Original Articles

POPULATION HEALTH MANAGEMENT Volume 17, Number 6, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/pop.2014.0008

Examining Incentive Design Strategies for Worksite Wellness Program Engagement Gregory J. Norman, PhD, Kevin J. Heltemes, MPH, and Joseph Drew, BA1

Abstract

The objective was to examine employee engagement in worksite wellness activities at 2 large US companies that differed in engagement strategy and incentive plan. Inclusion criteria were US employees aged 18 to 65 who were eligible to receive wellness benefits throughout 2012. Company B’s incentive was twice the dollar value of Company A’s and produced higher engagement rates for the health assessment (HA; 26.1% vs. 24.4%, P < .001), and biometric screening (32.8% vs. 25.4%, P < .001). Among the subgroup of employees who completed the HA and the biometric screening, 44.6% (N = 2,309) at Company A engaged in at least 1 coaching session compared to 8.9% (N = 594) at Company B. Fewer employees at Company A with high-risk cholesterol engaged in coaching compared to Company B (44.6% vs. 54.9%, P = .009). However, more Company A employees with high-risk blood pressure engaged in coaching compared to Company B (41.3% vs. 34.8%, P = .053). Company A engaged more obese employees compared to Company B (43.7% vs. 13.9%, P < .001), although obesity was not directly targeted at either company. Predictors of enrolling in coaching included being female, older age, higher education, and those not at high risk for stress, diet, and tobacco for Company A, and older age, and high risk for blood pressure, cholesterol, and obesity for Company B. A population approach to incentive design for program engagement engaged high-risk employees in coaching, and engaged a high proportion of employees not at high risk, but who can still be at risk for chronic diseases. (Population Health Management 2014;17:324–331)

Introduction

C

hronic diseases are 5 of the 10 leading causes of death in the United States with heart disease, cancer, chronic lower respiratory diseases, stroke, and diabetes accounting for 64.5% of all US deaths in 2010.1 These chronic diseases are associated with medical condition risk factors of hypertension, hyperlipidemia, and obesity, all of which can be affected by lifestyle choices such as physical activity, diet, and tobacco use. Unfortunately, 50% of American adults do not meet public health guidelines for physical activity,2 up to 75% of adults report low fruit and vegetable intake,3 and about 20% of US adults smoke cigarettes. The 5 most costly and preventable chronic conditions cost the United States nearly $347 billion in 2010, which was 30% of total health spending.4 These costs are expected to continue to increase, highlighting the need for preventive interventions to slow the progression from risk to disease.5 Worksite health and wellness programs provide an intervention channel to help employees and spouses make healthy lifestyle changes.6–8 Employers are invested in 1

employees’ health-related measures, such as absenteeism, presenteeism, and disability,9 because they directly affect business productivity. In addition, employees’ lifestyle and biometric risks have been positively associated with health care costs.10 Worksite health and wellness program activities include health assessments (HAs), biometric screenings, and various intervention programs such as telephonic coaching and online health education courses. Lifestyle coaching, defined as educating and motivating individuals to address their health risks and make lifestyle modifications, has been associated with positive outcomes regarding dietary choices, sedentary behavior, obesity, body fat percentage, cholesterol, blood pressure, and achieving health goals.6,7,11 Greater incentives and organizational outreach have been associated with higher rates of HA completion12 and better engagement in telephone-based lifestyle coaching.13 As a result, worksite health and wellness programs have begun to offer incentives to employees during a defined period each year to motivate engagement in wellness program activities. The incentive period is defined as the window during which

Healthyroads, San Diego, California.

