No. 3, July 2008
COMMUNITY CASE STUDY
Employer Adoption of Evidence-Based Chronic Disease Prevention Practices: A Pilot Study
Jeffrey R. Harris, MD, MPH, MBA, Jeffrey Cross, MPH, Peggy A. Hannon, PhD, MPH, Eustacia Mahoney, Sarah Ross-Viles, MPH,
Alan Kuniyuki, MS
Suggested citation for this article: Harris JR, Cross J, Hannon PA, Mahoney E, Ross-Viles S,
Kuniyuki A. Employer adoption of evidence-based chronic disease prevention practices: a pilot study. Prev Chronic Dis 2008;5(3).
jul/07_0070.htm. Accessed [date].
We conducted a pilot test of American Cancer Society Workplace Solutions, an intervention that takes a marketing approach to increasing employers’ adoption of evidence-based practices to prevent and control chronic diseases among their employees.
We delivered the intervention and assessed the changes in practices of 8 large employers in the Pacific Northwest.
Workplace Solutions recommends 15 employer practices in 5 categories: 1) health insurance benefits, 2) policies, 3) workplace programs, 4) health-promoting communication, and 5) tracking
of employee health behaviors to measure progress. The intervention includes
4 meetings with employers over 2 months and begins with a questionnaire-based assessment of employer practices. Tailored recommendations follow, along with practice-specific implementation assistance on requested topics. We tested
the intervention in a before–after study without a comparison group.
The employers ranged in size from 7500 to 115,522 employees and included private companies and public employers. Seven of the eight employers implemented more of the recommended practices at follow-up (an average of 13 months after the intervention) than at baseline. Overall, implementation of the practices increased from 38% at baseline to 61% at follow-up (P = .02).
Workplace Solutions is a promising new approach to bringing evidence-based best practices for preventing chronic disease to large numbers of adults.
Back to top
Employers are important community partners for preventing chronic diseases for
3 reasons. First, they have power over workplace environments that affect the lives of working-age adults, most of whom are in the workplace most days of the week (1). Second, employers face rapidly mounting health care and productivity costs attributable to chronic diseases experienced by their employees — many of whom are from the baby-boom generation and are now advancing through middle age (2)
— and have increasingly strong motivation to promote preventive practices aimed at these diseases. Third, employers control health-insurance coverage of preventive care for 59% of working adults and their dependents (3). Recent research shows that employers do a poor job of covering evidence-based preventive care. For example, less than 10% of employers of
any size offer optimal coverage for smoking cessation treatment (4).
For information about preventing chronic diseases among their employees, employers can draw on systematic reviews and recommendations from
4 expert groups: 1) the
U.S. Preventive Services Task Force (USPSTF), 2) the Advisory Committee on Immunization Practices (ACIP), 3) the Task Force on Community Preventive Services (TFCPS), and 4) the Partnership for Prevention (PFP). The USPSTF and the ACIP review the effectiveness of clinical preventive care (5,6). The TFCPS and its Guide to Community
Preventive Services review the effectiveness of policies, systems approaches, and community-based (including workplace-based) programs (7). The PFP builds upon the work of the USPSTF and the ACIP by prioritizing effective clinical preventive care services on the basis of cost-effectiveness
and their impact on health (8).
Identifying and recommending best practices in health insurance benefits and workplace policies and programs for employers is the first step; increasing the adoption of these practices by employers is the essential next step.
To ensure adoption, Maibach et al suggest using a marketing approach with 3
components: 1) conducting consumer research, 2) building sustainable
distribution channels, and 3) improving products and product selection and
reducing product price (9). Our research among workplace-related “consumers” has
found that large employers and their human resources staff are important targets both because of the substantial number of people
they employ and their potential to change norms regarding insurance coverage for
preventive care (10). Our sustainable distribution channel is a large, voluntary
public health organization, the American Cancer Society (ACS), which has
long-standing relationships with employers, thousands of staff throughout all 50
states, and one of the best-recognized “brands” for health in the country. We
followed Maibach’s recommendation for improving products and product selection and
for reducing product price in 3 ways. First, we selected a set of employer best practices that are evidence-based and focused on creating a supportive workplace environment for prevention. Second, we tailored the product to the needs and current practices of each employer. Third, we presented our information in face-to-face
sessions in the workplace, emphasized the business case for cost-effective prevention, and provided implementation assistance to ease adoption for employers.
