• Promoting a Healthier Workforce

    Health and Productivity Management

    landing-header-kc-hpm1_37748_17213
  • Health and Productivity Management Center

    Pharmaceuticals and Worker Productivity Loss: A Critical Review of the Literature

    Wayne N. Burton, MD; Alan Morrison, PhD; Albert I. Wertheimer, PhD

    Journal of Occupational and Environmental Medicine 2003; 45(6):610-621

    Many chronic illnesses that affect the working population can cause losses in productivity. The extent to which these productivity losses can be reduced by pharmacological treatment is of particular interest to employers, who bear the productivity costs and subsidize the cost of employees' health care. In the past several years, the effects of pharmaceuticals on productivity losses have been tested in numerous studies, including randomized, double-blind, placebo-controlled trials. In this article, we summarize and critically review these studies and, where appropriate, provide quantitative overviews. The evidence is very good for about a dozen drug classes that pharmaceuticals reduce productivity losses caused by respiratory illnesses (ie, asthma, allergic disorders, bronchitis, upper respiratory infections, and influenza) diabetes, depression, dysmenorrhea, and migraine. We also discuss the calculation of productivity costs, reductions in which may partially or completely offset the costs of treatment. This information should be helpful to occupational physicians who are increasingly providing recommendations on employer benefit plan designs and pharmaceutical benefits.


    From Bank One Corporation, Chicago, Illinois (Dr Burton); and the School of Pharmacy, Temple University, Philadelphia, Pennsylvania (Dr Morrison, Dr Wertheimer).

    Address correspondence to: Wayne N. Burton, MD, Corporate Medical Director, Bank One, One Bank One Plaza, Mail Code IL1-0006, Chicago, IL 60670-0006; e-mail:
    wayne n burtonmd@bankone.com.


    In the latter decades of the twentieth century, health economists calculated the economic impact of many common diseases. These cost-of-illness studies divided the costs into direct-the costs of medical treatment, generally including inpatient, outpatient, and pharmaceutical costs-and indirect costs-defined here as the cost of lost productivity. Indirect costs, or productivity costs, predominate for many chronic or recurrent conditions, such as migraine, which intermittently affect work performance but are rarely totally disabling. Pharmaceuticals, whether used for prevention or treatment, are effective against many of these diseases. The extent to which pharmaceuticals also reduce worker productivity losses is the subject of this article. This issue is of particular interest to those who bear the financial burden of productivity losses and subsidize health insurance-in the United States, typically the employer.

    In the past few years, concrete evidence of the effects of pharmaceuticals on worker productivity losses has emerged from clinical trials and other studies. This evidence provides the basis for calculating the savings in productivity costs and hence the value of the treatment to the employer. The purpose of this article is to critically review the empirical evidence of the effectiveness of pharmaceuticals in reducing worker productivity losses.

    Methods

    Literature Search

    The following databases were searched for relevant articles: Medline, International Pharmaceutical Abstracts, and the Cochrane Database of Systematic Reviews. The search was limited to articles in English, relating to human subjects, published between 1990 and September 2002. Medline was searched with keywords "absenteeism AND clinical trials," "work loss days," "work loss," "work-loss," "work lost," "productivity[title word] AND [asthma OR depression OR migraine OR hypertension OR hypercholesterolemia)[title word]," "productivity[title word] AND (influenza OR allerg* OR arthritis)[title word]," "productivity[title word] AND disorder[title word]," "productivity[title word] AND syndrome[title word]," "productivity[title word] AND anti-inflammatory[title word]," "fexofenadine AND work," "non-steroidal anti-inflammatory" AND (absenteeism OR work OR work-loss OR productivity)," "non-steroidal anti-inflammatory" AND work," "NSAID AND absenteeism," "NSAID AND "work loss," "NSAID AND productivity," "arthritis AND absenteeism," "arthritis AND work AND days," "osteoarthritis AND absenteeism," and "osteoarthritis AND work AND loss." The International Pharmaceutical Abstracts database was searched with keywords "productivity loss," "presenteeism," "productivity," "absenteeism," "work performance," "on the job performance," "productivity loss," and "indirect cost." The Cochrane Database was searched with the keyword "absenteeism" and with the title words "arthritis" and "osteoarthritis." In addition, the bibliographies of review articles1-8 and of articles identified in the literature searches were screened. Articles identified through these searches were supplemented with those known to the authors.

    Article Inclusion Criteria

    Articles were selected if they were published between 1990 and 2002, described prospective or retrospective studies of pharmaceuticals, included a control (placebo, no active treatment, or patients' usual therapy), and had a measure of productivity loss as an endpoint. As used here, the definition of productivity loss included loss of school time as well as loss of time from paid employment. Only articles reporting original productivity loss data or quantitative reviews (and references therein) of studies reporting such data were included; pharmacoeconomic modeling studies were excluded. Since the focus of the review is the impact of work loss on US employers, articles relating to drug classes unavailable in the United States or to productivity cost calculations in foreign countries were excluded.

    Analysis

    Articles are presented by disease category and then by drug class, and further discussed in terms of productivity loss and productivity cost. We defined productivity loss as a loss of work (or school) time and productivity cost as the monetary value of the productivity loss. We expressed productivity loss in terms of absenteeism (time off work or school) and "presenteeism," ie, work or school time lost because of a diminished capacity while at work.9, 10 In the description of studies of triptans for migraine, we defined presenteeism as the time worked with migraine symptoms corrected for the reduced efficiency while working with symptoms. Following previous authors, we calculated the total loss of work loss due to migraine as hours of absenteeism and presenteeism using the formula:

