidone (P < .001) (Table 3).
In the historic cohort, HCTZ utilization decreased from 90.2% to 83.5% (P < .001) and chlorthalidone utilization increased significantly from 9.3% to 16.0% (P < .001) (Figure). There was no significant change in the use of indapamide during this period (P = .73). Yearly trends from 2016 to 2021 are listed in Table 4.
Discussion
The findings of our evaluation demonstrate that despite the 2017 ACC/AHA BP guideline recommendations for using chlorthalidone, HCTZ predominates as the most prescribed thiazide diuretic within the VA. However, since the publication of this guideline, there has been an increase in chlorthalidone prescribing and a decrease in HCTZ prescribing within the VA.
A 2010 study by Ernst and colleagues revealed a similar trend to what was seen in our study. At that time, HCTZ was the most prescribed thiazide encompassing 95% of total thiazide utilization; however, chlorthalidone utilization increased from 1.1% in 2003 to 2.4% in 2008.8 In comparing our chlorthalidone utilization rates with these results, 9.3% in 2016 and 16.0% in 2021, the change in chlorthalidone prescribing from 2003 to 2016 represents a more than linear increase. This trend continued in our study from 2016 to 2021; the expected chlorthalidone utilization would be 21.2% in 2021 if it followed the 2003 to 2016 rate of change. Thus the trend in increasing chlorthalidone use predated the 2017 guideline recommendation. Nonetheless, this change in the thiazide prescribing pattern represents a positive shift in practice.
Our evaluation found a significantly higher rate of concomitant potassium or magnesium supplementation with chlorthalidone and indapamide compared with HCTZ in the active cohort. Electrolyte abnormalities are well documented adverse effects associated with thiazide diuretic use.9 A cross-sectional analysis by Ravioli and colleagues revealed thiazide diuretic use was an independent predictor of both hyponatremia (22.1% incidence) and hypokalemia (19% incidence) and that chlorthalidone was associated with the highest risk of electrolyte abnormalities whereas HCTZ was associated with the lowest risk. Their study also found these electrolyte abnormalities to have a dose-dependent relationship with the thiazide diuretic prescribed.10
While Ravioli and colleagues did not address the incidence of hypomagnesemia with thiazide diuretic use, a cross-sectional analysis by Kieboom and colleagues reported a significant increase in hypomagnesemia in patients prescribed thiazide diuretics.11 Although rates of electrolyte abnormalities are reported in the literature, the rates of concomitant supplementation are unclear, especially when compared across thiazide agents. Our study provides insight into the use of concomitant potassium and magnesium supplementation compared between HCTZ, chlorthalidone, and indapamide. In our active cohort, potassium was more commonly prescribed than magnesium. Interestingly, magnesium supplementation accounted for 25.9% of the total supplement use for HCTZ compared with rates of 22.4% and 21.0% for chlorthalidone and indapamide, respectively. It is unclear if this trend highlights a greater incidence of hypomagnesemia with HCTZ or greater clinician awareness to monitor this agent, but this finding may warrant further investigation. In addition, when considering the overall lower rate of supplementation seen with HCTZ in our study, the use of potassium-sparing diuretics should be considered. These agents, including triamterene, amiloride, eplerenone, and spironolactone, can be supplement-sparing and are available in combination products only with HCTZ.
Low chlorthalidone utilization rates are concerning especially given the literature demonstrating CVD benefit with chlorthalidone and the lack of compelling outcomes data to support HCTZ as the preferred agent.3,4 There are several reasons why HCTZ use may be higher in practice. First is clinical inertia, which is defined as a lack of treatment intensification or lack of changing practice patterns, despite evidence-based goals of care.12 HCTZ has been the most widely prescribed thiazide diuretic for years.7 As a result, converting HCTZ to chlorthalidone for a patient with suboptimal BP control may not be considered and instead clinicians may add on another antihypertensive or titrate doses of current antihypertensives.
There is also a consideration for patient adherence. HCTZ has many more combination products available than chlorthalidone and indapamide. If switching a patient from an HCTZ-containing combination product to chlorthalidone, adherence and patient willingness to take another capsule or tablet must be considered. Finally, there may be clinical controversy and questions around switching patients from HCTZ to chlorthalidone. Although the guidelines do not explicitly recommend switching to chlorthalidone, it may be reasonable in most patients unless they have or are at significant risk of electrolyte or metabolic disturbances that may be exacerbated or triggered with conversion.
When converting from HCTZ to chlorthalidone, it is important to consider dosing. Previous studies have demonstrated that chlorthalidone is 1.5 to 2 times more potent than HCTZ.13,14 Therefore, the conversion from HCTZ to chlorthalidone is not 1:1, but instead 50 mg of HCTZ is approximately equal to25 to 37.5 mg of chlorthalidone.14
Limitations
This study was limited by its retrospective design, gaps in data, duplicate active prescription data, and the assessment of concomitant electrolyte supplementation. As with any retrospective study, there is a potential for confounding and a concern for information bias with missing information. This study relied on proper documentation of prescription and demographic information in the Veterans Health Information Systems and Technology Architecture (VistA), as the CDW compiles information from this electronic health record. Strengths of the VistA include ease in clinical functions, documentation, and the ability for records to be updated from any VA facility nationally. However, there is always the possibility of user error and information to be omitted.
In our study, the documentation of BP values and subsequent analysis of overall BP control were limited. For BP values to be included in this study, they had to be recorded after the active thiazide prescription was written and from an in-person encounter documented in VistA. The COVID-19 pandemic shifted the clinical landscape and many primary care appointments during the active cohort evaluation period were conducted virtually. Therefore, patients may not have had formal vitals recorded. There may also be an aspect of selection bias regarding the chlorthalidone group. Although rates of thiazide switching were not assessed, some patients may have been switched from HCTZ or indapamide to chlorthalidone to achieve additional BP control. Thus, patients receiving chlorthalidone may represent a more difficult-to-control hypertensive population, making a finding of similar BP control rates between HCTZ and chlorthalidone an actual positive finding regarding chlorthalidone. Finally, this study did not assess adherence to medications. As the intent of the study was to analyze prescribing patterns, it is impossible to know if the patient was actively taking the medication at the time of assessment. When considering the rates of BP control, there were limited BP values, a potential for selection bias, and neither adherence nor patient self-reported home BP values were assessed. Therefore, the interpretation of overall BP control must be done with caution.
Additionally, duplicate prescriptions were noted in the active cohort. Rates of duplication were 0.2%, 0.08%, and 0.09% for HCTZ, chlorthalidone, and indapamide, respectively. With these small percentages, we felt this would not have a significant impact on the overall thiazide use trends seen in our study. Patients can receive prescriptions from multiple VA facilities and may have > 1 active prescriptions. This has been mitigated in recent years with the introduction of the OneVA program, allowing pharmacists to access any prescription on file from any VA facility and refill if needed (except controlled substance prescriptions). However, there are certain instances in which duplicate prescriptions may be necessary. These include patients enrolled and receiving care at another VA facility (eg, traveling for part of a year) and patients hospitalized at a different facility and given medications on discharge.
With the overall low rate of duplication prescriptions seen in each thiazide group, we determined that this was not large enough to cause substantial variation in the results of this evaluation and was unlikely to alter the results. This study also does not inform on the incidence of switching between thiazide diuretics. If a patient was switched from HCTZ to chlorthalidone in 2017, for example, a prescription for HCTZ and chlorthalidone would have been reported in this study. We felt that the change in chlorthalidone prescribing from January 1, 2016, to December 31, 2021, would reflect overall utilization rates, which may include switching from HCTZ or indapamide to chlorthalidone in addition to new chlorthalidone prescriptions.
Finally, there are confounders and trends in concomitant potassium or magnesium supplementation that were not accounted for in our study. These include concomitant loop diuretics or other medications that may cause electrolyte abnormalities and the dose-dependent relationship between thiazide diuretics and electrolyte abnormalities.10 Actual laboratory values were not included in this analysis and thus we cannot assess whether supplementation or management of electrolyte disturbances was clinically appropriate.
Conclusions
Thiazide utilization patterns have shifted possibly due to the 2017 ACC/AHA BP guideline recommendations. However, HCTZ continues to be the most widely prescribed thiazide diuretic within the VA. There is a need for future projects and clinician education to increase the implementation of guideline-recommended therapy within the VA, particularly regarding chlorthalidone use.
Hypertension is one of the most common cardiovascular disease (CVD) states, affecting nearly half of all adults in the United States.1 Numerous classes of antihypertensives are available for blood pressure (BP) management, including thiazide diuretics, which contain both thiazide and thiazide-like agents. Thiazide diuretics available in the US include hydrochlorothiazide (HCTZ), chlorthalidone, metolazone, and indapamide. These agents are commonly used and recommended as first-line treatment in the current 2017 American College of Cardiology/American Heart Association (ACC/AHA) guideline for the prevention, detection, evaluation, and management of high BP in adults.2
The ACC/AHA guideline recommends chlorthalidone as the preferred thiazide diuretic.2 This recommendation is based on its prolonged half-life compared with other thiazide agents, as well as the reduction of CVD seen with chlorthalidone in previous trials. The main evidence supporting chlorthalidone use comes from the ALLHAT trial, which compared chlorthalidone, amlodipine, and lisinopril in patients with hypertension. The primary composite outcome of fatal coronary artery disease or nonfatal myocardial infarction was not significantly different between groups. However, when looking at the incidence of heart failure, chlorthalidone was superior to both amlodipine and lisinopril.3 In the TOMHS trial, chlorthalidone was more effective in reducing left ventricular hypertrophy than amlodipine, enalapril, doxazosin, or acebutolol.4 Furthermore, both a systematic review and a retrospective cohort analysis suggested that chlorthalidone may be associated with improved CVD outcomes compared with HCTZ.5,6 However, prospective randomized trial data is needed to confirm the superiority of chlorthalidone over other thiazide diuretics.
HCTZ has historically been the most common thiazide diuretic.7 However, with the available evidence and 2017 ACC/AHA BP guideline recommendations, it is unclear whether this trend continues and what impact it may have on CVD outcomes. It is unclear which thiazide diuretic is most commonly used in the US Department of Veterans Affairs (VA) health care system. The purpose of this project was to evaluate current thiazide diuretic utilization within the VA.
