University of Southern California Dissertations and Theses

Documenting the impact of under-treatment for chronic disorders

(USC Thesis Other)

Documenting the impact of under-treatment for chronic disorders

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1
Documenting the Impact of Under-Treatment for Chronic Disorders

Xue Han
Faculty of the USC Graduate School
Doctor of Philosophy (HEALTH ECONOMICS)
University of Southern California
December 2018

2
Table of Contents
CHAPTER 1. Introduction ........................................................................................................ 4
CHAPTER 2. Primary Prevention Using Cholesterol-Lowering Medications in Patients Meeting
New Treatment Guidelines: A Retrospective Cohort Analysis ................................................. 9
Introduction .................................................................................................................................... 9
Methods ......................................................................................................................................... 10
Data sources ....................................................................................................................................... 10
Study cohort ....................................................................................................................................... 11
Exclusion criteria ................................................................................................................................ 11
Key explanatory variables .................................................................................................................. 12
Outcomes and other covariates ........................................................................................................... 12
Statistical analysis .............................................................................................................................. 13
Results ........................................................................................................................................... 13
Discussion ...................................................................................................................................... 16
Limitations ......................................................................................................................................... 17
Conclusion ..................................................................................................................................... 18
CHAPTER 3. Secondary Prevention Using Cholesterol-Lowering Medications in Patients with
Prior CVD Events: A Retrospective Cohort Analysis ................................................................ 27
Introduction .................................................................................................................................. 27
Methods ......................................................................................................................................... 28
Data sources ....................................................................................................................................... 28
Study cohort ....................................................................................................................................... 28
Inclusion and exclusion criteria .......................................................................................................... 29
Key explanatory variables and other covariates ................................................................................. 29
Outcomes ............................................................................................................................................ 30
Statistical analysis .............................................................................................................................. 30
Results ........................................................................................................................................... 31
Discussion ...................................................................................................................................... 34
Limitations ......................................................................................................................................... 34
Conclusions ................................................................................................................................... 35
CHAPTER 4. Impact of Timely Initiation of Antihypertensive Medications for Patients with
Hypertension or Elevated Blood Pressure .............................................................................. 47
Introduction .................................................................................................................................. 47
Methods ......................................................................................................................................... 48
Data sources ....................................................................................................................................... 48
Study cohort ....................................................................................................................................... 49
Exclusion criteria ................................................................................................................................ 49
Key Explanatory Variables and Other Covariates .............................................................................. 49
Outcomes ............................................................................................................................................ 50
Statistical methods .............................................................................................................................. 50
Results ........................................................................................................................................... 51
Discussion ...................................................................................................................................... 55
3
Limitations ......................................................................................................................................... 56
Conclusions ................................................................................................................................... 57
CHAPTER 5. Conclusions ........................................................................................................ 67

4
CHAPTER 1. Introduction
Chronic diseases are the leading cause of death and disability in the United States, resulting in
significant health and economics burden. It is reported that half of all adults, about 117 million
people, had at least one chronic health condition in 2012 and the number is projected to increase
by 1 percent each year through 2030. [1,2] Vogeli et al proposed that a patient with one chronic
illness has four physician visits per year, while those who had more than four chronic conditions
have average fourteen doctor office visits per year. [3] The chronic diseases increase the risk for
adverse health events, health resource utilization and health care costs. The total annual health
care expenditures are $2.7 trillion and 86% of those costs are for people with chronic and mental
health conditions. [1]
Mortality is the most common and severe consequence of chronic disorders. Premature death
reduces the labor supply and increases other economic burdens on society. Mortality risk
reduction due to interventions to control chronic diseases will reduce health care expenditures.
[4] Effective drug therapy is critical to prevent the disease progression and control complications
related to chronic disease. For example, anti-hypertensive drug therapy is the key intervention to
prevent stroke, heart failure, and coronary heart disease (CHD) in patients with hypertension.
These therapies include angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor
blockers (ARBs), calcium channel blocker (CCB) or diuretics. Patients with hyperlipidemia or
atherosclerotic cardiovascular diseases should initiate cholesterol-lowering therapy over the long
term to prevent further adverse events. There are multiple drug therapy options for
hyperlipidemia including statins, bile acid sequestrants, cholesterol absorption inhibitors, fibric
acid derivatives, niacin, omega-3 fatty acid ethyl esters.
Under-treatment with medications poses a major threat to an effective management of chronic
disease. Under-treated patients include patients who do not initiate prescribed medications in a
timely manner, thus going against medical advice [primary non-adherence], and patients who do
not adhere to treatment once initiated [secondary non-adherence]. Under-treatment because of
delayed prescribing and failure to fill prescribed therapy are difficult to differentiate using
retrospective analyses unless initial, unfilled prescriptions can be identified through electronic
5
prescribing records. Both forms of under-treatment are often ‘resolved’ only when patients
initiate treatment after the patient has experienced a related adverse clinical event. [5]
Secondary non-adherence is a well-recognized threat to the patient’s health and increased cost of
health care. [6] There is a large and growing research literature on the impact of secondary non-
adherence to prescribed therapies, [7-13] which has clearly documented the frequency and cost
of secondary nonadherence. [14] Unfortunately, much less research has been conducted
estimating the impact of patients failing to initiate treatment for chronic disease in a timely
manner, [15-17] due to the electronic prescribing data needed to identify patients who failed to
fill prescriptions written by their physicians. Even more difficult to measure is the extent to
which physicians fail to prescribe medications to treat chronic illnesses according to accepted
treatment guidelines.
The causes of primary and secondary non-adherence include a host of factors involving the
patient, treatment, payer, and health care provider. [18] Strategies designed to improve
medication adherence should reflect the nature of medical condition being treated, the
characteristics of the patient and physicians’ behavior. [19, 20] Evidence documenting the
relationship between the timing of drug therapy initiation and patient outcomes is lacking. This
evidence will guide efforts with which to encourage physician and patients to prescribe and
initiate medication in a timely manner. This dissertation focuses on estimating the impact of
delayed drug treatment and never having initiated therapy on patient outcomes and costs.
The three papers presented in this dissertation focus on prevention of cardiovascular disease
(CVD), which is the leading cause of death in the US. The total direct and indirect cost of CVD
in the United States for 2010 is estimated to be $315.4 billion. [21] Patients at high risk of
cardiovascular events due to hyperlipidemia, hypertension or diabetes are recommended for
treatment of cholesterol-lowering medications and/or anti-hypertensives. [22, 23] About one
third of US adults have hyperlipidemia and are at high risk for heart disease and stroke. Not only
patients with hyperlipidemia but also patients with diabetes and high ASCVD risk score are all
recommended for cholesterol-lowering medications in 2013 American College of Cardiology
and American Heart Association (ACC/AHA) guidelines. Approximately 29% of the U.S.
population are hypertensive [24]. In 2017, American College of Cardiology (ACC) issued new
guidelines that lowered the threshold of hypertension for recommending earlier intervention to
6
prevent cardiovascular disease and are expected to expand the size of the hypertensive
population designated for treatment.
The treatment of hyperlipidemia and hypertension is especially important in patients with co-
morbid diabetes. An estimated 19.7 million Americans had diagnosed diabetes mellitus,
representing 8.3% of the adult population in 2010. Such a large proportion of US population are
at high risk for cardiovascular events and are recommended for medications under new treatment
guidelines for both hyperlipidemia and hypertension.
This dissertation investigates the association between the timely initiation of drug treatment and
the risk of cardiovascular events for high risk patients and to document the benefits of timely
initiation of medications. Claims based administrative data from Humana and Optum will be
used in these analyses. The data include medical (inpatient, outpatient, ambulatory, ER),
pharmacy, demographics, eligibility, and laboratory test values, which are used to identify at-risk
patients using published treatment guidelines. Therapeutic regimens for hyperlipidemia and
hypertension will be investigated in separate analyses. These three dissertation papers address
the following questions:
1. The first paper investigates the impact of primary prevention using cholesterol-lowering
medications on cardiovascular events and healthcare costs for patients meeting the
treatment guidelines released by ACC/AHA in 2013.
2. The second paper examines the impact of initiating secondary prevention using
cholesterol-lowering medications on risk of CVD hospitalizations and healthcare costs in
patients with prior cardiovascular events.
3. The third paper estimates the association of timely initiation of anti-hypertensive
medications with cardiovascular events for patient with hypertension or elevated blood
pressure, who are the expanded target population for hypertension treatment under new
guidelines published by the ACC in 2017.

7
Reference
1. Centers for Disease Control and Prevention. Chronic Disease Prevention and Health
Promotion. Chronic Disease Overview. 2015.
2. Wu SY and Green A. Projection of Chronic Illness Prevalence and Cost Inflation. RAND
Corporation. 2000.
3. C. Vogeli et al. “Multiple Chronic Conditions: Prevalence, Health Consequences, and
Implications for Quality, Care Management, and Costs,” Journal of General Internal
Medicine 22, no. 3 Supp. (2007): 391–395.
4. Ross DeVol and Armen Bedroussian et al. An Unhealthy America: Economic Burden of
Chronic Disease. Milken Institute. 2007
5. Aurel O Iuga and Maura J McGuire et al. Adherence and health care costs. Risk
Management and Healthcare Policy 2014:7 35–44.
6. Beena Jimmy and Jimmy Jose. Patient Medication Adherence: Measures in Daily
Practice. Oman Med J. 2011 May; 26(3): 155–159.
7. Bender BG, Rand C. Medication non-adherence and asthma treatment cost. Curr Opin
Allergy Clin Immunol. 2004;4:191–195.
8. Haynes RB, McDonald H, Garg AX, et al. Interventions for helping patients to follow
prescriptions for medications. Cochrane Database Syst Rev. 2002;2:CD000011.
9. Mihalko SL, Brenes GA, Farmer DF, et al. Challenges and innovations in enhancing
adherence. Control Clinical Trials. 2004;25:447–457.
10. Aday LA, Begley CE, Lairson DR, et al. Evaluating the Healthcare System:
Effectiveness, Efficiency, and Equity. Chicago, IL: Health Administration Press; 2004.
11. Balkrishnan R. Predictors of medication adherence in the elderly. Clin Ther.
1998;20:764–771.
12. Ickovics JR, Meisler AW. Adherence in AIDS clinical trials: a framework for clinical
research and clinical care. J Clin Epidemiol. 1997;50:385–391.
13. Definitions, variants, and causes of nonadherence with medication: a challenge for
tailored interventions.
14. Spencer B Cherry, Joshua S. Benner, Mohamed A. Hussein et al. The Clinical and
Economic Burden of Nonadherence with Antihypertensive and Lipid-Lowering Therapy
in Hypertensive Patients. 2009. Value in Health. 12: 489-497.
15. Philip S. Wang, Patricia Berglund, Mark Olfson et al, Failure and Delay in Initial
Treatment Contact After First Onset of Mental Disorders in the National Comorbidity
Survey Replication. Arch Gen Psychiatry, 2005. 62(6):603-613.
16. R. M. G. Norman, A. K. Malla, M.B. Verdi et al. Understanding delay in treatment for
first-episode psychosis. Psychological Medicine. 2004. 34: 255-266.
17. Debra K. Moser, Laura P. Kimble, Mark J. Alberts et al. Reducing Delay in Seeking
Treatment by Patients With Acute Coronary Syndrome and Stroke.
Circulation. 2006;114:168-182.
18. Leslie R Martin, Summer L Williams, Kelly B Haskard et al. The challenge of patient
adherence. Ther Clin Risk Manag. 2005;1(3): 189–199.
8
19. RM DiMatteo, CD Sherbourne, RD Hays et al. Physicians' characteristics influence
patients' adherence to medical treatment: Results from the Medical Outcomes Study.
Health Psychology. 1993: 93-102.
20. KB Haskard Zolnierek and RM DiMatteo. Physician Communication and Patient
Adherence to Treatment: A Meta-analysis. Med Care. 2009; 47(8): 826–834.
21. American Health Association. Heart Disease and Stroke Statistics—2014 Update.
Circulation. 2014. 129(3):e28-e292.
22. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA Guideline on the
Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in
Adults: a Report of the American College of Cardiology/American Heart Association
Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25):2889-2934.
23. Paul A. James, MD; Suzanne Oparil, MD; Barry L. Carter et al. Evidence-Based
Guideline for the Management of High Blood Pressure in Adults Report From the Panel
Members Appointed to the Eighth Joint National Committee (JNC 8). 2014.
24. Center of Disease Control. Hypertension Prevalence and Control Among Adults: United
States, 2015–2016.

9
CHAPTER 2. Primary Prevention Using Cholesterol-Lowering Medications in Patients
Meeting New Treatment Guidelines: A Retrospective Cohort Analysis
Xue Han, MS,
1
D. Steven Fox, MD, MSc,
2
Michelle Chu, PharmD, CDE, BCACP,
3
J. Samantha
Dougherty, PhD,
4
Jeff McCombs, PhD
5

Introduction
Before 2013, cholesterol-lowering treatment, as defined by guidelines in the Third Adult
Treatment Panel (ATP-III), was indicated for patients who had abnormally high levels of low-
density lipoprotein (LDL), concomitant with other risk factors such as coronary heart disease,
diabetes, cigarette smoking and hypertension.
1
In 2013, the American College of Cardiology and
American Heart Association (ACC /AHA) issued new joint guidelines which recommended
cholesterol-lowering medication for primary or secondary prevention of atherosclerotic
cardiovascular disease (ASCVD) for four benefit groups:
Primary Prevention:
• Patients aged ≥21 years with low density lipoprotein cholesterol (LDL-C) levels ≥ 190
mg/dL [high-LDL group].
• Patients with either type 1 or type 2 diabetes, aged between 40-75 years, with LDL-C of
70-189 mg/dL [diabetes group].
• Patients with a 10-year ASCVD risk score ≥ 7.5%. (Note that patients falling into this
risk group were not included in our analytic sample, since insufficient data (e.g., on
smoking and blood pressure) were available in the paid claims database to calculate risk
scores).
Secondary Prevention:
• Patients aged ≥21 years with clinical evidence of atherosclerotic cardiovascular diseases.
1

The 2013 ACC/AHA guidelines were derived by synthesizing meta-analyses, cohort studies and
randomized controlled trials.
2
Those studies demonstrated that primary and secondary
prevention with cholesterol-lowering treatment, specifically statins, may reduce the incidence of
atherosclerotic cardiovascular events in patients previously diagnosed with cardiovascular
diseases (CVD), with diabetes, with a high LDL-C level, or an elevated ASCVD risk score.
Unsurprisingly, subsequent research has reported that the 2013 ACC/AHA cholesterol treatment
10
guidelines have significantly increased the number of high-intensity statin users.
3-7
Guideline
implementation may also improve patient outcomes by discouraging patients from delaying
treatment until after a cardiac or hospitalization event occurs.
8-14
However, other studies have
found that the overall rates of primary and secondary prevention using cholesterol-lowering
medication among the new benefit groups did not change.
15-17
Furthermore, some studies
indicated that the new guidelines’ risk-calculator may systematically overestimate ASCVD risk.
Those latter results raise questions whether these newly identified primary prevention patients
are truly “at risk”, and thus require treatment.
18-20
More research is needed using real world data
to clarify that debate, specifically by documenting whether additional clinical benefits can be
expected from the expanded primary prevention guidelines.
21-22

