Abstract
Objective:
To estimate absenteeism and presenteeism-related work loss due to herpes zoster (HZ) among working individuals of 50–64 years of age.
Methods:
This telephone survey included individuals with ≥1 insurance claim for HZ in the past year in administrative claims data from five US commercial health plans. Demographic information, characteristics of the HZ episode; impact of HZ on activities of daily living (ADL), and work days loss and productivity were surveyed.
Results:
Responses were obtained from 153 of 1654 individuals who were contacted and were eligible for the survey (9.3%). Most had moderate or severe HZ (72.6%). Close to two-thirds reported some impact of HZ on ADL such as shopping, housework/chores, and social engagement. About half (51%) reported missing work due to HZ, and about an equal percentage reported little or much worse productivity than usual due to HZ while at work. On average, age-adjusted absenteeism- and presenteeism-related work loss was estimated at 31.6 hours, and 84.4 hours, respectively, with a combined work loss of 116.0 hours per HZ episode in a working person of 50–64 years of age. Work loss tended to increase with age and the duration and severity of the HZ episode.
Conclusions:
The study documents a substantial societal burden of HZ-related work and productivity loss. This is important information to take into consideration, in addition to the direct medical burden, when making policy decisions around vaccine prevention of HZ.
Limitations:
The study may potentially be subject to selection bias due to low survey response rate and since only those cases who sought care for a HZ episode were captured. The study may also be subject to respondent recall bias. Finally, since some respondents could still be having the HZ episode at the time of survey, the study may potentially have under-estimated the work and productivity loss.
Background
Herpes zoster (HZ), commonly known as shingles, is caused by reactivation of varicella-zoster virus acquired during varicella infectionCitation1. Most adults have been infected with the varicella-zoster virus during childhood, but reactivation can occur as a result of aging-related immunosenescence and immunocompromising conditionsCitation2,Citation3. In the US, the estimated incidence in the general population is ∼3–4 cases per 1000 person-yearsCitation4,Citation5. The incidence increases with age peaking at ∼11–12 per 1000 person-years among those 80 years or older. Anti-viral treatment can help to reduce the severity and duration of HZ if diagnosed and treated earlyCitation6–8. Despite the availability of anti-viral treatment, chronic pain, also called postherpetic neuralgia (PHN), occurs in more than half of all older adults with HZCitation5,Citation9,Citation10.
HZ and its complications including PHN, pain, recurrence, reduced physical function, emotional distress, and reduced social and role function can negatively impact activities of daily living (ADL) and quality-of-lifeCitation5,Citation11–15. Prior studies have reported that HZ-associated pain interfered substantially with ADLs, resulting in poor health-related quality-of-lifeCitation16–18. These studies highlight the negative impact of HZ pain on functional status and health-related quality-of-life, but do not measure its impact on work-related activities.
Although HZ has been shown to lead to substantial morbidity and healthcare resource use among patients in the USCitation19, limited data are available on work loss related to HZ in the US. A recent study using absence and short-term disability records estimated the average work loss to be 3–6 days per working person of 18–65 years of age, and found that work loss increased with ageCitation20. Work loss costs were estimated to be US $500–1000 per working HZ patient in this study. However, this study only analyzed work loss due to time away from work (i.e., absenteeism) and did not assess work loss due to potentially lower productivity while employees were at work (i.e., presenteeism).
A shingles vaccine has been available in the US for the prevention of HZ among individuals of 60 years of age since May 2006. The vaccine was recently approved for use among those between 50–59 years of age. Policy-makers would seek to understand the value of the shingles vaccine in the younger age group.
The objective of the current study was to estimate absenteeism and presenteeism-related average work hours loss per episode of HZ among working individuals between 50–64 years of age in the US. The data should help policy-makers understand the impact of shingles and the potential value of the shingles vaccine from the societal perspective.
Methods
Data source
This study comprised a telephone survey of individuals of 50–64 years of age with a recent episode of incident HZ who were employed during their disease episode. Individuals with the disease were identified from the HealthCore Integrated Research Database, a Health Insurance Portability and Accountability Act (HIPAA) compliant administrative claims database consisting of five geographically dispersed health plans. The employment status during the HZ episode was confirmed via self-report during the telephone survey. The study was submitted to an independent institutional review board (IRB) and a partial waiver of HIPAA authorization was obtained for contacting study subjects. The waiver allowed for direct patient contact without receipt of prior consent; provided that the potential respondent was verbally consented during the phone call, prior to any data collection.