We present here the results of a pilot test of Workplace Solutions, a marketing approach to increase employers’ adoption of evidence-based practices to prevent and control chronic diseases among their employees. Our purpose in conducting the pilot
test was to assess the feasibility of our approach and to test whether employers adopted the recommended practices.
Back to top
We conducted the study as a joint project of the ACS Great West Division,
1 of 13 geographic divisions of the ACS, and the University of Washington Health Promotion Research Center (HPRC),
1 of 33 Prevention Research Centers supported by the Centers for Disease
Control and Prevention (CDC). ACS staff approached employers with whom they had
established fundraising relationships about participating in this new workplace health promotion project. Employers were eligible if they had more than 5000
employees and headquarters in Idaho, Oregon, or Washington.
Back to top
Selection of best practices
To select which best practices to include in our intervention, we reviewed the recommendations of the USPSTF, ACIP, and TFCPS
for interventions applicable to the workplace, working-age adults, and prevention of cancer and other chronic diseases. We included best practices applicable to the workplace even if they had not been evaluated in the workplace. Our review produced a set of 15 best practices that we categorized into
5 functional groups: 1) insurance benefits, 2) workplace
policies, 3) workplace programs, 4) tracking, and 5) communication (Table 1). Of the 15 best practices, 10 relate to creating a supportive environment for prevention.
We tested the intervention in a before–after study without a comparison group. Our intervention consisted of
4 face-to-face meetings with each employer during 2 months. Our intervention team consisted of ACS staff accompanied by
1 or 2 members of the HPRC (J.R.H. and J.C.). We met with the employers’ human resources staff in charge of purchasing health insurance benefits
at each employer’s headquarters. At the first recruitment meeting, we presented our general approach and emphasized
the importance of preventing cancer and chronic diseases for employers and employees. At the second meeting, we measured the employers’ baseline practices. After the second meeting, we
wrote a 5- to 8-page report of recommendations for improving all practices that the baseline survey indicated were not fully implemented. At the third meeting, we presented the recommendations, discussed the potential for adoption of best practices, and asked the employers to
choose 3 to 5 practices for adoption from the recommendations. At the fourth meeting, we presented Solution Sets for the practices employers selected. Solution Sets
consisted of 1 to 3 pages of implementation-oriented text (i.e., a summary of evidence detailing why the practice should be adopted and information about how to implement the practice) and other supporting materials, cost calculators that estimated first-year implementation costs and return on investment, lists of vendors
that could assist with the recommended practice, and information on relevant programs or materials available from the ACS (11).
Our intervention materials emphasized the business case for prevention of chronic diseases. For example, we highlighted that the PFP
review rated 3 of the 5 clinical preventive services we recommended as
either cost-saving (e.g., tobacco cessation treatment) or cost-neutral (e.g.,
colon cancer screening, influenza vaccination) (Table 1) (8) and that other
analyses show that providing tobacco-cessation treatment and influenza vaccination to employees is usually cost-saving,
particularly when productivity gains are counted (12,13).
The institutional review board of the University of Washington reviewed the study and classified it as
Employers completed 3 questionnaires during the intervention period: 2 at baseline and
1 after the intervention. At baseline, each employer completed a pre-assessment survey
of employer characteristics, employee demographics, and employers’ insurance providers. Employers also completed a baseline survey of best practices, a comprehensive survey of the employers’ health-related practices adapted from Golaszewski et al (14). The questionnaire included 115 items; 36 measured the 15 best practices that we included in our analyses. For follow-up at 1 year, we developed a streamlined version of the baseline survey and included only the questions relevant to the 15 best practices and a few items measuring employer satisfaction with the intervention.