Methods
This retrospective, observational study evaluated the prescribing pattern of thiazide diuretics from all VA health care systems from January 1, 2016, to January 21, 2022. Thiazide diuretic agents included in this study were HCTZ, chlorthalidone, indapamide, and any combination antihypertensive products that included these 3 thiazide diuretics. Metolazone was excluded as it is commonly used in the setting of diuretic resistance with heart failure. Data was obtained from the VA Corporate Data Warehouse (CDW) and divided into 2 cohorts: the active and historic cohorts. The active cohort was of primary interest and included any active VA thiazide diuretic prescriptions on January 21, 2022. The historic cohort included thiazide prescriptions assessed at yearly intervals from January 1, 2016, to December 31, 2021. This date range was selected to assess what impact the 2017 ACC/AHA BP guideline had on clinician preferences and thiazide diuretic prescribing rates.
Within the active cohort, demographic data, vital information, and concomitant potassium or magnesium supplementation were collected. Baseline characteristics included were age, sex, race and ethnicity, and BP. Patients with > 1 race or ethnicity reported were categorized as other. The first BP reading documented after the active thiazide diuretic initiation date was included for analysis to capture on-therapy BPs while limiting confounding factors due to other potential antihypertensive changes. This project was ruled exempt from institutional review board review by the West Palm Beach VA Healthcare System Research and Development Committee.
The primary outcome was the evaluation of utilization rates of each thiazide in the active cohort, reported as a proportion of overall thiazide class utilization within the VA. Secondary outcomes in the active thiazide cohort included concomitant potassium or magnesium supplement utilization rates in each of the thiazide groups, BP values, and BP control rates. BP control was defined as a systolic BP < 130 mm Hg and a diastolic BP < 80 mm Hg. Finally, the change in thiazide diuretic utilization patterns from January 1, 2016, to December 31, 2021, was evaluated in the historic cohort.
Statistical Analysis
Data collection and analysis were completed using the CDW analyzed with Microsoft SQL Server Management Studio 18 and Microsoft Excel. All exported data to Microsoft Excel was kept in a secure network drive that was only accessible to the authors. Protected health information remained confidential per VA policy and the Health Insurance Portability and Accountability Act.
Baseline demographics were evaluated across thiazide arms using descriptive statistics. The primary outcome was assessed and a χ2 test with a single comparison α level of 0.05 with Bonferroni correction to adjust for multiple comparisons when appropriate. For the secondary outcomes, analysis of continuous data was assessed using analysis of variance (ANOVA), and nominal data were assessed with a χ2 test with a single comparison α level of 0.05 and Bonferroni correction to adjust for multiple comparisons where appropriate. When comparing all 3 thiazide groups, after the Bonferroni correction, P < .01667 was considered statistically significant to avoid a type 1 error in a family of statistical tests.
Results
As of January 21, 2022, the active thiazide cohort yielded 628,994 thiazide prescriptions within the VA nationwide. Most patients were male, with female patients representing 8.4%, 6.6%, and 5.6% of the HCTZ, chlorthalidone, and indapamide arms, respectively (Table 1). Utilization rates were significantly different between thiazide groups (P < .001). HCTZ was the most prescribed thiazide diuretic (84.6%) followed by chlorthalidone (14.9%) and indapamide (0.5%) (Table 2).
BP values documented after prescription initiation date were available for few individuals in the HCTZ, chlorthalidone, and indapamide groups (0.3%, 0.2%, and 0.5%, respectively). Overall, the mean BP values were similar among thiazide groups: 135/79 mm Hg for HCTZ, 137/78 mm Hg for chlorthalidone, and 133/79 mm Hg for indapamide (P = .32). BP control was also similar with control rates of 26.0%, 27.1%, and 33.3% for those on HCTZ, chlorthalidone, and indapamide, respectively (P = .75). The use of concomitant potassium or magnesium supplementation was significantly different between thiazide groups with rates of 12.4%, 22.6%, and 27.1% for HCTZ, chlorthalidone, and indapamide, respectively (P < .001). When comparing chlorthalidone to HCTZ, there was a significantly higher rate of concomitant supplementation with chlorthalidone (P < .001) (Table 3).
In the historic cohort, HCTZ utilization decreased from 90.2% to 83.5% (P < .001) and chlorthalidone utilization increased significantly from 9.3% to 16.0% (P < .001) (Figure). There was no significant change in the use of indapamide during this period (P = .73). Yearly trends from 2016 to 2021 are listed in Table 4.
Discussion
The findings of our evaluation demonstrate that despite the 2017 ACC/AHA BP guideline recommendations for using chlorthalidone, HCTZ predominates as the most prescribed thiazide diuretic within the VA. However, since the publication of this guideline, there has been an increase in chlorthalidone prescribing and a decrease in HCTZ prescribing within the VA.
A 2010 study by Ernst and colleagues revealed a similar trend to what was seen in our study. At that time, HCTZ was the most prescribed thiazide encompassing 95% of total thiazide utilization; however, chlorthalidone utilization increased from 1.1% in 2003 to 2.4% in 2008.8 In comparing our chlorthalidone utilization rates with these results, 9.3% in 2016 and 16.0% in 2021, the change in chlorthalidone prescribing from 2003 to 2016 represents a more than linear increase. This trend continued in our study from 2016 to 2021; the expected chlorthalidone utilization would be 21.2% in 2021 if it followed the 2003 to 2016 rate of change. Thus the trend in increasing chlorthalidone use predated the 2017 guideline recommendation. Nonetheless, this change in the thiazide prescribing pattern represents a positive shift in practice.
Our evaluation found a significantly higher rate of concomitant potassium or magnesium supplementation with chlorthalidone and indapamide compared with HCTZ in the active cohort. Electrolyte abnormalities are well documented adverse effects associated with thiazide diuretic use.9 A cross-sectional analysis by Ravioli and colleagues revealed thiazide diuretic use was an independent predictor of both hyponatremia (22.1% incidence) and hypokalemia (19% incidence) and that chlorthalidone was associated with the highest risk of electrolyte abnormalities whereas HCTZ was associated with the lowest risk. Their study also found these electrolyte abnormalities to have a dose-dependent relationship with the thiazide diuretic prescribed.10
While Ravioli and colleagues did not address the incidence of hypomagnesemia with thiazide diuretic use, a cross-sectional analysis by Kieboom and colleagues reported a significant increase in hypomagnesemia in patients prescribed thiazide diuretics.11 Although rates of electrolyte abnormalities are reported in the literature, the rates of concomitant supplementation are unclear, especially when compared across thiazide agents. Our study provides insight into the use of concomitant potassium and magnesium supplementation compared between HCTZ, chlorthalidone, and indapamide. In our active cohort, potassium was more commonly prescribed than magnesium. Interestingly, magnesium supplementation accounted for 25.9% of the total supplement use for HCTZ compared with rates of 22.4% and 21.0% for chlorthalidone and indapamide, respectively. It is unclear if this trend highlights a greater incidence of hypomagnesemia with HCTZ or greater clinician awareness to monitor this agent, but this finding may warrant further investigation. In addition, when considering the overall lower rate of supplementation seen with HCTZ in our study, the use of potassium-sparing diuretics should be considered. These agents, including triamterene, amiloride, eplerenone, and spironolactone, can be supplement-sparing and are available in combination products only with HCTZ.
Low chlorthalidone utilization rates are concerning especially given the literature demonstrating CVD benefit with chlorthalidone and the lack of compelling outcomes data to support HCTZ as the preferred agent.3,4 There are several reasons why HCTZ use may be higher in practice. First is clinical inertia, which is defined as a lack of treatment intensification or lack of changing practice patterns, despite evidence-based goals of care.12 HCTZ has been the most widely prescribed thiazide diuretic for years.7 As a result, converting HCTZ to chlorthalidone for a patient with suboptimal BP control may not be considered and instead clinicians may add on another antihypertensive or titrate doses of current antihypertensives.
There is also a consideration for patient adherence. HCTZ has many more combination products available than chlorthalidone and indapamide. If switching a patient from an HCTZ-containing combination product to chlorthalidone, adherence and patient willingness to take another capsule or tablet must be considered. Finally, there may be clinical controversy and questions around switching patients from HCTZ to chlorthalidone. Although the guidelines do not explicitly recommend switching to chlorthalidone, it may be reasonable in most patients unless they have or are at significant risk of electrolyte or metabolic disturbances that may be exacerbated or triggered with conversion.
When converting from HCTZ to chlorthalidone, it is important to consider dosing. Previous studies have demonstrated that chlorthalidone is 1.5 to 2 times more potent than HCTZ.13,14 Therefore, the conversion from HCTZ to chlorthalidone is not 1:1, but instead 50 mg of HCTZ is approximately equal to25 to 37.5 mg of chlorthalidone.14
Limitations
This study was limited by its retrospective design, gaps in data, duplicate active prescription data, and the assessment of concomitant electrolyte supplementation. As with any retrospective study, there is a potential for confounding and a concern for information bias with missing information. This study relied on proper documentation of prescription and demographic information in the Veterans Health Information Systems and Technology Architecture (VistA), as the CDW compiles information from this electronic health record. Strengths of the VistA include ease in clinical functions, documentation, and the ability for records to be updated from any VA facility nationally. However, there is always the possibility of user error and information to be omitted.
In our study, the documentation of BP values and subsequent analysis of overall BP control were limited. For BP values to be included in this study, they had to be recorded after the active thiazide prescription was written and from an in-person encounter documented in VistA. The COVID-19 pandemic shifted the clinical landscape and many primary care appointments during the active cohort evaluation period were conducted virtually. Therefore, patients may not have had formal vitals recorded. There may also be an aspect of selection bias regarding the chlorthalidone group. Although rates of thiazide switching were not assessed, some patients may have been switched from HCTZ or indapamide to chlorthalidone to achieve additional BP control. Thus, patients receiving chlorthalidone may represent a more difficult-to-control hypertensive population, making a finding of similar BP control rates between HCTZ and chlorthalidone an actual positive finding regarding chlorthalidone. Finally, this study did not assess adherence to medications. As the intent of the study was to analyze prescribing patterns, it is impossible to know if the patient was actively taking the medication at the time of assessment. When considering the rates of BP control, there were limited BP values, a potential for selection bias, and neither adherence nor patient self-reported home BP values were assessed. Therefore, the interpretation of overall BP control must be done with caution.