    EQUATION

    Equation

    Statistics reported in the text are those given by the authors of the works cited, except for the meta-analyses of studies of influenza vaccination and naproxen. For the meta-analysis of influenza vaccine studies, we calculated a rate ratio (the "effect") for each study, representing the rate of days absent because of influenza in the vaccine group divided by the rate of days absent in the control group. The rate of days absent is the number of days absent from work (or school) divided by the total number of workdays in the study period. The effect size and upper and lower 95% confidence limits are presented. If the duration of the study was not stated, an influenza season from December to March was assumed, ie, 17 weeks. A working week of 5 days was assumed. We used a random effects model in computing the combined statistic and used the Q statistic to determine whether groups of effects sizes formed a statistically homogeneous set.11 We followed a similar procedure for trials of naproxen in women with primary dysmenorrhea, except that the rate calculated was the number of patients staying at home from work or school divided by the total number of patients in the treatment or placebo groups. (Note that, in some studies, time off school was reported as well as time off work; in the trials of naproxen, the endpoint was a composite of time off work or school.) No meta-analytical approach was attempted for trials of triptans in migraine because of lack of uniformity in the statistics reported in these studies. Instead, the median productivity loss across all studies was computed, disaggregated as absenteeism and presenteeism.

    Results

    Allergies

    Antihistamines.

    Antihistamines (H1-receptor antagonists) are used to treat the symptoms of allergies, including allergic rhinitis and urticaria (hives). In the case of allergic rhinitis, worker productivity is lowered both by the condition itself and by its treatment with non-prescription antihistamines-referred to as sedating antihistamines12, 13 Attention has turned, therefore, to the newer, nonsedating antihistamines. Comparative studies have shown that, unlike sedating antihistamines, nonsedating antihistamines do not impair cognitive performance or driving proficiency in simulated tests.14,15

    The two types of antihistamines were compared in a landmark study of employees of a large insurance company who processed insurance claims.16 The study was a retrospective analysis of a linked data set containing pharmaceutical claims and daily work output records, and the endpoint was the number of claims processed per day per employee who used antihistamines. Fifty-five percent of prescriptions for allergic rhinitis were for a recent-generation H1-receptor antagonist, ie, a nonsedating antihistamine, and the remaining 45% were for older generation drugs known to have sedating effects. Employees who used sedating antihistamines experienced, on average, about an 8% reduction in daily work output in the three days following receipt of the prescription. Employees who used nonsedating antihistamines experienced on average about a 5% increase in their daily work output in the same time period. The 13% net difference in daily work output between users of nonsedating and sedating antihistamines was statistically significant and economically valuable.

    The nonsedating antihistamine, fexofenadine, was tested against placebo in a 2-week randomized, double-blind trial.17 Work productivity was measured using the Work Productivity and Activity Impairment (WPAI) instrument, a self-administered questionnaire that measures overall work impairment caused by medical problems as a composite of time missed from work and impairment while at work (estimated using a visual analog scale).18,19Overall work (or classroom) productivity is reported as a percentage of the normal value of 100%. In the trial, patients with seasonal allergic rhinitis received fexofenadine 120 mg or 180 mg once daily.17 Overall work impairment decreased by 7.1% in the 120 mg fexofenadine group and 8.7% in the 180 mg fexofenadine group, compared with a 1.8% decrease in the placebo group (P < 0.005 for comparisons with placebo). This improvement was largely the result of improvements in impairment while at work. Comparable results were obtained in two similarly-designed trials of fexofenadine 60 mg twice daily.20

    Asthma

    Beta-Agonists.

    The effects of anti-asthma agents on worker productivity losses appear to have received relatively little study. This is surprising given the extent of the productivity losses attributed to asthma 21,22 and the central importance of drug treatment in the control of this illness. The effect of the beta-agonist, inhaled terbutaline, on productivity was tested in an open-label, single-group, pretest-posttest trial of asthmatics in primary care.23 The patient population consisted of adults (60%), school-aged children (30%), and preschool children (10%). Based on daily diaries in which the patients (or their parents) recorded the effects of their asthma on their daily life, the number of work or school days missed due to asthma decreased by 57%, from 1.0 per patient in the week before beginning inhaled terbutaline treatment to 0.4 per patient in the fourth and final week of treatment (P < 0.001).

    Leukotriene Receptor Antagonists

    Leukotriene receptor antagonists are a relatively new class of anti-asthma drugs. The clinical and economic effects of the leukotriene receptor antagonist, zafirlukast, were tested in a 13-week, randomized, double-blind, placebo-controlled trial in asthmatic outpatients 12 years of age and older.24 The time absent from work or school due to asthma, recorded by patients in a daily report form, was 0.16 days per patient per month for patients who received zafirlukast and 0.35 for patients treated with placebo (P = 0.04).

    Benign Prostatic Hyperplasia

    Alpha(1)-Adrenoceptor Antagonists.

    Benign prostatic hyperplasia is commonly associated with irritating or obstructive urinary symptoms that interfere with patients' daily activities. The alpha(1)-adrenoreceptor antagonist, terazosin, tested in a one-year randomized, double-blind, placebo-controlled trial, improved disease-specific functional status but did not have any effect on work loss. 25

    Depression and Anxiety

    Traditional Antidepressants.

    lthough traditional antidepressants, such as monamine oxidase inhibitors and tricyclics, have been largely superseded by selective serotonin-reuptake inhibitors (SSRIs) and newer compounds, studies with the older drugs provided the "proof of concept" that antidepressants could reduce work loss. Depression has been studied quite intensively from the point of view of its effects on patients' ability to function in everyday life. Work loss has typically been measured using psychometric scales that yield a score related to the patient's work history rather than a direct measure of the number of days of work lost. Several clinical trials of traditional antidepressants that used the work role items of the Social Adjustment Scale as a measure of work functioning were described in an overview by Mintz et al 26 In three double-blind, placebo-controlled trials of a tricyclic antidepressant (imipramine) and monoamine oxidase inhibitors (phenelzine and isocarboxazid), 27-31 there were large, statistically significant differences between the medication and placebo groups in work functioning. The median percentage of patients who were work-impaired, defined according to scores on the Social Adjustment Scale, was 45% (range 0-50%) in the antidepressant groups in these trials, compared with 81% (range 10-82%) in the placebo groups.