The objective of this study was to estimate the impact of initiating primary prevention treatment
with cholesterol-lowering medications on both the risk of CVD events, and on healthcare costs.
Two groups designated for primary prevention under the 2013 guidelines were included in this
analysis: patients with elevated LDL-C level, or diabetes. An additional analysis on the effects
of initiating secondary prevention treatment was undertaken separately. This study used
historical data that included both patients who never initiated cholesterol-lowering medication
for primary prevention, as well as patients who delayed therapy until after experiencing a CVD
event. Both groups of patients, neither of which initiated primary prevention using cholesterol
lowering medications, are needed to measure the potential benefits of initiating cholesterol-
lowering therapy for primary prevention, after reclassification under the 2013 Guidelines.
Methods
Data sources
This was a retrospective cohort analysis using data derived from the Humana claims database
covering January 1, 2007 to June 30, 2013, the seven-year period immediately prior to the
publication of the 2013 ACC/AHA guidelines. The Humana claims database is a national and
longitudinal database covering commercial and Medicare Advantage enrollees. The data include
medical claims, pharmacy claims and lab results for a subsample of individuals. Those data can
be used to derive relatively complete diagnostic and drug utilization profiles, fairly accurately
reflecting the patient’s health status at specific points in time.
11
Study cohort
The study population all met the 2013 ACC/AHA eligibility guidelines for primary prevention
using cholesterol-lowering medications, based on their diagnostic and laboratory lab value data.
Two primary prevention cohorts specified in the 2013 guidelines were identified and a third
hybrid ‘benefit’ group was created for patients who fit both risk group definitions:
• High-LDL group: Patients who had ≥1 lab claim of LDL-C, with LDL-C levels ≥190
mg/dL.
• Diabetes group: Patients aged between 40-75 years and with ≥1 medical claim with
ICD-9-CM diagnosis codes of diabetes.
• LDL-diabetes group: Patients with both elevated LDL-C and diabetes diagnosis within 9
months.
The earliest date at which the patient was identifiable as high-risk due to either elevated LDL-C
or diabetes was assigned as the patient’s index date.
Exclusion criteria
The study excluded those subjects with less than 12 months of continuous health plan enrollment
after the index date or less than 6 months of continuous enrollment before the index date. This
allowed both lead-in time to develop variables correlated with the patient’s health status at
baseline, and an at least one-year follow-up period to measure key outcome variables. Patients
with a major CVD event before the index date were defined as secondary prevention patients,
and were therefore excluded from the current analysis. Patients who filled or refilled a
prescription for any treatment targeting either CVD or high cholesterol before their index date
were likewise excluded. Patients with a coronary angioplasty procedure or coronary artery
bypass grafting (CABG) within 180 days following their index date were also excluded under the
assumption that patients with these procedures within this short period should more properly be
classified as secondary prevention candidates. Finally, patients younger than 21 years old were
excluded from the analysis.
12
Key explanatory variables
The purpose of this study was to estimate the effectiveness of initiating cholesterol-lowering
medications in primary prevention. Patients were defined as receiving primary prevention
treatment if a cholesterol-lowering medication was initiated before any post-index CVD event.
Primary prevention treatment was entered as a dichotomous variable in regressions. The
comparison group consisted of individuals identified as candidates for primary prevention but
not filling a cholesterol lowering prescription at all (or only following their first CVD event).
All available cholesterol-lowering drug therapies were used to determine if primary prevention
treatment was initiated. This included statins, bile acid sequestrants, cholesterol absorption
inhibitors, fibric acid derivatives, niacin, omega-3 fatty acid ethyl esters to define patient
treatment status. Additionally, the more specific effect of statin therapy was also analyzed as a
sensitivity analysis in which statin and non-statin medications are entered separately into the
multivariable analyses.
Outcomes and other covariates
This analysis investigated two primary outcomes: 1) The occurrence of CVD-related events [i.e.,
acute myocardial infarction (AMI), stroke, coronary angioplasty and CABG] at any time
following the index date, and 2) All-cause healthcare costs over the first year following the index
date. Cardiovascular events were selected because they reflect the expected consequences of
unmanaged hyperlipidemia. Health care costs were measured over the first year following the
patient’s index date and were specified by type of service [medical, pharmacy and total]. The
covariates used in the statistical models included age, gender, race, region, benefit group, health
plan type, types of cholesterol-lowering drug therapy [statin and non-statin], diagnostic history
[infection, endocrine disorders, diseases of blood, mental disorders, nervous system,
hypertension, circulatory system, respiratory system, digestive system, genitourinary system
anomaly], prescription drug use history [anti-obesity, anti-inflammatory, painkiller, antiviral,
anti-asthmatics, anti-seizure, antidepressants, antidiabetic, antihypertensive, antiparkinson,
antihistamines, antimetabolism, anticoagulants, hypnotic, ophthalmic, antithyroid, antiulcer and
anticholinergic], prior hospital admission, and health care cost at baseline, all as measured over
the 6 months prior to the patient’s index date.
13
Statistical analysis
All of the descriptive statistics comparing patient’s baseline characteristics across treatment
status used either the Pearson’s chi-squared test or a t-test. The association between primary
prevention and the risk of cardiovascular events was analyzed using Cox proportional hazards
models. Costs for medical, pharmacy and total health care services were analyzed using
generalized linear model [GLM] procedures. Three sensitivity analyses were also run to test the
robustness of results. The first sensitivity analysis estimated the effects of primary prevention
treatment using statin and non-statin drugs separately. The second sensitivity analysis added
dummy variables into the primary risk analyses for the AMI and stroke models which
specifically indicated the occurrence of CABG or coronary angioplasty procedures before the
occurrences of any other CVD event, on the assumption that these procedures are correlated with
the patient’s baseline severity of illness. The third sensitivity analysis reintroduced into the
analysis all of the patients with CABG or coronary angioplasty procedures within 180 day
following their index date. Analyses were performed using the SAS System for Unix, Version
9.3.
Results
Selection of patients for the study is outlined in Figure 1. It is difficult to project the extent to
which the new 2013 Guidelines expanded the total population of patients designated for primary
prevention, relative to prior guidelines. For example, all of the ‘high-LDL’ group found in Figure
1 meet the LDL requirements for primary prevention as specified under the ATP III guidelines.
However, this high-LDL group in Figure 1 may be under-estimated, since LDL-C is intended to
identify patients who likely have familial hypercholesterolemia, which cannot be identified here
(due to lack of genetic screening data). Conversely, diabetic patients were designated for
primary prevention in the 2013 ACC/AHA guidelines based solely on that diagnosis and age,
without necessarily having other CVD risk factors [family history of CVD, hypertension,
smoking, dyslipidemia, or albuminuria] as required under the prior ADA guidelines. Therefore,
the 2013 Guidelines likely increased the number of patients with diabetes designated for primary
prevention.
14
Data were available which somewhat clarify this issue: In our sample, 52.5% of diabetic patients
have at least one other CVD risk factor identifiable [i.e., prior history of hypertension,
dyslipidemia, or albuminuria] which would have designated these patients for treatment under
previous ADA guidelines. This proportion would likely increase if we had data on smoking
status. Even if 60% of the diabetes group would have been designated for treatment under older
guidelines, our results imply that full implementation of the 2013 Guidelines would significantly
increase the number of diabetic patients eligible for primary prevention using cholesterol
lowering drugs. The data in Figure 1 found that the high-LDL risk group experienced
significantly higher rates of primary prevention treatment during 2007-2013 than did the diabetes
group, 65% versus 35%, respectively. This low treatment rate in the diabetes group likely
reflects that the new 2013 guidelines expanded primary prevention to diabetic patients who were
not designated for primary prevention during the data period 2007-2013.
The unadjusted comparison of patient characteristics at baseline between patients with and
without primary prevention treatment, are shown in Table 1. Most patients were older than 65
years in each treatment group but patients tended to be clustered in the 65-74 age group. Men
were less likely than women to initiate primary prevention treatment, as were patients in the
diabetes benefit group. Racial and geographic differences were small, although in this study
most patients were Caucasian and from the Southern U.S.. Additionally, most patients who
initiated primary prevention treatment were enrolled in a HMO (Health Maintenance
Organization), whereas patients without primary prevention were covered by a PPO (Preferred
Provider Organization) insurance plan. Primary prevention patients also displayed lower rates of
prior hospitalization and lower costs over the 6-month period prior to being indexed as high risk.
Table 2 shows the frequency of patients’ initial prescription of cholesterol-lowering medications
by individual drugs. It is not surprising that 88% of patients initiated treatment using statins as
their cholesterol-lowering medication of choice. Half of those patients were specifically
prescribed simvastatin. In addition, patients treated with non-statins used fibronates (6.74%) as
their preferred drug for primary prevention treatment.
Table 3 presents unadjusted data for the study outcomes, stratified by primary prevention status.
The probability of having a stroke was significantly lower among patients initiating primary
prevention, whereas primary prevention patients did not exhibit a significantly different
15
unadjusted risk of AMI. Conversely, primary prevention patients had higher rates of coronary
angioplasty and CABG relative to patients without primary prevention, even after excluding
patients undergoing these procedures within 180 days following their index date. However,
primary prevention patients did experience a longer average delay between their index date and
an event date/end of data than did patients without primary prevention. The average time to
event ranged from 1130 to 1180 days for the four outcome events in the primary prevention
group, versus 900 to 980 days in the group without primary prevention. Finally, primary
prevention patients experienced lower hospital rates in the first post-index year [13%] compared
to ‘no primary prevention’ patients [16%]. Not surprisingly, the primary prevention population
experienced lower average overall costs than patients not receiving primary prevention, despite
higher pharmacy costs.
Results from the multivariable analyses, adjusting for other risk factors, are presented in Table 4.
Cox proportional hazard regression models were used to estimate hazard ratios of stroke, AMI,
coronary angioplasty and CABG, after adjusting for age, gender, race, region, benefit group,
health plan type, diagnostic history, prescription drug use history, prior hospital admission and
health service costs [top half of Table 4]. The results from Model 1 indicate that primary
prevention treatment was significantly associated with a reduced risk of stroke, AMI and
coronary angioplasty, whereas its association with CABG was insignificant. That is, initiating
cholesterol-lowering treatment before any cardiac event decreased the likelihood of stroke by
37% and AMI by 27%. Results from Model 1 also documented that the high-LDL and LDL-
diabetes risk groups both experienced a higher adjusted risk of all events, relative to patients with
diabetes only.
Additional results from three sensitivity analyses are presented as Model 2-4. Model 2
separately estimated the impact of initiating primary prevention treatment using statin versus
non-statin therapy. Unsurprisingly, statin therapies outperformed non-statin therapies with
respect to the risk reduction associated with primary prevention; however the differences
between the two drug classes was small and not statistically significant. Model 3 added dummy
variables that indicated post-index coronary angioplasty or CABG had occurred before other
CVD outcomes. These explanatory variables were used to control for possible differences in
baseline severity of illness. The association between primary prevention and the risk of stroke
16
and AMI was unchanged. Model 4 added back into the analysis those patients with a CABG or
angioplasty within 180 days following their index date (Those patients were excluded from
Model 1). The association of primary prevention with cardiovascular events, across all three risk
groups, remained consistent with the results from Model 1. Initiating primary prevention still
reduced the risk of stenting and CABG, and reduced total costs in the first year.
The relationship between primary prevention and healthcare costs was estimated using
generalized linear models [GLM], as reported in the lower half of Table 4. Primary prevention
was associated with a reduced total cost per patient in the first year following the index date of
$673, despite the prescription drug costs for these patients having increased by $219. Patients in
the high-LDL risk group were less costly to treat in the first post index year than patients in the
diabetes group, despite being at higher risk for CVD events.
Discussion
The 2013 ACC/AHA Guidelines still have significant untapped potential to improve clinical
outcomes and reduce health care costs. Several of our results are particularly worthy to note.
First, they imply that the ACC/AHA guidelines likely increased the total number of diabetic
patients meeting recommendations for primary prevention treatment with cholesterol-lowering
therapy. Although over 52% of our diabetic sample had other evidence of CVD risks, and some
additional diabetic patients in our sample may smoke,
23
we conclude that the 2013 ACC/AHA
guidelines will expand the total number of patients for whom primary prevention is
recommended. Moreover, those additional patients appear likely to benefit from primary
prevention, with better health outcomes, and lower costs.
Second, our results also document that the high-LDL group, while still benefiting significantly
from primary prevention, experienced a higher overall risk of CVD-related events, relative to the
diabetes group, whereas diabetic patients cost more to treat over the first post-index year. The
rate of primary prevention treatment in the high-LDL risk group was also only 65%, also
suggesting significant room for improvement. In our study, the primary prevention treatment
rate for the diabetes risk group was only 35%, likely due to diabetic patients with ‘normal’ LDL-
C levels having not yet been recommended for treatment under prior guidelines. However, the
finding that a significant proportion of patients with elevated ASCVD risk failed to initiate
17
primary prevention treatment is consistent with earlier studies.
9,13, 24
The 2013 Guidelines
designated additional candidates for cholesterol lowing therapy, and our results indicate that they
may indeed benefit from primary prevention treatment. Perhaps more importantly, this study
also demonstrated that primary prevention treatment was associated with both significant
reductions in CVD-event risk, and significantly lower health care costs. Those results proved
robust to sensitivity analyses that controlled for the severity of cardiac problems, even after the
index diagnosis.
In our analysis, primary prevention patients did not experience a significantly different
unadjusted risk of AMI, which conflicts with the conclusion from some clinical trials of CVD.
25-
26
However, after adjusting for patients’ other characteristics, primary prevention was associated
with a significant reduction in AMI risk, consistent with prior results. By contrast, both our
unadjusted and adjusted results demonstrated that primary prevention with cholesterol-lowering
medications can significantly reduce the risk of stroke.
27

Finally, these results likely represent an underestimate of the potential benefits of primary
prevention, since not all treated patients received statins, and the dose of statin therapy was not
assessed. Primary prevention patients were therefore likely undertreated, on average, by current
guidelines.
Limitations
Several limitations of this study require discussion. It was a retrospective cohort analysis.
Treatment selection bias may exist in any study that uses real world data to compare alternative
treatments. Patients who initiated primary prevention treatment may also differ in other
important but unmeasured ways from patients who delayed primary prevention until after an
event, or who did not initiate primary prevention treatment during the observation period.
Patients who were intolerant to cholesterol-lowering medications could also not be identified,
due to the lack of electronic medical records. The primary defense against potential treatment
selection bias, arising from the non-random initiation of primary prevention, is to more fully
document the observable characteristics of the patients in the study. This was the approach taken
here, as evidenced by the multiple baseline characteristics included in our statistical models.
Other more elegant statistical methods have also been developed such as instrumental variable
and Heckman selection regression methods. These approached were not applied here due to the
18
low probability that any unobserved factors could be a main source of the highly significant
effects estimated here for patients using primary prevention.
Another risk group, patients with high 10-year ASCVD risk scores, is also recommended for
primary prevention under 2013 guidelines. However, patients with high ASCVD risk scores
could not be identified, since smoking status and blood pressure was not reliably available in the
Humana data. Thus, that treatment group was not included in the analysis. Therefore, this study
cannot estimate the overall extent of users eligible for primary prevention according to the 2013
guidelines.
CVD is a leading cause of death in the United States, therefore, mortality is also an important
outcome that should be considered. We cannot estimate the association between the primary
prevention treatment and cardiac mortality since that mortality information was not available in
Humana data. Further analysis is also needed to document the impact of duration of therapy on
CVD outcomes and costs. For example, Grembowski, et al. found that the cumulative statin use
over time was correlated cumulatively with lower rates of cardiovascular death, stroke and MI,
but had no cumulative impact on costs.
21