324

INCENTIVE DESIGN STRATEGIES FOR WORKSITE WELLNESS PROGRAMS

employees can engage in these activities to receive specific incentive rewards (eg, gift cards, cash, prizes, reduced health insurance premiums). The behavioral economic theory behind extrinsic motivators, such as financial incentives to motivate employees to engage in wellness activities, is that they help people make better choices without actually limiting those choices.14,15 Extrinsic incentives motivate behavior by making it easier (ie, more valued, preferred) for people to choose short-term actions that are consistent with their own long-term interests (eg, living healthy, living without chronic disease).14 This study examined employee engagement in wellness programs’ activities (ie, HA, biometric screening, telephone coaching) in relation to the incentive design at 2 large US companies of similar employee size and geographic distribution. Company A’s wellness program was designed to encourage telephonic coaching for all employees. Company B’s wellness program encouraged all employees with medical coverage to take the HA and biometric screening, but targeted employees at highest risk (ie, cholesterol ‡ 240 mg/dL, blood pressure ‡ 140/90 mm Hg) to engage in telephonic coaching and used a punitive incentive strategy to encourage participation. The incentive value for Company A was $150 payable immediately in a cash gift card; the incentive value for Company B was $300 paid in the future benefit year via reduced employee contributions for medical coverage. Worksite wellness program engagement is an important leading indicator of whether a program will reach its full potential and low rates of participation can undermine a program’s impact.16,17 Understanding worksite wellness program incentive design strategies is crucial to sufficient participation and engagement in wellness programs. The aims of the current study were to: (1) examine engagement rates between companies for HA and biometric screening among entirely benefit eligible (EBE) members, (2) assess enrollment in telephonic coaching between companies for those employees who completed the HA and biometric screening and who are high risk and not at high risk for blood pressure, total cholesterol, and obesity, and (3) determine predictors of engaging in lifestyle coaching within each company. Methods Participants

Two US companies, with over 50,000 combined medical benefit eligible employees and locations distributed throughout the United States were included in this study. The two companies were selected based on their comparable size and geographic distribution, but contrasting wellness incentive designs. There were 86,437 EBE members employed at the 2 companies during 2012. EBE was defined as having access to the health and wellness benefit throughout all of 2012. Non-US employees and those without medical benefits were excluded (n = 29,485). Additionally, to achieve a comparable sample for each company, members older than age 65 or missing gender data were excluded (n = 425). The study population included 56,503 US employees ages 18 to 65. The study population was identified from employee eligibility records, which were then linked to existing databases

325

including the HA, biometric screenings, and electronic coaching records to obtain occupational data, health risks, and telephonic coaching session information. Wellness program and incentive designs

Both companies contracted with Healthyroads (a subsidiary of American Specialty Health, Inc.) for their health and wellness program. Healthyroads program components for these companies included HAs, biometric screenings, telephonic coaching, and online educational materials and resources. Telephonic lifestyle coaching consisted of oneon-one coaching for weight management, stress management, tobacco cessation, and healthy living. Details of the 2 companies’ incentive designs are presented in Table 1. The incentive period for Company A was from April through December of 2012 with April through May as the window to complete the HA and January through May as the window to complete the biometric screening. Completing these activities in the defined windows resulted in receiving a $50 incentive distributed as a check or gift card. Based on the HA, employees were stratified into low, moderate, and high health risk categories and eligible to receive $100 if they completed the minimum number of coaching sessions for their risk category; the number of required sessions to receive the incentive was 4, 7, or 9 sessions, respectively. In this incentive design employees could receive a total of $150 dollars for completing the 3 wellness program tasks. Company B’s incentive period was the full 2012 calendar year. Employees were eligible to receive a $300 reduction in their 2013 medical premiums for completing the HA, biometric screening, and having blood pressure and total cholesterol values below the defined high-risk thresholds (Table 1). If an employee’s biometric screening indicated blood pressure of at least 140/90 mm/Hg or total cholesterol of at least 240 mg/dl, then they needed to complete at least 4 sessions of telephone coaching to receive the incentive as a reasonable alternative to having the elevated biometric values. Employees could complete HA and biometric screenings anytime during the incentive period. If an employee was required to complete telephone coaching to receive the incentive but did not enroll in coaching prior to December 20th, then the employee was informed he or she would not be able to complete the 4 coaching sessions prior to the end of the incentive period. The $300 dollar incentive was distributed biweekly in the employee’s 2013 paychecks (ie, the year following the 2012 incentive period), and equaled $11.54 per paycheck. Measures