    SSRIs.

    he newer SSRIs were compared with tricyclic antidepressants in two retrospective, cross-sectional analyses of depressed French outpatients.32,33 Tollefson et al measured absenteeism from work in patients with either major or minor depression who had been employed at the beginning of their depressive episode. 32 The proportions of patients currently absent from work (ie, absent at the time of their office visit) were 45.3% of those not on antidepressants, 43.6% of those receiving tricyclic antidepressants, and 26.3% of those receiving the SSRI, fluoxetine (P = 0.001 for fluoxetine versus both other groups). In a similarly designed study reported by Souetre et al, 33 the rate of current absenteeism from work was 70.2% among those not treated with antidepressants, 57.7% among those treated with tricyclics (amitriptyline and clomipramine), and 39.8% among those treated with fluoxetine (P < 0.01 for all group differences). In summary, in both studies absenteeism was lower among patients treated with SSRIs than among untreated patients or patients treated with tricyclics; absenteeism was lower among patients treated with tricyclics than among untreated patients in only one of these studies (although earlier clinical trials proved that tricyclics could reduce work loss26).

    Benzodiazepines.

    The benzodiazepine, clonazepam, is an anxiolytic used to treat panic disorder. Jacobs et al tested the effects of clonazepam on worker productivity in a 6-week randomized, double-blind, placebo-controlled trial. 34 Worker productivity was measured using the WPAI. Employed patients who received clonazepam demonstrated a statistically significant 17% improvement in WPAI scores, while there was no corresponding improvement in the placebo group (scores increased by 2%). There is a potential for bias in these results, however, because of the low number of trial participants analyzed for worker productivity (79 of 266 clinically evaluatable subjects).

    Diabetes Mellitus

    Diabetes mellitus affects approximately 5.9% of Americans and was estimated in 1997 to account for $44 billion in direct and $54 billion in indirect costs.35 There are about 8 million Americans diagnosed with diabetes and another 8 million undiagnosed. Of the 8 million diagnosed Americans, up to 800,000 have type 1 and 7 to 7.5 million have type 2 diabetes.

    Sulfonylureas.

    The health economic benefits of glipizide, a second-generation sulfonylurea, in patients with type 2 diabetes were tested in a 12-week, randomized, double-blind, placebo-controlled trial. 36 Glipizide reduced absenteeism considerably: the rate ratio for days of work missed in the placebo and glipizide groups (ie, the proportion of days of work missed during the study in the placebo group divided by the proportion of days of work missed in the glipizide group) was 4.8 (95% CI = 2.0-11.9). In addition, the glipizide patient group had higher retained employment during the trial than the placebo group (97% versus 85%; P < 0.001). Productivity losses from absenteeism per worker per month by the end of the study were $24 for men in the glipizide group versus $115 in the placebo group (based on an average wage of $116 per day for men; Bureau of the Census.)

    Dysmenorrhea

    Nonsteroidal Anti-Inflammatory Drugs (NSAIDs).

    The effect of the NSAID naproxen on absenteeism from work or school in women with primary dysmenorrhea has been tested in a number of randomized, double-blind, placebo-controlled trials.37 We present an overview of these trials in Fig. 1.38-40 The data of Henzl et al represent a pooled analysis of 12 trials involving 431 women.40 (Some of these trials have been published separately.41-43) The rate ratio (rate of absenteeism with naproxen divided by rate of absenteeism with placebo) is 0.15 (95% CI = 0.10-0.22)-a large and highly statistically significant effect.40 The two other trials shown in Fig. 1represent research by a Swedish group.38,39 The effect sizes are smaller and the confidence intervals in the first trial are wide, due to a small sample size. 38 The Q statistic (not shown) for all three studies indicates heterogeneity of effect sizes. This, and inspection of Fig. 1, indicates that the effect of naproxen was significantly greater in the Henzl et al study than in the Dandenell et al study.39,40 Nevertheless, the consensus of the trials is that naproxen reduces absenteeism from work or school in women with primary dysmenorrhea.

    Fig. 1.
    Effect of Naproxen on work or school absenteeism in women with primary dysmenorrhea.

    Figure 1 June 03

    Dyspepsia

    Dyspepsia and other gastrointestinal diseases, such as peptic ulcer disease and gastroesophageal reflux disease, are generally treated with over-the-counter antacids, H2-receptor antagonists, such as ranitidine or cimetidine or, more recently, proton pump inhibitors, such as omeprazole. Although there is evidence that upper gastrointestinal diseases reduce productivity,44-46 the productivity losses as the result of dyspepsia itself have not been determined.

    Proton Pump Inhibitors.

    The effect of omeprazole on absenteeism in patients with functional dyspepsia was tested in a follow-up study of a randomized trial.47 The mean number of hours absent from work per employed patient over the 3-month follow-up period was not significantly lower with omeprazole than with placebo.

    Migraine

    Triptans.

    Triptans (serotonin receptor agonists) are a relatively new class of drugs used in the treatment of migraine headaches. The effects of triptans on productivity loss have been studied quite intensively in studies of two designs: randomized controlled trials and single-group pretest-posttest studies. The randomized controlled trials have established the efficacy of triptans in reducing productivity losses associated with migraine attacks at work; the pretest-posttest studies have determined productivity losses in more naturalistic settings. With the exceptions noted below, the results of both types of trials are consistent. Figure 2shows the results of randomized, controlled trials of subcutaneous sumatriptan48-50 and oral rizatriptan.51,52 (Three of these trials were double-blind, placebo-controlled 48,50,52 and two were open-label trials in which the triptan was compared with patients' usual therapy.49,51) The endpoint in Fig. 2 is the net average (per patient) hours of productivity loss per migraine attack, ie, the difference between the control and triptan patient groups). This endpoint is a composite of time absent from work during an attack and time lost because of working at lower than normal efficiency. These data-time absent, time worked with symptoms, and percent effectiveness while working with symptoms-were estimated by patients and recorded in diaries as they experienced their migraine attacks. While the results of the five trials varied, in each individual trial the difference between the triptan and control groups in average time lost (absenteeism plus presenteeism) was statistically significant. The median values for the time saved per migraine attack by using a triptan were 0.64 hours absenteeism, 0.6 hours presenteeism, and 1.1 hour total time.