Conclusion
In summary, fuller implementation of the 2013 Guidelines would expand primary prevention to a
large number of newly defined high-risk diabetes patients. Our results indicate that diabetic
patients within this expanded eligible population could benefit significantly from primary
prevention. Moreover, we found that primary prevention using cholesterol-lowering treatment
improved both high-LDL and diabetic patients’ 1-year post-index costs and clinical outcomes at
any post-index time, even without any consideration of long-term adherence to treatment. From
every stakeholder perspective: patient, payer, and policy, it seems clear that the benefits of
primary prevention justify greater efforts to treat a larger proportion of eligible patients.
Reference
1. National Cholesterol Education Program (NCEP). Third Report of the Expert Panel on
Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult
Treatment Panel III). NIH Publication. 2002; 02-5215.
2. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA Guideline on the
Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in
19
Adults: a Report of the American College of Cardiology/American Heart Association
Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25):2889-2934.
3. Ko MJ, Kim YJ, Park CM, et al. Applicability and Potential Clinical Effects of 2013
Cholesterol Guidelines on Major Cardiovascular Events. Am Heart J. 2015;170(3):598-
605.
4. Chia YC, Lim HM, Ching SM. Does Use of Pooled Cohort Risk Score Overestimate the
Use of Statin?: a Retrospective Cohort Study in a Primary Care Setting. BMC Fam Pract.
2014;15(172):1-6.
5. Pavlovic J, Greenland P, Deckers JW, et al. Comparison of ACC/AHA and ESC
Guideline Recommendations Following Trial Evidence for Statin Use in Primary
Prevention of Cardiovascular Disease Results from the Population-Based Rotterdam
Study. JAMA Cardiol. 2016;1(6):708-713.
6. Yeboah J, Sillau S, Delaney JC, et al. Implications of the New ACC/AHA Cholesterol
Guidelines for Primary Atherosclerotic Cardiovascular Disease Event Prevention in a
Multi- Ethnic Cohort: MESA. Am Heart J. 2015;169(3):387-395.
7. Maddox TM, Borden WB, Tang F et al. Implications of the 2013 ACC/AHA Cholesterol
Guidelines for Adults in Contemporary Cardiovascular Practice: Insights From the
NCDR PINNACLE Registry. Journal of the American college of cardiology.
2014:64(21):2183-2192.
8. Fonarow GC, Ballantyne CM. In-Hospital Initiation of Lipid-Lowering Therapy for
Patients with Coronary Heart Disease: the Time is Now. Circulation. 2001;103:2768-
2770.
9. Fonarow GC, French WJ, Parsons LS, Sun H, Malmgren JA. Use of lipid-lowering
medications at discharge in patients with acute myocardial infarction: data from the
National Registry of Myocardial Infarction 3. Circulation. 2001;103:38-44.
10. Sueta CA, Chowdhury M, Boccuzzi SJ, et al. Analysis of the degree of undertreatment of
hyperlipidemia and congestive heart failure secondary to coronary artery disease. Am J
Cardiol. 1999;83(9):1303-1307.
11. Jacobson TA, Griffiths GG, Varas C, Gause D, Sung JC, Ballantyne CM. Impact of
evidence-based “clinical judgment” on the number of American adults requiring lipid-
lowering drug therapy based on updated NHANES III data. Arch Intern
Med. 2000;160(9):1361-1369.
12. Pearson TA, Laurora I, Chu H, Kafonek S, et al. The Lipid Treatment Assessment Project
(L-TAP): a multicenter survey to evaluate the percentages of dyslipidemic patients
receiving lipid-lowering therapy and achieving low-density lipoprotein cholesterol
goals. Arch Intern Med. 2000;160(4):459-467.
13. Pattenden J, Watt I, Lewin RJP, Stanford N. Decision making processes in people with
symptoms of acute myocardial infarction: qualitative study. BMJ. 2002;324(1006):1-5.
14. Moser DK, Kimble LP, Alberts MJ, et al. Reducing Delay in Seeking Treatment by
Patients with Acute Coronary Syndrome and Stroke. Circulation. 2006;114:168-182.
15. Tran JN, Kao TC, Caglar T, et al. Impact of the 2013 Cholesterol Guideline on Patterns
of Lipid-Lowering Treatment in Patients with Atherosclerotic Cardiovascular Disease of
Diabetes after 1 Year. J Manag Care Spec Pharm. 2016;8(22):901-908.
16. Quek RG, Fox KM, Wang L, Li L, Gandra SR, Wong ND. A US Claims-Based Analysis
of Real-World Lipid-Lowering Treatment Patterns in Patients with High Cardiovascular
Disease Risk or a Previous Coronary Event. Am J Cardiol. 2016;117(4):495-500.
20
17. Smith SC. The 2013 ACC/AHA Guidelines on Treatment of Blood Cholesterol to
Improve Cardiovascular Outcomes: Implications for Statin Use and Dose vs. Target.
Cardiovascular Innovations and Applications. 2016: 1(4): 465-467.
18. DeFilippis AP, Young R, Carrubba CJ, et al. An analysis of calibration and
discrimination among multiple cardiovascular risk scores in a modern multiethnic
cohort. Ann Intern Med. 2015;162(4):266-275.
19. Kavousi M, Leening MJ, Nanchen D, et al. Comparison of application of the ACC/AHA
guidelines, Adult Treatment Panel III guidelines, and European Society of Cardiology
guidelines for cardiovascular disease prevention in a European cohort. JAMA.
2014;311(14):1416-1423.
20. Ridker PM, Cook NR. Statins: new American guidelines for prevention of cardiovascular
disease. Lancet. 2013;382(9907):1762-1765.
21. Grembowski D, Ralston JD, Anderson ML. Health Outcomes of Population-Based
Pharmacy Outreach to Increase Statin Use for Prevention of Cardiovascular Disease in
Patients with Diabetes. J Manag Care Spec Pharm. 2016;22(8):909-917.
22. US Preventive Services Task Force. Statin Use for the Primary Prevention of
Cardiovascular Disease in Adults: US Preventive Services Task Force Recommendation
Statement. JAMA. 2016;316(19):1997-2007.
23. Schipf S, Schmidt CO, Alte D. Smoking prevalence in Type 2 diabetes: results of the
Study of Health in Pomerania (SHIP) and the German National Health Interview and
Examination Survey (GNHIES). Diabet Med. 2009;26(8):791-7.
24. Moser DK, Kimble LP, Alberts MJ, et al. Reducing Delay in Seeking Treatment by
Patients with Acute Coronary Syndrome and Stroke. Circulation. 2006;114:168-182.
25. Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol
lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin
Survival Study (4S). The Lancet. 1994;344(8934):1383-1389.
26. Sacks FM, Pfeffer MA, Moye LA et al. The Effect of Pravastatin on Coronary Events
after Myocardial Infarction in Patients with Average Cholesterol Levels. N Engl J Med.
1996;335:1001-1009.
27. Everett BM, Glynn RJ, MacFadyen JG, et al. Rosuvastatin in the prevention of stroke
among men and women with elevated levels of C-reactive protein: justification for the
Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER).
Circulation. 2010;121:143–150.

21
Figure 1. Flow chart of patient selection based on exclusion criteria

Patient with elevated LDL-C
(N=96,478)

Patient with diabetes
(N= 1,625,838)
≥6 months pre-index & ≥1 year post-
index enrollment
(N=46,267) [48.0%]
≥6 months pre-index & ≥1 year post-
index enrollment
(N=344,295) [21.2%]

Aged ≥21
(N=46,135) [47.8%]

Aged 40-75
(N=254,721) [15.7%]

No cardiovascular diseases at baseline
(N=32,752) [33.9%]

No treatment of cardiovascular disease
or high cholesterol at baseline
(N=13,254) [13.7%]

No treatment of cardiovascular disease
or high cholesterol at baseline
(N=79,724) [8.5%]

No cardiovascular diseases at baseline
(N=194,292) [12.0%]

No angioplasty/CABG in 6 months
post
(N=13,012) [13.5%]
No angioplasty/CABG in 6 months
post
(N=78,054) [4.8%]
Only LDL risk group
(N=12,348)
Diabetes risk group
(N=77,728)
LDL&Diabetes risk group
(N=990)
Primary Prevention
(N=27,479) [35.4%]
No Primary Prevention
(N=50,249) [64.6%]
Primary Prevention
(N=776) [78.4%]
No Primary Prevention
(N=214) [21.6%]
Primary Prevention
(N=8,065) [65.3%]
No Primary Prevention
(N=4,283) [34.7%]
22
Table 1
Descriptive Demographics by Treatment Status:
Treated Before or After an Event and Never Treated

Characteristics Primary Prevention:
(N=36,320) [39.9%]
No Primary Prevention
(N=54,746) [60.1%]
P Value
Age at year of index date, [%]
21-44 years 7.3% 8.2% <0.0001
45-54 years 19.9% 20.7% 0.0062
55-64 years 22.0% 22.0% 0.9219
65-74 years 46.1% 45.0% 0.0020
Over 75 years 4.7% 4.2% <0.0001
Sex, [%]
Men 42.7% 43.7% 0.0020
Risk group, [%]
Only High-LDL 22.2% 7.8% <0.0001
Only Diabetes 75.7% 91.8% <0.0001
LDL-Diabetes 2.1% 0.4% <0.0001
Race, [%]
White 51.0% 49.6% <0.0001
Black 9.9% 9.4% 0.0084
Asian 0.6% 0.6% 0.2905
Hispanic 1.5% 1.2% 0.0002
Unknown 36.7% 39.1% <0.0001
Region, [%]
Midwest 21.8% 23.1% <0.0001
Northeast 1.2% 1.5% 0.0005
South 67.1% 64.7% <0.0001
West 8.3% 9.0% 0.0004
Unknown 1.6% 1.7% 0.3291
Health Plan, [%]
HMO 27.9% 23.6% <0.0001
PFFS 23.6% 23.7% 0.7191
POS 6.7% 6.8% 0.7375
PPO 29.1% 32.6% <0.0001
Other 12.7% 13.4% 0.0013
Prior Utilization [6 months]
Hospital Admission [%] 3.2% 5.4% <0.001
Medical Costs, $, mean [SD] 1,089 [3,235] 1,420 [4,403] <0.001
Pharmacy Costs, $, mean [SD] 563 [1,389] 646 [1,913] <0.001
TOTAL Costs, $, mean [SD] 2,128 [7,126] 3,198 [11,555] <0.001
23
Table 2
Frequency of Cholesterol-lowering Medications by Treatment Class
Cholesterol-lowering medication Percent
Statins (88.14%)
ATORVASTATIN 14.22%
FLUVASTATIN 0.10%
LOVASTATIN 6.63%
PITAVASTATIN 0.04%
PRAVASTATIN 15.00%
ROSUVASTATIN 9.12%
SIMVASTATIN 40.85%
COMBINATION 2.18%
Non-statins (11.86%)
Bile acid sequestrants
CHOLESTYRAMINE 1.25%
COLESEVELAM 1.16%
COLESTIPOL 0.24%
Fibric acid derivatives
FENOFIBRATE 4.29%
GEMFIBROZIL 2.45%
Ezetimibe 1.09%
Niacin 1.39%

24
Table 3
Patient Outcomes by Treatment Status

All risk groups
Primary Prevention:
(N=36,320) [39.9%]
No Primary Prevention
(N=54,746) [60.1%]
P value
Proportions of events, No. [%]
Stroke 8.0% 9.5% <0.0001
AMI 2.5% 2.6% 0.3223
Coronary angioplasty 2.0% 1.7% 0.0004
CABG 0.9% 0.6% <0.0001
Time to events, days, mean [SD]
Time to stroke 1,130 [569] 908 [523] <0.0001
Time to AMI 1,169 [569] 963 [524] <0.0001
Time to angioplasty 1,170 [568] 967 [523] <0.0001
Time to CABG 1,176 [569] 974 [525] <0.0001
Health Care Admission over 1st Year
Post

Hospital admission, No. [%] 13.4% 16.4% <0.0001
Health Care Cost over 1
st
Year Post,
$, Mean [SD]

Medical 6,013 [15,815] 7,865 [23,529] <0.0001
Pharmacy 1,732 [3,150] 1,598 [4,200] <0.0001
TOTAL 7,745 [16,638] 9,463 [24,540] <0.0001
25
Table 4
Summary of Estimated Effects of Primary Prevention and Risk Group Affiliation
CLINICAL OUTCOMES
Stroke AMI Coronary
angioplasty
CABG
HR 95% C.I. HR 95% C.I. HR 95% C.I. HR 95% C.I.
Cox Model 1 [N=91,066]
Primary
prevention
0.63* [0.60-0.66] 0.73* [0.66-0.79] 0.86* [0.78-0.95] 1.07 [0.92-1.26]
High-LDL
group [vs
Diabetes group]
1.38* [1.28-1.47] 1.37* [1.20-1.56] 1.58* [1.36, 1.84] 1.50* [1.17-1.92]
LDL-diabetes
group [vs
Diabetes group]
1.48* [1.23-1.79] 1.82* [1.31-2.53] 1.98* [1.37, 2.86] 2.53* [1.50-4.26]
Cox Model 2: [N=91,066] STATINS AND NON-STATINS EFFECTS ESTMATED SEPARATELY
Primary
prevention:
Statins
0.63* [0.60-0.66] 0.73* [0.66-0.80] 0.85* [0.76-0.94] 1.09 [0.93-1.29]
Primary
prevention:
Non-Statins
0.70* [0.63-0.77] 0.76* [0.63-0.92] 1.03 [0.84-1.25] 0.994 [0.71-1.39]
High-LDL
group [vs
Diabetes group]
1.38* [1.28-1.47] 1.37* [1.19-1.56] 1.59* [1.37-1.85] 1.49* [1.16-1.91]
LDL-diabetes
group [vs
Diabetes group]
1.48* [1.23-1.79] 1.81* [1.55-2.10] 1.99* [1.38-2.87] 2.52* [1.49-4.24]
Cox Model 3: Using CABG and Angioplasty Use as an Indicator of Severity of Illness at Baseline [N=91,066]
Primary
prevention
0.63* [0.60-0.66] 0.72* [0.66-0.79]
High-LDL
group [vs
Diabetes group]
1.38* [1.28- 1.47] 1.37* [1.20-1.57]
LDL-diabetes
group [vs
Diabetes group]
1.48* [1.22-1.79] 1.81* [1.30-2.52]
Cox Model 4: Analysis on All Patients [N=91,650]
Primary
prevention
0.63* [0.60-0.66] 0.68* [0.63-0.74] 0.78* [0.71-0.85] 1.00 [0.87-1.16]
High-LDL
group [vs
Diabetes group]
1.39* [1.31-1.50] 1.44* [1.27-1.63] 1.68* [1.47-1.92] 1.58* [1.28-1.96]
LDL-diabetes
group [vs
Diabetes group]
1.49* [1.24-1.80] 1.81* [1.31-2.49] 1.98* [1.41-2.78] 2.62* [1.65-4.16]
ALL-CAUSE COST OUTCOMES ONE YEAR POST
Medical Costs Drug Costs Total Costs
Estimated
Effect in
$$
95% C.I. Estimated
Effect in
$$
95% C.I. Estimated
Effect in
$$
95% C.I.
GLM Model [N=91,066]
26
(a). * represents the parameter is significant in a significance level of 0.05.