We scored employers on their responses to the questions related to each best practice. For questions concerning benefit coverage and tobacco-use restrictions, we used
3 possible scores: 1) a score of 0 if the practice was not in place at all, 2) a score of .75 if the practice was partially in place (i.e., covered with co-pay for benefits or smoking forbidden indoors), and 3) a score of 1 if the practice was fully in place (i.e., coverage with no co-pay for benefits or a campus-wide
ban on tobacco use). We used a score of .75 (rather than .50) for practices partially in place to reflect the fact that, by covering most of the costs associated with cancer screening and smoking
cessation medications or by forbidding smoking indoors,
employers are significantly aiding employees’ health. For all other questions, we scored dichotomously, using a score of 1 for a practice that was in place and a score of 0 for a practice that was not in place. For each best practice, we created a summary
score by summing the items measuring the practice and dividing by the number of items. Thus, we scored each best practice as being implemented from 0% to 100%. For each employer, we calculated an overall best practice score by summing the scores on the individual best practices and taking the mean.
Because the number of participating employers was small, we conducted primarily descriptive analyses. We calculated mean scores and 95% confidence intervals at baseline and follow-up, and mean change in score for 1) each best practice and 2) the total of all 15 best practices. Because of the small sample size and non-normal distribution of the data, we assessed the significance of the median change in scores with nonparametric sign tests. We present mean scores for ease of interpretation,
but all presented P values (α = .05) are from these sign tests of median change. We also examined change in baseline and follow-up scores separately for practices for which employers did
or did not receive Solution Sets.
Back to top
Of the 10 employers we approached, 9 agreed to participate in the intervention and completed the baseline surveys. One company later became ineligible to participate because it was purchased by another company, so the final sample
was 8 employers. We followed employers for an average of 13 months (range, 8–18 months) from baseline to follow-up assessment. Employers ranged in size from 7500 to 115,522 employees (median, 12,695; mean, 33,104; SD, 37,408), and represented various
industries (Table 2).
Baseline implementation of best practices
The employers’ total baseline best practice scores ranged from 23% to 58% (Table 2), with a mean of 38% (Table 3). Employers
varied considerably among practices at baseline; they were most likely to cover cancer screenings (78%) and impose smoking restrictions or bans (72%) (Table 3).
No employers had sun-protection policies, required insurance providers to track delivery of preventive services,
gave reminders for preventive services.
Follow-up implementation of best practices
Seven of eight employers improved their total best practice scores from baseline to follow-up (Table 2). Scores at follow-up ranged from 37% to 85% (Table 2), with a mean follow-up score of 61%
(Table 3), a significant increase from baseline (P = .02). Employers achieved significant improvement from baseline to follow-up in the areas of covering tobacco cessation treatment (31% mean change, P = .03) and covering cancer screening (18% mean change, P = .03) (Table 3). The
lowest mean change occurred for providing sun protection (0%) and providing physical activity facilities (8%).
Our duration of follow-up varied from 8 to 18 months, and benefit–design cycles for employers are often 12 months long; however, we found little difference in the change in practices for the
6 employers with at least 12 months of follow-up (25% change) and the 2 employers with fewer than 12 months of follow-up (20% change).
Impact of Solution Sets
Employers’ baseline scores were lower for practices for which they received Solution Sets (31%) than for practices for which they did not receive Solution Sets (41%), yet the follow-up scores for both groups of practices were essentially the same (63% for those given Solution Sets, 60% for those not given Solution Sets). Thus, employers’ scores improved 32% for practices with Solution Sets (P = .02) compared with 19% for practices with no Solution Sets (P = .45)
(data not shown). Employers were most likely to request Solution Sets for covering tobacco cessation treatment (n = 7) and for covering cancer screening (n = 7).
Feedback from employers
Employers were generally positive in their ratings of the intervention. Seven of the eight employers would recommend the intervention to other companies, and
5 employers intended to participate in additional programs offered by ACS.