    Fig. 2.
    Average reduction in productivity loss during a migraine attack treated with a triptan.

    Figure 2 June 03

    The results of nine pretest-posttest trials are shown in Fig. 3.53-61 In Fig. 3, absenteeism and presenteeism, expressed as the average days of productivity loss per patient per month, are presented for the pretest phase, when subjects used their usual therapy for migraine attacks, and the posttest phase, when subjects used sumatriptan. Each phase typically lasted from 3 to 12 months. Productivity losses were lower with sumatriptan in every trial, although there is considerable variation in the results. The median days of productivity loss per patient per month for all nine trials was 1.69 days with patients' usual therapy and 0.67 days with sumatriptan (median 0.75 days absenteeism and median 0.94 days presenteeism with patients' usual therapy and median 0.29 days absenteeism and median 0.38 days presenteeism with sumatriptan). (The median days of productivity loss in trials in which productivity loss was recorded as migraines occurred rather than at a later time-see Discussion-was 0.66 days per month with patients' usual therapy and 0.33 days per month with sumatriptan-median 0.42 days absenteeism and median 0.24 days presenteeism with patients' usual therapy, median 0.22 days absenteeism and median 0.11 days presenteeism with sumatriptan).53

    Fig. 3.
    Reduction in average days of productivity loss per patient per month with sumatriptan.

    Figure 3 June 03

    Schulman et al calculated productivity costs in a post hoc economic analysis of data from their randomized, placebo-controlled trial of sumatriptan.50 The hourly wage by occupational category was based on national averages obtained from the Bureau of Labor Statistics. Three migraine attacks per month were assumed. Expressed in $US 1996, the average cost per patient per month of lost productivity was $112.99 with placebo and $59.76 with sumatriptan, a saving of $53.23 per month.

    Respiratory Infection

    The estimated total cost to employers of 11 respiratory infections (including acute bronchitis, sore throat, pneumonia, and influenza, which are discussed here) in 1997 was an astronomical $112 billion.62 Work absences account for about one-third of the costs of respiratory illnesses to employers.62

    Antibiotics.

    Work loss was an endpoint in two trials of the use of antibiotics to treat respiratory infection. The effects of erythromycin in patients with acute bronchitis were tested in a randomized, double-blind, placebo-controlled trial conducted at three primary care centers in North Carolina.63 Patients treated with erythromycin missed an average of 0.81 days of work, compared with 2.16 for patients treated with placebo (P < 0.02). In contrast, antibiotic treatment of sore throat did not reduce work loss in an open, randomized trial of UK residents.64 This result was, perhaps, not surprising because prescribing antibiotics for sore throat only marginally affected the resolution of symptoms.

    Antiprotozoals.

    Pentamidine is an aromatic diamidine known to have activity against Pneumocystis carinii, a common cause of pneumonia in HIV-infected individuals. Aerosolized pentamidine, tested against no treatment in a randomized controlled trial, had prophylactic efficacy but no significant effect on absenteeism.65

    Antivirals.

    Zanamivir and oseltamivir are inhibitors of influenza virus neuraminidase that can be used to treat patients who have already developed symptoms of influenza. The effect of a 5-day course of inhaled zanamivir on absenteeism in people with influenza of less than 2 days duration was determined in a randomized, double-blind, placebo-controlled trial carried out in 14 countries in Europe and North America during the 1994-95 season.66 Workers in the placebo group missed an average of 3.3 days by the day 6 visit, compared with 2.8 days for workers treated with zanamivir twice daily and 2.5 days for workers treated with zanamivir four times daily. The difference between four-times-daily zanamivir and placebo was statistically significant (P = 0.031).

    In other studies of neuraminidase inhibitors, the endpoint was time to return to normal activities rather than a direct measure of work loss. In a pooled analysis of six studies (all randomized, double-blind, placebo-controlled trials) patients treated with inhaled zanamivir (10 mg twice daily for 5 days) returned to normal activities in 6.0 days compared with 7.0 days for those treated with placebo (P < 0.001). 67 In another pooled analysis, adults in all zanamivir treatment arms also returned to their normal activities in about one day less than with placebo: difference -0.874 days 95% CI = (-1.661 to -0.086; P < 0.05).68 Similar results for time to return to normal activities were reported in a randomized, double-blind, placebo-controlled trial of oseltamivir in healthy adults of working age.69

    Influenza Virus Vaccination.

    The effects of influenza vaccination on influenza symptoms and associated work loss have been subjected to the most rigorous and intensive testing. At least eight randomized, controlled trials have been conducted, most in the United States. We present an overview of the studies with adults in Fig. 4. Six of these studies are randomized, double-blind, placebo-controlled trials70-74 and two are case-control studies. 75,76 The endpoint depicted in Fig. 4 is the rate ratio of work loss days due to influenza (or influenza-like illness) during an influenza season. Use of an influenza vaccine resulted in a statistically significant reduction in lost workdays in every study, with the exception of a trial conducted in the 1997 to 1998 season.74 The ineffectiveness of the vaccine in this instance was attributed to a poor match between the specific vaccine used and the influenza strains then circulating. 74 Excluding this value, the median numbers of work-loss days per person in the vaccine and control groups were 0.32 and 0.63, respectively, difference -0.28 days (range -0.04 to -0.52). The Q statistic (not shown) indicates heterogeneity in study effect sizes even when the Bridges study 74 is excluded. This might reflect variation in the virulence of the influenza strains, the efficacy of the vaccines, and the employee populations studied.