Patient selection
The population of interest for the study was employed workers between the ages of 50–64 years who had an incident episode of HZ. Using the HealthCore Integrated Research Database, potential survey participants met the following inclusion criteria during the intake period between January 1, 2005 and November 30, 2005: ≥1 claim for HZ (ICD 9-CM codes 053.xx) in medical claims; age between 50–64 years at the index date defined as the date of first HZ diagnosis; ≥6 months of continuous eligibility in the health plan pre-index date; and continuously eligible through the end of the intake period. An incident episode of HZ was defined as the absence of a HZ diagnosis in the medical claims data during the 6-month pre-index period.
Telephone survey
An initial mailer was sent to potential participants informing them that a research agency representative would be contacting them regarding a research study, and that participants would receive a nominal honorarium for completing a telephone survey. The mailer was followed by a telephone call for the survey during the month of January in 2006. To help minimize recall bias, potential participants with an index date closest to date of the telephone call were contacted for the survey first. Verbal informed consent was obtained over the telephone prior to initiating the survey. Patient age and a recent HZ episode (i.e., self-reported HZ episode within ±2 months of the index date) were confirmed at the beginning of the survey. Surveys for those not meeting these inclusion criteria were terminated. Finally, only those who self-reported as being employed during the recent HZ episode were continued for the survey.
In order to have representation across the age groups, an a priori sample target of 100 patients in the 50–59 year old age group, and 50 patients in the 60–64 year old age group was defined for the study. No formal sample size calculation was conducted since findings were intended to be mostly descriptive.
The survey collected information across several domains including socio-demographics; characteristics of HZ episode; employment characteristics at the time of HZ episode; the number of full and partial work days missed (referred to as absenteeism) due to HZ and primary reasons for missed days; percent reduction in productivity at work during the HZ episode compared to usual (referred to as presenteeism) and primary reasons for lower productivity; and the impact of HZ episode on a few key activities of daily living (ADL).
Data analysis
Descriptive statistics included means and standard errors of the means for continuous variables, and frequencies and percentages for categorical variables. Differences in means were assessed using t-tests or ANOVA. Number of hours of work loss due to absenteeism was calculated using number of hours worked per day, and the number of full and partial days missed due to HZ. It was assumed that those missing a partial day missed half of a work day. Number of hours of work loss due to presenteeism was calculated in several steps. First, the number of potential work hours during HZ episode was calculated using duration of HZ episode in weeks and number of hours worked per week. The number of potential work hours was adjusted for any full/partial days missed due to HZ episode. Number of hours loss due to presenteeism was then calculated using the percent reduction in productivity. Combined work loss was calculated by adding number of hours of work loss from absenteeism and presenteeism.
Following is an example to describe the calculation of number of work hours loss due to absenteeism and presenteeism. If a person worked 30 hours per week for 5 days (i.e., 6 hours per day) and missed 2 full days and 1 partial day due to HZ, then potential work hours loss due to absenteeism was 15 hours (2 * 6 + 0.5 * 6). If the HZ episode lasted 17 days, i.e., 2.43 weeks, then potential work hours during HZ episode after adjusting for full/partial missed days was 57.9 hours (2.43 * 30 − 15). If the person reported 70% productivity during the HZ episode compared to usual (i.e., 30% productivity loss), then the number of work hours loss due to presenteeism was 17.4 hours (57.9 * 0.3). The combined work hours loss due to absenteeism and presenteeism was 32.4 hours (15 + 17.4).
Finally, the overall work loss was age-adjusted for the employed civilian non-institutional population in the US between the ages of 50 and 64 in 2010Citation21.