Participant sex and age as of January 1, 2012 was determined from company benefit eligibility records. Age was dichotomized into 18 to 44 and 45 to 65 years old. Additional demographic information obtained from the HA included education level, job tenure, job type, and marital status. The Healthyroads HA measures lifestyle risks to identify high-risk lifestyle categories including sedentary behavior, poor diet, high stress, and tobacco use. Sedentary was defined as £ 10 minutes of moderate and no minutes of vigorous exercise per week. Poor diet was derived as £ 1 serving per

326

NORMAN, HELTEMES, AND DREW

Table 1. Incentive Design Characteristics by Company Company A Incentive type Timing to complete activities Dispensing incentives Incentive contingencies

Number of lifestyle coaching sessions

Probability of distribution

Company B

Gift card or check 2-month window to complete HA and biometric screening. 9-month window to complete coaching sessions. Discreet reward for completion of each task. HA completion is gateway to receiving incentive for other tasks. Total of $150: $25 for HA $25 for biometric screening $100 for minimum number of coaching sessions Low risk = 4 sessions Moderate risk = 7 sessions High risk = 9 sessions

Assured

Reduction in 2013 medical premiums 12-month window to complete HA, biometric screening, and coaching sessions. Aggregate reward for completing all tasks. Reward given in monthly increments in following year paychecks. Total of $300 for completing HA and biometric screening and meeting values of blood pressure < 140/90 mm/Hg and total cholesterol < 240 mg/dL. If blood pressure < 140/90 mm/Hg and total cholesterol < 240 mg/dL, then no sessions required. If blood pressure > 140/90 mm/Hg or total cholesterol > 240 mg/dL, then 4 or more sessions required. Contingent on being eligible employee in subsequent benefit year

HA, health assessment.

day of fruit, vegetables, or grains. High stress was defined as endorsing at least a level of 8 out of 10 for health, home, or work stress. Tobacco use was assessed by asking, ‘‘Are you currently using tobacco?’’ Biometric screenings were conducted by an independent laboratory and included measured height, weight, and blood pressure, and a fasting venipuncture blood draw. Completion of the HA and the biometric screenings within the incentive design windows was determined from Healthyroads’ electronic database. Enrollment in telephone coaching and number of telephone coaching sessions completed was determined from electronic coaching records at Healthyroads. Statistical analysis

v2 Tests were used to assess the association of categorical variables with coaching enrollment status. Logistic regression was used to assess risk factors of lifestyle coaching enrollment. For the multivariate models, covariates with bivariate associations at the P < .10 level were entered into the final model. All tests were 2-tailed, and a P value of < .05 determined statistical significance. All statistical analyses were conducted using SAS software, version 9.3 (SAS Institute Inc., Cary, NC). Results Demographic characteristics

Of the 56,503 Healthyroads members fully eligible for wellness program benefits in 2012, 28,605 (50.6%) were employees of Company A and 27,898 (49.4%) were employed by Company B. Company A had a significantly higher proportion of males (73.9% vs. 29.9%, P < .001), and

members 45 to 65 years old (51.8% vs. 31.9%, P < .001) compared to Company B. HA and biometric screening completion

Figure 1 shows the monthly number of HA completions for the 2 companies. Company A had a 2-month window (April–May) to complete the HA compared to employees at Company B, who could complete the HA at any time during the benefit year. Similarly, Figure 2 shows the monthly number of employees completing biometric screenings where Company A had a defined window ( January–May) in which to complete the screening, while Company B had the entire benefit year to complete the screening. Overall, Company A had a lower engagement rate for the HA (24.4% vs. 26.1%, v2 = 21.5, df = 1, P < .001) and biometric screening (25.4% vs. 32.8%, v2 = 379.3, df = 1, P < .001,) compared to Company B. Overall, Company A had 18.1% of all benefit eligible employees (N = 5174) complete both an HA and biometric screening, whereas Company B had 23.8% (N = 6635) completing both an HA and biometric screening. (Data not shown.) Enrollment in coaching