Primary
prevention
-673* [-936 to -411] 219* [170 to 267] -454* [-725 to -184]
High-LDL
group [vs
Diabetes group]
-674* [-1,064 to -284] -154* [-226 to -82] -828* [-1,230 to -427]
LDL-diabetes
group [vs
Diabetes group]
1,676* [470 to 2,881] 96 [-127 to 319] 1,772* [531 to 3,012]
27
CHAPTER 3. Secondary Prevention Using Cholesterol-Lowering Medications in Patients
with Prior CVD Events: A Retrospective Cohort Analysis
Xue Han, MS,
1
D. Steven Fox, MD, MPhil
1
, Michelle Chu, PharmD, CDE, BCACP,
2
J.
Samantha Dougherty, PhD,
3
Jeff McCombs, PhD
1

Introduction
Cardiovascular disease (CVD) is the leading cause of death in the United States, resulting in
significant health and economic burdens.
1
In 2013, the American College of Cardiology (ACC)
and the American Heart Association (AHA) updated the treatment guidelines for primary and
secondary prevention using cholesterol-lowering medications.
2
We previously estimated the
impact of primary prevention with cholesterol-lowering medications on reducing CVD events,
since treatment rates among high-risk populations remained low, even after the release of the
new guidelines.
3
We found that treating patients who meet the new 2013 guidelines for primary
prevention was associated with both a reduced CVD event risk and lower healthcare costs.
4
The
purpose of this paper is to estimate the impact of treating patients who meet guidelines for
secondary prevention with cholesterol-lowering medications.
Clinical trials have previously documented that treatment with cholesterol-lowering medications
significantly decreases cardiovascular risks, and mortality, when used for secondary prevention
in patients who had already experienced a CVD related event.
5-9
These results were reflected in
the ATP III guidelines, which defined secondary prevention patients as having elevated Low
Density lipoprotein-Cholesterol (LDL-C) levels and a previous coronary heart disease event.
10

The 2013 guidelines expanded the definition of secondary prevention to include people with any
prior diagnosis of atherosclerotic cardiovascular disease (ASCVD), even with normal LDL-C
levels.
2,5-10
This change expanded eligibility for secondary prevention to a larger group of
patients, newly defined as high risk under the recent guidelines.
Some studies have reported that secondary prevention for CVD is poorly implemented in clinical
practice.
8,11,12
Piepoli, et al, found that only a small proportion of heart attack survivors and
patients after revascularization surgery initiated secondary prevention.
13,14
Conversely, Bellows,
et al, found that commercial plan patients with ASCVD were more adherent to statin therapy
after the release of new guidelines.
15
More generally, the outcomes associated with secondary
28
prevention are not well documented for patients newly eligible under the 2013 guidelines.
16

Also, some recent research indicates that pre-treatment LDL levels do impact the cardiovascular
risk reduction benefit that statin treatment confers.
17,18
This study addresses these gaps in the
literature by documenting the likely impact of expanding secondary prevention treatment under
the 2013 guidelines.
This study used historical, pre-2013 claims-based data to identify high-risk patients who would
qualify for secondary prevention under the 2013 guidelines. We then used the observed
treatment history of those patients to estimate the impact of initiating secondary prevention
(using cholesterol-lowering medications) on the risks of both ASCVD hospitalizations and
healthcare costs.
Methods
Data sources
This was a retrospective cohort analysis using paid claims data from a large health insurance
provider in the U.S. The data covered January 1, 2007 to June 30, 2013, a six and a half-year
period prior to the 2013 ACC/AHA guidelines. The overall database covers 22 million persons
residing in all regions of the US and includes patients covered under Medicare Advantage,
Medicare Part D Medicaid, and commercial plans (HMO & PPO). Data elements include
administrative claims from medical encounters (inpatient, outpatient, ambulatory, ER), pharmacy
dispensing, demographics, enrollment eligibility, and laboratory test values (for a subsample of
members). These data can be used to derive relatively complete diagnostic and drug utilization
profiles, which reflect the patient’s health status at a given point in time.
Study cohort
Patients included in the analytic sample met the requirements for secondary prevention, as
specified in the 2013 ACC/AHA guidelines, based on their diagnostic and procedures data.
Specifically, secondary prevention patients were identified if they had ≥1 medical claim with a
Qualifying International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-
9-CM) diagnosis, or a paid claim for a CPT-4/ICD-9-DM procedure service directly related to
atherosclerotic cardiovascular disease (ASCVD). ASCVD diagnoses included: acute myocardial
infarction (AMI), unstable angina, subarachnoid and intracerebral hemorrhage, stroke, transient
29
cerebral ischemia (TIA), cerebrovascular disease, atherosclerosis, peripheral vascular disease,
arterial embolism, and arterial disorder.
11
ASCVD-related procedures included: coronary artery
bypass graft surgery (CABG), coronary angioplasty, revascularization surgery and peripheral
bypass. [Appendix 1] The earliest date at which the patient qualified for secondary prevention
was assigned as that patient’s index date.
Inclusion and exclusion criteria
Once identified, study patients were also required to have at least 6 months of continuous
enrollment before their index date, with both medical and pharmacy coverage continuing 12+
months following their index date. This allowed the analysis to develop variables correlated
with their health status at baseline, and to measure each patient’s subsequent events and costs.
Patients who filled or refilled a prescription for any treatment for high cholesterol before their
index date were excluded. Finally, patients younger than 21 years old were also excluded from
the analysis.
Key explanatory variables and other covariates
Patients were catagorized as having received secondary prevention if they initiated cholesterol-
lowering medication at any point following their index ASCVD diagnosis, but before
experiencing any subsequent ASCVD-related hospitalization. Secondary prevention was entered
as a dichotomous variable in the analysis. The comparison group consisted of individuals not
filling any cholesterol-lowering prescriptions, or filling an initial prescription after a new (post-
index) ASCVD-related hospitalization. All available cholesterol-lowering treatments were used
to define secondary prevention status (statins, bile acid sequestrants, cholesterol absorption
inhibitors, fibric acid derivatives, niacin, omega-3 fatty acid ethyl esters).
The covariates used to adjust baseline risk in the statistical models included: age, gender, race,
region, benefit group, health plan type [Medicare, Medicaid, health maintain organization
(HMO), point of service (POS), preferred provider organization (PPO)], diagnostic and
prescription histories, prior hospital admissions, and baseline health care costs (as measured over
6 months prior to the patient’s index date). The specific ASCVD diagnosis or event triggering
the index date was entered as a covariate in the analysis; a patient could have more than one such
30
‘index’ diagnosis. Baseline LDL-C levels were also collected for the sub-group of patients with
available lab results.
Outcomes
This analysis investigated two primary outcomes: 1) ASCVD-related hospitalization at any time
following the index date, and 2) healthcare costs over the first year following the index date.
ASCVD-related hospital admission was selected to capture any new ASCVD-related events
which occurred following the index date. An ASCVD-related admission was defined as a patient
having any of the following admission diagnoses or procedures: AMI, unstable angina,
hemorrhage, stroke, TIA, cerebrovascular disease, atherosclerosis, peripheral vascular disease,
arterial embolism, arterial disorder, revascularization surgery, coronary angioplasty, peripheral
bypass and CABG. Inpatient services were used to define subsequent ASCVD events in order to
more easily distinguished new events from follow-up services related just to the index event
which identified the patient as a candidate for secondary prevention. Health care costs were
measured over the first year following the patient’s index date, and were classified by type of
service [medical, pharmacy and total].
A sub-analysis, which only included patients with baseline LDL-C lab values (specified as an
independent variable in the analysis), was used to test the association between treatment effect
and baseline LDL-C. The LDL-C lab results collected closest to index date (either pre-index date
or within 5 days following index date) was assigned as patients’ baseline LDL. The LDL levels
were classified into four categories: LDL-C<101 mg/dL, 101-130 mg/dL, 131-160 mg/dL and
>160 mg/dL.
Statistical analysis
Descriptive statistics comparing patients’ baseline characteristics across treatment status used
either the Pearson’s chi-square test or Student’s t-test to identify significant differences.
19,20
The
impact of secondary prevention on the risk of cardiovascular-related hospitalization was
estimated using Cox proportional hazards models.
21
The impact of secondary prevention on the
costs for medical, pharmacy and total health care services was estimated using generalized linear
model [GLM] procedures.
22

31
Three sub-analyses were also performed on the hazard models. The first sub-analysis estimated
the effect of secondary prevention on CVD hospitalization at any time following the index,
adjusting for different baseline LDL levels. The second sub-analysis further investigated possible
differential treatment effects of secondary prevention on the risk of CVD hospitalization across
different LDL-C levels by adding an interaction term between the dummy treatment variable and
patients’ baseline LDL levels. The last sub-analysis assessed the effect of specific statin drugs
among patients who were initially treated with a statin. All analyses were performed using the
SAS System for Unix, Version 9.3.
Results
A funnel diagram outlining patient selection for the study is presented in Figure 1. Just over 1.4
million patients were identified with ASCVD, of whom 593,958 had sufficient data duration
surrounding their index date. The final sample comprised 272,899 patients eligible for
secondary prevention under the 2013 guidelines. Nearly 64% of those patients never filled any
prescriptions for a cholesterol-lowering medication, and 5% only filled a prescription after a
second (i.e., post-index) CVD-related hospitalization. This low treatment rate may be due,
partially, to some patients not meeting secondary prevention criteria under the earlier ATP III
guidelines (in place during the study period), which required that patients’ LDL-C also exceed
130 mg/dL. The low rate of treatment under the old guidelines implies that full compliance with
the new 2013 guidelines would likely increase dramatically the number of patients designated for
secondary prevention. Determining how much clinical benefit will result from the expansion in
eligibility is the purpose of the present study.
Unadjusted comparisons of baseline characteristics between patients who did and did not receive
secondary prevention treatment are shown in Table 1. Patients who initiated secondary
prevention treatment were mostly male and between 65-74 years, which likewise explains the
high Medicare coverage prevalence. Most patients qualified for secondary prevention due to a
diagnosis of AMI, unstable angina, stroke or peripheral vascular disease. Somewhat
surprisingly, patients who initiated secondary prevention treatment displayed lower unadjusted
rates of hospitalization, and lower all-cause health costs over the 6-month period prior to
becoming eligible for secondary prevention. For the subgroup patients who had baseline LDL
data available, treated patients had higher LDL-C levels relative to untreated patients, which
32
reflects that prior to 2013 patients were indicated for cholesterol-lowering medications mainly
depending on LDL-C level. Most patients who initiated treatment used statins and about 68% of
statin patients used moderate intensity statin treatments. In addition, 2.9% of treated patients
used a combination of statin and non-statin as their regimen for secondary prevention [See
Appendix Table 1].
Table 2 presents unadjusted data for the study outcomes, which must be viewed with caution
given the significant differences in the two study populations, and also possible differences in
their follow-up period durations. The unadjusted probabilities of having a CVD-related
hospitalization were lower, both in the first post-index year, or at any enrollment time, for
patients initiating secondary prevention, even though treated patients had a longer average
follow-up duration, as compared to patients without secondary prevention treatment. Conversely,
patients receiving treatment had a higher probability of experiencing non-CVD hospitalization
during any time following their index date (which may be due to their longer average follow-up
duration). Those differing times to event, and to the end of enrollment data, were accounted for
in the Cox proportional hazard regression, reported below. The secondary prevention population
also had higher medical costs, inpatient costs and pharmacy costs over the first year following
index date, versuspatients not receiving secondary prevention.
Results of multivariable analyses adjusting for the patient’s risk factors are in Table 3. Cox
models were estimated for both the likelihood of a cardiovascular-related hospitalization, and for
non-CVD hospitalization, compared between groups with/without secondary prevention. GLMs
were used to model post-index costs over 1 year. All models adjusted for age, gender, race,
region, index ASCVD event, health plan type, diagnostic history, prescription drug use history,
prior hospital admission and health service costs.
The Baseline Cox model results for hospitalization indicated that initiating secondary prevention
treatment was associated with significantly reduced risks of CVD-related hospitalization, by 33%
[H.R. 0.67]. Also, the results indicated that initiating secondary prevention treatment was not
associated with an altered risk of non-CVD hospitalization [HR is 1.00 (C.I. 0.98-1.01)].
This estimate of the impact of secondary prevention on the risk of CVD-related hospitalization
did not change significantly when re-estimated using only those patients with a known baseline
LDL level, entering the LDL level into the Cox model as a covariate. Note that the relative risk
33
of a CVD-related hospitalization increases monotonically with LDL level, while the risk of non-
CVD related hospitalizations is significantly higher in the two middle-LDL risk groups.
Since the absolute risk of a CVD-related hospitalization also varies by LDL level, the second
sensitivity analysis reported in Table 4 calculated the effect of secondary prevention in each LDL
risk group. These results found that the treatment effect is not independent of the patient’s
baseline LDL level, ranging from relative risk reductions of 26% for patients with baseline LDL
of 101-130 mg/dL, increasing to 39% for patients with a baseline LDL > 160 mg/dL - after
adjusting for other risk factors. Meanwhile, the associated absolute risk reduction increased
monotonically from 0.55% in patients with LDL <101 mg/dL to 3.21% in patients with LDL
>160 mg/dL.
The final sensitivity analysis documented the effects of various statin therapies on CVD
hospitalization risk (Table 3). This was accomplished by analyzing patients who filled statin
medications and had baseline LDL lab data [N=12,803]. The pattern of relative impacts clearly
indicated that Fluvastatin, Pitavastatin, and Pravastatin did not differ significantly in their impact
on CVD hospitalization risk, relative to Lovastatin. Atorvastatin, Rosuvastatin and Simvastatin
did not perform as well as the other statin therapies; they were associated with higher rates of
CVD hospitalization relative to Lovastatin.
The impact of secondary prevention on healthcare costs was estimated using GLM. (Table 3)
Secondary prevention was significantly associated with a higher total cost over one year of $509
[p<0.05] primarily due to increased prescription drug costs per patient of $272. Specifically,
patients with an index diagnosis of hemorrhage, coronary angioplasty, peripheral bypass and
CABG as their reason for secondary prevention eligibility were costlier to treat in the first post-
index year than patients without those diagnoses.
Finally, we estimated whether the impact of secondary prevention on CVD hospitalization risk
was uniform across all patients, depending on their index ASCVD event. The treatment effects
across different ASCVD risk groups are displayed graphically in Figure 2. The overall treatment
effect reported in Table 3 is indicated at the bottom of the figure [H.R. = 0.67]. These results
indicated that secondary prevention had a more significant protective effect on the likelihood of a
future CVD-related hospitalization among certain ASCVD risk groups. In particular, secondary
34
prevention treatment was associated with larger reductions in CVD hospitalization if patients had
already experienced unstable angina, hemorrhage, arterial disorder, CABG or revascularization.
Discussion
The criteria for identifying patients for secondary prevention specified in 2013 ACC/AHA
guidelines dropped the requirement that secondary prevention patients exhibit an LDL-C level >
130 mg/dL (drug optional for 100-129 mg/dL). Our results report that only 31% of patients
meeting new guideline criteria for secondary prevention were treated with cholesterol-lowering
medications during the pre-guideline period. This gap in treatment may be partially attributable
to the fact that 75% of at risk ASCVD patients with normal LDL-C levels did not meet treatment
recommendations under the ATP III guidelines. The new guidelines likely increased the number
of patients with ASCVD treated with cholesterol-lowering medications.
Secondary prevention was estimated to have a larger impact on reducing hospitalization risk in
the groups with unstable angina, hemorrhage, arterial disorder, or CABG. These patients should
therefore receive particular encouragement to initiate cholesterol-lowering medication as
secondary prevention to prevent future hospitalizations.
As expected, patients whose LDL-C levels were between 130-160 mg/dL, >160 mg/dL had a
higher risk of experiencing CVD-related hospitalization, compared to patients in lower LDL
level. While ASCVD patients in high LDL categories exhibited larger absolute risk reductions
for CVD hospitalization than lower LDL patients if they initiated cholesterol-lowering treatment,
secondary prevention was found to be effective across all baseline LDL levels. This appears to
reinforce recently published results from randomized controlled clinical trials, which suggested
the largest benefit from statin treatment was in higher LDL groups.17
Limitations
The non-randomized initiation of treatment may result from important but unmeasured factors
that preferentially divided patients into different treatment groups. This study did document the
existence of many significant differences in the observable characteristics of treated and
untreated patients. While our statistical analyses used this extensive set of characteristics as
35
covariates, it remains impossible to fully control for treatment selection using any non-
randomized approach.
The number of patients receiving secondary prevention may also be under-estimated due to some
missing statin claims for cash pay/$4 lists. This means that some patients in the “no secondary
prevention” group may have actually initiated cholesterol-lowering medications. Thus, the
effectiveness of treatment in reducing CVD-related hospitalization may be somewhat under-
estimated.
Sensitivity analyses were performed to clearly understand the relative risk of CVD
hospitalization and associated treatment effect across each LDL-C level group. Since lab results
reporting some patients’ LDL-C levels were not available in the data, these analyses were only
conducted in the smaller sub-population with baseline LDL-C available. The secondary
prevention effect on CVD-related hospitalization did not change in this sub-group analysis, at
least partially validating the results from the whole-sample analysis.
Patient adherence to cholesterol lowering medications was also not directly considered in this
analysis, only initial prescription fills. Future analyses should account for the patients’ long-term
adherence to medications, estimating the impact of duration of therapy, rather than just the act of
initiating treatment.
23

Conclusions
Implementation of the 2013 guidelines expanded secondary prevention treatment to all ASCVD
patients regardless of their LDL-C levels. Patients within this expanded eligible population will
benefit significantly from secondary prevention. Secondary prevention is associated with the
largest risk reduction in hospitalization for patients who had highest risk of experiencing CVD
related hospitalization. While secondary treatment benefits all patients, patients with higher
LDL levels suffer from a higher absolute risk of experiencing CVD-related hospitalization, and
also experienced larger absolute risk reductions after initiating cholesterol-lowering medications.
However, secondary prevention does show an increase in all-cause medical costs during the first
post-index year.