Results
Selection of respondents
A total of 3859 individuals met the study inclusion criteria with a recent episode of incident HZ, of which 2062 individuals could not be contacted due to various reasons such as disconnected telephone numbers, patients being deceased, or other reasons. Of the 1797 patients contacted, a total of 143 respondents were excluded during the interview for various reasons including no self-reported diagnosis of HZ within ±2 months of the index date (77), no employment at the time of HZ episode (62), and age not between 50–64 years at the index date (3). One person was excluded because of concurrent cancer chemotherapy and HZ. Of the 1654 remaining who were contacted and eligible, a total of 866 refused the survey, and 635 were lost to follow-up (e.g., accepted to be called at an agreed-upon later time but could not be reached again, answering machine). Completed questionnaires were obtained from 153 respondents, which represented a response rate of ∼9.3% among those who were contacted for the survey and were eligible. Of the 153 who completed the survey, 104 were in the 50–59 year age group and 49 in the 60–64 year age group. Those who completed the survey were generally similar to those who refused or were lost to follow-up with respect to demographics such as gender (males 36.6% vs 44.1%, respectively; p > 0.05) and annual income (less than or equal to $50,000 56.2% vs 52.9%, 50,000–100,000 41.2% vs 44.0%, and over 100,000 2.6% vs 3.1%, respectively; p > 0.05), the number of physician visits (mean + SD = 3.31 ± 3.88 vs 3.31 ± 3.69, respectively; p > 0.05), and hospitalizations in the pre-index period (mean ± SD = 0.07 ± 0.3 vs 0.08 ± 0.49, respectively; p > 0.05). However, those who refused or were lost to follow-up were slightly older in age (years) compared to those who completed the survey (mean ± SD = 57.4 ± 4.3 vs 56.6 ± 4.2, respectively; p < 0.05).
Respondent characteristics
summarizes respondent socio-demographics and HZ episode characteristics. The majority of respondents were female (63.4%), married (68.6%), and White (88.9%). The average respondent typically worked 5 days per week for ∼8 hours each day. In terms of characteristics of the HZ episode, over one-half reported HZ rash on the chest, stomach, or back (53.7%), and HZ-related pain lasting >1 month (51.6%). Close to three-quarters (72.5%) reported a moderate-to-severe HZ episode.
Table 1. Socio-demographic and herpes zoster (HZ) disease episode characteristics of the respondents (n = 153).
Impact on activities of daily living, and work and productivity
presents the impact of the HZ episode on activities of daily living (ADL), and work and productivity. Close to two-thirds of the respondents reported some impact of the HZ episode on ADL such as shopping, housework/chores, and social engagement. Little over half (51%) reported missing some time due to HZ. Over a third (42%) and a quarter (29%) reported missing full and partial work days, respectively. About 8.5% reported missing work for the entire duration of the HZ episode.
Table 2. Impacts of herpes zoster (HZ) disease episode on activities of daily living, and work and productivity (n = 153).
Of those who worked during the HZ episode, close to half (46%) reported some or greater difficulty performing usual tasks while at work. Over half (51%) reported little or much worse productivity than usual due to HZ while at work. Overall work productivity during the HZ episode was 80.3% (±22.7%) compared to prior to the HZ episode.
Over half of the respondents reported pain or discomfort as the primary reason for missing work or absenteeism (58%) or for lower productivity or presenteeism due to HZ (64%). Other primary reasons for missing work or absenteeism included medical appointment for HZ (10%), doctor recommendation to stay home (8%), side-effects of HZ medications (3%), fear of spreading HZ (6%), inability to concentrate on job (1%), and other reasons (14%). Other primary reasons for lower productivity or presenteeism included feeling tired or less energetic (11%), inability to concentrate (10%), side-effects of HZ medications (5%), and other reasons (10%).
presents the estimated mean (SE) number of hours of work loss related to absenteeism, presenteeism, and the combined hours of work loss by selected respondent characteristics. On average, age-adjusted absenteeism- and presenteeism-related work loss was estimated at 31.6 (unadjusted 33.7 ± 4.8) hours and 84.4 (unadjusted 90.0 ± 11.1) hours, respectively, with a combined loss of 116.0 (unadjusted 123.7 ± 12.0) hours per working person due to the episode of HZ. Work loss tended to vary significantly by the duration and severity of the HZ episode. For example, combined work loss was as high as 231.3 ± 25.2 hours among those with a severe episode compared to just 25.2 ± 8.4 hours among those with a mild episode of HZ.
Table 3. Estimated hours of absenteeism- and presenteeism-related work loss due to herpes zoster (HZ) disease episode by selected characteristics.
Discussion
To the authors’ knowledge, this is the first US study estimating both absenteeism- and presenteeism-related work loss due to HZ. The study found that about half of the respondents with HZ suffered some work loss as a result of the disease. On average, the combined hours of work loss from absenteeism and presenteeism was estimated at ∼116.0 hours per working individual or 14.5 days when assuming an 8-hour work day. The work loss was greater in older age, with longer duration of the episode, and with greater severity of the episode.
