Abstract
Hypertension is a costly disease; however, the investment needed for a cost‐neutral hypertension management program (HMP) is unknown. A Markov decision analytic model simulated the outcomes of a hypothetical HMP. Patients were between the ages of 25 and 65 years, had existing hypertension, and were newly diagnosed with diabetes. The control group received standard care. The HMP group received standard care and were enrolled in an HMP. Data regarding rates of disease states and costs were gathered from the literature. A third‐party payer can invest as much as $159, $109, and $41 per person per month in an HMP for a neutral return on investment in the 5‐year, 3‐year, and 1‐year time horizon, respectively. The HMP group achieved greater gains in quality‐adjusted life‐years and lower total health‐related costs. As the time horizon increases, more money can be invested. HMPs can be a cost‐effective and cost‐neutral proposition.
Hypertension is a disease that affects more than 65 million persons in the United States, about 1 in 4 individuals. 1 Numerous sequelae can result from hypertension including stroke, ischemic heart disease, and renal disease. Approximately 23% of these individuals, or 15 million, also have diabetes. 2 Individuals with diabetes alone are at an increased risk for cardiovascular disease and nephropathy. The combination of diabetes and hypertension increases the risk for end‐organ damage. To avoid these complications, recent guidelines from the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC7) have suggested that the blood pressure (BP) treatment goal in individuals with diabetes be <130/80 mm Hg rather than <140/90 mm Hg. 2
To achieve these goals in patients with diabetes, aggressive treatment of hypertension must be pursued. Many patients, including patients with diabetes, may require ≥2 hypertensive medications to control their BP. 2 A decrease of 20 mm Hg in systolic BP for persons between the ages of 40 and 69 years may decrease death rates due to vascular disease by approximately 50%. 3
Previous studies have examined the role of hypertension management programs (HMPs) in controlling BP. 4 , 5 , 6 , 7 , 8 Rudd and colleagues9 determined that an intensive nurse management program lowered systolic BP by 14.2 mm Hg compared with 5.7 mm Hg in the usual care group (P<.01). Roumie and colleagues8 examined the effectiveness of physician education, electronic reminders, and patient education, and noted that the combination of these 3 interventions had the greatest decrease in systolic BP (16 mm Hg). Lee and colleagues6 randomized patients with coronary risk factors to a comprehensive pharmacy program and was able to decrease systolic BP from 133.2 mm Hg to 120 mm Hg (P=.02). The PREMIER trial demonstrated that when a programmatic intervention such as losing weight, exercising, decreasing sodium and alcohol intake, along with adopting the Dietary Approaches to Stop Hypertension (DASH) diet, mean systolic BP was lowered by 4.3 mm Hg (P<.001) compared with the group who were just given advice. 4 A previous study has demonstrated the cost‐effectiveness of reducing hypertension in patients with diabetes. 10
The cost of treating the complications of hypertension alone in the United States is approximately $22 billion. Aggressive treatment of hypertension in a population with diabetes as shown in the United Kingdom Prospective Diabetes Study (UKPDS)11 can substantially reduce morbidity and mortality. Thus far, studies have shown that pharmacologic approaches to reducing BP have not only been cost‐effective but also cost‐saving. 10 The addition of an HMP through various modalities may further reduce BP and decrease morbidity and mortality. However, it is not known whether these programs represent an additional burden of cost to a third‐party payer. The adoption of HMPs by third‐party payers may increase if the programs are not associated with an additional cost. The aim of our study is to determine the cost of a hypothetical HMP that will result in a neutral return on investment.
Methods
Overview
A Markov decision analytic model was created on TreeAge Pro Suite 2008 (TreeAge Software, Inc, Williamstown, MA) to simulate 2 hypothetical cohorts of patients with hypertension and diabetes in the United States to determine their health‐related costs and health outcomes. This model determined the health and cost outcomes of an HMP based on the data associated with UKPDS 38 via a third‐party payer perspective. 11 The HMP found in the literature has different modalities each with its own associated costs and health outcomes. These programs achieved a mean decrease in systolic BP between 4.3 mm Hg and 16 mm Hg in the intervention groups compared with the nonintervention groups. 4 , 6 , 9 This study determined the cost of a hypothetical HMP that will result in a neutral return on investment.