Back to top
Workplace Solutions was associated with a large and significant increase in implementation of evidence-based best practices aimed at prevention of cancer and other chronic diseases by large employers. At baseline, only 38% of our recommended practices were in place, so there was substantial room for improvement. At follow-up, 61% of recommended practices were in place. Seven of eight employers improved their implementation of best practices after the intervention.
Of the 15 prevention practices we addressed, 2 (covering tobacco cessation treatment and covering cancer screening) improved significantly. From these results, we speculate that employers find it easiest to change practices that can be outsourced, such as health insurance coverage, but we need to test larger numbers of employers to be certain. The large employers in our study were all self-insured, and changing health insurance coverage may be more difficult for midsized
and small employers that are not self-insured.
We found a large change among best practices for which we provided implementation-oriented assistance via Solution Sets. Furthermore, we found that employers were more likely to request this assistance for practices on which they scored poorly at baseline. These results suggest that more intensive intervention may have been associated with greater effect. Alternatively, the request for a Solution Set could have represented an intention by the employer to adopt the practice.
Our more subjective assessment of contributors and barriers to success raised
3 important points. First, age-appropriateness of our focus on chronic diseases could have been an issue for the one employer who made no change in response to the intervention. This employer had a much younger employee population than the other employers in our study. Second, the opportunity to work with high-ranking human-resources staff appeared to affect adoption of recommended practices positively. Third,
our 2 government employers had relatively low rates of adoption of recommended practices, which may
be caused by the long decision chain within unionized governmental bureaucracies.
The change in tobacco-related practices merits special mention. Like other researchers, we found that employer coverage of tobacco cessation treatment was low at baseline (4). Tobacco cessation treatment remains a missed opportunity for employers, because it is cost-saving and valuable to smoking employees. Unfortunately, it is not provided by most employers.
The limitations of this pilot study include its small sample, its design, and our focus on employer practices rather than employee behaviors. Our small pilot study included only
8 employers. Nonetheless, our intervention was associated with a meaningful and statistically significant change in the implementation of recommended practices.
Our study design, before–after without comparison, raises the possibility of historical effects, socially desirable responses, and interviewer bias. Changes of this magnitude
during this short period seem likely to be due to the intervention rather than to historical effects. The objective nature of the practices we measured makes report of better practices because of social desirability unlikely. Interviewer bias is possible, because the intervention team measured implementation
of practices both before and after the intervention. However, during interviewer training, we emphasized a consistent approach to measurement.
Our focus on employer practices and not employee behaviors was another limitation. Our best practices were recommended by the USPSTF or the TFCPS because they
are effective in increasing healthy behaviors. We can reasonably expect that employees will improve the behaviors targeted by the intervention practices, but this remains to be proven.
We have developed and pilot-tested a marketing-oriented approach to improving large employers’ practices for preventing chronic diseases among their employees. The behaviors we targeted are tied to the leading causes of death in the United States (15). The practices we targeted have a strong evidence base and rank highly on effectiveness and cost-effectiveness (8). Strengths of our marketing-oriented approach include its focus on employers’ need to control health-related costs
and its emphasis on environmental approaches to behavior change.
How generalizable is a resource-intensive intervention like ours to midsized and small-sized employers? In Washington, a state of average population size, there are approximately 206,000 employers (10). Approximately 200 large employers (i.e.,
those with more than 1000 employees) employ 18% of the workforce, and 3500 midsized employers (i.e., with 100 to 999 employees) employ 30% of the workforce. Approximately 202,000 small employers (i.e., with 99 or fewer employees) employ the remaining 52% of the
workforce. Working with a partner like the ACS, reaching large and midsized companies with this type of an intervention or a streamlined version we are now testing in partnership with the ACS, might be feasible. However, for small employers we need to consider other broad-reach approaches, such as an interactive tool on the Web, or work through powerful intermediaries, such as health insurance brokers.
In the future, we plan to confirm the results of this pilot study with a larger employer sample and a more robust study design. Future studies should test the effects of the intervention on employee behaviors, employee productivity, and employer health care costs.