36

Reference
1. Benjamin EJ, Blaha MJ, Chiuve SE et al. Heart Disease and Stroke Statistics—2017 Update A
Report From the American Heart Association. Circulation. 2017;135:e146–e603.
2. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA Guideline on the Treatment
of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults: a Report of the
American College of Cardiology/American Heart Association Task Force on Practice
Guidelines. J Am Coll Cardiol. 2014;63(25):2889-2934.
3. Bucholze EM, Rodday AM, Kolor K et al. Prevalence and Predictors of Cholesterol
Screening, Awareness, and Statin Treatment Among US Adults With Familial
Hypercholesterolemia or Other Forms of Severe Dyslipidemia (1999-2014). Circulation 2018;
137(16).
4. Han X, Fox DS, McCombs J. Secondary Prevention Using Cholesterol-Lowering Medications
in Patients with Prior CVD Events: a Retrospective Cohort Analysis. Journal of Managed Care &
Specialty Pharmacy.2018.
5. L Wei, et al. Statin use in the secondary prevention of coronary heart disease in primary care:
cohort study and comparison of inclusion and outcome with patients in randomised trials. British
Medical Journal. 2005;330(7495):821-824.
6. Nakamura, M., Yamashita, T., Yajima, J. et al. Impact of early statin initiation on secondary
prevention in Japanese patients with coronary artery disease. 2011; Journal of Cardiology.
2011;57(2):172-180.
7. Fleg, JL., Forman, DE, Berra, K, et al. Secondary Prevention of Atherosclerotic
Cardiovascular Disease in Older Adults: A Scientific Statement From the American Heart
Association. Circulation.2013;128(22):2422–46.
8. Piepoli, MF., Corra, U, Adamopoulos, S. Secondary prevention in the clinical management of
patients with cardiovascular diseases. Core components, standards and outcome measures for
referral and delivery. Eur J Prev Cardiol. 2014;21(6): 664-81.
9. Mega, JL, Stitziel, NO, Smith, JG. Genetic risk, coronary heart disease events, and the clinical
benefit of statin therapy: an analysis of primary and secondary prevention trials. Lancet.
2015;385:2264-71.
10. Expert Panel on Detection, Evaluation. Executive Summary of the Third Report of the
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and
Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA.
2011;285(19):2486–97.
11. Tran JN., Kao TZ, Caglar T et al. Impact of the 2013 Cholesterol Guideline on Patterns of
Lipid-Lowering Treatment in Patients with Atherosclerotic Cardiovascular Disease or Diabetes
After 1 Year. Journal of Managed Care & Specialty Pharmacy.2016;22: (8): 901–8.
12. Bucholz EM, Rodday AM, Kolor K et al. Prevalence and Predictors of Cholesterol
Screening, Awareness, and Statin Treatment Among US Adults With Familial
37
Hypercholesterolemia or Other Forms of Severe Dyslipidemia (1999-2014). Circulation. 2018;
137:00-00.
13. Piepoli MF, Corra U, Abreu A.et al. Challenges in secondary prevention of cardiovascular
diseases A review of the current practice. International Journal of Cardiology. 2015;180:114-
119.
14. Cheng, K, Ingram, N, Keenan, J, et al. Evidence of Poor Adherence to Secondary Prevention
after Acute Coronary Syndromes: Possible Remedies through the Application of New
Technologies. BMJ. 2014;2:1-7.
15. Bellows BK, Olsen CJ, Voelker J et al. Antihyperlipidemic Medication Treatment Patterns
and Statin Adherence Among Patients with ASCVD in a Managed Care Plan After Release of
the 2013 ACC/AHA Guideline on the Treatment of Blood Cholesterol. Journal of Managed Care
& Specialty Pharmacy. 2016;22: (8): 892–900.
16. Okerson, Ted, Jeetvan Patel, Stefan DiMario, Tanya Burton, Jerald Seare, and David J.
Harrison. Effect of 2013 ACC/AHA Blood Cholesterol Guidelines on Statin Treatment Patterns
and Low-Density Lipoprotein Cholesterol in Atherosclerotic Cardiovascular Disease Patients.
Journal of the American Heart Association.2017;6(3):1-8.
17. Navarese NP, Robinson JG, Kowalewski M et al. Association between baseline LDL-C level
and total and cardiovascular mortality after LDL-C lowering. JAMA. 2018. 319(15):1566-1579.
18. Navar AM, Peterson ED. Challenges in Interpreting the Lipid-Lowering Trials Biology vs
Ecology. JAMA. 2018; 319(15):1549-1551.
19. Plackett RL. Karl Pearson and the Chi-Squared Test. International Statistical Review / Revue
Internationale de Statistique. 1983.51(1);59-72.
20. DeCoster J. Testing Group Differences using T-tests, ANOVA, and Nonparametric
Measures. Accessed November. 2006.30(2010); 202006-0.
21. Cox DR. Regression Models and Life-Tables. Journal of the Royal Statistical Society. Series
B (Methodological). 1972. 34(2);187-220.
22. Moran JL, Solomon PJ, Peisach AR. New models for old questions: generalized linear
models for cost prediction. Journal of Evaluation in Clinical Practice. 2007. 13(3); 381-389.
23. Grembowski D, Ralston JD, Anderson ML. Health Outcomes of Population-Based Pharmacy
Outreach to Increase Statin Use for Prevention of Cardiovascular Disease in Patients with
Diabetes. J Manag Care Spec Pharm. 2016;22(8):909-917.

38
Figure 1. Funnel chart of patient selection, based on exclusion criteria

Patient with clinical ASCVD
(N=1,415,594)

≥6 months pre-index & ≥1 year post-index enrollment
& both medical/pharmacy enrollee
(N=563,958) [39.8%]
No treatment of high cholesterol at baseline
(N=275,264) [19.4%]

Age>21 years old
(N=272,899) [19.3%]
Early Secondary Prevention
(N=85,343) [31.3%]
Late Secondary Prevention
(N=13,669) [5.0%]
No Secondary Prevention
(N=173,887) [63.7%]

39
Table 1
Descriptive Demographics by Treatment Status:
Treated Before CVD-related Hospitalization, versus Late/Never Treated
Characteristics Early Secondary
Prevention
(N=85,343) [31.3%]
Late/No Secondary
Prevention
(N=187,556) [68.7%]
P Value
Age at Year of Index Date, [%]
22-44 years 2.4% 5.5%

<0.0001

45-54 years 7.9% 7.9%
55-64 years 14.8% 12.0%
65-74 years 45.3% 37.8%
Over 75 years 29.6% 36.8%
Sex, [%]
Men 51.3% 46.2% <0.0001
Index Diagnosis, [%]
AMI 21.1% 11.9% <0.0001
Unstable angina 11.4% 5.9% <0.0001
Hemorrhage 1.3% 2.1% <0.0001
Stroke 28.8% 27.6% <0.0001
Transient cerebral ischemia 10.6% 10.4% 0.0558
Cerebrovascular disease 11.7% 16.4% <0.0001
Atherosclerosis 7.4% 8.7% <0.0001
Peripheral vascular disease 25.6% 32.0% <0.0001
Arterial embolism 0.5% 0.4% 0.0438
Arterial disorder 0.9% 0.9% 0.3029
CABG 3.2% 0.5% <0.0001
Coronary angioplasty 9.8% 2.0% <0.0001
Revascularization surgery 0.03% 0.01% <0.0001
Peripheral bypass 0.16% 0.17% 0.3844
Race, [%]
White 70.0% 68.9%

<0.0001

Black 10.4% 9.2%
Asian 0.5% 0.4%
Hispanic 1.8% 1.3%
Unknown 17.4% 20.2%
Region, [%]
Midwest 23.5% 29.1%

<0.0001

Northeast 1.8% 2.3%
South 64.5% 59.2%
West 9.1% 8.7%
Unknown 1.2% 0.7%
Health Plan, [%]
Medicare 85.8% 83.5%

<0.0001

Medicaid 0.4% 0.4%
HMO 3.8% 4.0%
POS 2.8% 3.1%
PPO 5.7% 5.6%
Other plan 1.6% 3.4%
40

Table 2
Unadjusted Patient Outcomes, by Treatment Status

All risk groups
Secondary
Prevention
(N=85,343) [31.3%]
No Secondary
Prevention
(N=187,556) [68.7%]
P value
Proportions of events, No. [%]
CVD-related hospitalization 16.7% 17.0% <0.0001
Non-CVD hospitalization 43.5% 40.0% <0.0001
CVD hospitalization in 1
st
year 6.8% 9.6% <0.0001
Non-CVD hospitalization in 1
st
year 22.7% 22.5% 0.3507
Time to events, days, mean [SD]
Time to CVD-related
hospitalization
1,042 [574] 830 [515] <0.0001
Time to non-CVD hospitalization 815 [586] 700 [507] <0.0001
Time to end of enrollment 1,150 [549] 943 [506] <0.0001
Health Care Cost over 1
st
Year Post, $, Mean [SD]
Medical
7,177 [11,933] 7,019 [13,211] <0.0001
Inpatient
12,071 [27,612] 9,474 [28,898] <0.0001
Pharmacy
2,262 [3,706] 1,631 [4,566] <0.0001
TOTAL 21,511 [33,386] 18,124 [35,645] <0.0001

41
Table 3
Summary of Estimated Effects of Secondary Prevention

CLINICAL OUTCOMES
Cardiovascular-related
hospitalization
Non-CVD related hospitalization
HR 95% C.I. HR 95% C.I.
Baseline Cox Model of Hospitalization Risk [N=272,899]
Secondary prevention 0.67* [0.66-0.69] 1.00 [0.98-1.01]
Change in Absolute Risk -0.26%
Adding Baseline LDL Levels for Subpopulation [N=49,224]
Secondary prevention 0.68* [0.64-0.72] 1.05* [1.01-1.08]
101-130 mg/dL (vs<101
mg/dL)
1.02 [0.96-1.08] 0.91* [0.88-0.94]
131-160 mg/dL (vs<101
mg/dL)
1.16* [1.08-1.25] 0.87* [0.83-0.92]
>160 mg/dL(vs<101) 1.45* [1.31-1.60] 0.95 [0.88-1.02]
Treatment Effect (Relative Risk) in Each LDL-C Level Group for Subpopulation [N=49,224]
LDL <101 mg/dL 0.66* [0.60-0.72] 1.04 [0.98-1.10]
Change in Absolute Risk -0.55%
LDL 101-130 mg/dL 0.74* [0.67-0.82] 1.12* [1.06-1.19]
Change in Absolute Risk 0.92%
LDL 131-160 mg/dL 0.65* [0.57-0.73] 0.97 [0.89-1.05]
Change in Absolute Risk -1.14%
LDL >160 mg/dL 0.61* [0.51-0.73] 0.94 [0.83-1.07]
Change in Absolute Risk -3.21%
Treatment Effect (Relative Risk) by Statin Class for Treated Group [N=12,803]
Comparison Group [vs. Lovastatin]
Atorvastatin 1.21* [1.10-1.32]
Fluvastatin 1.32 [0.82-2.14]
Pitavastatin 0.49 [0.12-1.97]
Pravastatin 1.04 [0.94-1.14]
Rosuvastatin 1.22* [1.10-1.35]
Simvastatin 1.20* [1.10-1.30]
ALL-CAUSE COST OUTCOMES ONE YEAR POST
Medical+Inpatient
Costs
Drug Costs Total Costs
Effect
in $$
95% C.I.
Effect
in $$
95% C.I.
Effect
in $$
95% C.I.
GLM Model [N=272,899]
Secondary prevention
147 [-63, 356] 272* [246, 297] 509* [293, 726]

42
Figure 2. Summary Adjusted Secondary Prevention Effect across
Index ASCVD Risk Groups

a
CABG=Coronary Artery Bypass Grafting surgery; AMI=Acute Myocardial Infarction; TIA=Transient Ischemic Attack;
b
Blue vertical line represents the general secondary prevention treatment effect for all ASCVD risk groups.

Secondarly prevention
TIA
Atherosclerosis
Cerebrovascular disease
Coronary angioplasty
Peripheral artery disease
Stroke
Peripheral bypass
AMI
Arterial embolism
Unstable angina
Arterial disorder
Hemorrhage
CABG
Revascularization
Index ASCVD diagnosis
0.67 [0.66, 0.69]
0.81 [0.75, 0.87]
0.77 [0.70, 0.85]
0.75 [0.71, 0.81]
0.73 [0.67, 0.80]
0.72 [0.69, 0.75]
0.70 [0.67, 0.73]
0.67 [0.46, 0.95]
0.67 [0.63, 0.70]
0.66 [0.51, 0.84]
0.64 [0.60, 0.68]
0.63 [0.51, 0.77]
0.48 [0.40, 0.57]
0.44 [0.38, 0.51]
0.27 [0.07, 0.95]
Hazard ratio [95%]
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
Secondarly Prevention Effect on Patients with ASCVD Diagnosis

43
Appendix 1. A List of ICD-9-CM Codes Used
Description Code Type Code
AMI ICD-9-CM diagnosis 410.xx, 412.xx
Unstable angina ICD-9-CM diagnosis 411.xx
Hemorrhage ICD-9-CM diagnosis 430.x, 431.x, 432.x
Stroke ICD-9-CM diagnosis 433.xx, 434.xx, 436.xx
Transient cerebral
ischemia
ICD-9-CM diagnosis 435.xx
Cerebrovascular disease ICD-9-CM diagnosis 437.xx,438.xx, 439.xx
Atherosclerosis ICD-9-CM diagnosis 440.xx
Peripheral vascular disease ICD-9-CM diagnosis 443.xx, 440.2
Arterial embolism ICD-9-CM diagnosis 444.xx, 445.xx, 449.xx
Arterial disorder ICD-9-CM diagnosis 446.xx, 447.xx, 448.xx
CABG ICD-9-CM procedure 36.10, 36.11, 36.12, 36.13, 36.14, 36.15,
36.16, 36.19
CPT codes 335.10-335.19, 335.21-335.23, 335.33-
335.36, S2005-S2009
Coronary angioplasty ICD-9-CM procedure 00.61, 00.62, 00.64, 00.65, 00.66, 00.40-
00.48, 36.04, 36.06, 36.07, 36.09, 38.22
CPT codes 372.05, 372.06, 372.36, 372.37, 372.46,
372.47, 354.52, 354.72, 354.76, 929.20-
929.29, 929.33-929.38, 929.44, 929.73,
929.82, 929.84, 929.95, 929.96
Peripheral bypass ICD-9-CM procedure 39.2x
CPT codes 00.63
Revascularization
procedure
ICD-9-CM procedure 36.2x, 36.3x, 36.9x

44
Appendix 2. Patient-Level Use in Cholesterol-Lowering Medications

Total Number of Treated Patients (N=85,343)
Number Percentage
Cholesterol-lowering agents (Rx=86,526)
Any statin 77,315 89.35%
Low intensity 13,918 18.00%
Moderate intensity 52,420 67.80%
High intensity 10,977 14.20%
Bile acid binding resins 2,540 2.94%
Fibrates 3,781 4.37%
Ezetimibe 2,286 2.64%
Niacin 1,332 1.54%
Omega-3-fatty acid 737 0.85%
Treatment regiments (N=85,343)
Statin 74,779 87.62%
Non-statin 8,066 9.45%
Combinations
(statin and non-statin)
2,498 2.93%
45
Appendix 3
Patient Characteristics Predicting the Likelihood of Initiating Secondary Prevention: A Logistic Analysis
Odds Ratio 95% Confidence Interval
Baseline LDL [vs. >160 mg/dL)
< 101 mg/dL 0.28* [0.26, 0.30]
101-130 mg/dL 0.40* [0.38, 0.43]
131-160 mg/dL 0.69* [0.65, 0.73]
Age [vs. > 75 years]
21-44 years 0.60* [0.54, 0.67]
45-54 years 1.05 [0.98, 1.13]
55-64 years 1.19* [1.12, 1.26]
65-74 years 1.24* [1.20, 1.28]
Sex [vs female]
Male 1.19* [1.15, 1.23]
Index diagnosis
AMI 1.82* [1.71, 1.93]
Unstable angina 1.64* [1.53, 1.76]
Hemorrhage 0.93 [0.81, 1.06]
Stroke 1.43* [1.36, 1.51]
Transient cerebral ischemia 1.36* [1.28, 1.44]
Cerebrovascular disease 1.05 [0.99, 1.11]
Atherosclerosis 0.94* [0.89, 0.99]
Peripheral vascular disease 1.18* [1.12, 1.25]
Arterial embolism 0.85 [0.65, 1.11]
Arterial disorder 0.99 [0.84, 1.16]
CABG 5.22* [4.45, 6.11]
Coronary angioplasty 4.24* [3.89, 4.62]
Revascularization 0.63 [0.17, 2.37]
Peripheral bypass 0.78 [0.50, 1.20]
Race [vs. white]
Black 1.08 [0.89, 1.30]
Asian 1.03 [0.98, 1.08]
Hispanic 1.24* [1.13, 1.37]
Other 0.81* [0.75, 0.87]
Region [vs. West]
South 0.97 [0.92, 1.02]
Midwest 0.77* [0.72, 0.82]
North East 0.69* [0.59, 0.81]
Other 1.39 [0.92, 2.09]
Health Plan [vs. PPO]
HMO 0.84* [0.76, 0.93]
Medicaid 0.96 [0.80, 1.16]
Medicare 0.87* [0.78, 0.96]
POS 0.87* [0.77, 0.97]
Other 0.73* [0.63, 0.84]
Prior Utilization
Hospital admission 1.05 [0.99, 1.11]
46