Model Strategy and Assumptions
The model was based on 2 hypothetical and identical groups of 1000 individuals with both hypertension (mean BP of 159/94 mm Hg prior to intervention) and diabetes. Patients in this study were between the ages of 25 and 65 years, had newly diagnosed diabetes, and existing diagnoses of hypertension. The control group received standard of care. The HMP group received standard of care and a hypothetical program to reduce hypertension. Standard of care for the patient’s diabetes care included outpatient visits, self‐testing, laboratory tests, and prescription medication. 11 , 12 Standard of care for the patient’s hypertension included 3 office visits, laboratory tests, and prescription medication. BP was measured as noted in previous studies. 11 Although physician practices typically overestimate BP by 5 mm Hg, the BP methodology utilized in UKPDS is the gold standard. 13 , 14 On average, patients with less tight control received one antihypertensive medication, which was a generic angiotensin‐converting enzyme inhibitor such as lisinopril, but some also received a second medication. 11 , 12 As a result of the intervention, the control and HMP groups had a mean BP of 154/87 mm Hg and 144/82 mm Hg, respectively. 11
The control and HMP groups had health outcomes associated with this difference in mean BP as previously described in UKPDS. 11 Thus, both groups (control and HMP) had a simultaneous probability to advance to 1 of 6 disease states each year, including uncomplicated hypertension and diabetes, heart failure, stroke, coronary artery disease (myocardial infarction), nephropathy, and death from other causes (Figure). In these disease states, individuals may die from 1 of the 6 above causes or continue to live within their disease states. Each of these disease states and substates have associated costs and effectiveness. We examined the costs and health outcomes following a 1‐year, 3‐year, or 5‐year time horizon.
Figure.
Markov model. HF indicates heart failure; HTN, hypertension; DM, diabetes mellitus; MI, myocardial infarction.
When individuals entered the heart failure state, they either progressed through the state or subsequently died from heart failure. The incidence of heart failure differed between the control and HMP groups. After an individual in either group was diagnosed with heart failure, individuals faced a risk of death due to heart failure, as noted in Table I. When individuals entered the stroke state, they may have experienced a fatal or nonfatal stroke. Individuals who experienced a nonfatal stroke were at greater risk for having a recurrent stroke than individuals who did not experience a prior stroke. When individuals entered the coronary artery disease state, they will have had a fatal or nonfatal myocardial infarction. Treatment options such as percutaneous coronary intervention, coronary artery bypass graft surgery, or medical treatment were available if the patient had a nonfatal myocardial infarction. Risk of recurrent myocardial infarction rose if one had a nonfatal myocardial infarction. When individuals entered the nephropathy state they were separated into 2 substates: renal failure or fatal renal failure. The disease state (deaths from other causes) was created because individuals may die from other causes not related to the above disease states.
Table I.
Incidence Rates
| Disease State | Disease Substate | Hypertension Management Program | Control | Reference |
|---|---|---|---|---|
| HF | ||||
| HF | 3.6/1000 pya | 8.1/1000 py | 11 | |
| Death from HF (year 1) | 0.305 | 20 | ||
| Death from HF (year 2) | 0.140 | 20 | ||
| Death from HF (year 3) | 0.286 | 20 | ||
| Death from HF (year 4) | 0.341 | 20 | ||
| Death from HF (>year 5) | 0.457 | 20 | ||
| Stroke | ||||
| Nonfatal stroke | 5/1000 py | 8.9/1000 py | 11 | |
| Fatal stroke | 1.5/1000 py | 3.6/1000 py | 11 | |
| Recurrent stroke | 0.0266 | 0.0383 | 21 | |
| MI | ||||
| Nonfatal MI | 8.9/1000 py | 9.9/1000 py | 11 | |
| Fatal MI | 9.8/1000 py | 13.8/1000 py | 11 | |
| Fatal MI outside of hospital | 0.57 | 22 | ||
| Fatal MI inside of hospital | 0.43 | 22 | ||
| Recurrent MI | 0.03 | 23 | ||
| Nephropathy | ||||
| Renal failure | 1.4/1000 py | 2.3/1000 py | 11 | |
| Fatal renal failure | 0.3/1000 py | 1/1000 py | 11 | |
| Other cause of death | ||||
| Other cause of death | 24.4/1000 py | 29.1/1000 py | 11 | |
Abbreviations: HF, heart failure; MI, myocardial infarction. aCases per person‐years (py). All numbers are reported as incidence per year unless noted otherwise. Person‐years.
