Cost and Utilization Avoidance With Mail Prompts: A Randomized Controlled Trial

Objective: To study the medical service utilization changes and return on investment from a health plan’s direct mailings that either encouraged members to receive influenza vaccinations or encouraged members to call a nurse advice service.

Study Design: Randomized controlled trial with 2 intervention groups and 1 control group consisting of all members over age 65 years who were enrolled in 5 states in the Blue Cross and Blue Shield Government-wide Service Benefit Plan. Sample size was 134,791 individuals.

Methods: Administrative claims–based influenza, pneumonia, heart failure, and respiratory inpatient bed days, emergency department (ED) visits, physician evaluation and management visits, other outpatient visits, and nurse advice call rates were compared between the intervention and control groups.

Results: The influenza mailing intervention group experienced 2.87% (P = .033) fewer conditionrelated inpatient bed days and 7.25% (P = .101) fewer condition-related ED visits. The nurse advice service mailing intervention group experienced 7.65% (P <.001) fewer condition-related inpatient bed days and 6.75% (P = .125) fewer condition-related ED visits. Per dollar spent, the return on investment was estimated to be $2.51 for the influenza mailing intervention and $24.24 for the nurse advice mailing intervention.

Conclusions: Administrative claims data suggest that members respond to health plan mailings. By mailing information to their members, health plans can affect rates of medical service utilization and generate cost savings.

(Am J Manag Care. 2008;14(11):748-754)

This study provides convincing evidence that a relatively simple mail-delivered prompt to encourage flu vaccination or to encourage the use of a nurse advice line can have measurable effects on health services utilization rates and generate cost savings.

• Mailing information to their members is a cost-effective way for health plans to affect condition-related medical service utilization with a positive return on investment.
• This study tested an intervention designed to support the Advisory Committee of Immunization Practices recommendation to increase vaccination levels and an intervention designed to measure the effect of a nurse advice service.

Influenza is a highly contagious viral disease, with widespread outbreaks and epidemics typically occurring in the United States during the winter months. Influenza was responsible for an average of 36,000 deaths per year in the United States from 1990-1999.¹ Although rates of infection are highest among children, serious illness and death are most common in individuals over the age of 65 years and those with medical conditions that place them at higher risk for complications,^2-5 with people over the age of 65 years accounting for 63% of all influenza-related hospitalizations.^6,7 Influenza vaccination has been found to be associated with reductions in hospitalizations for heart disease, cerebrovascular disease, pneumonia, and influenza, while the risk of death from all causes during the influenza season is reduced.⁸

Even though vaccination rates increased during the 1990s, the Advisory Committee of Immunization Practices from the Centers for Disease Control and Prevention (CDC) recommends using strategies to increase vaccination levels, including reminder systems and standard orders programs.^9-12

Economic studies of influenza vaccinations for people over the age of 65 years have shown overall cost savings and reductions in hospitalization and death.^13-16 Similarly, studies of adults under the age of 65 years also show that influenza vaccinations can reduce direct medical costs as well as indirect costs from worker absenteeism.^17-22 The Task Force on Community Preventive Services has shown costeffectiveness studies to range between $3 and $46 per additional vaccine administered,²³ whereas client recall and reminder systems have been found to be effective in increasing community demand for vaccines.^24-26

This study tested an intervention designed to support the Advisory Committee of Immunization Practices recommendation to increase vaccination level. We estimated the benefits of influenza vaccines for the elderly^7,27-29 and also studied the effect of a nurse advice service, with the goal of reducing inpatient bed days and emergency department (ED) visits for members over the age of 65 years in a preferred provider organization health plan (the Blue Cross and Blue Shield Government-wide Service Benefit Plan).