Table 2 shows that among the subgroup of employees who completed the HA and the biometric screening, 44.6% (N = 2,309) at Company A engaged in at least 1 coaching session compared to 8.9% (N = 594) at Company B. More Company A employees in the high-risk blood pressure group engaged in coaching compared to Company B. However, a smaller proportion of Company A employees in the high-risk cholesterol group engaged in coaching compared to Company B. Table 2 shows that a much greater proportion of obese employees engaged in coaching at

INCENTIVE DESIGN STRATEGIES FOR WORKSITE WELLNESS PROGRAMS

FIG. 1.

327

Frequency of health risk assessment (HRA) completions by month.

Company A compared to Company B, although this risk factor was not directly targeted by either program. Table 2 also shows that, compared to Company B, greater proportions of employees at Company A who were not categorized as ‘‘high risk’’ engaged in coaching for all 3 biometric measures. Predictors of lifestyle coaching enrollment stratified by company among those who completed the HA and biometric screening

Table 3 presents logistic regression models of predictors of engaging in at least 1 coaching session stratified by

FIG. 2.

company. For Company A in the covariate adjusted model, significant predictors of enrolling in lifestyle coaching included female sex, older age category, and the higher education categories (ie, at least a bachelor’s degree, and those with some college coursework). Employees in Laborer, Managerial, Professional, and Technical job categories were significantly less likely to enroll in coaching when compared to Clerical employees at Company A. Company A employees categorized as high risk for stress, diet, tobacco, and blood pressure were less likely to enroll in coaching compared to those without these health risks. For Company B in the covariate adjusted model, employees in the older age category as

Frequency of biometric screening completions by month.

328

NORMAN, HELTEMES, AND DREW

Table 2. Enrollment in Coaching by Blood Pressure, Total Cholesterol, and BMI Risk Level Company A (N = 5174) Coaching N (%)

No Coaching N (%)

Coaching N (%)

No Coaching N (%)

P

OR

95% CI

2309 (44.6)

2865 (55.4)

594 (8.9)

6041 (91.1)

< .001

8.2

(7.4, 9.1)

318 (41.4) 1887 (44.6)

452 (58.6) 2348 (55.4)

102 (34.8) 490 (7.7)

191 (65.2) 5841 (92.3)

.053 < .001

1.3 9.6

(1.0, 1.7) (8.6, 10.7)

112 (44.6) 2191 (44.6)

139 (55.4) 2721 (55.4)

241 (54.9) 351 (5.7)

198 (45.1) 5830 (94.3)

.009 < .001

.66 13.4

(.49, .90) (11.8, 15.1)

511 (43.7) 1693 (44.3)

658 (56.3) 2131 (55.7)

171 (13.9) 420 (7.8)

1056 (86.1) 4979 (92.2)

< .001 < .001

4.8 9.4

(3.9, 5.8) (8.4, 10.6)

Biometric Risk Overall Blood Pressure High Risk Not High Risk Cholesterol High Risk Not High Risk BMI High Risk Not High Risk

Company B (N = 6635)

BMI, body mass index; CI, confidence interval; OR, odds ratio.

Table 3. Regression Analysis Assessing Predictors of Lifestyle Coaching Enrollment Among Those Completing an HA and Biometric Screening Company A (n = 5174) Predictors of Lifestyle Coaching Enrollment Gender Female Male Age Category 45–65 18–44 Job Category Laborer Managerial Other Professional Sales Technical Clerical Education Bachelor’s degree or higher Some college High school or less Job tenure >5 years 2 to 5 years

Examining incentive design strategies for worksite wellness program engagement.

The objective was to examine employee engagement in worksite wellness activities at 2 large US companies that differed in engagement strategy and ince...
226KB Sizes 0 Downloads 3 Views