Additionally, duplicate prescriptions were noted in the active cohort. Rates of duplication were 0.2%, 0.08%, and 0.09% for HCTZ, chlorthalidone, and indapamide, respectively. With these small percentages, we felt this would not have a significant impact on the overall thiazide use trends seen in our study. Patients can receive prescriptions from multiple VA facilities and may have > 1 active prescriptions. This has been mitigated in recent years with the introduction of the OneVA program, allowing pharmacists to access any prescription on file from any VA facility and refill if needed (except controlled substance prescriptions). However, there are certain instances in which duplicate prescriptions may be necessary. These include patients enrolled and receiving care at another VA facility (eg, traveling for part of a year) and patients hospitalized at a different facility and given medications on discharge.
With the overall low rate of duplication prescriptions seen in each thiazide group, we determined that this was not large enough to cause substantial variation in the results of this evaluation and was unlikely to alter the results. This study also does not inform on the incidence of switching between thiazide diuretics. If a patient was switched from HCTZ to chlorthalidone in 2017, for example, a prescription for HCTZ and chlorthalidone would have been reported in this study. We felt that the change in chlorthalidone prescribing from January 1, 2016, to December 31, 2021, would reflect overall utilization rates, which may include switching from HCTZ or indapamide to chlorthalidone in addition to new chlorthalidone prescriptions.
Finally, there are confounders and trends in concomitant potassium or magnesium supplementation that were not accounted for in our study. These include concomitant loop diuretics or other medications that may cause electrolyte abnormalities and the dose-dependent relationship between thiazide diuretics and electrolyte abnormalities.10 Actual laboratory values were not included in this analysis and thus we cannot assess whether supplementation or management of electrolyte disturbances was clinically appropriate.
Conclusions
Thiazide utilization patterns have shifted possibly due to the 2017 ACC/AHA BP guideline recommendations. However, HCTZ continues to be the most widely prescribed thiazide diuretic within the VA. There is a need for future projects and clinician education to increase the implementation of guideline-recommended therapy within the VA, particularly regarding chlorthalidone use.
1. Centers for Disease Control and Prevention. Hypertension cascade: hypertension prevalence, treatment and control estimates among U.S. adults aged 18 years and older applying the criteria from the American College of Cardiology and American Heart Association’s 2017 Hypertension Guideline—NHANES 2015–2018. Updated May 12, 2023. Accessed October 12, 2023. https://millionhearts.hhs.gov/data-reports/hypertension-prevalence.html
2. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. doi:10.1161/HYP.0000000000000065
3. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288(23):2981-2997. doi:10.1001/jama.288.23.2981
4. Liebson PR, Grandits GA, Dianzumba S, et al. Comparison of five antihypertensive monotherapies and placebo for change in left ventricular mass in patients receiving nutritional-hygienic therapy in the Treatment of Mild Hypertension Study (TOMHS). Circulation. 1995;91(3):698-706. doi:10.1161/01.cir.91.3.698
5. Roush GC, Holford TR, Guddati AK. Chlorthalidone compared with hydrochlorothiazide in reducing cardiovascular events: systematic review and network meta-analyses. Hypertension. 2012;59(6):1110-1117. doi:10.1161/HYPERTENSIONAHA.112.191106
6. Dorsch MP, Gillespie BW, Erickson SR, Bleske BE, Weder AB. Chlorthalidone reduces cardiovascular events compared with hydrochlorothiazide: a retrospective cohort analysis. Hypertension. 2011;57(4):689-694. doi:10.1161/HYPERTENSIONAHA.110.161505
7. Vongpatanasin W. Hydrochlorothiazide is not the most useful nor versatile thiazide diuretic. Curr Opin Cardiol. 2015;30(4):361-365. doi:10.1097/HCO.0000000000000178
8. Ernst ME, Lund BC. Renewed interest in chlorthalidone: evidence from the Veterans Health Administration. J Clin Hypertens (Greenwich). 2010;12(12):927-934. doi:10.1111/j.1751-7176.2010.00373.x
9. Greenberg A. Diuretic complications. Am J Med Sci. 2000;319(1):10-24. doi:10.1016/S0002-9629(15)40676-7
10. Ravioli S, Bahmad S, Funk GC, Schwarz C, Exadaktylos A, Lindner G. Risk of electrolyte disorders, syncope, and falls in patients taking thiazide diuretics: results of a cross-sectional study. Am J Med. 2021;134(9):1148-1154. doi:10.1016/j.amjmed.2021.04.007
11. Kieboom BCT, Zietse R, Ikram MA, Hoorn EJ, Stricker BH. Thiazide but not loop diuretics is associated with hypomagnesaemia in the general population. Pharmacoepidemiol Drug Saf. 2018;27(11):1166-1173. doi:10.1002/pds.4636
12. O’Connor PJ, Sperl-Hillen JAM, Johnson PE, et al. Clinical Inertia and Outpatient Medical Errors. In: Henriksen K, Battles JB, Marks ES, et al, editors. Advances in Patient Safety: From Research to Implementation (Volume 2: Concepts and Methodology). Rockville (MD): Agency for Healthcare Research and Quality (US); 2005. https://www.ncbi.nlm.nih.gov/books/NBK20513/
13. Carter BL, Ernst ME, Cohen JD. Hydrochlorothiazide versus chlorthalidone: evidence supporting their interchangeability. Hypertension. 2004;43(1):4-9. doi:10.1161/01.HYP.0000103632.19915.0E
14. Liang W, Ma H, Cao L, Yan W, Yang J. Comparison of thiazide-like diuretics versus thiazide-type diuretics: a meta-analysis. J Cell Mol Med. 2017;21(11):2634-2642. doi:10.1111/jcmm.13205
1. Centers for Disease Control and Prevention. Hypertension cascade: hypertension prevalence, treatment and control estimates among U.S. adults aged 18 years and older applying the criteria from the American College of Cardiology and American Heart Association’s 2017 Hypertension Guideline—NHANES 2015–2018. Updated May 12, 2023. Accessed October 12, 2023. https://millionhearts.hhs.gov/data-reports/hypertension-prevalence.html
2. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. doi:10.1161/HYP.0000000000000065
3. ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288(23):2981-2997. doi:10.1001/jama.288.23.2981
4. Liebson PR, Grandits GA, Dianzumba S, et al. Comparison of five antihypertensive monotherapies and placebo for change in left ventricular mass in patients receiving nutritional-hygienic therapy in the Treatment of Mild Hypertension Study (TOMHS). Circulation. 1995;91(3):698-706. doi:10.1161/01.cir.91.3.698
5. Roush GC, Holford TR, Guddati AK. Chlorthalidone compared with hydrochlorothiazide in reducing cardiovascular events: systematic review and network meta-analyses. Hypertension. 2012;59(6):1110-1117. doi:10.1161/HYPERTENSIONAHA.112.191106
6. Dorsch MP, Gillespie BW, Erickson SR, Bleske BE, Weder AB. Chlorthalidone reduces cardiovascular events compared with hydrochlorothiazide: a retrospective cohort analysis. Hypertension. 2011;57(4):689-694. doi:10.1161/HYPERTENSIONAHA.110.161505
7. Vongpatanasin W. Hydrochlorothiazide is not the most useful nor versatile thiazide diuretic. Curr Opin Cardiol. 2015;30(4):361-365. doi:10.1097/HCO.0000000000000178
8. Ernst ME, Lund BC. Renewed interest in chlorthalidone: evidence from the Veterans Health Administration. J Clin Hypertens (Greenwich). 2010;12(12):927-934. doi:10.1111/j.1751-7176.2010.00373.x
9. Greenberg A. Diuretic complications. Am J Med Sci. 2000;319(1):10-24. doi:10.1016/S0002-9629(15)40676-7
10. Ravioli S, Bahmad S, Funk GC, Schwarz C, Exadaktylos A, Lindner G. Risk of electrolyte disorders, syncope, and falls in patients taking thiazide diuretics: results of a cross-sectional study. Am J Med. 2021;134(9):1148-1154. doi:10.1016/j.amjmed.2021.04.007
11. Kieboom BCT, Zietse R, Ikram MA, Hoorn EJ, Stricker BH. Thiazide but not loop diuretics is associated with hypomagnesaemia in the general population. Pharmacoepidemiol Drug Saf. 2018;27(11):1166-1173. doi:10.1002/pds.4636
12. O’Connor PJ, Sperl-Hillen JAM, Johnson PE, et al. Clinical Inertia and Outpatient Medical Errors. In: Henriksen K, Battles JB, Marks ES, et al, editors. Advances in Patient Safety: From Research to Implementation (Volume 2: Concepts and Methodology). Rockville (MD): Agency for Healthcare Research and Quality (US); 2005. https://www.ncbi.nlm.nih.gov/books/NBK20513/
13. Carter BL, Ernst ME, Cohen JD. Hydrochlorothiazide versus chlorthalidone: evidence supporting their interchangeability. Hypertension. 2004;43(1):4-9. doi:10.1161/01.HYP.0000103632.19915.0E
14. Liang W, Ma H, Cao L, Yan W, Yang J. Comparison of thiazide-like diuretics versus thiazide-type diuretics: a meta-analysis. J Cell Mol Med. 2017;21(11):2634-2642. doi:10.1111/jcmm.13205
Evaluation of Interventions by Clinical Pharmacy Specialists in Cardiology at a VA Ambulatory Cardiology Clinic
Integration of CPSs into an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for other cardiology health care providers.