The average absenteeism-related work loss due to HZ of 31.6 hours (i.e., ∼4 work days) per working individual found in this study is consistent with a previous US estimate of 3–6 days per person with HZCitation20. However, the current study also found that presenteeism-related work loss was over 2.5-times higher than the work loss related to absenteeism alone. This suggests that the work loss impact of HZ on employers and employees could be much greater than previously documented.
The work loss appeared to be greater among older respondents. The mean respondent age in this study was 56.6 years. The incidence of HZ at that average age is 4.6 per 1000 person-yearsCitation4. At 4.6 cases of HZ per 1000, and assuming 14.5 days of work loss per case, employers could potentially lose 67 days of work per 1000 employees between 50–64 years of age every year. In addition, if employees continue to maintain employment later in life, the impact of presenteeism and absenteeism due to HZ is likely to increase due to the increased severity of HZ in older age groups.
An early estimate of the cost of absenteeism-related work loss at US$500–1000 per working HZ patient was based on employees missing 3–6 days of work, and did not include productivity lossCitation20. A UK study of 65 patients with acute HZ reported that being employed was associated with increased costs of HZ to society (p = 0.0001)Citation22. The authors noted that the societal costs of HZ complications for patients of less than 65 years may be higher due to absence from work. The current study’s findings support these reports for absenteeism. However, work loss due to presenteeism in people going to work may present an even higher economic burden on society than the costs associated with absenteeism alone. Applying the same hourly wage rate (US$20.32) as used in a previous study and assuming 116.0 hours of combined absenteeism- and presenteeism-related, work loss is estimated to potentially cost an average of ∼$2,350 per working HZ patient between 50–64 years of age—three quarters of this economic burden being due to presenteeismCitation20.
The results of the current study demonstrate that work loss due to HZ is substantial in the 50–64 year age group in the US. The data should help policy-makers understand the value of HZ vaccination for the younger age group of 50–59 years. Employers may also use these data for their own internal decisions regarding employee coverage for HZ vaccine.
The study had several limitations. The major limitation was a quite low response rate for the survey (∼9%). A selection bias in terms of the type of respondents willing to complete the survey was possible, even though those who completed the survey mostly had similar characteristics as those who refused or were lost to follow-up. For example, it was possible that people with severe HZ episodes were more likely to participate in the survey although there is no way to determine this due to lack of data from those who declined to participate or were lost to follow-up. The results may have been subject to respondent recall bias. For example, it is possible that those with severe episodes were more likely to recall their episodes accurately. However, it was interesting to note that even those with moderate episodes had an average of 97 hours (∼12 work days) of combined work and productivity loss. Further, respondents who self-reported that they did not have HZ at the time an HZ diagnosis code was found in the healthcare database may have not remembered their HZ episode, or the diagnosis listed in the database may have been inaccurate or be a ‘rule out’ diagnosis. It is unclear how this potential bias may have influenced the results. Another limitation was that only those cases that sought care for their HZ episode were captured by the study. Finally, since some of the respondents could potentially still be having the HZ episode at the time of the interview, the estimates related to work loss in this study may be slightly under-estimated.
The results of this study indicate that HZ causes substantial disease burden from the societal perspective for individuals of 50–64 years of age in the form of work and productivity loss. The additional societal burden of work and productivity loss found in this study should be considered, along with the direct medical costs burden, to obtain a more complete picture of the impact of HZ. Policy decisions around the prevention of HZ should consider both the direct medical and indirect costs of HZ.
Conclusions
The study documents a substantial societal burden of HZ-related work and productivity loss. This is important information to take into consideration, in addition to the direct medical burden, when making policy decisions around vaccine prevention of HZ.
Declaration of interest
The research described in this manuscript was supported by funding from Merck & Co., Inc. Merck is the manufacturer of a shingles vaccine.
Transparency
Declaration of funding
The research described in this manuscript was supported by funding from Merck & Co., Inc. Merck is the manufacturer of a shingles vaccine. Merck employees (mentioned below) were integrally involved in designing the study methodology and analysis plan.
Declaration of financial/other relationships
P.K.S, J.P., R.W., & P.S. are currently employed by Merck. L.S. was a past employee of Merck.