The model assumed (1) that the health outcomes associated with decreasing BP from pharmacologic therapy would result in the same health outcomes as nonpharmacologic therapy, if the decrease in BP were the same; (2) that the individuals in the groups would not be affected by outside forces or affected equally for both HMP and control groups; (3) that the difference in the BP between the control and HMP group was due to the HMP; (4) that both the control and HMP groups received similar health outcomes from standard of care; and (5) an incremental cost‐effectiveness ratio of $50,000/quality‐adjusted life‐years (QALY) was used as the threshold of cost‐effectiveness.
Incidence Rates
Each of the 6 disease states (uncomplicated hypertension and diabetes, heart failure, stroke, coronary artery disease [myocardial infarction], nephropathy, and death from other causes) have an associated incidence for the control and HMP groups. Incidence rates leading to these 6 disease states were gathered from the literature and are shown in Table I.
Costs
For this model, we utilized a third‐party payer perspective. The direct health‐related costs that appear in Table II are from previously published estimates of Medicare reimbursements. All costs and results have been adjusted to 2007 US dollars using the consumer price index and have been rounded to the nearest $1. A discount rate of 3% per year is incorporated into the model for all costs and outcomes.
Table II.
Costs
| Cost | Medicare Costa | Reference |
|---|---|---|
| Diabetes care/year | $2250 | 12 |
| Hypertension care/year | $587 | 11 |
| Heart failure year 1 | $34,110 | 24 |
| Heart failure year 2 | $10,154 | 24 |
| Heart failure year 3 | $8426 | 24 |
| Heart failure year 4 | $6085 | 24 |
| Heart failure year ≥5 | $3648 | 24 |
| Heart failure death | $5000 | Estimate |
| Stroke | $45,244 | 25 |
| Stroke follow‐up/year | $2612 | 25 |
| Stroke death | $5000 | Estimate |
| Myocardial infarction | $10,894 | 26 |
| Myocardial infarction death | $19,523 | 26 |
| Renal failure/year | $51,282 | 27 |
| Renal failure death | $64,619 | 27 |
| Other death | $5000 | Estimate |
a2007 US dollars.
Quality of Life
The quality of life (QOL) of a person depends on their state of health. For example, an individual who is dead has a QOL of 0. A person in perfect health has a QOL of 1. Disease states that are intermediate of perfect health and death are listed in Table III.
Table III.
Quality of Life
Analysis
The main results are reported as QALYs, total health‐related costs, incremental effectiveness, and cost of the HMP per month per person. The reported QALYs are the total amount of QALYs per group for the period shown. The total health‐related costs include costs for standard of care, morbidity, and mortality. The total health‐related costs do not include programmatic costs for the HMP since the aim is to determine the cost of the HMP. The cohort was not further stratified into different age groups, as third‐party payers may disseminate HMP to all of their members who fit this disease profile. Conducting the analyses in cohorts that are more at risk (eg, higher age ranges) would provide for less conservative results.
The control and HMP group results yield both costs and QALYs. In order to determine the cost of an HMP based on a neutral return on investment, the maximum cost of the HMP is equal to the incremental benefits (total health‐related costs and QALYs) derived from the HMP compared with the control program. The benefit from total health‐related costs is calculated by finding the difference of the costs between the HMP and control groups. The benefit from QALY is calculated by multiplying the incremental effectiveness or difference in QALYs by an incremental cost‐effectiveness ratio of $50,000/QALY. Thus the incremental benefit of the HMP group compared with the control group consisted of the differences in the total health‐related costs and QALYs. To determine the cost of the HMP, the incremental benefit is divided by the number of people in the model and the number of months in the time horizon.
Sensitivity analyses were conducted according to published literature on the various disease states and substates, costs, and QOL (Table IV) to increase generalizability of this model. For the disease states and substates, the confidence interval from the relative risk was used to generate the best‐case and worse‐case values for the HMP group. One‐way sensitivity analysis was conducted on the base case model on the variables indicated in Table IV. In these analyses, only one variable at a time was modified for the 5‐year time horizon. If the model was sensitive to the changes that were made in the sensitivity analyses, threshold analysis and 2‐way sensitivity analysis was conducted.