Health care providers face many challenges in utilizing cardiovascular therapies, such as anticipated shortages in physicians, patients with more complicated conditions, shifting medication regimens, management needs, and increased accountability for quality and performance measures.1 To meet the potential increase in service demand, cardiology practices are embracing cardiovascular team-based care.1 Advanced practice providers, such as advanced practice registered nurses (APRNs), physician assistants (PAs), and clinical pharmacy specialists (CPSs), have education, training, and experience to extend the team’s capability to meet these complex management needs.1
The role of CPSs within a cardiovascular care team includes providing a variety of patient-specific services, such as collaborating with other cardiology providers, to optimize evidence-based pharmacotherapy, preventing medication-related adverse events/errors, improving patient understanding of their medication regimen, and ultimately, improving patient outcomes.2 Health care systems, such as Kaiser Permanente of Colorado, have demonstrated improved clinical outcomes for patients with coronary artery disease (CAD) by implementing a multidisciplinary collaborative cardiac care service, including a clinical pharmacy cardiac risk service, in which CPSs assisted with management of cholesterol-lowering, hypertension, diabetes mellitus (DM), and smoking-cessation therapies, which resulted in a 76% to 89% reduction in all-cause mortality associated with CAD in multiple evaluations.3,4
Pharmacists providing medication therapy management (MTM) services in Minnesota had higher goal attainment for patients with hypertension and hyperlipidemia than did pharmacists who did not provide MTM services.5 MTM services provided by pharmacists led to an improvement in clinical outcomes for patients as well as a reduction in overall health care expenditures compared with that of a control group of patients who did not receive MTM services.5 Furthermore, CPS integration in the heart failure (HF) setting has led to improvements in utilization and optimization of guideline-directed medical therapies, an area in which recent data have suggested deficiencies exist.6-8 A full review of the outcomes associated with CPS involvement in cardiovascular care is beyond the scope of this article; but the recent review by Dunn and colleagues provides more detail.2
With the increasing number of patients with cardiovascular disease,expanding integration of CPSs in the cardiovascular team providing MTM services may reduce the burden of other providers (MD, PA, APRN, etc), thereby increasing access for not only new patients, but also diagnostic and interventional work, while potentially improving clinical and economic outcomes.2 The value of integrating CPSs as members of the cardiovascular care team is recognized in a variety of inpatient and ambulatory practice settings.2-6 However, data are limited on the number and types of interventions made per encounter as direct patient care providers. Expanded granularity regarding the effect of CPSs as active members of the cardiovascular team is an essential component to evaluate the potential benefit of CPS integration into direct patient care.
Methods
The West Palm Beach (WPB) Veteran Affairs Medical Center (VAMC) outpatient cardiology clinic consists of 6 full-time employee (FTE) cardiologists, 4 PAs or APRNs, 10 other cardiology health care staff members (registered/license practical nurses and technicians), and 2 cardiology CPSs providing direct patient care
The cardiology pharmacotherapy clinic is open 20.5 hours per week with 41 appointment slots (30 minutes each), of which 7 appointments are delivered via clinic video telehealth and 34 appointments are traditional face-to-face visits.9 The remaining CPS time is assigned to other clinical care and administrative areas to fit facility need, including oversight of the CPS-run 24-hour ambulatory blood pressure clinic, postgraduate year 2 cardiology pharmacy practice residency program directorship, and other administrative activities for the facility.10
The cardiology CPSs practice under an advanced scope of practice in which they independently manage medications (initiate, modify, discontinue), order diagnostic testing (laboratory, monitoring, imaging, etc) needed for medication management, and create monitoring and treatment plans for patients referred to the cardiology pharmacotherapy clinic by other cardiology providers. The diseases managed within the clinic vary based on patient-specific needs, but may include HF, dyslipidemia, hypertension, anticoagulation, CAD, arrhythmias, cardiovascular risk factor assessment and reduction, and medication reconciliation and teaching. Patients are referred for CPS management directly from facility cardiologist and cardiology clinic PAs and APRNs. Workload and interventions carried out are captured in the Pharmacists Achieve Results with Medications Demonstration (PhARMD) tool and patient care encounter tracking.9
Data Collection
Using local data from workload tracking, the number of CPS encounters was determined from July 6, 2015, to October 1, 2015. Data were collected on the types and volume of interventions made by CPSs in the cardiology pharmacotherapy clinic using the PhARMD tool (Figure). 
The PhARMD tool was initially developed and implemented for CPSs in primary care pharmacotherapy clinics and was used to evaluate the types and volume of CPS interventions made in this setting.11 Since this initial evaluation, the tool has been updated, standardized nationally by the Department of Veterans Affairs (VA) Pharmacy Benefits Management Clinical Pharmacy Practice Office, and integrated across numerous VAMCs and associated outpatient clinics. The tool remains embedded within the VA electronic health record (EHR) and allows the capture of specific CPS interventions of several types (ie, both pharmacologic and nonpharmacologic interventions, including adjust dose or frequency; change or discontinue medication; initiate medication; monitor medication; counsel on adherence, contraindications, drug interactions, and drugs not indicated; reconcile medication; and prevent or manage adverse drug events [ADEs]) specific to certain diseases, such as anemia, anticoagulation, HF, type 2 DM (T2DM), hypertension, dyslipidemia, and tobacco cessation.
Given that the interventions captured by the PhARMD tool are based on self-report of the CPS performing the intervention, a quality assurance (QA) measure was taken to audit a random sample of interventions to validate the accuracy of reported data. A Pharmacy Benefits Management PhARMD Project QA report provided the 20% random sample of encounters for each cardiology CPS to be reviewed. This percentage was determined by VAMC Clinical Pharmacy Program Office (CPPO) directives on implementation of the PhARMD tool. During the QA period, the provided sample was reviewed to determine whether the intervention(s) recorded with the PhARMD tool matched the actions documented in the EHR. The QA review was done through a manual chart review by an author not involved in recording the original interventions. Both WPB VAMC cardiology CPSs passed the QA review (> 80% concurrence with tool logged and chart documented interventions as required by VA CPPO directive), with a 90.9% concurrence between the EHR and PhARMD tool documentation.
Statistical Analyses
Data on intervention type and encounter number were evaluated with descriptive statistics. The information was characterized and diagrammed with Excel (Microsoft, Redmond, WA) charts and graphs.
Cost-avoidance calculations were done using previously described methods and are included for exploratory analysis.11,12 Briefly, published estimates of cost avoidance associated with various interventions from the outpatient setting within a VAMC setting were applied as appropriate to the various interventions captured with the PhARMD tool.11,12 These estimates from Lee and colleagues were derived from detailed chart review of interventions made and the potential harm prevented.12 Costs or cost avoidances associated with interventions were calculated from pooled examination of 600 interventions in a VAMC with drug costs before and after the intervention, costs associated with harms prevented by the intervention, as well as the VAMC hourly pharmacist wages associated with making an intervention and processing and filling original vs recommended therapies.
The costs presented represent a “best-case” scenario in which all interventions made are expected to prevent patient harms. The costs related to avoided outcomes, facility overhead, and auxiliary staff cannot be included but highlight the many considerations that must be considered when examining potential cost-avoidance calculations. The estimates and methods at hand were chosen because, to our knowledge, no other consensus model exists that would be more appropriate for use in the situation and health care system at hand. Cost-avoidance estimates were calculated by extrapolating the 88-day study period values to a yearly estimate. All cost estimates were adjusted for inflation using the consumer price index calculator as per convention in previous analyses using the cost-avoidance estimates at hand.11-13
Results
From July 6, 2015, through October 1, 2015, 301 patient encounters occurred, and 529 interventions were documented with the PhARMD tool. The mean number of interventions per encounter was 1.8. Interventions were 65.2% pharmacologic and 34.8% nonpharmacologic. Of pharmacologic interventions, 27.1% were for HF, 12.7% for hypertension, 8.8% for dyslipidemia, 2.8% for anticoagulation, 1.4% for tobacco cessation, 1.1% for T2DM, 0.3% for anemia, and 45.8% for other conditions (Table 1). 
The main types of pharmacologic interventions across all diseases were related to adjustments in medication dose or frequency (42.3%) and change or discontinuation of medications (20.0%). 
Discussion
Evaluation of the interventions and encounters at the WPB VAMC ambulatory cardiology pharmacotherapy clinic suggests that CPSs are able to contribute to direct patient care independently of interventions performed by other cardiology providers. Specifically, 1.8 interventions per encounter were made by CPSs in this study. In a prior evaluation of CPS interventions recorded with the PhARMD tool in a VAMC primary care setting, 2.3 interventions per encounter were recorded.11 In comparing the present volume of interventions with the volume recorded in the study by Hough and colleagues, the difference in practice setting may account for differences seen.11
The primary care medication management setting would capture a broader array of clinical interventions than would the ambulatory cardiology clinic of the present study, so it is reasonable that more interventions would be captured per encounter in the primary care clinic. The difference in practice settings affecting the character of collected interventions can be seen because most interventions in this study at an ambulatory cardiology clinic were related to HF, whereas in Hough and colleagues 39.2% of the disease-specific interventions were related to DM, and only 2.9% were related to HF.11 The differences inherent in the intervention populations can also be seen by comparing the percentage of interventions related to hypertension and dyslipidemia: 30% and 28% in the study by Hough and colleagues compared with 13% and 9%, respectively, in the present study.11
Comparison of the present evaluation and Hough and colleagues is also hindered by the PhARMD tool used. The PhARMD tool used in the initial evaluation has been modified on a national level to improve the granularity of intervention data collected.
Our cost-avoidance estimate of $433,324.06 per year seems lower than that estimated in the previous evaluation when all applicable interventions were included.11 However, this study had several differences compared with those of previous VAMC studies looking at clinical interventions performed by CPSs. The main differences are the volume and setting in which interventions were being made. For example, in comparison with Hough and colleagues, the studies include different practice settings (primary care vs cardiology specialty clinic) and number of FTEs involved in the study (4.65 vs 1). If the cost avoidance is distributed evenly per FTE in the previous study, the following calculation is observed: $649,551.99 per FTE, which is closer to this study’s estimation. Given that primary care is a broader setting than is ambulatory cardiology, it is not surprising that more types of interventions and the overall volume/absolute number of interventions would be higher. Thus, the lower estimated cost avoidance in our study may be attributed to the lower volume of intervention opportunities availed to the cardiology CPS. Another difference is that detailed types of interventions related to hypertension, DM, dyslipidemia, and HF were not included in Hough and colleagues, whereas our study included all applicable interventions regardless of relation to diseases, which may account for a degree of the variation in intervention breakdown between the 2 studies.11 However, as noted previously, some interventions for these particular diseases may not fully capture the rationale for pharmacotherapy interventions, such as drug dose changes or discontinuations, which may misrepresent the potential cost avoidance associated with them in reality.