METHODS

All subscribers (households) and their dependents over the age of 65 years enrolled in the Blue Cross and Blue Shield Government-wide Service Benefit Plan in the states of Oklahoma, Rhode Island, Kentucky, California, Arizona, Utah, and Colorado in October 2002 were eligible for the study. Subscribers were either current federal employees or retired federal employees with the Blue Cross and Blue Shield Government-wide Service Benefit Plan as their primary insurance payer.

Study Design

Because there were 2 mailed interventions, there were 2 separate intervention groups. The objective of 1 of the mailings was to promote receiving influenza vaccination (influenza mailing group). The purpose of the other mailing was to promote the use of a telephonic nurse advice service (nurse advice service mailing group). The goal of each intervention was to reduce influenza-related inpatient bed days and ED visits.

The nurse advice service was sponsored by the Blue Cross and Blue Shield Government-wide Service Benefit Plan and provided through a contract with a third-party vendor. The nurse advice service employed approximately 400 registered nurses through a distributed call center model. A standardized triage system using proprietary symptom-based algorithms and caller education content was used. Members of all 3 groups (control, influenza vaccination mailer, and nurse advice service mailer) had unlimited access to the nurse advice service.

The mailings for the first arm of the intervention (influenza mailing group) were sent at the beginning of the study period (October 1, 2002, and November 1, 2002) to increase the likelihood that a person would read at least 1 of the 2 identical mailings and subsequently receive an influenza vaccination.

The influenza mailer was based on CDC influenza and influenza vaccination clinical content, and was reviewed and approved by physicians board certified in internal medicine and emergency medicine. The mailer included descriptions of high-risk populations and of the beneficial effects of vaccination, the recommended timing for receiving the vaccine, and a recommendation for frequent hand washing.

The mailings for the second arm of the intervention (nurse advice service mailing group) were sent at the beginning and middle of the study period (October 1, 2002, and January 1, 2003, respectively) to increase the likelihood that a person would read at least 1 of the 2 identical mailings and subsequently call the nurse advice service if symptoms developed.

The nurse advice service mailer included a description of influenza symptoms (fever, chills, rhinorrhea, myalgias, and headache) and a brief description of high-risk populations. It also included an invitation to call the nurse advice service at the earliest sign of symptoms and stated that nurse advice service would provide 24/7 telephonic access to a registered nurse who would help persons understand what they could do to start feeling better, when they should see a doctor, and what new or worsening symptoms should prompt a call back to the service or their physician.

The control group was not sent either mailing.

Hypotheses

The outcome measures included influenza, pneumonia, heart failure, and other respiratory inpatient bed days, ED visits, physician evaluation and management visits, and other outpatient visits. The diagnosis codes for the outcomes were defined following those used by Nichol et al¹⁴ and Davis et al.¹⁶ Utilization for pneumonia and influenza was determined by International Classification of Diseases, Ninth Revision, Clinical Modification³¹ (ICD-9-CM) codes 480-487. Utilization for congestive heart failure was determined by ICD-9-CM code 428. Utilization for all respiratory conditions was determined by ICD-9-CM codes 460-462, 465, 466, 480-487, and 500-518. Because no diagnosis-related group information was available, the reason for each admission or visit was determined by the most frequent first-listed diagnoses for each admission or visit. Physician evaluation and management visits were determined by looking at all outpatient claims that did not occur on the same day as an inpatient admission or ED visit using Clinical Procedural Terminology codes 99201-99215 or 99241-99245. Other outpatient visits represented the remaining outpatient visits that did not occur on the same day as an inpatient admission or ED visit.

Statistical AnalysisTo check the randomization, demographic and baseline variables were compared to ensure that there were no significant differences among the 3 groups, as was indicated by the fact that all measured variables were within 1 standard deviation of one another.