    Fig. 4.
    Relative risk of work-loss days in trials of influenza vaccination in adults. Bridges 2000a refers to the 1997 to 1998 influenza season and Bridges 2000b to the 1998 to 1999 season reported by Bridges et al.74

    Figure 4 June 03

    In addition to studies of adults, two trials showed that influenza vaccination reduced absenteeism in preschool or school children.77,78 Colombo et al randomly assigned healthy preschool children to either influenza vaccination or no treatment.78 During the 5-month follow-up period (1995-1996), the mean duration of absenteeism from day care was 2.3 days in unvaccinated children and 0.5 days in vaccinated children (P < 0.001). Khan et al. reported a randomized, single-blind, placebo-controlled trial of two different vaccines in Russian schoolchildren.77 The risk of school absence because of respiratory illness during the period of peak influenza activity was reduced by 56% with a US inactivated split-virus vaccine (P = 0.06 versus placebo) and by 47% with a Russian live, attenuated virus vaccine (P = 0.06 versus placebo).

    Productivity costs were estimated for several of the influenza vaccine trials. In the trial reported by Nichol et al., the enrollees were employees in the Minneapolis-St. Paul area.70 Work-loss costs were based on the 1994 median earnings of full-time U.S. workers ($93.40 per day; Bureau of Labor Statistics). The productivity costs per employee were $5.84 for work time lost for vaccination (30 minutes per worker), $1.87 for work loss as a result of side effects from vaccination, and -$48.57 for the work loss avoided by vaccination: a net productivity saving of $40.86. There was also a net saving in direct costs. The direct costs per employee were $10 for vaccination, $0.70 for medical care of side effects, and -$16.68 for the medical care avoided for respiratory illness.

    Conversely, in the trial reported by Bridges et al, vaccination incurred a net productivity cost.74 The subjects were full-time employees of Ford Motor Company in Dearborn, Michigan. Direct and indirect costs were computed in essentially the same way as in the study by Nichol et al described above. An 8-hour workday was valued at $235.12 as wages plus benefits for professional specialty and technical civilians in goods-producing industries in large U.S. companies in 1999. Economic analysis of vaccination in the 1998-1999 season, when the vaccine was effective against circulating influenza strains, indicated that both the indirect and direct costs in the vaccinated group exceeded those in the control group. The cost of lost workdays was $19.40 per employee in the vaccine group and $28.43 in the control group, but the time taken for vaccination (costing $14.70) meant that productivity costs were greater in the vaccinated group. Similarly, the direct costs of influenza-like illness were $6.22 in the vaccine group and $9.71 in the control group, but this was offset by the cost of the vaccination ($10). The difference in productivity cost outcome between the studies reported by Nichol et al. and by Bridges et al. might be related to the severity of the influenza and/or the propensity of the employees to be absent from work due to illness: absenteeism due to influenza-like illness in the placebo group-and, hence, the potential for reduction in productivity costs-was considerably greater in the former than in the latter trial.70,74

    Vaccination was cost-saving in the two case-control studies of influenza vaccination.75,76 In the first of these studies, the subjects were employees of a nuclear facility in Washington State.76 The productivity costs, based on the average salary of $175.24 (in $US 1994) per 8-hour day for employees at the nuclear facility, were estimated to be $38.12 lower per vaccinated employee than per unvaccinated employee. Direct costs were also estimated to be lower in the vaccinated group. In the study reported by Campbell et al, the subjects were employees of a North Carolina textile corporation.75 Based on the average hourly wage of $6.50, the saving in lost productivity due to influenza was $37.78 per employee. This exceeded the $22.13 cost of vaccination (which included: $3.50 per employee for vaccine, syringes, and alcohol preparations; $15.38 per employee in nursing time to contact employees, administer the vaccine, and complete all paperwork; and $3.25 per employee for the half-hour of work time taken for the vaccination).

    Discussion

    Productivity Study Designs

    There is convincing evidence in the literature of a reduction in worker productivity losses caused by illness for over a dozen drug classes (Table 1). Evidence is particularly extensive in the cases of influenza vaccination, triptans for migraine, and non-sedating antihistamines for allergic disorders. Much of the evidence relating the use of pharmaceuticals to reductions in productivity losses comes from randomized, double-blind, placebo-controlled trials and can be accepted as valid; this applies, with some trivial exceptions, to all of the randomized controlled trials listed in Table 1. Some studies had less rigorous designs, eg, case controlled studies or single-group, pretest-posttest studies. Pretest-posttest studies may produce spurious results because of time trends in the data unrelated to the intervention, particularly when a high-risk patient group is selected for study. In several instances, however, where less rigorous study designs were used-influenza vaccination, 75,76 non-sedating antihistamines,16 and triptans for migraine53-61-the basic results were borne out by those of randomized trials. The only exceptions to this, ie, studies without corroboration from randomized controlled trials, are a pretest-posttest trial of a beta-agonist for asthma23 and two retrospective, cross-sectional analyses of SSRIs for depression.32,33

    Table 1 June 03

    Not all drug trials produced positive results: a null result was seen in trials of an alpha(1)-adrenoceptor antagonist for benign prostatic hyperplasia,25 antibiotic treatment of sore throat,64 an antiprotozoal for HIV/AIDs,65 and a proton-pump inhibitor for dyspepsia.47 These studies were not necessarily designed, however, to have the statistical power to detect a significant difference in absenteeism, which was often a secondary study endpoint. Absenteeism was, however, one of the main study endpoints in trials of influenza vaccination, triptans, and NSAIDs, where work loss was reduced.