Table IV.
Sensitivity Analysis
| Variable | Sensitivity Analysis | References | |
|---|---|---|---|
| Best‐Case | Worst‐Case | ||
| Discount rate | 0 | 10% | Estimate |
| Disease states | |||
| Heart failure in HMP | 1.62/1000 pya | 7.61/1000 py | 11 |
| Nonfatal stroke in HMP | 2.49/1000 py | 9.97/1000 py | 11 |
| Fatal stroke in HMP | 0.47/1000 py | 4.79/1000 py | 11 |
| Nonfatal MI in HMP | 4.85/1000 py | 16.14/1000 py | 11 |
| Fatal MI in HMP | 5.93/1000 py | 16.70/1000 py | 11 |
| Renal failure in HMP | 0.35/1000 py | 5.08/1000 py | 11 |
| Death from renal failure in HMP | 0.03/1000 py | 3.66/1000 py | 11 |
| Other cause of death in HMP | 2.44/1000 py | 48.8/1000 py | Estimate |
| Quality of life | |||
| Stroke (moderate) | 0.81 | 0.12 | 31 |
| Heart failure | 0.91 | 0.3 | 30 |
| MI | 0.99 | 0.3 | 30 |
| Renal failure | 0.84 | 0.43 | 30 |
| Cost | |||
| Heart failure year 1 | $17,869 | $39,026 | 32, 33, 34 |
| Heart failure death | $13,700 | $17,100 | 34, 35, 36 |
| Stroke | $7945 | $90,488 | 37, estimate |
| Stroke follow‐up | $261 | $10,655 | 38 |
| Stroke death | $13,700 | $17,100 | 34, 35, 36 |
| MI | $6100 | $15,800 | 38 |
| MI death | $1952 | $39,046 | Estimate |
| Renal failure | $5128 | $98,100 | Estimate, 39, 40, 41 |
| Renal failure death | $6462 | $115,200 | Estimate, 34, 35, 36, 39, 40, 41 |
| Other death | $13,700 | $17,100 | 34, 35, 36 |
Abbreviations: HMP, hypertension management program; MI, myocardial infarction. aCases per person‐years (py).
Results
Baseline Results
Our Markov decision analytic model generates costs and QALYs in both the HMP and control group. The total health‐related costs, incremental costs, and QALYs of 1000 patients with hypertension and diabetes in the HMP and control group for the 1‐year, 3‐year, and 5‐year time horizon are listed in Table V. The results show that the HMP group dominated the control group in all specified time horizons when programmatic costs were excluded. That is, the HMP group compared with the control group was associated with greater gains in effectiveness (QALYs) and lower total health‐related costs in the 1‐year, 3‐year, and 5‐year time horizon.
Table V.
Baseline Results
| Strategy | Effectiveness (QALY) | Incremental Effectiveness | Total Health‐Related Costs, $a | Incremental Cost, $ | Cost of HMP per Month per Person, $ |
|---|---|---|---|---|---|
| 5 Years | |||||
| HMP | 4571 | 6055.73 | 159 | ||
| Control | 4426 | −145 | 8325.60 | 2269.87 | |
| 3 Years | |||||
| HMP | 2848 | 4513.54 | 109 | ||
| Control | 2796 | −52 | 5827.66 | 1314.12 | |
| 1 Year | |||||
| HMP | 984 | 3090.00 | 41 | ||
| Control | 979 | −5 | 3337.56 | 247.56 | |
Abbreviations: HMP, hypertension management program; QALY, quality‐adjusted life‐years. a2007 US dollars (thousands).
With a cohort of 1000 individuals in the HMP group for a 5‐year time horizon, there were 6 strokes, 19 myocardial infarctions, 4 heart failures, 1 case of end‐stage renal disease, and a total of 24 deaths across all groups. Similarly, in the control group for a 5‐year time horizon, there were 12 strokes, 23 myocardial infarctions, 8 heart failures, 2 cases of end‐stage renal disease, and a total of 29 deaths across all groups.
Cost of HMP
The cost of the HMP per month per person is calculated based on a neutral return on investment. Therefore, a third‐party payer could invest as much as $159, $109, and $41 per person per month in an HMP for the 5‐year, 3‐year, and 1‐year time horizon, respectively. At 5 years, if the third‐party payer invests $159 per person per month, the intervention will not incur any additional costs due to the benefits of the program. In addition, at that time we will have hypothetically prevented 6 strokes, 4 myocardial infarctions, 4 heart failures, and 1 case of end‐stage renal disease. The HMP would provide a favorable return on investment when a third‐party payer invests any amount less than shown and is able to achieve the same health outcomes.
Sensitivity Analysis
I, II illustrate the base case values used for the model. The ranges for the sensitivity analyses for selected incidence rates, costs, discount rate, and QOL are reported in Table IV. The HMP group continued to have greater gains in effectiveness (QALYs) and lower total health‐related costs than the control group when sensitivity analyses were conducted on costs, discount rate, and QOL at the 5 year time horizon. When sensitivity analysis was conducted on the rates for the health states and substates listed in Table IV, the Markov model was not sensitive to these changes except for the variable of other causes of death.
Threshold analysis was conducted on the variable of other causes of death. While keeping other variables constant at their base case values, when the variable in the HMP group was >41.7/1000 person‐years, the HMP group was not cost‐effective compared with the control group using $50,000/QALY as the threshold. Similarly, when other causes of death in the control group was <11/1000 person‐years, the HMP group was not cost‐effective as compared with the control group. A 2‐way sensitivity analysis of other causes of death showed that the control group becomes cost‐effective as compared with the HMP group only when other causes of death in the HMP group was >30/1000 person‐years and other causes of death in the control group was <20/1000 person‐years.
Discussion
HMPs in various forms, such as intense case management, dietary changes, physical fitness, and electronic medical reminders have been studied as a means to lower BP. 4 , 6 , 8 Many of these aforementioned methods have been used in combination with pharmacologic medications to lower BP. Decreases in systolic BP have been associated with decreased morbidity and mortality. The aim of our study was to determine the costs of a hypothetical HMP in a population with newly diagnosed diabetes and previously diagnosed hypertension that would result in a neutral return on investment.
At the 1‐year, 3‐year, and 5‐year time horizons, the HMP group dominated the control group when programmatic costs were not included. The HMP group had greater gains in effectiveness (QALYs) and lower total health‐related costs than the control group. The HMP group was thus associated with lower mortality and lower morbidity. For example, in the 5‐year time horizon, the HMP group was associated with a total of 24 deaths compared with 29 deaths in the control group. A third‐party payer could invest as much as $41, $109, and $159 per person per month in an HMP for the 1‐year, 3‐year, and 5‐year time horizon, respectively. When less money is invested than shown into the HMP, yet results in the same gains in effectiveness, the HMP group dominates the control group and results in cost‐savings.
In most cases, open enrollment for the selection of health plans occurs once a year. Because of this, many third‐party payers would like to recoup their investments within the same year. The time horizon of 1 year is selected to reflect the concern that investments into an HMP may not be recouped if there is a substantial attrition rate from that third‐party payer. However, not all individuals leave their insurance plans after 1 year, and therefore a 3‐year and 5‐year time horizon was created. Depending on the attrition rate of the third‐party payer, a different time horizon may be selected. This model shows that as the time horizon increases, more money can be invested into the HMP for a neutral return on investment.
The baseline Markov model was not sensitive to changes that were made in costs, discount rate, or QOL. In addition, the model was not sensitive to all of the disease states and substates, except for other causes of death. The 2‐way sensitivity analysis revealed that there must have been a greater number of deaths that were attributed to other causes in the HMP group compared with the control group in order for the HMP group to become less favorable. However, there is no valid rationale for this to be true.
This model demonstrates that HMPs are economically feasible. Previous studies have identified declines in BP in nonpharmacologically based hypertension programs. It appears that programs aimed at lowering BP with a variety of methods may have the best potential. 8 , 9 Substantial benefits accrued with a program that provided physician education through an educational e‐mail regarding BP, electronic reminders for specific patients, and patient education. 8 Programs such as this do not appear to require a high investment in capital. However, trials that have comprehensive plans from experts for losing weight, exercising, and decreasing alcohol and sodium intake may require a substantial amount of capital to achieve these results. 4 Programs that are able to lower BP and will not necessitate a substantial amount of capital may be more favorable. These programs may include nurse telemonitoring of a patient’s BP, pharmacist advisement programs, fitness prescriptions, and culturally sensitive hypertension programming. 5 , 6 , 7 , 9 , 15 , 16
A limitation of this study was that it was based on data from the UKPDS 38 and not from current trials that follow the JNC7 guidelines. However, the UKPDS studies are some of the most comprehensive studies that look at major morbidity and mortality; more recent studies have not yet been completed that follow the current guidelines. Thus, it is not known whether the investment strategy that we determined for third‐party payers can be a cost‐neutral proposition in the era of JNC7.
The second limitation of the study is that we assume health outcomes associated with decreasing BP from antihypertensive medication would result in the same outcomes as an HMP program. Recent successes of non‐pharmacologic interventions such as salt reduction programs have decreased cardiovascular events. 15 A meta‐analysis observed that different drug regimens for lowering BP do not appear to have differing effects on major cardiovascular events. 17 If cardiovascular outcomes are not dependent upon drug class, then there may be no significant difference in health outcomes between pharmacologic and non‐pharmacologic reductions of BP. However, nonpharmacologic interventions usually do not reduce BP to as great a degree or in as many people as pharmacologic therapy.
The third limitation of the study is that it did not focus on the potential adverse effects of the pharmacologic agents used in the standard of care or change in QOL associated with an HMP. However, recent studies have noted that individuals taking medication for BP, cholesterol, and diabetes did not have a significantly different QOL than individuals who were not taking medications. 18 In fact, QOL actually improves while taking prescribed drugs if BP is reduced. Patients who were randomized to intensive therapy for BP, diabetes, and dyslipidemia consisting of diet and physical exercise also experienced an improvement in QOL compared with controls. 19 Our model is conservative in that it did not account for an improved QOL for patients in the HMP.
The last limitation of this study is that the data are focused on a limited population who were newly diagnosed with diabetes and had existing diagnoses of hypertension. It is not known whether these data are readily generalizable to a patient with long‐standing diabetes and hypertension. However, patients with long‐standing diabetes or hypertension are more likely to have increased risk for morbidity and mortality. If the study were to focus on a population with increased risk for mortality and morbidity, this would most likely result in a less conservative model.
The study showed that the HMP group dominated the control group in most scenarios; the HMP group had greater gains in QALYs and lower total health‐related costs. The model demonstrates that as the time horizon increases, the favorability of the HMP group increases. That is, as the attrition rate decreases, more money can be invested into the HMP for a neutral return on investment. Less money can be invested in the HMP, if the HMP continues to yield similar health outcomes. Therefore, as the attrition rate of a third‐party payer decreases, the HMP may result in cost‐savings. Third‐party payers may wish to focus on programs that require less capital; such programs may include diet and exercise advice, culturally sensitive hypertension programming, case management, pharmacist advisement, and fitness prescriptions. Despite investments into HMP, third‐party payers must also institute strategies to administer cost‐effective pharmacologic therapies to lower BP. Overall, investment in an HMP by third‐party payers can be a cost‐effective and cost‐neutral proposition.
References
- 1. Fields LE, Burt VL, Cutler JA, et al. The burden of adult hypertension in the united states 1999 to 2000: a rising tide. Hypertension. 2004;44:398–404. [DOI] [PubMed] [Google Scholar]
- 2. Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure. Hypertension. 2003;42:1206–1252. [DOI] [PubMed] [Google Scholar]
- 3. Lewington S, Clarke R, Qizilbash N, et al. Age‐specific relevance of usual blood pressure to vascular mortality: a meta‐analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002;360:1903–1913. [DOI] [PubMed] [Google Scholar]
- 4. Appel LJ, Champagne CM, Harsha DW, et al. Effects of comprehensive lifestyle modification on blood pressure control: main results of the PREMIER clinical trial. JAMA. 2003;289:2083–2093. [DOI] [PubMed] [Google Scholar]
- 5. Artinian NT, Flack JM, Nordstrom CK, et al. Effects of nurse‐managed telemonitoring on blood pressure at 12‐month follow‐up among urban african americans. Nurs Res. 2007;56:312–322. [DOI] [PubMed] [Google Scholar]
- 6. Lee JK, Grace KA, Taylor AJ. Effect of a pharmacy care program on medication adherence and persistence, blood pressure, and low‐density lipoprotein cholesterol: a randomized controlled trial. JAMA. 2006;296:2563–2571. [DOI] [PubMed] [Google Scholar]
- 7. Petrella RJ, Koval JJ, Cunningham DA, et al. Can primary care doctors prescribe exercise to improve fitness? The step test exercise prescription (STEP) project. Am J Prev Med. 2003;24:316–322. [DOI] [PubMed] [Google Scholar]
- 8. Roumie CL, Elasy TA, Greevy R, et al. Improving blood pressure control through provider education, provider alerts, and patient education: a cluster randomized trial. Ann Intern Med. 2006;145:165–175. [DOI] [PubMed] [Google Scholar]
- 9. Rudd P, Miller NH, Kaufman J, et al. Nurse management for hypertension. A systems approach. Am J Hypertens. 2004;17:921–927. [DOI] [PubMed] [Google Scholar]
- 10. Elliott WJ, Weir DR, Black HR. Cost‐effectiveness of the lower treatment goal (of JNC VI) for diabetic hypertensive patients. Joint national committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med. 2000;160:1277–1283. [DOI] [PubMed] [Google Scholar]
- 11. UK prospective diabetes study group . Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ. 1998;317:703–713. [PMC free article] [PubMed] [Google Scholar]
- 12. CDC diabetes cost‐effectiveness group . Cost‐effectiveness of intensive glycemic control, intensified hypertension control, and serum cholesterol level reduction for type 2 diabetes. JAMA. 2002;287:2542–2551. [DOI] [PubMed] [Google Scholar]
- 13. Mitka M. Many physician practices fall short on accurate blood pressure measurement. JAMA. 2008;299:2842–2844. [DOI] [PubMed] [Google Scholar]
- 14. Houweling ST, Kleefstra N, Lutgers HL, et al. Pitfalls in blood pressure measurement in daily practice. Fam Pract. 2006;23:20–27. [DOI] [PubMed] [Google Scholar]
- 15. Cook NR, Cutler JA, Obarzanek E, et al. Long term effects of dietary sodium reduction on cardiovascular disease outcomes: observational follow‐up of the trials of hypertension prevention (TOHP). BMJ. 2007;334:885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Paschal AM, Lewis RK, Martin A, et al. Evaluating the impact of a hypertension program for african americans. J Natl Med Assoc. 2006;98:607–615. [PMC free article] [PubMed] [Google Scholar]
- 17. Turnbull F, Neal B, Algert C, et al. Effects of different blood pressure‐lowering regimens on major cardiovascular events in individuals with and without diabetes mellitus: results of prospectively designed overviews of randomized trials. Arch Intern Med. 2005;165:1410–1419. [DOI] [PubMed] [Google Scholar]
- 18. U.K. prospective diabetes study group . Quality of life in type 2 diabetic patients is affected by complications but not by intensive policies to improve blood glucose or blood pressure control (UKPDS 37). Diabetes Care. 1999;22:1125–1136. [DOI] [PubMed] [Google Scholar]
- 19. Menard J, Payette H, Dubuc N, et al. Quality of life in type 2 diabetes patients under intensive multitherapy. Diabetes Metab. 2007;33:54–60. [DOI] [PubMed] [Google Scholar]
- 20. Owan TE, Hodge DO, Herges RM, et al. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med. 2006;355:251–259. [DOI] [PubMed] [Google Scholar]
- 21. PROGRESS Collaborative Group . Randomised trial of a perindopril‐based blood‐pressure‐lowering regimen among 6105 individuals with previous stroke or transient ischaemic attack. Lancet. 2001;358:1033–1041. [DOI] [PubMed] [Google Scholar]
- 22. Rosamond WD, Chambless LE, Folsom AR, et al. Trends in the incidence of myocardial infarction and in mortality due to coronary heart disease, 1987 to 1994. N Engl J Med. 1998;339:861–867. [DOI] [PubMed] [Google Scholar]
- 23. Abizaid A, Costa MA, Centemero M, et al. Clinical and economic impact of diabetes mellitus on percutaneous and surgical treatment of multivessel coronary disease patients: insights from the arterial revascularization therapy study (ARTS) trial. Circulation. 2001;104:533–538. [DOI] [PubMed] [Google Scholar]
- 24. Liao L, Jollis JG, Anstrom KJ, et al. Costs for heart failure with normal vs reduced ejection fraction. Arch Intern Med. 2006;166:112–118. [DOI] [PubMed] [Google Scholar]
- 25. Brown DL, Boden‐Albala B, Langa KM, et al. Projected costs of ischemic stroke in the united states. Neurology. 2006;67:1390–1395. [DOI] [PubMed] [Google Scholar]
- 26. Kaul P, Armstrong PW, Cowper PA, et al. Economic analysis of the assessment of the safety and efficacy of a new thrombolytic regimen (ASSENT‐3) study: costs of reperfusion strategies in acute myocardial infarction. Am Heart J. 2005;149:637–644. [DOI] [PubMed] [Google Scholar]
- 27. Matas AJ, Schnitzler M. Payment for living donor (vendor) kidneys: a cost‐effectiveness analysis. Am J Transplant. 2004;4:216–221. [DOI] [PubMed] [Google Scholar]
- 28. Wexler DJ, Grant RW, Wittenberg E, et al. Correlates of health‐related quality of life in type 2 diabetes. Diabetologia. 2006;49:1489–1497. [DOI] [PubMed] [Google Scholar]
- 29. Mathias SD, Bates MM, Pasta DJ, et al. Use of the health utilities index with stroke patients and their caregivers. Stroke. 1997;28:1888–1894. [DOI] [PubMed] [Google Scholar]
- 30. Tengs TO, Wallace A. One thousand health‐related quality‐of‐life estimates. Med Care. 2000;38:583–637. [DOI] [PubMed] [Google Scholar]
- 31. Tengs TO, Yu M, Luistro E. Health‐related quality of life after stroke a comprehensive review. Stroke. 2001;32:964–972. [DOI] [PubMed] [Google Scholar]
- 32. Kasper EK, Gerstenblith G, Hefter G, et al. A randomized trial of the efficacy of multidisciplinary care in heart failure outpatients at high risk of hospital readmission. J Am Coll Cardiol. 2002;39:471–480. [DOI] [PubMed] [Google Scholar]
- 33. Delea TE, Vera‐Llonch M, Richner RE, et al. Cost effectiveness of carvedilol for heart failure. Am J Cardiol. 1999;83:890–896. [DOI] [PubMed] [Google Scholar]
- 34. Emanuel EJ, Ash A, Yu W, et al. Managed care, hospice use, site of death, and medical expenditures in the last year of life. Arch Intern Med. 2002;162:1722–1728. [DOI] [PubMed] [Google Scholar]
- 35. Hoover DR, Crystal S, Kumar R, et al. Medical expenditures during the last year of life: findings from the 1992–1996 medicare current beneficiary survey. Health Serv Res. 2002;37:1625–1642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Levinsky NG, Yu W, Ash A, et al. Influence of age on medicare expenditures and medical care in the last year of life. JAMA. 2001;286:1349–1355. [DOI] [PubMed] [Google Scholar]
- 37. Lee WC, Christensen MC, Joshi AV, et al. Long‐term cost of stroke subtypes among medicare beneficiaries. Cerebrovasc Dis. 2007;23:57–65. [DOI] [PubMed] [Google Scholar]
- 38. Chan PS, Nallamothu BK, Gurm HS, et al. Incremental benefit and cost‐effectiveness of high‐dose statin therapy in high‐risk patients with coronary artery disease. Circulation. 2007;115:2398–2409. [DOI] [PubMed] [Google Scholar]
- 39. Boulware LE, Jaar BG, Tarver‐Carr ME, et al. Screening for proteinuria in US adults: a cost‐effectiveness analysis. JAMA. 2003;290:3101–3114. [DOI] [PubMed] [Google Scholar]
- 40. Lee H, Manns B, Taub K, et al. Cost analysis of ongoing care of patients with end‐stage renal disease: the impact of dialysis modality and dialysis access. Am J Kidney Dis. 2002;40:611–622. [DOI] [PubMed] [Google Scholar]
- 41. Mohr PE, Neumann PJ, Franco SJ, et al. The case for daily dialysis: its impact on costs and quality of life. Am J Kidney Dis. 2001;37:777–789. [DOI] [PubMed] [Google Scholar]