Total costs [in $1,000] 1.00 [1.00, 1.00]
Medication Treatment History
Anti-obesity 0.91 [0.80, 1.03]
Anti-inflammatory 1.10* [1.06, 1.14]
Painkiller 1.04 [1.00, 1.07]
Antiviral 1.15* [1.11, 1.19]
Anti-asthmatics 1.08* [1.04, 1.13]
Anti-seizure 0.99 [0.95, 1.04]
Antidepressants 1.04 [1.00, 1.08]
Antidiabetic 1.67* [1.59, 1.76]
Antihypertensive 1.50* [1.44, 1.56]
Antiparkinson 0.96 [0.88, 1.05]
Antihistamines 1.06* [1.02, 1.10]
Antimetabolism 1.04 [0.99, 1.10]
Anticoagulants 1.10* [1.04, 1.16]
Hypnotic 0.99 [0.95, 1.04]
Ophthalmic 1.06* [1.02, 1.10]
Antithyroid 1.08* [1.03, 1.13]
Antiulcer 1.11* [1.07, 1.15]
Anticholinergic 0.97 [0.93, 1.01]
Medical History
Infection 0.89* [0.86, 0.93]
Neoplasm 0.92* [0.89, 0.95]
Endocrinology 0.88* [0.85, 0.91]
Diabetes 1.04* [1.00, 1.09]
Blood disease 0.84* [0.81, 0.87]
Mental disorders 0.93* [0.90, 0.97]
Nervous system 0.86* [0.83, 0.89]
Hypertension 0.94* [0.90, 0.98]
Circulation 1.00 [0.97, 1.04]
Respiratory system 0.89* [0.86, 0.93]
Digestive system 0.95* [0.92, 0.99]
Genitourinary system 0.94* [0.91, 0.97]
47
CHAPTER 4. Impact of Timely Initiation of Antihypertensive Medications for Patients
with Hypertension or Elevated Blood Pressure
Xue Han, MS,
1
Jeff McCombs, PhD,
1
Michelle Chu, PharmD, CDE, BCACP,
2
J. Samantha
Dougherty, PhD,
3
D. Steven Fox, MD, MPhil
1

Introduction
Hypertension, which significantly increases the risk of cardiovascular disease (CVD), affects
one-third of the U.S. population.
1,2,3
The economic burden of hypertension was estimated to be
$73.4 billion in 2009.
4
Many patients with hypertension also suffer from other risk-enhancing
comorbidities. Those illnesses, concomitant with hypertension, are estimated to increase the
CVD risk.
5-8
The relationship between hypertension and coexisting diseases is generally bi-
directional. For example, diabetes is frequently concomitant with hypertension, and the overlap
between hypertension and diabetes substantially increases the risk of vascular complications.
9

Similarly, both hypertension and diabetes predispose patients to renal functional abnormalities.
10

Lifetime therapy with antihypertensive medications is strongly recommended for patients with
other risk-enhancing comorbidities to achieve blood pressure control to prevent cardiovascular
events.
11

Since hypertension is normally a chronic disease without acute clinical symptoms, it is common
for hypertensive patients to delay initiating drug treatment until after they have experienced a
cardiovascular event.
12
That treatment delay is a major challenge for delivering successful
disease management. Earlier studies have reported that under-treatment not only results in
adverse health outcomes, but also increases the economic burden to society.
13-15
Therefore, the
latest American College of Cardiology/American Heart Association (ACC/AHA) High Blood
Pressure Guidelines lowered the recommended blood pressure thresholds, recommending earlier
intervention to prevent CVD events.
16
While the new guidelines are designed to significantly
expand treatment with antihypertensive medications for US adults at high CVD risk, this
expansion must also address the fact that patients frequently delay filling prescribed
antihypertensive medications once diagnosed.
17
More evidence is needed to encourage clinicians
to prescribe antihypertensive therapy early and encourage patients to comply with both earlier
intervention and other forms of prevention.
48
This study estimated the impact of timely initiation of medications on CVD risk reduction for
patients who meet antihypertensive medication treatment indications under the new guidelines.
Outcomes included both CVD-related event risk, and healthcare costs during the first year after
meeting antihypertensive treatment indications. We also estimated the additional impact of
antihypertensive treatment on the risk of cardiovascular disease and health care costs in patients
who have hyperlipidemia, diabetes, chronic kidney disease or obesity, as well as in patients only
diagnosed with elevated blood pressure.
Methods
This was a retrospective cohort analysis using paid claims data from a large health insurance
provider in the U.S. covering from January 1, 2007 to September 30, 2016. These data were
used to identify high-risk patients who qualify for antihypertensive treatment under the 2017
ACC/AHA High Blood Pressure guidelines. The study sample included both patients who never
initiated antihypertensive medication once diagnosed, as well as patients who delayed therapy
until after experiencing a CVD event.
Data sources
The Optum claims database is a national and longitudinal database that covers 47.8 million
persons residing in all regions of the US, and includes patients covered under both commercial
plans and Medicare Advantage plans. The data include administrative claims from medical
(physician, outpatient care such as laboratory tests, other services), inpatient, pharmacy,
demographics, enrollment eligibility, and mortality. These data can be used to derive relatively
complete diagnostic and drug utilization profiles, which reflect the patient’s health status over
time.

49
Study cohort
Patients who received a diagnosis of hypertension or elevated blood pressure were assumed to
meet the clinical requirements for treatment with antihypertensive medications as specified in the
2017 ACC/AHA High Blood Pressure guidelines and were identified for possible inclusion in
this study. Specifically, patients with ≥2 medical claims with a diagnosis of hypertension or
elevated blood pressure were identified [codes 401.XX, 796.2 in the International Classification
of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)]. Patients with elevated blood
pressure were an addition to the population recommended for treatment under the new 2017
guidelines and likely represent the large expansion of treated patients if the 2017 guidelines are
fully implemented. The day of the earliest hypertension/elevated blood pressure diagnosis was
assigned as the patient’s index date. Once patients were identified for possible inclusion in the
analysis, their baseline drug use and diagnostic profiles were summarized using data from 6
months prior to their index date.
Exclusion criteria
The identified patients were required to have continuous health insurance coverage for the 6
months prior to and for at least 12 months after their index date. The requirement for a minimum
of 6 month of pre-index data allowed the analysis to summarize each patient’s prescription drug
and diagnostic profile, which are a measure of their health status at baseline. Since there is no
clear consensus on the management of hypertension in pregnant women, these women were
excluded from the analysis. Patients were also excluded if they filled or refilled a prescription
for any antihypertensive medication before their diagnosis-based index date, or experienced a
stroke, acute myocardial infarction (AMI) or any other significant CVD event before their index
date.
Key Explanatory Variables and Other Covariates
Patients were considered to initiate timely antihypertensive treatment if they initiated treatment
before experiencing an AMI or stroke event. Timely treatment was entered as a dummy variable
in the analysis. The comparison group consisted of individuals not filling any antihypertensive
medications or filling the prescription only after an AMI or stroke event. All available
antihypertensive therapies used as initial therapy were included in this analysis: angiotensin-
50
converting enzyme (ACE) inhibitors, angiotensin-receptor blockers (ARBs), β-blockers (BB),
calcium channel blockers (CCBs), thiazide-type diuretics, other diuretics, and alpha blockers
(AA). The drug class of the patient’s initial therapy was entered as a dummy variable in a
sensitivity analysis used to test for differential effects by drug class. The diagnoses of
hypertension and elevated blood pressure were specified as dummy variables and the differential
impact of timely treatment was investigated for patients with the elevated blood pressure
diagnosis.
Other study variables included gender, age, race, region, health plan type [Exclusive Provider
Organizations (EPO), Indemnity (IND), Health Maintenance Organization (HMO), Point of
Service (POS), Preferred Provider Organization (PPO)]. Baseline comorbidities indicators for
diabetes, chronic kidney disease and obesity were created as dummy variables for special
attention in the analysis as these three comorbidities are known risk factors for CVD. Other
baseline variables included diagnostic and prescription drug profiles. The therapeutic classes of
drug used in define the patient’s profile include anti-obesity, anti-inflammatory, analgesic,
antiviral, anti-asthmatics, anti-seizure, antidepressants, antidiabetic, antihypertensive, anti-
Parkinson, antihistamines, antimetabolic agents, anticoagulants, hypnotic, ophthalmic,
antithyroid, antiulcer and anticholinergic. Patient’s hospital admission and health care cost in the
prior 6 months were also included in the analysis as indicators of patients’ health status.
Outcomes
This analysis investigated two classes of outcomes: 1) clinical events [AMI, stroke and death]
allowing the index date, and 2) all-cause healthcare costs over the first year following the index
date. All-cause health cost in the first year following index date will be estimated by types of
service: medical, pharmacy and total health services.
Statistical methods
Descriptive statistics comparing patient’s baseline characteristics across treatment status used
either the Pearson’s chi-squared test or a t-test to identify significant differences.
18-19
The
association between antihypertensive drug therapy and the risk of cardiovascular-related events
was estimated using Cox proportional hazards regressions. This approach allows the analysis to
take into account the temporal relationship between time to treatment and the time to any future
51
events.
20
The impact of timely antihypertensive medication use on the costs for medical,
pharmacy and total health care services were estimated using GLM procedures.
21

Three sensitivity analyses were performed in this study to provide more detailed information
concerning the impact of timely treatment. The first sensitivity analysis estimated treatment
effects separately for patients with diabetes, chronic kidney disease and obesity. The second
sensitivity analysis was conducted using only treated patients to assess the association between
the patient’s selection of an initial anti-hypertensive treatment class and outcomes. Patients with
combination therapy will have separate dichotomous variables for each of their initial drug class
sections. A dichotomous variable indicating combination therapy is also included in this
analysis. Finally, we re-defined ‘timely treatment’ using a 120-day grace period following the
patients index date, during which antihypertension therapy could be initiated. Analyses were
performed using Version 9.3 of the SAS System for Unix.
Results
Figure 1 summarizes the patient selection for the study. Over 8 million patients were identified
with at least 2 recorded diagnoses for hypertension or elevated blood pressure, of which 2.5
million were found to have sufficient continuous enrollment for inclusion in the study and 1.8
million had no prior CVD diagnosis or were not pregnant in the baseline period. Screening for
prior antihypertensive prescription drug use and use of CVD related medications resulted in a
study population of 916,633 patients. That is, approximately half of the eligible patients
exhibited evidence of treatment related to hypertension at baseline.
Nearly 66% of patients who were identified here as being newly eligible for anti-hypertensive
treatment under the new guidelines filled a prescription prior to experiencing an (i.e., post-index)
AMI or stroke. That is, only one third of newly diagnosed hypertensive patients delay initiation
of anti-hypertensive medication therapy until after experiencing a significant CVD event.
Table 1 presents the unadjusted comparisons of baseline characteristics between patients with
and without timely anti-hypertensive treatment. Most patients with elevated blood pressure were
never treated with any anti-hypertensive treatment. Patients who did not initiate timely treatment
displayed higher rates of hyperlipidemia, diabetes, chronic kidney disease and obesity at baseline
52
and were more likely to have a prior inpatient admission and higher health care costs during the
6 months pre-index period.
A logistic regression model of treatment initiation was estimated to untangle the multiple
competing risk factors associated with timely treatment initiation [Table 2]. A few key results
may be of interest to the reader. Patients aged ≥75 years were the least likely to initiate anti-
hypertensive treatment. Male patients were less likely to initiate treatment. Patients who were
diagnosed with elevated blood pressure rarely initiate treatment relative to patients with a
hypertension diagnosis. Patients with hyperlipidemia, diabetes and chronic kidney disease at
baseline had lower probability of filling an antihypertensive prescription. Patients living in the
Northeast were less likely than patients living in the South, West and Midwest to initiate anti-
hypertensive medications on time. Each patient’s pre-index medication and diagnostic profiles
were also included in the analysis. In general, patients with a history of prescription drug use in
the pre-index period were more likely to initiate treatment, while patients in concomitant with
morbidities at baseline did not tend to initiate treatment with anti-hypertensive medications.
Table 3 shows the frequency of patients’ initial prescription of anti-hypertensive medications by
individual drugs. A large proportion of patients (30%) initiated single treatment therapy using
angiotensin-converting enzyme (ACE) inhibitors as their initial anti-hypertensive medication of
choice. Only 2% of patients would choose an alpha blocker (AA) as their mono-treatment. In
addition, 25% of treated patients used combination of different medication classes as their
preferred anti-hypertensive treatment regimen. Particularly, 82% of those patients were
prescribed thiazide and 54% of patients used ACE in their combination therapies.
Results from the multivariable analyses of event risk and cost are reported in Table 4. These
models adjust for the patient’s risk factors including patient's characteristics and prior diagnostic
and treatment history. Cox proportional hazard regression models were estimated to take better
advantage of available post-index data, which vary in length across observations. Cox models
were estimated for the likelihood of an AMI, stroke and death between groups with/without
treatment. These results are presented in the top four panels of Table 4. Generalized linear
models were used to model post-index cost over 1 year and results are presented in the bottom
half of Table 4. All models were adjusted for age, gender, region, type of diagnoses, co-
53
morbidity, health plan type, diagnostic history, prescription drug use history, prior hospital
admission and health service costs.
The Cox model results indicated that initiating timely anti-hypertensive treatment was associated
with significantly reduced risk of AMI, stroke and death. That is, initiating anti-hypertensive
treatment before having any events decreased the likelihood of having AMI by 59%, stroke by
60% and death by 9%. Results from Cox model also documented that the patients who were
diagnosed with elevated blood pressure rather than hypertension experienced significant lower
risk of outcome events. In addition, if patients had diabetes, they experienced 41%, 22% and
51% higher risk of having AMI, stroke and death, respectively. Chronic kidney disease was
associated with increased risk of AMI, stroke and death but its association with stroke was not
statistically significant. However, hypertensive patients with hyperlipidemia experienced lower
probability of having AMI, stroke or death, which may reflect a a clearer understanding of their
high CVD risk which impacted on both pattern of healthy lifestyle and good medication
adherence. Generally, obesity was not associated with the increased risk of outcome events for
hypertensive patients.
Sensitivity analyses were estimated to document the relationship between timely anti-
hypertensive treatment and event outcomes for patients with elevated blood pressure, diabetes,
chronic kidney disease and obesity at baseline using the full sample patients. The results are
summarized in the second panel of results in Table 4. The timely initiation of anti-hypertensive
drug therapy had significant protective effects on reducing AMI, stroke especially for patients
with hyperlipidemia, diabetes, chronic kidney disease or obesity. The risk reductions of AMI or
stroke for patients with diabetes or obesity were not significantly different from general
treatment effects, while treatment was associated with a significantly smaller reduced risk of
AMI or stroke than general treatment for patients with hyperlipidemia or chronic kidney disease.
Furthermore, the risk reductions in stroke associated with timely treatment were more
pronounced for patients with chronic kidney disease than AMI effects. Patients with any co-
morbidity experienced insignificant reductions in the risk of death due to timely treatment.
Finally, timely treatment in patient with elevated blood pressure only reduced the risk of stroke.
The second sensitivity analysis used data for treated patients to document the effects of the
timely initiation of alternative anti-hypertensive treatment classes on the risk of outcome events
54
controlling for combination of therapy. This was accomplished by analyzing only those patients
who filled a prescription for anti-hypertensive medications in a timely manner. The results
measured drug class effects for treated patients relative to ACE inhibitors. ARBs were
associated with larger risk reduction of AMI and death as initial prescription but had no
significantly different effect on reducing stroke than using ACE as treatment therapy. Patients
who initiated thiazides experienced larger reduction in risk of stroke and AMI relative to
initiating ACE. Other treatment options are all inferior to ACE inhibitors. The pattern of relative
impacts clearly indicate that ACE inhibitors and ARB classes of anti-hypertensive drugs were
associated with lower rates of AMI, stroke and death relative to other treatment classes like
CCB, BB, other diuretics and AA.
The sensitivity analysis of the definition of timely treatment is also presented in Table 4. When
the timely treatment was defined as initiating medications within 120 days following the
hypertension diagnosis, treatment was still associated with risk reduction in AMI and stroke, but
was associated with to a higher death rate.
Finally, the estimated impact of the timely initiation anti-hypertensive treatment on healthcare
costs was estimated using generalized linear models [GLM]. These results are reported at the
bottom of Table 4. Anti-hypertensive treatment was significantly associated lower total cost
over one year of $934 [p<0.05] primarily due to decreased medical costs per patient of $1,224
which were partially offset by increased prescription drug costs per patient of $451. Patients
with an index diagnosis of elevated blood pressure spent less on health care in the first post-
index year than patients with hypertension. Patients with hyperlipidemia also had lower health
cost than other patients. Moreover, patients with diabetes, chronic kidney disease or obesity
displayed higher one-year post-treatment cost.
Table 5 presents unadjusted data for the study outcomes, stratified by treatment status. These
results must be viewed with caution given the significant differences in the two study
populations and possible differences in the length of the follow-up period across subpopulations.
Treated patients have higher unadjusted risk of death relative to untreated patients even though
untreated patients are more likely to be over age 75. Conversely, the unadjusted probabilities of
having either an AMI or stroke were lower in patients initiating anti-hypertensive medications.
Treated patients experienced a longer average time gap between their index date and an event
55
date/end of enrollment than did patients without treatment. Finally, the treated population had
lower inpatient cost but higher pharmacy cost over the first year following index date than did
patients not initiating early treatment.
Discussion
The 2017 guidelines expand the population of patients recommended for treatment with anti-
hypertensive drug therapy. First, the blood pressure threshold for treatment was reduced to
130/80 mm Hg. Second, the new guidelines added patients with a diagnosis of elevated blood
pressure to the population of patients recommended for treatment. These patients are likely the
significantly expand the size of the population recommended for treatment according to new
guidelines. Only 3% treated patients in this study had a baseline diagnosis of elevated blood
pressure while 25% of untreated patients fell into this category. The estimated number exceeds
the range of increased treatment-eligible population published in the existing study.21 It may due
to the fact that not all patients with elevated blood pressure in our analysis qualified for treatment
under new guidelines if they are not at risk of ASCVD. This treatment pattern is not unexpected
as patients with elevated blood pressure experienced significant lower risk of outcome events
than did patients with hypertension, which may also result from inaccuracy of clinical blood
pressure data for defining elevated blood pressure. Despite their lower baseline risk, the
initiation of treatment by patients with elevated blood pressure was associated with reduced risk
of AMI and stroke, though these treatment effects were much smaller than for patients with a
baseline hypertension diagnosis.
Patients with a baseline diagnosis of diabetes, chronic kidney disease experienced significantly
higher risk of having AMI, stroke, death and higher costs than patients without these
comorbidities. The results reinforced the main findings in recently published studies.
22,23