    Measurement of Productivity Loss

    In the key study of the effects of antihistamines on productivity loss, the subjects were insurance claims processors and their productivity was measured directly as the number of claims processed per day.16 In very few studies, however, have the decreases in worker productivity associated with illness been measured directly. Instead, productivity is typically measured indirectly in terms of loss of work-time, because objective worker productivity data is available only for occupations such as medical claims processors, telephone customer service operators, and bank check processors-the productivity of most employees, particularly knowledge workers, is difficult to measure objectively. Furthermore, in most of the studies reviewed here, the method of measuring productivity loss was not only indirect but was also subjective, ie, by patient self-report.

    Several instruments have been developed to quantify self-reported decrements in worker productivity associated with chronic disease or the use of pharmaceuticals. 79 Examples are the WPAI, a general instrument that was used in the trials of non-sedating antihistamines and clonazepam,17,20,34 and the Migraine Work and Productivity Loss Questionnaire (MWPLQ), a disease-specific instrument developed to measure productivity loss in migraine.51 Nevertheless, patients' estimates of their percent effectiveness while working with symptoms, which appear in these instruments and, in particular, in the calculation of presenteeism in migraine studies, remain subjective approximations.

    Recall Bias

    Patient report is likely to be most reliable when the period of recall is short. Many studies used patient diaries for daily recording of study events, eg, in trials of beta-agonists for asthma,23 influenza virus vaccines,71,72 leukotriene receptor agonists for asthma,24 and NSAIDs for dysmenorrhea.39 And, in the randomized trials of triptans for migraine, events were recorded in diaries on an hourly basis48,50 or when a migraine headache resolved. 51, 52 Absenteeism recorded in this way is likely to be reliable. In other studies, patients were asked to recall absenteeism for prior periods of weeks or months-a less reliable procedure. Thus, in the un-blinded, pretest-posttest studies of sumatriptan, it may not be coincidental that the study with the longest recall period (12 to 24 months) resulted in the greatest average monthly work loss during the usual-therapy phase (6.2 days), 60 greater than in studies with a 3-month recall period (2.4 days 54 and 2.0 days 61), and greater in turn than in studies in which a daily migraine diary was kept (mean, range of 3 studies, 0.89 days, 0.35-1.36 days);53,55,57 the results of the last three studies are the most reliable.

    Estimation of Productivity Costs

    Although absenteeism is a relatively concrete endpoint that can be directly observed, productivity costs are more difficult to measure. There are two main approaches to estimating the monetary value of productivity losses-the human capital approach and the friction cost approach.80 Lofland et al 81 compared the two approaches in an analysis of the posttest period of a pretest-posttest trial of sumatriptan.61 The productivity cost of migraine calculated by Lofland et al using the human capital approach was $110.40 per employee per month (in $US 1996)-the same value as calculated under the base case of the friction approach, under which it was assumed that the friction period was equal to the time of employee absence, that the value of output production was the same as the employee's occupational wages plus benefits, and that the cost of hiring, replacing or training workers was zero. If, as has been suggested, absences shorter than those that lead to worker replacement were valued at only 80% of the production value lost in that period,82 then the productivity cost was $88.32 per employee per month. If the work lost because of short-term absences did not incur productivity costs-because the lost productivity might be canceled, postponed, or made up after return to work-Lofland et al estimated the productivity cost to be as low as $26.53 per employee per month, which is about one-quarter of the human capital estimate.

    Future Research

    There is solid evidence that pharmaceuticals reduce worker productivity losses due to illness for over a dozen drug classes available in the United States. Upper respiratory infections, particularly influenza, are heavily represented among the illnesses, but asthma, allergic disorders, depression, diabetes, dysmenorrhea, and migraine are also represented. Many drug classes, however, and several important illnesses are underrepresented. For instance, there appears to be little research into the effects of pharmaceuticals in reducing work loss due to arthritis, a common cause of disability that imposes significant indirect cost on employers.83,84

    There is also little research into the relationship between work loss and pharmacological treatment of workers' dependents, including children. Treatment of children's or other family members' infectious illnesses may prevent the spread of infection among family members, 85 especially to employed family members, and hence contribute to lowering of worker productivity. Although several studies have demonstrated that pharmaceuticals can reduce school absenteeism in children, the associated work loss reduction by their parents, who are no longer obliged to care for them at home, has not been quantified.

    Estimates of productivity losses have usually been based on subjective measurement and their reliability would be improved if objective measures of productivity loss were used. This is particularly the case with presenteeism, which is an important component of productivity loss that, except in the case of migraine, has been underresearched. In all of the studies reviewed, productivity costs were calculated rather than being directly observed. These calculations are to some extent speculative because there is disagreement about their theoretical underpinnings.86,87 In addition, productivity cost calculations applied to specific groups of employees cannot necessarily be generalized: like politics, all productivity costs are local. Finally, we have ignored the role of pharmaceuticals in reducing losses in non-work time, which were estimated in several of the studies, because our theme is the productivity costs to the employer.

    Conclusions

    This critical review has shown that pharmaceuticals can reduce productivity losses in several key illnesses of the employed population. The studies of influenza vaccination, treatment of migraine with triptans, and treatment of allergic rhinitis with non-sedating antihistamines in particular represent significant contributions to the field, although notable work has been reported for other drug classes. Purchasers of health care, including employers, should consider the cost savings that result when worker productivity losses are reduced-savings that offset the cost of the pharmacological treatment. The costs of some classes of medications may best be viewed as an investment in the improved health and productivity of the workforce.

    Acknowledgment

    The authors thank Batoul Senhaji for her assistance in researching the productivity literature and W. Zhang for providing unpublished meta-analytical data.

    Preparation of this article was supported by the National Pharmaceutical Council.

    References

    1. A closer look at allergies. Asthma and Allergy Foundation of America and the National Pharmaceutical Council, Reston, VA; 2001.

    2. Goetzel RZ, Ozminkowski RJ, Meneades L, Stewart M, Schutt DC. Pharmaceuticals-cost or investment? An Employer's Perspective. J Occup Environ Med. 2000; 42: 338-351.