Limitations
Of general importance, the PhARMD tool may underestimate the number of interventions made such that multiple interventions for a medical condition may have been completed but only captured as 1 intervention, which may represent a limitation of the tool when multiple interventions are made for the same disease (eg, titration of both β-blocker and angiotensin-converting enzyme inhibitor doses at a single appointment in a patient with HF with reduced left ventricular ejection fraction). Improved clarity about interventions made would require laborious chart review, which was not feasible. The evaluation at hand included a preliminary QA review, adding confidence that overdocumentation was not being done and the values represented at worst an underestimation of actual CPS intervention impact. Because this study was an initial evaluation of interventions made by CPSs in an ambulatory cardiology pharmacotherapy setting, whether these same outcomes would exist in other patient cohorts is unclear. However, these data do provide a foundational understanding of what may be expected from CPS integration into a cardiovascular care team.
These findings may be limited in generalizability to other health care systems and situations in which CPSs are afforded the regulatory opportunity to practice independently within an established scope of practice or collaborative practice agreements. The Veterans Health Administration system has been a leader in integrating CPSs into direct patient care roles and serves as a potential model for application by other groups. This evaluation’s data support continued efforts to create such independent practice environments as they allow for qualified CPSs to practice to their full clinical potential and have the fullest possible effect on cardiovascular outcomes.
Previous studies looking at cost savings in MTM programs have established a substantial return in economic investment with patients being managed by pharmacists.5,14 Given that the interventions made in this study were not tied to attainment of clinical outcomes, a limitation to our study, the cost-avoidance estimates should be interpreted cautiously. However, we know of no such tool that is available to allow accurate capture of clinical event reduction in a single center with consistent CPS involvement in care. A clear opportunity exists regarding design of a model that measures clinical, economic, and humanistic outcomes related to the interventions performed by cardiology CPSs, but developing and deploying such a model may be challenging because guideline-directed medical therapies vary significantly based on many patient-specific issues, and identifying optimal or truly optimized medical therapy is at times a subjective task, especially in a single center. Using the types and volumes of interventions made by CPSs as a surrogate for these higher-level outcomes is still of value in order to understand the effect and role of CPSs in cardiovascular care. At present, the cost-avoidance estimates presented in this evaluation are based on the most appropriate system-specific data at hand, with the realization that actual cost avoidance in practice may vary widely and should be the topic of future research.
Conclusion
As cardiovascular team-based care continues to expand with the support of large organizations, such as the American College of Cardiology Foundation, Heart Failure Society of America, and American College of Clinical Pharmacy Cardiology Practice and Research Network, the need for understanding the effect of CPSs on patient care measures and health care costs becomes more pronounced.2,15 The results of this study demonstrate how integration of CPSs in an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for cardiology physicians and providers, allowing more availability for diagnostic testing and care.
Interventions made by CPSs functioning as independent providers
1. Brush JE Jr, Handberg EM, Biga C, et al. 2015 ACC health policy statement on cardiovascular team-based care and the role of advanced practice providers. J Am Coll Cardiol. 2015;65(19):2118-2136.
2. Dunn SP, Birtcher KK, Beavers CJ, et al. The role of the clinical pharmacist in the care of patients with cardiovascular disease. J Am Coll Cardiol. 2015;66(19):2129-2139.
3. Sandoff BG, Kuca S, Rasmussen J, Merenich JA. Collaborative cardiac care service: a multidisciplinary approach to caring for patients with coronary artery disease. Perm J. 2008;12(3):4-11.
4. Merenich JA, Olson KL, Delate T, Rasmussen J, Helling DK, Ward DG; Clinical Pharmacy Cardiac Risk Service Study Group. Mortality reduction benefits of a comprehensive cardiac care program for patients with occlusive coronary disease. Pharmacotherapy. 2007;27(10):1370-1378.
5. Isetts BJ, Schondelmeyer SW, Artz MB, et al. Clinical and economic outcomes of medication therapy management services: the Minnesota experience.
6. Martinez AS, Saef J, Paszcuzuk A, Bhatt-Chugani H. Implementation of a pharmacist-managed heart failure medication titration clinic. Am J Health Syst Pharm. 2013;70(12):1070-1076.
7. Roth GA, Poole JE, Zaha R, Zhou W, Skinner J, Morden NE. Use of guideline-directed medications for heart failure before cardioverter-defibrillator implantation. J Am Coll Cardiol. 2016;67(9):1062-1069.
8. Noschese LA, Bergman CL, Brar CK, Kansal MM. The pharmacist’s role in medication optimization for patients with chronic heart failure. Fed Pract. 2017;34(suppl 10):S10-S15.
9. Coakley C, Hough A, Dwyer D, Parra D. Clinical video telehealth in a cardiology pharmacotherapy clinic. Am J Health Syst Pharm. 2013;70(22):1974-1975.
10. Khazan E, Anastasia E, Hough A, Parra D. Pharmacist-managed ambulatory blood pressure monitoring service. Am J Health Syst Pharm. 2017;74(4):190-195.
11. Hough A, Vartan CM, Groppi JA, Reyes S, Beckey NP. Evaluation of clinical pharmacy interventions in a Veterans Affairs medical center primary care clinic. Am J Health Syst Pharm. 2013;70(13):1168-1172.
12. Lee AJ, Boro MS, Knapp KK, Meier JL, Korman NE. Clinical and economic outcomes of pharmacist recommendations in a Veterans Affairs medical center. Am J Health Syst Pharm. 2002;59(21):2070-2077.
13. US Department of Labor. CPI inflation calculator. www.bls.gov/data/inflation_calculator.htm. Accessed January 18, 2019.
14. Perez A, Doloresco F, Hoffman JM, et al. Economic evaluations of clinical pharmacy services: 2001-2005. Pharmacotherapy. 2008;29(1):128.
15. Milfred-LaForest SK, Chow SL, DiDomenico RJ, et al. Clinical pharmacy services in heart failure: an opinion paper from the Heart Failure Society of America and American College of Clinical Pharmacy Cardiology Practice and Research Network. Pharmacotherapy. 2013;33(5):529-548.
Integration of CPSs into an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for other cardiology health care providers.
Integration of CPSs into an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for other cardiology health care providers.
Health care providers face many challenges in utilizing cardiovascular therapies, such as anticipated shortages in physicians, patients with more complicated conditions, shifting medication regimens, management needs, and increased accountability for quality and performance measures.1 To meet the potential increase in service demand, cardiology practices are embracing cardiovascular team-based care.1 Advanced practice providers, such as advanced practice registered nurses (APRNs), physician assistants (PAs), and clinical pharmacy specialists (CPSs), have education, training, and experience to extend the team’s capability to meet these complex management needs.1
The role of CPSs within a cardiovascular care team includes providing a variety of patient-specific services, such as collaborating with other cardiology providers, to optimize evidence-based pharmacotherapy, preventing medication-related adverse events/errors, improving patient understanding of their medication regimen, and ultimately, improving patient outcomes.2 Health care systems, such as Kaiser Permanente of Colorado, have demonstrated improved clinical outcomes for patients with coronary artery disease (CAD) by implementing a multidisciplinary collaborative cardiac care service, including a clinical pharmacy cardiac risk service, in which CPSs assisted with management of cholesterol-lowering, hypertension, diabetes mellitus (DM), and smoking-cessation therapies, which resulted in a 76% to 89% reduction in all-cause mortality associated with CAD in multiple evaluations.3,4
Pharmacists providing medication therapy management (MTM) services in Minnesota had higher goal attainment for patients with hypertension and hyperlipidemia than did pharmacists who did not provide MTM services.5 MTM services provided by pharmacists led to an improvement in clinical outcomes for patients as well as a reduction in overall health care expenditures compared with that of a control group of patients who did not receive MTM services.5 Furthermore, CPS integration in the heart failure (HF) setting has led to improvements in utilization and optimization of guideline-directed medical therapies, an area in which recent data have suggested deficiencies exist.6-8 A full review of the outcomes associated with CPS involvement in cardiovascular care is beyond the scope of this article; but the recent review by Dunn and colleagues provides more detail.2
With the increasing number of patients with cardiovascular disease,expanding integration of CPSs in the cardiovascular team providing MTM services may reduce the burden of other providers (MD, PA, APRN, etc), thereby increasing access for not only new patients, but also diagnostic and interventional work, while potentially improving clinical and economic outcomes.2 The value of integrating CPSs as members of the cardiovascular care team is recognized in a variety of inpatient and ambulatory practice settings.2-6 However, data are limited on the number and types of interventions made per encounter as direct patient care providers. Expanded granularity regarding the effect of CPSs as active members of the cardiovascular team is an essential component to evaluate the potential benefit of CPS integration into direct patient care.
Methods
The West Palm Beach (WPB) Veteran Affairs Medical Center (VAMC) outpatient cardiology clinic consists of 6 full-time employee (FTE) cardiologists, 4 PAs or APRNs, 10 other cardiology health care staff members (registered/license practical nurses and technicians), and 2 cardiology CPSs providing direct patient care
The cardiology pharmacotherapy clinic is open 20.5 hours per week with 41 appointment slots (30 minutes each), of which 7 appointments are delivered via clinic video telehealth and 34 appointments are traditional face-to-face visits.9 The remaining CPS time is assigned to other clinical care and administrative areas to fit facility need, including oversight of the CPS-run 24-hour ambulatory blood pressure clinic, postgraduate year 2 cardiology pharmacy practice residency program directorship, and other administrative activities for the facility.10
The cardiology CPSs practice under an advanced scope of practice in which they independently manage medications (initiate, modify, discontinue), order diagnostic testing (laboratory, monitoring, imaging, etc) needed for medication management, and create monitoring and treatment plans for patients referred to the cardiology pharmacotherapy clinic by other cardiology providers. The diseases managed within the clinic vary based on patient-specific needs, but may include HF, dyslipidemia, hypertension, anticoagulation, CAD, arrhythmias, cardiovascular risk factor assessment and reduction, and medication reconciliation and teaching. Patients are referred for CPS management directly from facility cardiologist and cardiology clinic PAs and APRNs. Workload and interventions carried out are captured in the Pharmacists Achieve Results with Medications Demonstration (PhARMD) tool and patient care encounter tracking.9
Data Collection
Using local data from workload tracking, the number of CPS encounters was determined from July 6, 2015, to October 1, 2015. Data were collected on the types and volume of interventions made by CPSs in the cardiology pharmacotherapy clinic using the PhARMD tool (Figure).