However, patients had lower probability of having cardiovascular events if they were diagnosed
with hyperlipidemia and hypertension, which may be partially due to their improved lifestyles.
Early treatment with anti-hypertensive drug therapy was estimated to significantly reduce the
risk of AMI, stroke and death in these patients though the treatment effect for these patients did
not differ significantly from other patients. Furthermore, we found that anti-hypertensive
treatment was associated with a significant reduction in health care costs. Our results are not
56
sensitive to a change in the definition of timely treatment when we identified the treatment
initiated within 120 days as timely treatment.
Finally, our analysis of treatment effects by class of anti-hypertensive medication used found
clear evidence that ACE inhibitors, ARBs, and thiazide medications outperform the other classes
of anti-hypertensives including other diuretics, BB, CCB and AA. ACE inhibitor medications
were the most frequently used class of anti-hypertensive medications in monotherapy. The
second most common class of monotherapy was beta blockers [13.4% of all patients, Table 3]
while ARB medications came in fourth in frequency of use as monotherapy [7.5%]. The
appropriateness of this pattern of prescribing warrants more investigation.
Limitations
Several limitations of this study require discussion. Treatment selection bias may exist in any
study using real world data that compares across alternative treatments. The non-random
initiation of treatment may result from important but unmeasured differences across treated and
untreated patients as documented in Table 1. While our statistical analyses entered an extensive
set of independent variables as covariates, it remains impossible to fully control for treatment
selection using this approach. Other more sophisticated statistical methods (e.g. instrumental
variable) were not applied here, leaving some residual uncertainty that unobserved factors could
be the true driver for the results.
Our identification of patients for inclusion in the study was based solely on recorded diagnoses
as blood pressure measurements are not included in the data. The 2017 guidelines recognize that
clinic BP value may not be the most accurate diagnostic tool and should be replaced by home BP
or ambulatory BP monitoring. The inaccuracy of recorded diagnoses introduces an unknown
degree of mis-classification of non-hypertensive patients into the analysis.
The number of patients receiving timely anti-hypertensive treatment may be under-estimated due
to some missing pharmacy claims if the patient paid with cash. This is particularly important as
most anti-hypertensive medications are generically available at low cost. Therefore, some
patients who actually initiated anti-hypertensive medications were designated in no treatment
group. Given that fact, the treatment effect on reducing cardiovascular-related events is likely to
be under-estimated as well.
57
Patient adherence to anti-hypertensive medications once diagnosed was also not directly
considered in this analysis, only initial prescription fills. We found that simply initiating timely
anti-hypertensive medications improved clinical outcomes regardless of adherence to treatment.
Future analyses should account for the patients’ long-term adherence to medication and estimate
the impact of duration of therapy rather than the act of initiating treatment.
Hypertension can lead to other clinical events such as heart value problems and heart failure.
The extent to which a patient delay therapy until after these events, they would still be defined as
treated ‘on time.’ This potential mis-classification of the patient would tend to under-estimate
the benefits from early treatment.
Conclusions
In summary, initiating timely anti-hypertensive medications was associated with reduced risk of
having AMI, stroke and death for all patients with a diagnosis of hypertension or elevated blood
pressure as identified in new guidelines. Patients with an elevated blood pressure diagnosis, the
primary source of any expanded eligible population under new guidelines, experienced lower
risk of stroke once treated. Furthermore, initiating ACE, ARB and thiazide among treated
population were associated with larger risk reduction of AMI and stroke than initiating other
anti-hypertensive classes. The anti-hypertensive treatment resulted in larger risk reduction of
AMI and stroke for patients who had diabetes or obesity relative to patients with chronic kidney
disease. Moreover, treatment did show a significant effect on lowering all-cause health care
costs in the first post-index year.

58
References
1. Nwankwo T, Yoon SS, Burt V et al. Hypertension among adults in the United States:
National Health and Nutrition Examination Survey, 2011–2012. NCHS Data Brief. 2013.
No.133.
2. Psaty BM, Furberg CD, Kuller LH et al. Association between blood pressure level and the
risk of myocardial infarction, stroke, and total mortality: the cardiovascular health study.
Arch Intern Med. 2001.161:1183–1192.
3. Kannel WB. Blood pressure as a cardiovascular risk factor prevention and treatment.
JAMA. 1996.275(20):1571-1576.
4. Cohen JD. Hypertension epidemiology and economic burden: refining risk assessment to
lower costs. Manag Care. 2009.18(10):51-8.
5. Staessen JA, Thijs L, Fagard R et al. Predicting cardiovascular risk using conventional vs
ambulatory blood pressure in older patients with systolic hypertension. JAMA. 1999.
282(6):539-546.
6. Luchsinger JA, Honig LS, Tang MX. Depressive symptoms, vascular risk factors, and
Alzheimer's disease. Int. J. Geriatr. Psychiatry. 2008.23:922–928
7. Rahman M, Pressel S, Davis BR et al. Renal outcomes in high-risk hypertensive patients
treated with an angiotensin-converting enzyme inhibitor or a calcium channel blocker vs a
diuretic. Arch Intern Med. 2005.165(8):936-46.
8. Sowers JR, Epstein M, Frohlich ED. Diabetes, hypertension, and cardiovascular disease
an update. Hypertension. 2001.37:1053-1059.
9. Lago RM, Singh PP, Nesto RW. Diabetes and hypertension. Nature Clinical Practice
Endocrinology & Metabolism. 2007.3(10): 667.
10. Rahmouni K, Correia LG, Haynes WG. Obesity-associated hypertension.
Hypertension. 2005.45:9-14.
11. James PA, Oparil S, Carter BL. Evidence-based guideline for the management of high
blood pressure in adults report from the panel members appointed to the Eighth Joint
National Committee (JNC 8). 2014.
12. Staessen JA, Thijs L, Fagard R. Effects of immediate versus delayed antihypertensive
therapy on outcome in the Systolic Hypertension in Europe Trial. Journal of
Hypertension. 2004. 22(4):847-857.
13. McCombs JS, Nichol MB, Newman CM et al. The costs of interrupting antihypertensive
drug therapy in a Medicaid population. Med Care. 1994.32(3):214- 226.
14. Andersson F, Cline C, Ryden-Bergsten, T et al. Angiotensin converting enzyme (ACE)
inhibitors and heart failure: the consequences of underprescribing. Pharmacoeconomics.
1999;15(6):535-550.
15. Han X, Fox DS, Chu LM, Dougherty JS, McCombs J. Secondary prevention using
cholesterol-lowering medications in patients with prior CVD events: a retrospective
cohort analysis. Journal of Managed Care & Specialty Pharmacy.2018.
16. 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.
Hypertension. 2017.71(3):1-481.
17. Hales CM, Carroll MD, Simon PA et al. Hypertension prevalence, awareness, treatment,
and control among adults aged ≥18 years - Los Angeles County,1999-2006 and 2007-
2014. CDC Morbidity and Mortality Weekly Report. 2017.66(32):846-849.
59
18. Plackett RL. Karl Pearson and the Chi-Squared Test. International Statistical Review.
Revue Internationale de Statistique. 1983.51(1):59-72.
19. DeCoster J. Testing group differences using T-tests, ANOVA, and nonparametric
measures. 2006.30(2010):202006-0.
20. Cox DR. Regression models and life-tables. Journal of the Royal Statistical Society.
Series B (Methodological). 1972.34(2):187-220.
21. Munter P, Carey RM, Gidding S et al. Potential US population impact of the 2017
ACC/AHA high blood pressure guideline. Circulation. 2018.137(2):109-118.
22. Landsberg L, Aronne LJ, Beilin LJ et al. Obesity ‐related hypertension: pathogenesis,
cardiovascular risk, and treatment. Journal of Clinical Hypertension. 2013.15(1):14-33.
23. Lorber Daniel. Importance of cardiovascular disease risk management in patients with
type 2 diabetes mellitus. Diabetes Metab Syndr Obes. 2014. 7:169–183.

60
Figure 1. Flow chart of patient selection based on exclusion criteria

Patient with at lease 2 hypertension diagnoses
(N=8,276,077)
≥6 months pre & ≥1 year post index enrollment
(N=2,537,510) [30.7%]
No cardiovascular diseases/pregnancy in 6 months pre-index
(N=1,804,597) [21.8%]
STUDY POPULATION: No anti-hypertensives/cardiac drug in 6
months pre-index
(N=916,633) [11.1%]
Late/No treatment
(N=312,239) [34.1%]
Early treatment
(N=604,394) [65.9%]
61
Table 1. Descriptive Demographics for Sample Population
Treatment status Treatment
[N= 604,394][65.9%]
No Treatment
[N=312,239][34.1%]
P-value
Types of Hypertension Diagnoses, [%]
Hypertension 97.3% 74.9%
<.0001
Elevated blood pressure 2.7% 25.1%
Age at year of index date, [%]
<45 28.8% 32.2%
<.0001
45-54 29.9% 24.8%
55-64 22.4% 20.4%
65-74 13.4% 15.5%
Over 75 5.5% 7.2%
Sex, [%]
Male 52.8% 54.7% <.0001
Region, [%]
West 17.2% 17.9%
<.0001
Northeast 8.5% 13.5%
Midwest 23.4% 20.7%
South 50.9% 47.9%
Health plan [%]
EPO 11.1% 10.9%
<.0001
HMO 14.8% 14.2%
IND 1.0% 0.9%
POS 59.0% 55.9%
PPO 5.0% 6.2%
Other 9.1% 12.0%
Comorbidity
Hyperlipidemia 12.8% 17.6% <.0001
Diabetes 5.2% 5.4% 0.001
Chronic Kidney disease 0.6% 0.7% <.0001
Obesity 2.9% 3.2% <.0001
Drug at baseline
Antihyperlipidemics 8.6% 10.6% <.0001
Antidiabetics 4.4% 3.9% <.0001
Antiobesity 1.0% 1.3% <.0001
Prior Utilization [6 months]
Hospital Admission [%] 1.73% 2.25% <.0001
Medical Costs, $, mean [SD] 3,368 [16,535] 4,070 [20,012] <.0001
Inpatient Costs, $, mean [SD] 858 [15,785] 1,278 [30,412] <.0001
Pharmacy Costs, $, mean [SD] 586 [1,929] 618 [2,297] <.0001
TOTAL Costs, $, mean [SD] 4,812 [29,719] 5,966 [46,021] <.0001

62
Table 2. Patient Characteristics Predicting the Likelihood of Initiating Anti-Hypertensive Medications:
A Logistic Analysis
Odds Ratio 95% Confidence Interval
Age [vs. ≥ 75 years]
< 45 years 1.32* [1.29, 1.36]
45-54 years 1.58* [1.54, 1.62]
55-64 years 1.40* [1.36, 1.43]
65-74 years 1.22* [1.19, 1.24]
Sex [vs. female]
Male 0.87* [0.86, 0.88]
Disease [vs. hypertension]
Elevated Blood Pressure 0.07* [0.07, 0.08]
Comorbidity
Hyperlipidemia 0.74* [0.73, 0.75]
Diabetes 0.96* [0.94, 0.99]
Chronic Kidney Disease 0.97 [0.91, 1.03]
Obesity 1.02 [1.00, 1.05]
Region [vs. West]
South 1.08* [1.07, 1.09]
Midwest 1.20* [1.19, 1.22]
North East 0.63* [0.62, 0.64]
Health Plan [vs. HMO]
EPO 1.02* [1.00, 1.04]
IND 1.24* [1.18, 1.30]
POS 1.07* [1.05, 1.08]
PPO 0.88* [0.86, 0.90]
Other 0.81* [0.79, 0.82]
Prior Utilization
Hospital admission 0.87* [0.83, 0.90]
Total costs [in $1,000] 1.00 [1.00, 1.00]
Medication Treatment History
Anti-hyperlipidemic 0.82* [0.80, 0.83]
Anti-obesity 0.96 [0.92, 1.01]
Anti-inflammatory 1.11* [1.09, 1.13]
Painkiller 1.14* [1.12, 1.15]
Antiviral 1.07* [1.05, 1.08]
Anti-asthmatics 1.01 [0.99, 1.04]
Anti-seizure 0.94* [0.92, 0.97]
Antidepressants 1.03* [1.01, 1.04]
Antidiabetic 1.10* [1.07, 1.12]
Antiparkinson 1.00 [0.93, 1.06]
Antihistamines 1.07* [1.04, 1.09]
Antimetabolism 0.90* [0.87, 0.93]
Anticoagulants 0.94 [0.87, 1.02]
Hypnotic 0.99 [0.96, 1.02]
Ophthalmic 1.16* [1.12, 1.19]
Antithyroid 0.84* [0.82, 0.86]
Antiulcer 1.06* [1.04, 1.08]
Anticholinergic 1.01 [0.98, 1.05]
Medical History
Infection 0.92* [0.90, 0.94]
Neoplasm 0.93* [0.91, 0.94]
Endocrinology 0.92* [0.90, 0.94]
63

Blood disease 1.03* [1.02, 1.04]
Mental disorders 0.96* [0.93, 0.99]
Nervous system 0.92* [0.90, 0.93]
Congenital 0.99 [0.94, 1.04]
Circulation 0.96* [0.93, 0.99]
Respiratory system 0.91* [0.90, 0.93]
Digestive system 0.88* [0.87, 0.90]
Genitourinary system 0.91* [0.90, 0.93]
64
Table 3. Patient-Level Use in Anti-Hypertensive Medications

Initial prescription
Total Number of Treated Patients [N=604,394]
Number Percentage
Mono therapy
Angiotensin-converting enzyme (ACE) 183,196 30.3%
Angiotensin-receptor blockers (ARB) 45,605 7.5%
Beta-blockers (BB) 81,198 13.4%
Calcium channel blockers (CCB) 55,997 9.3%
Diuretics 11,320 1.9%
Thiazide 59,865 9.9%
Alpha blockers (AA) 13,788 2.3%
Combination therapy 153,425 25.4%
Angiotensin-converting enzyme (ACE) 82,452 53.7%
Angiotensin-receptor blockers (ARB) 37,128 24.2%
Beta-blockers (BB) 22,417 14.6%
Calcium channel blockers (CCB) 27,575 18.0%
Diuretics 22,228 14.5%
Thiazide 125,477 81.8%
Alpha blockers (AA) 3,624 2.4%
65
Table 4. Summary of Estimated Effects of Anti-Hypertensive Medications
1. * represents ≤0.05, ** represents ≤0.01, *** represents ≤0.001.
2. “a” means that the value is statistically significant but not statistically different from overall treatment
effect.