The rates of condition-related inpatient bed days, ED visits, physician evaluation and management visits, and other outpatient visits were compared between the intervention and the control groups in the intervention time period. Because randomization was based on clusters (subscribers were randomized rather than individuals), P values were calculated with a &#967;² statistic by using the SAS software (version 9; SAS Institute Inc, Cary, NC) clustering option. Statistical significance of differences of utilization and vaccinations between each group was calculated by using the &#967;² statistic generated by &#8220;proc genmod&#8221; using the &#8220;repeated&#8221; option in SAS to account for the clustering effect on the variance.³¹

Savings Calculation

Because the mailings were sent out in bulk, no information was available on undeliverable pieces. Subscribers (households), not individual members, were randomized and sent the 2 mailings to reduce cross-contamination between household members, avoiding the possibility of a husband and wife pair being randomized into different groups.

Table 1

The total sample size was 134,791 individuals, of whom 26,474 in the intervention group were sent the influenza mailing, and 26,864 in the intervention group were sent the nurse advice mailing; 81,453 were in the control group. The unbalanced study design was influenced by a fixed budget for mailings for the intervention group. Characteristics of intervention and control individuals are shown in . The randomization appeared to be appropriate for the available demographic variables. Differences between the intervention groups and control group all were less than 1 standard deviation.

and

Table 2 Table 3

show the utilization rates per 10,000 people for each intervention group compared with the control group over the 5-month study period of October 15, 2002, through March 15, 2003. Compared with the control group, the influenza mailing intervention group experienced 2.87% (P = .033) fewer condition-related inpatient bed days and 7.25% (P = .101) fewer condition-related ED visits compared with the control group. No significant differences between groups for physician evaluation and management visits, other outpatient visits, or influenza vaccinations were observed. Influenza vaccinations often are given in settings that do not generate claims, thus limiting the reliability of evidence of influenza vaccinations as seen via administrative claims.

Compared with the control group as shown in Table 3, the nurse advice service mailing intervention group experienced 7.65% (P <.001) fewer condition-related inpatient bed days, 6.75% (P = .125) fewer condition-related ED visits, 16.13% (P = .049) more nurse advice service general health information calls, and 18.43% (P <.001) more nurse advice service symptomatic calls, of which there was an increase of 54.14% (P = .002) for influenza and upper respiratory illness compared with the control group. No significant differences between groups for physician evaluation and management visits were observed, although there was a lower rate for the intervention group for other outpatient visits: 4.63% (P = .001). No significant differences between groups for influenza vaccinations were observed. Influenza vaccines were not encouraged in the nurse advice service mailing intervention group.

For nurse advice symptomatic calls, the distributions of nurse recommendations for the nurse advice mailing intervention group and the control group were not statistically different from each other and were as follows: (1) access urgent or emergent care was 16.0% for the intervention group and 18.2% for the control group, (2) speak to their provider was 32.1% for the intervention group and 32.2% for the control group, (3) make an appointment with their provider was 14.7% for the intervention group and 14.9% for the control group, and (4) access self-care instructions with monitoring was 37.2% for the intervention group and 34.7% for the control group.

Regression analysis was performed on inpatient bed days, which included age and sex as covariates, as part of a sensitivity analysis even though a regression framework was not needed as an analytic tool given the randomized design. The results of the regression analysis showed that the estimated impact of the bed-day rate differed by less than 1% for both intervention groups, thus lending support to an appropriate randomization process.

Table 4 shows the overall financial results. For the influenza mailing intervention group, incremental savings totaled $122,656 for the unit cost method and $63,096 for the PMPM method. For the nurse advice mailing intervention group, incremental savings totaled $309,824 for the unit cost method and $387,783 for the PMPM method. The total cost of the intervention (cost of the mailings) was $32,000, which was allocated evenly for each arm for purposes of calculating a return on investment (ROI) for each arm of the intervention. The ROI for the influenza mailing intervention was calculated to be $4.88 and $2.51 for the unit cost and PMPM methodologies, respectively, with net savings per person of $3.68 and $1.43, respectively. The ROI for the nurse advice intervention was calculated to be $19.36 and $24.24 for the unit cost and PMPM methodologies, respectively, with net savings per person of $10.94 and $13.84, respectively.