The PhARMD tool was initially developed and implemented for CPSs in primary care pharmacotherapy clinics and was used to evaluate the types and volume of CPS interventions made in this setting.11 Since this initial evaluation, the tool has been updated, standardized nationally by the Department of Veterans Affairs (VA) Pharmacy Benefits Management Clinical Pharmacy Practice Office, and integrated across numerous VAMCs and associated outpatient clinics. The tool remains embedded within the VA electronic health record (EHR) and allows the capture of specific CPS interventions of several types (ie, both pharmacologic and nonpharmacologic interventions, including adjust dose or frequency; change or discontinue medication; initiate medication; monitor medication; counsel on adherence, contraindications, drug interactions, and drugs not indicated; reconcile medication; and prevent or manage adverse drug events [ADEs]) specific to certain diseases, such as anemia, anticoagulation, HF, type 2 DM (T2DM), hypertension, dyslipidemia, and tobacco cessation.
Given that the interventions captured by the PhARMD tool are based on self-report of the CPS performing the intervention, a quality assurance (QA) measure was taken to audit a random sample of interventions to validate the accuracy of reported data. A Pharmacy Benefits Management PhARMD Project QA report provided the 20% random sample of encounters for each cardiology CPS to be reviewed. This percentage was determined by VAMC Clinical Pharmacy Program Office (CPPO) directives on implementation of the PhARMD tool. During the QA period, the provided sample was reviewed to determine whether the intervention(s) recorded with the PhARMD tool matched the actions documented in the EHR. The QA review was done through a manual chart review by an author not involved in recording the original interventions. Both WPB VAMC cardiology CPSs passed the QA review (> 80% concurrence with tool logged and chart documented interventions as required by VA CPPO directive), with a 90.9% concurrence between the EHR and PhARMD tool documentation.
Statistical Analyses
Data on intervention type and encounter number were evaluated with descriptive statistics. The information was characterized and diagrammed with Excel (Microsoft, Redmond, WA) charts and graphs.
Cost-avoidance calculations were done using previously described methods and are included for exploratory analysis.11,12 Briefly, published estimates of cost avoidance associated with various interventions from the outpatient setting within a VAMC setting were applied as appropriate to the various interventions captured with the PhARMD tool.11,12 These estimates from Lee and colleagues were derived from detailed chart review of interventions made and the potential harm prevented.12 Costs or cost avoidances associated with interventions were calculated from pooled examination of 600 interventions in a VAMC with drug costs before and after the intervention, costs associated with harms prevented by the intervention, as well as the VAMC hourly pharmacist wages associated with making an intervention and processing and filling original vs recommended therapies.
The costs presented represent a “best-case” scenario in which all interventions made are expected to prevent patient harms. The costs related to avoided outcomes, facility overhead, and auxiliary staff cannot be included but highlight the many considerations that must be considered when examining potential cost-avoidance calculations. The estimates and methods at hand were chosen because, to our knowledge, no other consensus model exists that would be more appropriate for use in the situation and health care system at hand. Cost-avoidance estimates were calculated by extrapolating the 88-day study period values to a yearly estimate. All cost estimates were adjusted for inflation using the consumer price index calculator as per convention in previous analyses using the cost-avoidance estimates at hand.11-13
Results
From July 6, 2015, through October 1, 2015, 301 patient encounters occurred, and 529 interventions were documented with the PhARMD tool. The mean number of interventions per encounter was 1.8. Interventions were 65.2% pharmacologic and 34.8% nonpharmacologic. Of pharmacologic interventions, 27.1% were for HF, 12.7% for hypertension, 8.8% for dyslipidemia, 2.8% for anticoagulation, 1.4% for tobacco cessation, 1.1% for T2DM, 0.3% for anemia, and 45.8% for other conditions (Table 1).
The main types of pharmacologic interventions across all diseases were related to adjustments in medication dose or frequency (42.3%) and change or discontinuation of medications (20.0%).
Discussion
Evaluation of the interventions and encounters at the WPB VAMC ambulatory cardiology pharmacotherapy clinic suggests that CPSs are able to contribute to direct patient care independently of interventions performed by other cardiology providers. Specifically, 1.8 interventions per encounter were made by CPSs in this study. In a prior evaluation of CPS interventions recorded with the PhARMD tool in a VAMC primary care setting, 2.3 interventions per encounter were recorded.11 In comparing the present volume of interventions with the volume recorded in the study by Hough and colleagues, the difference in practice setting may account for differences seen.11
The primary care medication management setting would capture a broader array of clinical interventions than would the ambulatory cardiology clinic of the present study, so it is reasonable that more interventions would be captured per encounter in the primary care clinic. The difference in practice settings affecting the character of collected interventions can be seen because most interventions in this study at an ambulatory cardiology clinic were related to HF, whereas in Hough and colleagues 39.2% of the disease-specific interventions were related to DM, and only 2.9% were related to HF.11 The differences inherent in the intervention populations can also be seen by comparing the percentage of interventions related to hypertension and dyslipidemia: 30% and 28% in the study by Hough and colleagues compared with 13% and 9%, respectively, in the present study.11
Comparison of the present evaluation and Hough and colleagues is also hindered by the PhARMD tool used. The PhARMD tool used in the initial evaluation has been modified on a national level to improve the granularity of intervention data collected.
Our cost-avoidance estimate of $433,324.06 per year seems lower than that estimated in the previous evaluation when all applicable interventions were included.11 However, this study had several differences compared with those of previous VAMC studies looking at clinical interventions performed by CPSs. The main differences are the volume and setting in which interventions were being made. For example, in comparison with Hough and colleagues, the studies include different practice settings (primary care vs cardiology specialty clinic) and number of FTEs involved in the study (4.65 vs 1). If the cost avoidance is distributed evenly per FTE in the previous study, the following calculation is observed: $649,551.99 per FTE, which is closer to this study’s estimation. Given that primary care is a broader setting than is ambulatory cardiology, it is not surprising that more types of interventions and the overall volume/absolute number of interventions would be higher. Thus, the lower estimated cost avoidance in our study may be attributed to the lower volume of intervention opportunities availed to the cardiology CPS. Another difference is that detailed types of interventions related to hypertension, DM, dyslipidemia, and HF were not included in Hough and colleagues, whereas our study included all applicable interventions regardless of relation to diseases, which may account for a degree of the variation in intervention breakdown between the 2 studies.11 However, as noted previously, some interventions for these particular diseases may not fully capture the rationale for pharmacotherapy interventions, such as drug dose changes or discontinuations, which may misrepresent the potential cost avoidance associated with them in reality.
Limitations
Of general importance, the PhARMD tool may underestimate the number of interventions made such that multiple interventions for a medical condition may have been completed but only captured as 1 intervention, which may represent a limitation of the tool when multiple interventions are made for the same disease (eg, titration of both β-blocker and angiotensin-converting enzyme inhibitor doses at a single appointment in a patient with HF with reduced left ventricular ejection fraction). Improved clarity about interventions made would require laborious chart review, which was not feasible. The evaluation at hand included a preliminary QA review, adding confidence that overdocumentation was not being done and the values represented at worst an underestimation of actual CPS intervention impact. Because this study was an initial evaluation of interventions made by CPSs in an ambulatory cardiology pharmacotherapy setting, whether these same outcomes would exist in other patient cohorts is unclear. However, these data do provide a foundational understanding of what may be expected from CPS integration into a cardiovascular care team.
These findings may be limited in generalizability to other health care systems and situations in which CPSs are afforded the regulatory opportunity to practice independently within an established scope of practice or collaborative practice agreements. The Veterans Health Administration system has been a leader in integrating CPSs into direct patient care roles and serves as a potential model for application by other groups. This evaluation’s data support continued efforts to create such independent practice environments as they allow for qualified CPSs to practice to their full clinical potential and have the fullest possible effect on cardiovascular outcomes.
Previous studies looking at cost savings in MTM programs have established a substantial return in economic investment with patients being managed by pharmacists.5,14 Given that the interventions made in this study were not tied to attainment of clinical outcomes, a limitation to our study, the cost-avoidance estimates should be interpreted cautiously. However, we know of no such tool that is available to allow accurate capture of clinical event reduction in a single center with consistent CPS involvement in care. A clear opportunity exists regarding design of a model that measures clinical, economic, and humanistic outcomes related to the interventions performed by cardiology CPSs, but developing and deploying such a model may be challenging because guideline-directed medical therapies vary significantly based on many patient-specific issues, and identifying optimal or truly optimized medical therapy is at times a subjective task, especially in a single center. Using the types and volumes of interventions made by CPSs as a surrogate for these higher-level outcomes is still of value in order to understand the effect and role of CPSs in cardiovascular care. At present, the cost-avoidance estimates presented in this evaluation are based on the most appropriate system-specific data at hand, with the realization that actual cost avoidance in practice may vary widely and should be the topic of future research.
Conclusion
As cardiovascular team-based care continues to expand with the support of large organizations, such as the American College of Cardiology Foundation, Heart Failure Society of America, and American College of Clinical Pharmacy Cardiology Practice and Research Network, the need for understanding the effect of CPSs on patient care measures and health care costs becomes more pronounced.2,15 The results of this study demonstrate how integration of CPSs in an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for cardiology physicians and providers, allowing more availability for diagnostic testing and care.