    3. Heaney CA, Goetzel RZ. A review of health-related outcomes of multi-component worksite health promotion programs. Am J Health Promot. 1997; 11: 290-307.

    4. McCunney RJ. Health and productivity: a role for occupational health professionals. J Occup Environ Med. 2001; 43: 30-35.

    5. Simon GE, Barber C, Birnbaum HG, et al. Depression and work productivity: the comparative costs of treatment versus nontreatment. J Occup Environ Med. 2001; 43: 2-9.

    6. Solomon GD, Santanello N. Impact of migraine and migraine therapy on productivity and quality of life. Neurology 2000; 55 ( 9 Suppl 2): S29-S35.

    7. Stang P, Cady R, Batenhorst A, Hoffman L. Workplace productivity. A review of the impact of migraine and its treatment. Pharmacoeconomics. 2001; 19: 231-244.

    8. Foster RH, Plosker GL. Glipizide. A review of the pharmacoeconomic implications of the extended-release formulation in type 2 diabetes mellitus. Pharmacoeconomics. 2000; 18: 289-306.

    9. Burton WN, Conti DJ, Chen CY, Schultz AB, Edington DW. The role of health risk factors and disease on worker productivity. J Occup Environ Med. 1999; 41: 863-877.

    10. Burton WN, Conti DJ, Chen CY, Schultz AB, Edington DW. The economic burden of lost productivity due to migraine headache: a specific worksite analysis. J Occup Environ Med. 2002; 44: 523-529.

    11. Petitti DB. Meta-Analysis, Decision Analysis, and Cost-Effectiveness Analysis. Methods for Quantitative Synthesis in Medicine. Monographs in Epidemiology and Biostatistics, Volume 24. New York: Oxford University Press; 1994.

    12. Crystal-Peters J, Crown WH, Goetzel RZ, Schutt DC. The cost of productivity losses associated with allergic rhinitis. Am J Manage Care. 2000; 6: 373-378.

    13. Burton WN, Conti DJ, Chen CY, Schultz AB, Edington DW. The impact of allergies and allergy treatment on worker productivity. J Occup Environ Med. 2001; 43: 64-71.

    14. Kay GG, Berman B, Mockoviak SH, et al. Initial and steady-state effects of diphenhydramine and loratadine on sedation, cognition, mood, and psychomotor performance. Arch Intern Med. 1997; 157: 2350-2356.

    15. Weiler JM, Bloomfield JR, Woodworth GG, et al. Effects of fexofenadine, diphenhydramine, and alcohol on driving performance. A randomized, placebo-controlled trial in the Iowa driving simulator. Ann Intern Med. 2000; 132: 354-363.

    16. Cockburn IM, Bailit HL, Berndt ER, Finkelstein SN. Loss of work productivity due to illness and medical treatment. J Occup Environ Med. 1999; 41: 948-953.

    17. Meltzer EO, Casale TB, Nathan RA, Thompson AK. Once-daily fexofenadine HCl improves quality of life and reduces work and activity impairment in patients with seasonal allergic rhinitis. Ann Allergy Asthma Immunol. 1999; 83: 311-317.

    18. Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics. 1993; 4: 353-365.

    19. Reilly MC, Tanner A, Meltzer EO. Work, classroom and activity impairment instruments. Validation studies in allergic rhinitis. Clin Drug Invest. 1996; 11: 278-288.

    20. Tanner LA, Reilly M, Meltzer EO, Bradford JE, Mason J. Effect of fexofenadine HCl on quality of life and work, classroom, and daily activity impairment in patients with seasonal allergic rhinitis. Am J Managed Care 1999; 5 ( Suppl): S235-S247.

    21. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. N Engl J Med. 1992; 326: 862-866.

    22. Smith DH, Malone DC, Lawson KA, Okamoto LJ, Battista C, Saunders WB. A national estimate of the economic costs of asthma. Am J Respir Crit Care Med. 156: 787-793, 1997.

    23. Northfield M, Patel RK, Richardson A, Taylor MD, Richardson PD. Lifestyle changes in mild asthma during intermittent symptom-related use of terbutaline inhaled via 'Turbohaler'. Curr Med Res Opin. 1991; 12: 441-449.

    24. Suissa S, Dennis R, Ernst P, Sheehy O, Wood-Dauphinee S. Effectiveness of the leukotriene receptor antagonist zafirlukast for mild-to-moderate asthma: a randomized, double-blind, placebo-controlled trial. Ann Intern Med. 1997; 126: 177-183.

    25. Hillman AL, Schwartz JS, Willian MK, et al. The cost-effectiveness of terazosin and placebo in the treatment of moderate to severe benign prostatic hyperplasia. Urology. 1996; 47: 169-178.

    26. Mintz J, Mintz LI, Arruda MJ, Hwang SS. Treatments of depression and the functional capacity to work. Arch Gen Psychiatry. 1992; 49: 761-768.

    27. Stewart JW, Quitkin FM, McGrath PJ, et al. Social functioning in chronic depression: effect of 6 weeks of antidepressant treatment. Psychiatry Res. 1988; 25: 213-222.

    28. Quitkin FM, McGrath PJ, Stewart JW, et al. Phenelzine and imipramine in mood reactive depressives. Further delineation of the syndrome of atypical depression. Arch Gen Psychiatry. 1989; 46: 787-793.

    29. Davidson JR, Giller EL, Zisook S, Overall JE. An efficacy study of isocarboxazid and placebo in depression, and its relationship to depressive nosology. Arch Gen Psychiatry 1988; 45: 120-127.

    30. Giller E Jr, Bialos D, Riddle MA, Waldo MC. MAOI treatment response: multiaxial assessment. J Affect Disord. 1988; 14: 171-175.