CLINICAL OUTCOMES
AMI Stroke Mortality
HR 95% C.I. HR 95% C.I. HR 95% C.I.
Cox Model [N=916,633]
Treatment 0.41*** [0.40-0.42] 0.40*** [0.40-0.41] 0.91*** [0.87-0.94]
Elevated BP [vs.
Hypertension]
0.18*** [1.16-1.20] 0.26*** [0.24-0.27] 0.47*** [0.42-0.53]
Hyperlipidemia 0.79*** [0.75- 0.83] 0.92*** [0.90-0.95] 0.66*** [0.62-0.70]
Diabetes 1.41*** [1.32-1.50] 1.22*** [1.18-1.27] 1.51*** [1.42-1.61]
Chronic Kidney 1.35*** [1.16-1.57] 1.03 [0.94-1.13] 1.48*** [1.30-1.69]
Obesity 0.90 [0.80-1.01] 0.88*** [0.83-0.94] 0.81*** [0.70-0.94]
Sensitivity Analysis of Estimating Treatment Effect on Patients with Co-morbidity [N=916,633]
Overall Treatment
Effect
0.40*** [0.39-0.42] 0.40*** [0.39-0.41] 0.90*** [0.86-0.94]
Elevated BP 0.86 [0.64-1.15] 0.47*** [0.40-0.56] 1.17 [0.91-1.51]
Hyperlipidemia 0.45* [0.41-0.49] 0.44*** [0.42-0.46] 0.93 [0.84-1.03]
Diabetes 0.38***,a [0.34-0.42] 0.38***,a [0.35-0.40] 0.89 [0.79-1.00]
Chronic Kidney 0.72*** [0.54-0.96] 0.47*** [0.39-0.56] 0.84 [0.65-1.07]
Obesity 0.37***,a [0.29-0.46] 0.40*,a [0.35-0.45] 0.96 [0.70-1.31]
Sensitivity Analysis of Estimating Treatment Class Effect on Treated Patients [N=604,394]
ARB [vs ACE] 0.89*** [0.81-0.98] 0.98 [0.92-1.03] 0.73*** [0.66-0.82]
BB [vs ACE] 1.15*** [1.07-1.24] 1.22*** [1.17-1.27] 1.15*** [1.07-1.23]
CCB [vs ACE] 1.02 [0.93-1.10] 1.12*** [1.07-1.17] 1.18*** [1.10-1.28]
DURETICS [vs ACE] 1.43*** [1.26-1.63] 1.08 [0.99-1.16] 2.20*** [2.01-2.41]
THIAZIDE [vs ACE] 0.92 [0.83-1.01] 0.90* [0.85-0.95] 1.18*** [1.10-1.28]
AA [vs ACE] 1.18* [1.04-1.35] 1.24*** [1.15-1.33] 1.40*** [1.25-1.56]
Sensitivity Analysis of an Alternative Definition of Timely Treatment using 120 Days Grace Period
[N=916,633]
Treatment 0.41*** [0.40-0.42] 0.40*** [0.40-0.41] 0.91*** [0.87-0.94]
Treatment < 120 days 0.55*** [0.53-0.57] 0.80*** [0.78-0.81] 2.17*** [2.09-2.25]
ALL-CAUSE COST OUTCOMES ONE YEAR POST
Medical+Inpatient Costs Drug Costs Total Costs
Effect in $ 95% C.I. Effect in $ 95% C.I. Effect in $ 95% C.I.
Treatment -1,224 *** [-1,743, -705] 451*** [428, 475] -934*** [-1,458, -410]
Elevated BP [vs.
Hypertension]
-
12,922***
[-13,711, -
12,132]
-184*** [-220, -148]
-
12,949***
[-13,718, -
12,180]
Hyperlipidemia -3,966 [-4,622, -3,311] 20 [-13, 52] -3,934*** [-4,586,-3,281]
Diabetes 1,490*** [374, 2605] 143*** [90, 195] 1,447*** [340, 2,554]
Chronic Kidney 7,365*** [5,495, 9,234] 729*** [594, 863] 7,936*** [5,973, 9,900]
Obesity 1,841*** [701, 2982] -226*** [-289, -166] 1,667*** [489, 2,803]
66
Table 5. Descriptive Patient Outcomes by Treatment Status

Treatment Status
Treatment
[N=604,394]
No Treatment
[N=312,239]
P-value
Absolute Risk
Reduction
Proportion of events, No. [%]
Stroke 4.0% 7.5% <.0001 -3.5%
AMI 1.3% 2.4% <.0001 -1.1%
Death 1.5% 1.4% <.0001 0.1%
Time to events, days, mean [SD]
Time to Stroke 1,341 [781] 1,142 [708] <.0001 NA
Time to AMI 1,361 [784] 1,199 [721] <.0001 NA
Time to Death 1,373 [785] 1,228 [727] <.0001 NA
Health Care Cost over 1st Year Post, $, Mean [SD]
Medical 16,471 [51,611] 16,342 [50,591] 0.2514 NA
Inpatient 5,922 [44,630] 6,281 [46,027] 0.0003 NA
Pharmacy 2,134 [5,273] 1, 648 [5,211] <.0001 NA
Total 24,525 [89,596] 24,270 [89,813] 0.1978 NA

67
CHAPTER 5. Conclusions
The three papers presented in this dissertation provide evidence concerning the impact of timely
treatment for hyperlipidemia or hypertension on the risk of cardiovascular disease (CVD). Even
though the latest hyperlipidemia and hypertension treatment guidelines expanded preventive
treatment to a larger group of population, which triggers the debates about treatment expansion,
our studies demonstrated that patients newly recommended for treatment could benefit from
initiation of medication therapy according to guidelines. It is suggested that patients should
initiate cholesterol-lowering treatment or anti-hypertensive medications once they qualify for
treatment under updated guidelines.
Moreover, our studies summarized the current issues and documented the following conclusions.
First, delaying treatment or failing to initiate treatment is very common among disease area of
hyperlipidemia and hypertension. Particularly, the non-initiation rates of cholesterol-lowering
medications for patients qualified for primary and secondary prevention of cardiovascular events
are nearly 60%, 70%, respectively. The non-initiation rate of anti-hypertensive medications is
relatively lower, which is about 30% among all treatment eligible population under updated
guidelines.
The impact of delayed therapy is well documented in this dissertation. It is estimated that the
consequences of delayed treatment are severe in terms of increased risk and increased health care
costs. We found that patients qualified for secondary prevention of cardiovascular disease would
experience 33% of lower probability of having cardiovascular hospitalization if they initiated
cholesterol-lowering medications after they were identified as high risk patients. Anti-
hypertensive medications are associated with 60% risk reductions in AMI and stroke, and are
associated with 10% risk reduction in mortality. Moreover, we have the similar finding in
patients eligible for primary prevention treatment that treated patients have 14%-37% lower risk
of having cardiovascular events including AMI, stroke, coronary angioplasty compared to
untreated patients. Furthermore, the primary prevention patients also had lower health care costs
in the whole year compared to those patients without cholesterol-lowering treatment.
Hypertensive patients could save up to $1000 dollars if they were prescribed anti-hypertensive
medications before having any cardiovascular events. In conclusion, delaying or forgoing the
68
initiation of appropriate medication therapy likely result in larger cardiovascular event risk and
higher health care expenses.
What can be done about delayed treatment? Delayed treatment is associated with many factors
including large co-payment, pills burden and lack of confidence to medications.
1,2
Patients tend
to delay or avoid some necessary health utilization because of their perceived costly copayment
amount. The existing literature demonstrated the number of pills has a significant effect on
medication adherence.
3
Patients with hyperlipidemia or hypertension always have more than one
co-morbidity and have to manipulate multiple types of prescription. A limitation to medication
adherence may due to lack of belief or expectations for both physicians and patients that
medication works well.
4
The limited information about the specific treatment effect for different
subgroups would interfere with medication prescription. For example, the ACC/AHA guidelines
expand both primary and secondary cardiovascular prevention treatment to more patients which
raised concerns for physicians about whether all expansive patients would benefit from
treatment.
Many interventions have been proposed to address the issue of delayed treatment. For example,
electronic prescriptions could be used to track patient’s primary non-compliance. Prescribers
would encourage patients to be adherent to treatment if they have the feedback about whether
patients filled the treatment medication in a timely manner.
Also, educate prescribers the importance of medication initiation is another approach to improve
the compliance to medications. The evidence from our papers could help physicians prescribe
medications incorporating with their own perceived competencies and standards of care in their
professional field. For example, we documented that treatment rates in hyperlipidemia and
hypertension are low. However, all those patients who are recommended for cholesterol-
lowering medication or hypertension prevention could benefit from treatment. Physicians should
prescribe cholesterol-lowering medications or anti-hypertensive medications once patients are
high risk of cardiovascular events under new guidelines to reduce the risk of cardiovascular
events and lower health care costs.

69
Reference
1. Goldman DP, Joyce GF et al. Pharmacy benefits and the use of drugs by the chronically ill.
JAMA. 2004 May 19;291(19):2344-50.
2. John Hsu, Mary Reed et al. Cost-sharing: patient knowledge and effects on seeking emergency
department care. Medical Care.2004.42. 290-296.
3. Leslie R Martin,1 Summer L Williams et al. The challenge of patient adherence. Ther Clin
Risk Manag. 2005; 1(3): 189–199.
4. Beena Jimmy and Jimmy Jose. Patient medication adherence: measures in daily practice.
Oman Med J. 2011; 26(3): 155–159.

Abstract (if available)

Abstract The three papers presented in this dissertation focus on prevention of cardiovascular disease (CVD), which is the leading cause of death in the US. The total direct and indirect cost of CVD in the United States for 2010 is estimated to be $315.4 billion. Patients at high risk of cardiovascular events due to hyperlipidemia, hypertension or diabetes are recommended for treatment of cholesterol-lowering medications and/or anti-hypertensives. About one third of US adults have hyperlipidemia and are at high risk for heart disease and stroke. Not only patients with hyperlipidemia but also patients with diabetes and high ASCVD risk score are all recommended for cholesterol-lowering medications in 2013 American College of Cardiology and American Heart Association (ACC/AHA) guidelines. In 2017, American College of Cardiology (ACC) issued new guidelines that lowered the threshold of hypertension for recommending earlier intervention to prevent cardiovascular disease and are expected to expand the size of the hypertensive population designated for treatment. Such a large proportion of US population are at high risk for cardiovascular events and are recommended for medications under new treatment guidelines for both hyperlipidemia and hypertension. However, people tend to delay their suggested medications until they experienced some medical events. ❧ This dissertation investigates the association between the timely initiation of drug treatment and the risk of cardiovascular events for high risk patients and to document the benefits of timely initiation of medications. Claims based administrative data from Humana and Optum will be used in these analyses. The data include medical (inpatient, outpatient, ambulatory, ER), pharmacy, demographics, eligibility, and laboratory test values, which are used to identify at-risk patients using published treatment guidelines. Therapeutic regimens for hyperlipidemia and hypertension will be investigated in separate analyses. These three dissertation papers address the following questions: ❧ 1. The first paper investigates the impact of primary prevention using cholesterol-lowering medications on cardiovascular events and healthcare costs for patients meeting the treatment guidelines released by ACC/AHA in 2013. ❧ 2. The second paper examines the impact of initiating secondary prevention using cholesterol-lowering medications on risk of CVD hospitalizations and healthcare costs in patients with prior cardiovascular events. ❧ 3. The third paper estimates the association of timely initiation of anti-hypertensive medications with cardiovascular events for patient with hypertension or elevated blood pressure, who are the expanded target population for hypertension treatment under new guidelines published by the ACC in 2017.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The value of novel antihyperlipidemic treatments in the U.S. healthcare system: Reducing the burden of cardiovascular diseases and filling the gap of low adherence in statins

PDF

Understanding primary nonadherence to medications and its associated healthcare outcomes: a retrospective analysis of electronic medical records in an integrated healthcare setting

PDF

Pharmacotherapy among patients with type 2 diabetes

PDF

Impact of pharmacy-based transitional care on healthcare utilization and costs

PDF

An evaluation of risk predictors for disease progression in Veteran Affairs patients with chronic hepatitis C

PDF

The impact of treatment decisions and adherence on outcomes in small hereditary disease populations

PDF

Effects of a formulary expansion on the use of atypical antipsychotics and health care services by patients with schizophrenia in the California Medicaid Program

PDF

Assessment of the impact of second-generation antipscyhotics in Medi-Cal patients with bipolar disorder using panel data fixed effect models

PDF

The development of the prognostic propensity score: an introduction to a method to identify optimal treatment according to individual tailoring variables when heterogeneous treatment effects are ...

PDF

Three essays on comparative economic and clinical outcomes of antipsychotics using retrospective claims data

PDF

Three essays in health economics

PDF

Selected papers on the evaluation of healthcare costs of prematurity and necrotizing enterocolitis using large retrospective databases

PDF

Defining intrarenal sexual dimorphisms to obesity and hypertension: understanding the female advantage

PDF

Renin-angiotensin system modulation for the prevention and treatment of metabolic dysfunction

PDF

Development of ivermectin as a platform for the treatment and/or prevention of alcohol use disorders

Asset Metadata

Creator Han, Xue (author)

Core Title Documenting the impact of under-treatment for chronic disorders

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Health Economics

Publication Date 12/03/2018

Defense Date 11/30/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag anti-hypertensive medication,cardiovascular disease,cholesterol-lowering medication,healthcare cost,OAI-PMH Harvest,under-treatment

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor McCombs, Jeffrey (committee chair), Fox, Steven (committee member), Myerson, Rebecca (committee member)

Creator Email xuehan@usc.edu,yulinshiwo@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-109153

Unique identifier UC11675856

Identifier etd-HanXue-6989.pdf (filename),usctheses-c89-109153 (legacy record id)

Legacy Identifier etd-HanXue-6989.pdf

Dmrecord 109153

Document Type Dissertation

Format application/pdf (imt)

Rights Han, Xue

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA