Long-Term Outcomes of a Cardiovascular and Diabetes Risk-Reduction Program Initiated by a Self-Insured Employer

Abstract

Background

Cardiovascular disease remains the leading cause of death in America and poses a significant challenge for self-insured employers attempting to improve employee health and well-being while controlling healthcare costs. Disease state management programs can be an effective means of achieving these outcomes, but the durability and long-term effects of such programs have limited evaluation.

Objective

To assess the 5-year health, economic, and quality-of-life patient outcomes of an employer-sponsored disease state management program.

Methods

This was a longitudinal, 5-year, quasi-experimental, pre-/postenrollment study. Self-insured health plan members with hypertension, hyperlipidemia, diabetes, or a combination of these conditions met with a pharmacist regularly (monthly for the first year, then varied by participant) to implement lifestyle medicine programs, optimize medication therapy, and facilitate the coordination of care. Biometric markers, lifestyle behaviors, quality of life, and work productivity were assessed on an annual basis.

Results

The significant biometric improvements (mean) seen after 5 years of program participation compared with pre-enrollment included decreased low-density lipoprotein cholesterol levels (96.71 mg/dL vs 84.83 mg/dL, respectively), increased high-density lipoprotein cholesterol levels (39.32 mg/dL vs 46.12 mg/dL), and decreased systolic blood pressure (132.04 mm Hg vs 123.63 mm Hg) and diastolic blood pressure (85.75 mm Hg vs 75.83 mm Hg). The average exercise time increased (50 minutes weekly vs 156.04 minutes weekly), as did fruit and vegetable consumption (3.98 servings daily vs 5.27 servings daily). The program participants reported improved general health and a reduced number of unhealthy days. The combined healthcare and productivity return on investment for the program at 5 years was $9.64 for every $1 invested.

Conclusions

Significant changes in employees' health, well-being, and health-related costs are possible through sustained participation in an employer-sponsored disease state management program.

Cardiovascular disease (CVD) remains the leading cause of death in America, resulting in approximately 610,000 deaths each year.1 The prevalence of CVD in the United States is increasing. The American Heart Association projects that by 2030, approximately 40.5% of the US population will have CVD.2 The costs associated with the treatment of heart disease and other chronic conditions are increasing. In the previous decade, medical costs associated with CVD have increased by 6% annually, and currently comprise approximately 17% of the nation's healthcare expenditure.2 People with chronic conditions report lower quality of life (QOL) and more unhealthy days compared with individuals without chronic disease.3 These statistics pose a significant challenge for employers attempting to control healthcare costs and improve productivity within their organization.

The Cardiovascular and Diabetes Risk-Reduction Program is a pharmacist-led disease management program designed to decrease the risk for expensive adverse health outcomes and improve the overall health, QOL, and productivity of employees of a self-insured employer. Health risk assessment data for the employer were used to determine the most common and costly conditions to target through program intervention, namely, CVD (ie, hypertension and dyslipidemia) and diabetes.

KEY POINTS

• ▸Disease management programs can improve employee health and control costs, but their long-term effects have not been evaluated.
• ▸This longitudinal study assessed the 5-year health, economic, and quality-of-life outcomes of an employer-sponsored disease management program.
• ▸After 5 years (but not after 1 year) of program participation, average LDL levels were significantly lower (–11.88; P = .04) and HDL levels significantly higher (+6.8; P <.01).
• ▸Similarly, after 5 years, systolic and diastolic blood pressure levels were significantly lower (–8.41 mm Hg; P = .01; –9.92 mm Hg; P <.01, respectively).
• ▸Average exercise time increased by 106 minutes weekly (P <.01), and fruit and vegetable consumption by 1.29 servings daily (P = .05).
• ▸Participants reported better general health and a reduced number of unhealthy days after 5 years in the program.
• ▸The combined healthcare and productivity ROI for the program was $9.64 for every $1 spent at 5 years.
• ▸These improvements were durable, or were enhanced, after 5 years, which supports long-term use of such programs.

The program employs a pharmacist in the position of “ambulatist” in charge of daily operations and regular monthly check-ups with the participating employees. The ambulatist works in collaboration with an interprofessional care team consisting of a dietitian, exercise physiologist, health educator, licensed mental health provider, and the employee's primary care physician and specialty physicians (ie, endocrinologist and/or cardiologist, if applicable).

Employees voluntarily enroll in the program with the incentive of receiving personalized health and wellness coaching and free medications on entry. Eligibility requirements include full-time employment with the program sponsor; a standing diagnosis of diabetes, hyperlipidemia, hypertension, or a combination of these; and currently receiving healthcare benefits from the employer.

The participants in the risk-reduction program attend regular (ie, at least once monthly in the first year, spread out to no less than quarterly as control of conditions is established) one-on-one appointments with a pharmacist. The visits consist of medication therapy management, implementation and adherence to 7 personalized lifestyle medicine programs (ie, physical activity, healthy eating, stress management, restorative sleep, moderate alcohol consumption, tobacco abstinence/cessation, and weight control), and chronic disease care coordination practices.

To achieve the highest level of program adherence and success, each participant is provided with educational materials, a home blood pressure monitor, a pedometer, lifestyle behavior tracking tools, free access to employer exercise facilities, monthly support group meetings, and access to a licensed mental healthcare provider. In addition, employees with diabetes received an initial consultation with a dietitian, 6 hours of American Diabetes Association–approved education classes, and access to point-of-care glycated hemoglobin (HbA_1c) analyses as appropriate.

The program was initiated in the fall of 2008 at a private Midwestern university, and is offered at no charge to all eligible employees. The health effects of program participation were assessed after 1 year of participation and have been reported previously.4–7 Specifically, the 1-year health-related QOL (HRQoL) improved by 20.6% (P <.001), and the number of self-reported unhealthy days (physical and mental) decreased by 42.5% (P <.01). Cardiovascular risk (ie, general 10-year) decreased by 2.02% (P = .017), and a correlated heart and vascular age estimation decreased by 2.7 years (P = .004). Participation in the lifestyle medicine activities of exercise, fruit and vegetable intake, and stress reduction significantly improved (P <.01). Medication adherence improved 15% (P <.001), and the financial return on investment (ROI) was $4.02:$1. The current study aims to assess the durability of these outcomes based on additional data representing 5 years of program intervention.

Methods

The risk-reduction program began enrollment in August 2008. At baseline and annually thereafter for the next 5 years, the participants were assessed for changes in biometric data (ie, weight, waist circumference, blood pressure, lipid panel, HbA_1c), lifestyle habits (ie, tobacco use, physical activity, fruit and vegetable intake, sleep quantity, stress rating), HRQoL, and productivity.

Contemporary methods of measurement were utilized to obtain outcomes data. Employee self-reported HRQoL is measured using a validated survey tool from the Centers for Disease Control and Prevention.8 This survey tool has been used since 1993 to measure HRQoL in the Behavior Risk Factor Surveillance System, and since 2000 in the National Health and Nutrition Examination Survey.8 The 4-question survey assesses self-reported general health and the number of physically and mentally unhealthy days experienced in the past month.

Cardiovascular risk was calculated using the general CVD (10-year risk) calculator developed by the Framingham Heart Study.9 This tool predicts the risk for experiencing one of several CVD outcomes (ie, coronary death, myocardial infarction, coronary insufficiency, angina, ischemic stroke, hemorrhagic stroke, transient ischemic attack, peripheral arterial disease, or heart failure) within the next 10 years. The tool can also be used to estimate heart and vascular age. The variables that are used in the calculation include age, sex, systolic blood pressure, treatment for hypertension, smoking status, diabetes diagnosis, high-density lipoprotein (HDL) cholesterol, and total cholesterol.

Lifestyle habits were collected via participant self-reporting in a lifestyle journal that was provided to each participant by the program.10 The participants use the journal to regularly track their amount of exercise, fruit and vegetable consumption, stress rating, amount of sleep, alcohol consumption, and tobacco use (in addition to other measures). The participants were required to track these behaviors regularly, and the lifestyle journal was reviewed at each monthly visit.

Work productivity was assessed in the study through the use of the Work Productivity and Activity Impairment questionnaire for general health.11 This validated tool tracks time away from work (ie, absenteeism), as well as decreased job performance while at work (ie, presenteeism).11

Finally, the ROI analysis was completed by a third-party company, Health Improvement Solutions.12 The Health Improvement Solutions ROI calculator helps to determine the effectiveness of programs by evaluating cost-savings from risk reduction and program investment within the organization. The tool uses medical and productivity risk factor costs identified through a combination of research and an extensive proprietary risk factor cost database. Along with risk prevalence, Health Improvement Solutions includes program costs, such as staffing, administrative fees, consulting, and other investment costs. The median annual compensation is used to determine the health-related work productivity portion of the ROI. The program's ROI is provided in 4 outputs: medical only, absenteeism only, presenteeism only, and overall ROI (medical and productivity combined).

Statistical Analysis

A statistical analysis was completed using the Wilcoxon signed-rank test as a nonparametric test to compare the median difference between the baseline and 5-year time points. A P value of <.05 was considered statistically significant. Descriptive analyses were also used to compare the number and percentage of employees participating in each of the lifestyle medicine activities.

Results

As of May 2016, 25 employees had completed at least 5 years of program participation and were included for assessment in the current study. The participants' baseline data are described in Table 1.

Table 1.

5-Year Program Participant Baseline Data

Cohort information	Participants (N = 25)
Female sex, N (%)	13 (52)
Male sex, N (%)	12 (48)
Average age at program initiation, yrs	53.04
Hypertension, N (%)	20 (80)
Hyperlipidemia, N (%)	18 (72)
Diabetes, N (%)	11 (44)

After 5 years of program intervention, significant improvements were achieved in various biometric measures. Specifically, reductions in low-density lipoprotein (LDL) and non-HDL cholesterol, as well as systolic and diastolic blood pressure, were demonstrated. A significant increase in HDL was observed. Cholesterol changes in the participants' biometric measures are summarized in Table 2.

Table 2.

Changes in Participant Biometric and Lifestyle Measures

Biometric and lifestyle data	Baseline	After 5 years in program	Change from baseline	P value
Total cholesterol, mg/dL	174.44	160.96	–13.48	.09
LDL cholesterol, mg/dL	96.71	84.83	–11.88	.04
HDL cholesterol, mg/dL	39.32	46.12	6.8	<.01
Triglycerides, mg/dL	171.40	136.28	–35.12	.11
Non-HDL, mg/dL	134.79	111.50	–23.29	<.01
Fasting blood sugar, mg/dL	127.68	109.56	–18.12	.23
HbA1c, %	7.42	7.36	–0.06	.90
Systolic BP, mm Hg	132.04	123.63	–8.41	.01
Diastolic BP, mm Hg	85.75	75.83	–9.92	<.01
Weight, lbs	212.12	209.71	–2.41	.35
Height, in	67.19	66.96	–0.23	.08
Body mass index, kg/m2	33.01	32.28	–0.73	.24
Cardiovascular event risk in the next 10 years	14.25	12.67	–1.58	.34
Heart age, yrs	64.56	62.56	–2	.51
Exercise, min/wk	50.00	156.04	106.04	<.01
Fruit and vegetable consumption, servings/day	3.98	5.27	1.29	.05
Sleep, hrs	6.95	6.85	–0.1	.68
Stress (1–5)	3.25	2.42	–0.83	<.01

BP indicates blood pressure; HbA_1c, glycated hemoglobin; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Several lifestyle medicine–related cardiovascular risk factors were also significantly improved. The participants' exercise time increased, as did their fruit and vegetable consumption. Furthermore, stress levels were measured on a scale of 1 to 5, with 1 indicating low stress (ie, feeling calm and in control), 3 indicating moderate stress, and 5 indicating high stress (ie, feeling frantic and out of control), and the participants' scores decreased significantly. Changes in participant lifestyle behavior are summarized in Table 2.

The program participants reported a significantly better general health rating after 5 years of participation, using a simple rating scale (1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor). Similarly, participants reported statistically significant changes in the average number of unhealthy days they experienced in 1 month. The changes in participant-reported HRQoL are summarized in Table 3.

Table 3.

Changes in Program Participant Quality of Life

Quality-of-life data	Baseline	After 5 years in program	Change from baseline	P value
Self-reported general health ratinga	3.09	2.55	–0.54	.01
Physically unhealthy days in the past month, N	4.19	1.61	–2.58	.07
Mentally unhealthy days in the past month, N	6.50	2.83	–3.67	.04
Total unhealthy days (physical + mental) in the past month, N	9.39	4.35	–5.04	.02
Days with usual activities, self-care, and recreation impeded by physical/mental health in the past month, N	1.83	0.39	–1.44	.16

^a1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor.

The healthcare cost ROI for the 5-year cohort was 3.85:1 across the 5-year period. When looking at the ROI for productivity, the ratio was 1.36:1 for absenteeism, 4.43:1 for presenteeism, and 5.79:1 for total productivity (ie, absenteeism plus presenteeism). When healthcare and productivity savings were combined, the ROI was $9.64:$1, meaning that over the 5 years of program participation, this cohort eliminated health risks that led to a savings of $9.64 for every $1 invested in their health through the program. The Figure depicts the results from the 5-year ROI analysis.

Figure. Return on Investment 5-Year Analysis.

Discussion

The participants in the risk-reduction program displayed various improvements in health, well-being, and work productivity after 5 years. Improvements in HRQoL, exercise, and fruit and vegetable intake, as well as in financial ROI, were demonstrated after 1 year of program participation and were durable after 5 years of follow-up.

Statistically significant improvements in LDL cholesterol, HDL and non-HDL cholesterol, as well as systolic and diastolic blood pressure, were demonstrated at the 5-year follow-up, but not after 1 year. This may suggest that the benefits of sustained lifestyle modification (ie, increased physical activity, improved nutrition) may require more than 1 year to manifest.

In addition to the significant findings previously mentioned, the risk-reduction program also showed considerable noteworthy improvements in health that were not statistically significant. For example, the average body weight decreased slightly (from 212.12 lbs at baseline to 209.71 lbs after 5 years). These findings, although not statistically significant, are noteworthy, considering that the average person gains approximately 3.35 pounds over a 4-year period.13

Another notable example includes the decrease in cardiovascular risk and heart age. After 5 years, the program participants displayed a decrease in calculated heart age from an average of 64.56 years to 62.56 years and a decrease in the calculated 10-year general CVD risk from 14.25% to 12.67%. Although neither reduction was statistically significant, both calculations are highly influenced by age and would therefore tend to increase over the course of 5 years.

Finally, participants in the risk-reduction program are more likely to meet disease and lifestyle behavior targets after participating in the program for 5 years. The changes in the achievement of guideline-based treatment goals are described in Table 4.14–19

Table 4.

Guideline-Based Chronic Condition Treatment Goal Achievement Rates

Chronic condition treatment goal	Patients achieving goal at baseline, N (%)	Patients achieving goal after 5-year program, N (%)	Change from baseline, N (%)
JNC 8 BP treatment goal for persons <60 yrs14: <140/90 mm Hg	13:24 (54)	23:24 (96)	10 (42)
ADA fasting plasma glucose diagnostic threshold15: <126 mg/dL	17:25 (68)	19:25 (76)	2 (8)
NLA desirable LDL cholesterol16: <100 mg/dL	12:25 (48)	15:25 (60)	3 (12)
BMI indicating normal weight17: <25 kg/m2	0:24 (0)	3:24 (12)	3 (12)
HHS recommended exercise time/week18: ≥150 minutes of moderate activity weekly	4:24 (17)	14:24 (58)	10 (41)
HHS recommended daily servings of combined fruits and vegetables19: ≥5 combined servings daily	9:23 (39)	14:23 (61)	5 (22)

ADA indicates American Diabetes Association; BMI, body mass index; BP, blood pressure; HHS, Department of Health & Human Services; JNC 8, Eighth Joint National Committee; LDL, low-density lipoprotein; NLA, National Lipid Association.

Limitations

These findings are limited by the small sample size and the noncontrolled study design.

Additional limitations include the volunteer status of participants and reliance on self-reported data. Individuals who actively seek out such programs are likely more inclined to make changes and adhere to therapy than the population as a whole.

Furthermore, lifestyle behavior data were obtained from a self-reported lifestyle journal.

The ROI data are uncharacteristically high compared with other disease management programs that have been published.20,21 The authors attempted to be as conservative as possible with the data included in the calculation and felt that having a third-party company perform the calculation was best to calculate an accurate assessment of ROI.

Conclusions

The results of the current study suggest that participation in a cardiovascular and diabetes risk-reduction program may improve participants' health, QOL, and productivity, while saving money for self-insured employers. Improvements in HRQoL, exercise, fruit and vegetable consumption, and financial ROI were evident after 1 year of participation and were maintained or further improved after 5 years of program participation.

Notably, statistically significant improvements in LDL, HDL and non-HDL cholesterol, as well as systolic and diastolic blood pressure, although not observed after 1 year of follow-up, were demonstrated at the 5-year follow-up. These results suggest that sustained participation in such a program has durable and additional benefits lasting during and after program participation. These outcomes support the long-term administration of such employer-based programs.