The nurse advice service already was an existing benefit for all members, so the intervention incurred no incremental nurse advice service costs for the health plan. In other words, the ROI was for the mailing that encouraged use of the nurse advice service, not for implementing a nurse advice service for a membership with no already-existing nurse advice service. For the influenza mailing intervention group, the entire cost of influenza vaccines for the influenza mailing group was added to the cost of the intervention.

DISCUSSION

2. Monto AS, Kioumehr F. The Tecumseh study of respiratory illness, IX: occurrence of influenza in the community, 1966-1971. Am J Epidemiol. 1975;102(6):553-563.

4. Barker WH, Mullooly JP. Impact of epidemic type A influenza in a defined adult population. Am J Epidemiol. 1980;112(6):798-811.

6. Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalizations in the United States. JAMA. 2004;292(11):1333-1340.

8. Nichol KL, Nordin J, Mullooly J, Lask R, Fillbrandt K, Iwane M. Influenza vaccination and reduction in hospitalizations for cardiac disease and stroke among the elderly. N Engl J Med. 2003;348(14):1322-1332.

10. CDC. Use of standing orders programs to increase adult vaccination rates. MMWR Recomm Rep. March 24, 2000;49(RR-1):15-16.

12. Ndiaye SM, Hopkins DP, Shefer AM, et al. Interventions to improve influenza, pneumococcal polysaccharide, and hepatitis B vaccination coverage among high-risk adults: a systematic review. Am J Prev Med. 2005;28(5 suppl):248-279.

14. Nichol KL, Wuorenma J, von Sternberg T. Benefits of influenza vaccination for low-, intermediate-, and high-risk senior citizens. Arch Intern Med. 1998;158(16):1769-1776.

16. Davis JW, Lee E, Taira DA, Chung RS. Influenza vaccination, hospitalizations,

17. Office of Technology Assessment. Cost Effectiveness of Influenza Vaccination. Washington, DC: US Government Printing Office; December 1981.

19. Campbell DS, Rumley MH. Cost-effectiveness of the influenza vaccine in a healthy, working-age population. J Occup Environ Med. 1997;39(5):408-414.

21. Demicheli V, Jefferson T, Rivetti D, Deeks J. Prevention and early treatment of influenza in healthy adults. Vaccine. 2000;18(11-12):957-1030.

23. The Task Force on Community Preventive Services. Vaccine— Preventable Diseases: Improving Coverage in Children, Adolescents, and Adults. http://www.thecommunityguide.org/vaccine/vpd-econ.pdf. Accessed February 27, 2008.

25. Lieu TA, Capra AM, Makol J, Black SB, Shinefield HR. Effectiveness and cost-effectiveness of letters, automated telephone messages, or both for underimmunized children in a health maintenance organization. Pediatrics. 1998;101(4):E3.

27. Simonsen L, Reichert TA, Viboud C, Blackwelder WC, Taylor RJ, Miller MA. Impact of influenza vaccination on seasonal mortality in the US elderly population. Arch Intern Med. 2005;165(3):265-272.

29. Jackson LA, Nelson JC, Benson P, et al. Functional status is a confounder of the association of influenza vaccine and risk of all cause mortality in seniors. Int J Epidemiol. 2006;35(2):345-352.

31. International Classification of Diseases, Ninth Revision, Clinical Modification. Washington, DC; US Department of Health and Human Services; 1996.

33. CDC. Influenza and pneumococcal vaccination coverage among persons aged >65 years and persons aged 18-64 years with diabetes or asthma—United States, 2003. MMWR. 2004;53(43):1008-1012.

35. Gilman BH, Bonito AJ, Eicheldinger C. Impact of influenza immunization on medical expenditures among Medicare elderly, 1999-2003. Am J Prev Med. 2007;32(2):107-115.

Download PDF: Cost and Utilization Avoidance With Mail Prompts: A Randomized Controlled Trial