Interventions made by CPSs functioning as independent providers
Health care providers face many challenges in utilizing cardiovascular therapies, such as anticipated shortages in physicians, patients with more complicated conditions, shifting medication regimens, management needs, and increased accountability for quality and performance measures.1 To meet the potential increase in service demand, cardiology practices are embracing cardiovascular team-based care.1 Advanced practice providers, such as advanced practice registered nurses (APRNs), physician assistants (PAs), and clinical pharmacy specialists (CPSs), have education, training, and experience to extend the team’s capability to meet these complex management needs.1
The role of CPSs within a cardiovascular care team includes providing a variety of patient-specific services, such as collaborating with other cardiology providers, to optimize evidence-based pharmacotherapy, preventing medication-related adverse events/errors, improving patient understanding of their medication regimen, and ultimately, improving patient outcomes.2 Health care systems, such as Kaiser Permanente of Colorado, have demonstrated improved clinical outcomes for patients with coronary artery disease (CAD) by implementing a multidisciplinary collaborative cardiac care service, including a clinical pharmacy cardiac risk service, in which CPSs assisted with management of cholesterol-lowering, hypertension, diabetes mellitus (DM), and smoking-cessation therapies, which resulted in a 76% to 89% reduction in all-cause mortality associated with CAD in multiple evaluations.3,4
Pharmacists providing medication therapy management (MTM) services in Minnesota had higher goal attainment for patients with hypertension and hyperlipidemia than did pharmacists who did not provide MTM services.5 MTM services provided by pharmacists led to an improvement in clinical outcomes for patients as well as a reduction in overall health care expenditures compared with that of a control group of patients who did not receive MTM services.5 Furthermore, CPS integration in the heart failure (HF) setting has led to improvements in utilization and optimization of guideline-directed medical therapies, an area in which recent data have suggested deficiencies exist.6-8 A full review of the outcomes associated with CPS involvement in cardiovascular care is beyond the scope of this article; but the recent review by Dunn and colleagues provides more detail.2
With the increasing number of patients with cardiovascular disease,expanding integration of CPSs in the cardiovascular team providing MTM services may reduce the burden of other providers (MD, PA, APRN, etc), thereby increasing access for not only new patients, but also diagnostic and interventional work, while potentially improving clinical and economic outcomes.2 The value of integrating CPSs as members of the cardiovascular care team is recognized in a variety of inpatient and ambulatory practice settings.2-6 However, data are limited on the number and types of interventions made per encounter as direct patient care providers. Expanded granularity regarding the effect of CPSs as active members of the cardiovascular team is an essential component to evaluate the potential benefit of CPS integration into direct patient care.
Methods
The West Palm Beach (WPB) Veteran Affairs Medical Center (VAMC) outpatient cardiology clinic consists of 6 full-time employee (FTE) cardiologists, 4 PAs or APRNs, 10 other cardiology health care staff members (registered/license practical nurses and technicians), and 2 cardiology CPSs providing direct patient care
The cardiology pharmacotherapy clinic is open 20.5 hours per week with 41 appointment slots (30 minutes each), of which 7 appointments are delivered via clinic video telehealth and 34 appointments are traditional face-to-face visits.9 The remaining CPS time is assigned to other clinical care and administrative areas to fit facility need, including oversight of the CPS-run 24-hour ambulatory blood pressure clinic, postgraduate year 2 cardiology pharmacy practice residency program directorship, and other administrative activities for the facility.10
The cardiology CPSs practice under an advanced scope of practice in which they independently manage medications (initiate, modify, discontinue), order diagnostic testing (laboratory, monitoring, imaging, etc) needed for medication management, and create monitoring and treatment plans for patients referred to the cardiology pharmacotherapy clinic by other cardiology providers. The diseases managed within the clinic vary based on patient-specific needs, but may include HF, dyslipidemia, hypertension, anticoagulation, CAD, arrhythmias, cardiovascular risk factor assessment and reduction, and medication reconciliation and teaching. Patients are referred for CPS management directly from facility cardiologist and cardiology clinic PAs and APRNs. Workload and interventions carried out are captured in the Pharmacists Achieve Results with Medications Demonstration (PhARMD) tool and patient care encounter tracking.9
Data Collection
Using local data from workload tracking, the number of CPS encounters was determined from July 6, 2015, to October 1, 2015. Data were collected on the types and volume of interventions made by CPSs in the cardiology pharmacotherapy clinic using the PhARMD tool (Figure).
The PhARMD tool was initially developed and implemented for CPSs in primary care pharmacotherapy clinics and was used to evaluate the types and volume of CPS interventions made in this setting.11 Since this initial evaluation, the tool has been updated, standardized nationally by the Department of Veterans Affairs (VA) Pharmacy Benefits Management Clinical Pharmacy Practice Office, and integrated across numerous VAMCs and associated outpatient clinics. The tool remains embedded within the VA electronic health record (EHR) and allows the capture of specific CPS interventions of several types (ie, both pharmacologic and nonpharmacologic interventions, including adjust dose or frequency; change or discontinue medication; initiate medication; monitor medication; counsel on adherence, contraindications, drug interactions, and drugs not indicated; reconcile medication; and prevent or manage adverse drug events [ADEs]) specific to certain diseases, such as anemia, anticoagulation, HF, type 2 DM (T2DM), hypertension, dyslipidemia, and tobacco cessation.
Given that the interventions captured by the PhARMD tool are based on self-report of the CPS performing the intervention, a quality assurance (QA) measure was taken to audit a random sample of interventions to validate the accuracy of reported data. A Pharmacy Benefits Management PhARMD Project QA report provided the 20% random sample of encounters for each cardiology CPS to be reviewed. This percentage was determined by VAMC Clinical Pharmacy Program Office (CPPO) directives on implementation of the PhARMD tool. During the QA period, the provided sample was reviewed to determine whether the intervention(s) recorded with the PhARMD tool matched the actions documented in the EHR. The QA review was done through a manual chart review by an author not involved in recording the original interventions. Both WPB VAMC cardiology CPSs passed the QA review (> 80% concurrence with tool logged and chart documented interventions as required by VA CPPO directive), with a 90.9% concurrence between the EHR and PhARMD tool documentation.
Statistical Analyses
Data on intervention type and encounter number were evaluated with descriptive statistics. The information was characterized and diagrammed with Excel (Microsoft, Redmond, WA) charts and graphs.
Cost-avoidance calculations were done using previously described methods and are included for exploratory analysis.11,12 Briefly, published estimates of cost avoidance associated with various interventions from the outpatient setting within a VAMC setting were applied as appropriate to the various interventions captured with the PhARMD tool.11,12 These estimates from Lee and colleagues were derived from detailed chart review of interventions made and the potential harm prevented.12 Costs or cost avoidances associated with interventions were calculated from pooled examination of 600 interventions in a VAMC with drug costs before and after the intervention, costs associated with harms prevented by the intervention, as well as the VAMC hourly pharmacist wages associated with making an intervention and processing and filling original vs recommended therapies.
The costs presented represent a “best-case” scenario in which all interventions made are expected to prevent patient harms. The costs related to avoided outcomes, facility overhead, and auxiliary staff cannot be included but highlight the many considerations that must be considered when examining potential cost-avoidance calculations. The estimates and methods at hand were chosen because, to our knowledge, no other consensus model exists that would be more appropriate for use in the situation and health care system at hand. Cost-avoidance estimates were calculated by extrapolating the 88-day study period values to a yearly estimate. All cost estimates were adjusted for inflation using the consumer price index calculator as per convention in previous analyses using the cost-avoidance estimates at hand.11-13
Results
From July 6, 2015, through October 1, 2015, 301 patient encounters occurred, and 529 interventions were documented with the PhARMD tool. The mean number of interventions per encounter was 1.8. Interventions were 65.2% pharmacologic and 34.8% nonpharmacologic. Of pharmacologic interventions, 27.1% were for HF, 12.7% for hypertension, 8.8% for dyslipidemia, 2.8% for anticoagulation, 1.4% for tobacco cessation, 1.1% for T2DM, 0.3% for anemia, and 45.8% for other conditions (Table 1).
The main types of pharmacologic interventions across all diseases were related to adjustments in medication dose or frequency (42.3%) and change or discontinuation of medications (20.0%).
Discussion
Evaluation of the interventions and encounters at the WPB VAMC ambulatory cardiology pharmacotherapy clinic suggests that CPSs are able to contribute to direct patient care independently of interventions performed by other cardiology providers. Specifically, 1.8 interventions per encounter were made by CPSs in this study. In a prior evaluation of CPS interventions recorded with the PhARMD tool in a VAMC primary care setting, 2.3 interventions per encounter were recorded.11 In comparing the present volume of interventions with the volume recorded in the study by Hough and colleagues, the difference in practice setting may account for differences seen.11
The primary care medication management setting would capture a broader array of clinical interventions than would the ambulatory cardiology clinic of the present study, so it is reasonable that more interventions would be captured per encounter in the primary care clinic. The difference in practice settings affecting the character of collected interventions can be seen because most interventions in this study at an ambulatory cardiology clinic were related to HF, whereas in Hough and colleagues 39.2% of the disease-specific interventions were related to DM, and only 2.9% were related to HF.11 The differences inherent in the intervention populations can also be seen by comparing the percentage of interventions related to hypertension and dyslipidemia: 30% and 28% in the study by Hough and colleagues compared with 13% and 9%, respectively, in the present study.11
Comparison of the present evaluation and Hough and colleagues is also hindered by the PhARMD tool used. The PhARMD tool used in the initial evaluation has been modified on a national level to improve the granularity of intervention data collected.