Authors’ contributions
P.K.S. contributed to the design of the study, data analysis plan, writing of the methods and results sections of the manuscript, and final revisions of the manuscript. C.M. contributed to the design of the study, data analysis plan, conducting the analysis, and commenting on the drafts of the manuscript. J.P. contributed to the design of the study, the data analysis plan, and commenting on the drafts of the manuscript. R.W. participated in data analysis, and writing of the introduction and discussion sections of the manuscript. P.S. contributed to the design of the study, the data analysis plan, and commenting on the drafts of the manuscript. L.S. contributed to the design of the study, the data analysis plan, and commenting on the drafts of the manuscript. All authors read and approved the final manuscript.
Acknowledgments
The authors thank Dr. Paul Stang for assisting with the study design, Ms. Wendy Horn for medical writing support, and Ms. Jennifer Pawlowski and Ms. Pia L Graham for technical support with the manuscript. Paul Stang was a paid consultant of HealthCore Inc. on the study. Funding for Wendy Horn, Insight Communications Group LLC was provided by Merck & Co., Inc. Jennifer Pawlowski and Pia L Graham are employees of Merck & Co., Inc.
References
- Cohen JI, Brunell PA, Straus SE, et al. Recent advances in varicella-zoster virus infection. Ann Intern Med 1999;130:922–32
- Choo PW, Donahue JG, Manson JE, et al. The epidemiology of varicella and its complications. J Infect Dis 1995;172:706–12
- Arvin A. Aging, immunity, and the varicella-zoster virus. N Engl J Med 2005;352:2266–7
- Insinga R, Itzler RF, Pellissier JM, et al. The incidence of herpes zoster in a United States administrative database. J Gen Intern Med 2005;20:748–53
- Yawn B, Saddier P, Wollan PC, et al. A population-based study of the incidence and complication rates of herpes zoster before zoster vaccine introduction. Mayo Clin Proc 2007;82:1341–9
- Huff JC, Bean B, Balfour HH, et al. Therapy of herpes zoster with oral acyclovir. Am J Med 1988;85(Suppl 2A):84–8
- Morton P, Thomson AN. Oral acyclovir in the treatment of herpes zoster in general practice. N Z Med J 1989;102:93–5
- McKendrick MW, McGill JI, White JE, et al. Oral acyclovir in acute herpes zoster. Br Med J 1986;293:1529–32
- Johnson R, McElhaney J, Pedalino B, et al. Prevention of herpes zoster and its painful and debilitating complications. Int J Infect Dis 2007;11:S42–S7
- Gnann J, Whitley RJ. Herpes zoster. N Engl J Med 2002;347:340–6
- Schmader K. Epidemiology and impact on quality of life of postherpetic neuralgia and painful diabetic neuropathy. Clin J Pain 2002;18:350–4
- Dworkin R, Schmader KE. Treatment and prevention of postherpetic neuralgia. Clin Infect Dis 2003;36:877–82
- Katz J, Cooper EM, Walther RR, et al. Acute pain in herpes zoster and its impact on health-related quality of life. Clin Infect Dis 2004;39:342–8
- Yawn BP, Wollan PC, Kurland MJ, et al. Herpes zoster recurrences more frequent than previously reported. Mayo Clin Proc 2011;86:88–93
- Schmader K, Sloane R, Pieper C, et al. The impact of acute herpes zoster pain and discomfort on functional status and quality of life in older adults. Clin J Pain 2007;23:490–6
- Mauskopf J, Austin R, Dix L, et al. The Nottingham Health Profile as a measure of quality of life in zoster patients: convergent and discriminant validity. Qual Life Res 1994;3:431–5
- Oster G, Harding G, Dukes E, et al. Pain, medication use, and health-related quality of life in older persons with postherpetic neuralgia: results from a population-based survey. J Pain 2005;6:356–63
- Lydick E, Epstein RS, Himmelberger D, et al. Herpes zoster and quality of life: a self-limited disease with severe impact. Neurology 1995;45:52–3
- Insinga R, Itzler RF, Pellissier JM. Acute/subacute herpes zoster; healthcare resource utilsation and costs in a group of US health plans. Pharmacoeconomics 2007;25:155–69
- White R, Lenhart G, Singhal PK, et al. Incremental 1-year medical resource utilization and costs for patients with herpes zoster from a set of US health plans. PharmacoEconomics 2009;27:781–92
- Employment status of the civilian noninstitutional population by age, sex, and race. United States Bureau of Labor Statistics, 2010. Available at: http://www.bls.gov/cps/cpsaat3.pdf, accessed July 19, 2011
- Scott F, Johnson RW, Leedham-Green M, et al. The burden of herpes zoster: a prospective population based study. Vaccine 2006;24:1308–14