Back to top
This publication was supported in part by CDC and the National Cancer Institute through the Cancer Prevention and Control Research Network, a network within the CDC’s Prevention Research Centers Program (Grant no. 1-U48-DP-000050). Additional support was provided by the ACS.
We thank Alan Kuniyuki for statistical review and advice and Coleen Pedack for working with the intervention team.
Back to top
Corresponding Author: Jeffrey R. Harris, MD, MPH, MBA, Health Promotion Research Center, Department of Health Services, University of Washington, 1107 NE 45th St, Seattle, WA 98105. Telephone: 206-616-8113. E-mail: email@example.com.
Author Affiliations: Jeffrey Cross, National Home Office, American Cancer Society, Atlanta, Georgia; Peggy A. Hannon,
Sarah Ross-Viles, Alan Kuniyuki, Health Promotion Research Center, Department of Health Services, University of Washington, Seattle, Washington; Eustacia
Mahoney, Great West Division, American Cancer Society, Seattle, Washington, and
National Home Office, American Cancer Society, Atlanta, Georgia.
Back to top
- Employment and unemployment. Washington (DC): U.S. Department of Labor, Bureau of Labor Statistics; 2006.
- Galvin RS, Delbanco S.
Between a rock and a hard place: understanding the employer mind-set. Health Aff (Millwood) 2006;25(6):1548-55.
- Employer health benefits 2006 annual survey. Menlo Park (CA): Kaiser Family Foundation; 2006.
- Bondi MA, Harris JR, Atkins D, French ME, Umland B.
Employer coverage of clinical preventive services in the United States. Am J Health Promot 2006;20(3):214-22.
- U.S. Preventive Services Task Force (USPSTF). Rockville (MD): Agency for
Healthcare Research and Quality. http://www.ahrq.gov/clinic/uspstfix.htm.
Accessed December 20, 2006.
- Advisory Committee on Immunization Practices (ACIP). Atlanta (GA): Centers
for Disease Control and Prevention. http://www.cdc.gov/nip/acip/. Accessed
Dec 8, 2006.
- The Community Guide. Atlanta (GA): Centers for Disease Control and
Prevention. http://www.thecommunityguide.org. Accessed Dec 13, 2006.
- Maciosek MV, Coffield AB, Edwards NM, Flottemesch TJ, Goodman MJ, Solberg LI.
Priorities among effective clinical preventive services: results of a systematic review and analysis. Am J Prev Med 2006;31(1):52-61.
- Maibach EW, Van Duyn MAS, Bloodgood B. A marketing perspective on
disseminating evidence-based approaches to disease prevention and health
promotion. Prev Chronic Dis 2006;3(3).
- Lichiello P, Harris JR, Cross J, O'Neill MK, Gardner M. Employment-based prevention of chronic disease in Washington State, 2005. Seattle
(WA): University of Washington Health Promotion Research Center; 2005.
- Workplace solutions: building a healthy workforce. Atlanta (GA): American
Cancer Society. http://www.acsworkplacesolutions.com. Accessed December 15,
- Fellows JF, Rehm B, Hornbrook M, Hollis J, Haswell TC, Dickerson J, Volk
C. Making the Business Case for Smoking Cessation and ROI Calculator.
Portland (OR): Kaiser Permanente Center for Health Research; 2004. http://www.businesscaseroi.org.
Accessed December 15, 2006.
- Nichol KL.
Cost-benefit analysis of a strategy to vaccinate healthy working adults against influenza. Arch Intern Med 2001;161(5):749-59.
- Golaszewski T, Barr D, Pronk N.
Development of assessment tools to measure organizational support for employee health. Am J Health Behav 2003;27(1):43-54.
- Mokdad AH, Marks JS, Stroup DF, Gerberding JL.
Actual causes of death in the United States, 2000.
[Published erratum in: JAMA 2005;293(3):293-4, JAMA 2005;293(3):298]. JAMA 2004;291(10):1238-45.
Back to top