Our cost-avoidance estimate of $433,324.06 per year seems lower than that estimated in the previous evaluation when all applicable interventions were included.11 However, this study had several differences compared with those of previous VAMC studies looking at clinical interventions performed by CPSs. The main differences are the volume and setting in which interventions were being made. For example, in comparison with Hough and colleagues, the studies include different practice settings (primary care vs cardiology specialty clinic) and number of FTEs involved in the study (4.65 vs 1). If the cost avoidance is distributed evenly per FTE in the previous study, the following calculation is observed: $649,551.99 per FTE, which is closer to this study’s estimation. Given that primary care is a broader setting than is ambulatory cardiology, it is not surprising that more types of interventions and the overall volume/absolute number of interventions would be higher. Thus, the lower estimated cost avoidance in our study may be attributed to the lower volume of intervention opportunities availed to the cardiology CPS. Another difference is that detailed types of interventions related to hypertension, DM, dyslipidemia, and HF were not included in Hough and colleagues, whereas our study included all applicable interventions regardless of relation to diseases, which may account for a degree of the variation in intervention breakdown between the 2 studies.11 However, as noted previously, some interventions for these particular diseases may not fully capture the rationale for pharmacotherapy interventions, such as drug dose changes or discontinuations, which may misrepresent the potential cost avoidance associated with them in reality.
Limitations
Of general importance, the PhARMD tool may underestimate the number of interventions made such that multiple interventions for a medical condition may have been completed but only captured as 1 intervention, which may represent a limitation of the tool when multiple interventions are made for the same disease (eg, titration of both β-blocker and angiotensin-converting enzyme inhibitor doses at a single appointment in a patient with HF with reduced left ventricular ejection fraction). Improved clarity about interventions made would require laborious chart review, which was not feasible. The evaluation at hand included a preliminary QA review, adding confidence that overdocumentation was not being done and the values represented at worst an underestimation of actual CPS intervention impact. Because this study was an initial evaluation of interventions made by CPSs in an ambulatory cardiology pharmacotherapy setting, whether these same outcomes would exist in other patient cohorts is unclear. However, these data do provide a foundational understanding of what may be expected from CPS integration into a cardiovascular care team.
These findings may be limited in generalizability to other health care systems and situations in which CPSs are afforded the regulatory opportunity to practice independently within an established scope of practice or collaborative practice agreements. The Veterans Health Administration system has been a leader in integrating CPSs into direct patient care roles and serves as a potential model for application by other groups. This evaluation’s data support continued efforts to create such independent practice environments as they allow for qualified CPSs to practice to their full clinical potential and have the fullest possible effect on cardiovascular outcomes.
Previous studies looking at cost savings in MTM programs have established a substantial return in economic investment with patients being managed by pharmacists.5,14 Given that the interventions made in this study were not tied to attainment of clinical outcomes, a limitation to our study, the cost-avoidance estimates should be interpreted cautiously. However, we know of no such tool that is available to allow accurate capture of clinical event reduction in a single center with consistent CPS involvement in care. A clear opportunity exists regarding design of a model that measures clinical, economic, and humanistic outcomes related to the interventions performed by cardiology CPSs, but developing and deploying such a model may be challenging because guideline-directed medical therapies vary significantly based on many patient-specific issues, and identifying optimal or truly optimized medical therapy is at times a subjective task, especially in a single center. Using the types and volumes of interventions made by CPSs as a surrogate for these higher-level outcomes is still of value in order to understand the effect and role of CPSs in cardiovascular care. At present, the cost-avoidance estimates presented in this evaluation are based on the most appropriate system-specific data at hand, with the realization that actual cost avoidance in practice may vary widely and should be the topic of future research.
Conclusion
As cardiovascular team-based care continues to expand with the support of large organizations, such as the American College of Cardiology Foundation, Heart Failure Society of America, and American College of Clinical Pharmacy Cardiology Practice and Research Network, the need for understanding the effect of CPSs on patient care measures and health care costs becomes more pronounced.2,15 The results of this study demonstrate how integration of CPSs in an ambulatory cardiology clinic may translate to cost avoidance and a reduction in workload burden for cardiology physicians and providers, allowing more availability for diagnostic testing and care.
Interventions made by CPSs functioning as independent providers
1. Brush JE Jr, Handberg EM, Biga C, et al. 2015 ACC health policy statement on cardiovascular team-based care and the role of advanced practice providers. J Am Coll Cardiol. 2015;65(19):2118-2136.
2. Dunn SP, Birtcher KK, Beavers CJ, et al. The role of the clinical pharmacist in the care of patients with cardiovascular disease. J Am Coll Cardiol. 2015;66(19):2129-2139.
3. Sandoff BG, Kuca S, Rasmussen J, Merenich JA. Collaborative cardiac care service: a multidisciplinary approach to caring for patients with coronary artery disease. Perm J. 2008;12(3):4-11.
4. Merenich JA, Olson KL, Delate T, Rasmussen J, Helling DK, Ward DG; Clinical Pharmacy Cardiac Risk Service Study Group. Mortality reduction benefits of a comprehensive cardiac care program for patients with occlusive coronary disease. Pharmacotherapy. 2007;27(10):1370-1378.
5. Isetts BJ, Schondelmeyer SW, Artz MB, et al. Clinical and economic outcomes of medication therapy management services: the Minnesota experience.
6. Martinez AS, Saef J, Paszcuzuk A, Bhatt-Chugani H. Implementation of a pharmacist-managed heart failure medication titration clinic. Am J Health Syst Pharm. 2013;70(12):1070-1076.
7. Roth GA, Poole JE, Zaha R, Zhou W, Skinner J, Morden NE. Use of guideline-directed medications for heart failure before cardioverter-defibrillator implantation. J Am Coll Cardiol. 2016;67(9):1062-1069.
8. Noschese LA, Bergman CL, Brar CK, Kansal MM. The pharmacist’s role in medication optimization for patients with chronic heart failure. Fed Pract. 2017;34(suppl 10):S10-S15.
9. Coakley C, Hough A, Dwyer D, Parra D. Clinical video telehealth in a cardiology pharmacotherapy clinic. Am J Health Syst Pharm. 2013;70(22):1974-1975.
10. Khazan E, Anastasia E, Hough A, Parra D. Pharmacist-managed ambulatory blood pressure monitoring service. Am J Health Syst Pharm. 2017;74(4):190-195.
11. Hough A, Vartan CM, Groppi JA, Reyes S, Beckey NP. Evaluation of clinical pharmacy interventions in a Veterans Affairs medical center primary care clinic. Am J Health Syst Pharm. 2013;70(13):1168-1172.
12. Lee AJ, Boro MS, Knapp KK, Meier JL, Korman NE. Clinical and economic outcomes of pharmacist recommendations in a Veterans Affairs medical center. Am J Health Syst Pharm. 2002;59(21):2070-2077.
13. US Department of Labor. CPI inflation calculator. www.bls.gov/data/inflation_calculator.htm. Accessed January 18, 2019.
14. Perez A, Doloresco F, Hoffman JM, et al. Economic evaluations of clinical pharmacy services: 2001-2005. Pharmacotherapy. 2008;29(1):128.
15. Milfred-LaForest SK, Chow SL, DiDomenico RJ, et al. Clinical pharmacy services in heart failure: an opinion paper from the Heart Failure Society of America and American College of Clinical Pharmacy Cardiology Practice and Research Network. Pharmacotherapy. 2013;33(5):529-548.
1. Brush JE Jr, Handberg EM, Biga C, et al. 2015 ACC health policy statement on cardiovascular team-based care and the role of advanced practice providers. J Am Coll Cardiol. 2015;65(19):2118-2136.
2. Dunn SP, Birtcher KK, Beavers CJ, et al. The role of the clinical pharmacist in the care of patients with cardiovascular disease. J Am Coll Cardiol. 2015;66(19):2129-2139.
3. Sandoff BG, Kuca S, Rasmussen J, Merenich JA. Collaborative cardiac care service: a multidisciplinary approach to caring for patients with coronary artery disease. Perm J. 2008;12(3):4-11.
4. Merenich JA, Olson KL, Delate T, Rasmussen J, Helling DK, Ward DG; Clinical Pharmacy Cardiac Risk Service Study Group. Mortality reduction benefits of a comprehensive cardiac care program for patients with occlusive coronary disease. Pharmacotherapy. 2007;27(10):1370-1378.
5. Isetts BJ, Schondelmeyer SW, Artz MB, et al. Clinical and economic outcomes of medication therapy management services: the Minnesota experience.
6. Martinez AS, Saef J, Paszcuzuk A, Bhatt-Chugani H. Implementation of a pharmacist-managed heart failure medication titration clinic. Am J Health Syst Pharm. 2013;70(12):1070-1076.
7. Roth GA, Poole JE, Zaha R, Zhou W, Skinner J, Morden NE. Use of guideline-directed medications for heart failure before cardioverter-defibrillator implantation. J Am Coll Cardiol. 2016;67(9):1062-1069.
8. Noschese LA, Bergman CL, Brar CK, Kansal MM. The pharmacist’s role in medication optimization for patients with chronic heart failure. Fed Pract. 2017;34(suppl 10):S10-S15.
9. Coakley C, Hough A, Dwyer D, Parra D. Clinical video telehealth in a cardiology pharmacotherapy clinic. Am J Health Syst Pharm. 2013;70(22):1974-1975.
10. Khazan E, Anastasia E, Hough A, Parra D. Pharmacist-managed ambulatory blood pressure monitoring service. Am J Health Syst Pharm. 2017;74(4):190-195.
11. Hough A, Vartan CM, Groppi JA, Reyes S, Beckey NP. Evaluation of clinical pharmacy interventions in a Veterans Affairs medical center primary care clinic. Am J Health Syst Pharm. 2013;70(13):1168-1172.
12. Lee AJ, Boro MS, Knapp KK, Meier JL, Korman NE. Clinical and economic outcomes of pharmacist recommendations in a Veterans Affairs medical center. Am J Health Syst Pharm. 2002;59(21):2070-2077.
13. US Department of Labor. CPI inflation calculator. www.bls.gov/data/inflation_calculator.htm. Accessed January 18, 2019.
14. Perez A, Doloresco F, Hoffman JM, et al. Economic evaluations of clinical pharmacy services: 2001-2005. Pharmacotherapy. 2008;29(1):128.
15. Milfred-LaForest SK, Chow SL, DiDomenico RJ, et al. Clinical pharmacy services in heart failure: an opinion paper from the Heart Failure Society of America and American College of Clinical Pharmacy Cardiology Practice and Research Network. Pharmacotherapy. 2013;33(5):529-548.