    31. Elkin I, Shea MT, Watkins JT, et al. National Institute of Mental Health Treatment of Depression Collaborative Research Program. General effectiveness of treatments. Arch Gen Psychiatry. 1989; 46: 971-982;discussion 983.

    32. Tollefson GD, Souetre E, Thomander L, Potvin JH. Comorbid anxious signs and symptoms in major depression: impact on functional work capacity and comparative treatment outcomes. Int Clin Psychopharmacol. 1993; 8: 281-293.

    33. Souetre E, Lozet H, Cimarosti I. Predicting factors for absenteeism in patients with major depressive disorders. Eur J Epidemiol. 1997; 13: 87-93.

    34. Jacobs RJ, Davidson JR, Gupta S, Meyerhoff AS. The effects of clonazepam on quality of life and work productivity in panic disorder. Am J Manage Care. 1997; 3: 1187-1196.

    35. Musich SA, Burton WN, Edington DE. Costs and benefits of prevention and disease management. Dis Manage Health Outcomes. 1999; 5: 153-166.

    36. Testa MA, Simonson DC. Health economic benefits and quality of life during improved glycemic control in patients with type 2 diabetes mellitus: a randomized, controlled, double-blind trial. JAMA. 1998; 280: 1490-1496.

    37. Zhang WY, Li Wan Po A. Efficacy of minor analgesics in primary dysmenorrhoea: a systematic review. Br J Obstet Gynaecol. 1998; 105: 780-789.

    38. Jacobson J, Cavalli-Bjorkman K, Lundstrom V, Nilsson B, Norbeck M. Prostaglandin synthetase inhibitors and dysmenorrhea. A survey and personal clinical experience. Acta Obstet Gynecol Scand Suppl. 1979; 87: 73-79.

    39. Dandenell LO, Lalos O, Lisciak J, Sandstrom B, Barany S, Nilsson B. Clinical experience of naproxen in the treatment of primary dysmenorrhea. Acta Obstet Gynecol Scand Suppl. 1979; 87: 95-100.

    40. Henzl MR, Massey S, Hanson FW, Buttram VC, Rosenwaks Z, Pauls FD. Primary dysmenorrhea: the therapeutic challenge. J Reprod Med. 1980; 25 ( 4 Suppl): 226-235.

    41. Hanson FW, Izu A, Henzl MR. Naproxen sodium in dysmenorrhea. Its influence in allowing continuation of work/school activities. Obstet Gynecol 1978; 52: 583-587.

    42. Rosenwaks Z, Jones GS, Henzl MR, Dubin NH, Ghodgaonkar RB, Hoffman S. Naproxen sodium, aspirin, and placebo in primary dysmenorrhea. Reduction of pain and blood levels of prostaglandin F2-alpha metabolite. Am J Obstet Gynecol 1981; 140: 592-598.

    43. Hanson FW. Naproxen sodium, ibuprofen and a placebo in dysmenorrhea. J Reprod Med 1982; 27: 423-427.

    44. Eggleston A, Farup C, Meier R. The Domestic/International Gastroenterology Surveillance Study (DIGEST): design, subjects and methods. Scand J Gastroenterol Suppl. 1999; 231: 9-14.

    45. Haycox A, Einarson T, Eggleston A. The health economic impact of upper gastrointestinal symptoms in the general population: results from the Domestic/International Gastroenterology Surveillance Study (DIGEST). Scand J Gastroenterol Suppl. 1999; 231: 38-47.

    46. Henke CJ, Levin TR, Henning JM, Potter LP. Work loss costs due to peptic ulcer disease and gastroesophageal reflux disease in a health maintenance organization. Am J Gastroenterol. 2000; 95: 788-792.

    47. Meineche-Schmidt V, Talley NJ, Pap A, et al. Impact of functional dyspepsia on quality of life and health care consumption after cessation of antisecretory treatment. A multicentre 3-month follow-up study. Scand J Gastroenterol. 1999; 34: 566-574.

    48. Cady RC, Ryan R, Jhingran P, O'Quinn S, Pait DG. Sumatriptan injection reduces productivity loss during a migraine attack: results of a double-blind, placebo-controlled trial. Arch Intern Med. 1998; 158: 1013-1018.

    49. Laloux P, Vakaet A, Monseu G, Jacquy J, Bourgeois P, van der Linden C. Subcutaneous sumatriptan compared with usual acute treatments for migraine: clinical and pharmacoeconomic evaluation. Acta Neurol Belg. 1998; 98: 332-341.

    50. Schulman EA, Cady RK, Henry D, et al. Effectiveness of sumatriptan in reducing productivity loss due to migraine: results of a randomized, double-blind, placebo-controlled clinical trial. Mayo Clin Proc. 2000; 75: 782-789.

    51. Davies GM, Santanello N, Gerth W, Lerner D, Block GA. Validation of a migraine work and productivity loss questionnaire for use in migraine studies. Cephalalgia. 1999; 19: 497-502.

    52. Dasbach EJ, Carides GW, Gerth WC, Santanello NC, Pigeon JG, Kramer . Work and productivity loss in the rizatriptan multiple attack study. Cephalalgia. 2000; 20: 830-834.

    53. Adelman JU, Sharfman M, Johnson R, et al. Impact of oral sumatriptan on workplace productivity, health-related quality of life, healthcare use, and patient satisfaction with medication in nurses with migraine. Am J Manage Care. 1996; 2: 1407-1416.

    54. Cohen JA, Beall D, Beck A, et al. Sumatriptan treatment for migraine in a health maintenance organization: economic, humanistic, and clinical outcomes. Clin Ther. 1999; 21: 190-204.

    55. Cortelli P, Dahlof C, Bouchard J, et al. A multinational investigation of the impact of subcutaneous sumatriptan. III: Workplace productivity and non-workplace activity. Pharmacoeconomics. 1997; 11 ( Suppl 1): 35-42.

    56. Dahlof CG. Health-related quality of life under six months' treatment of migraine-an open clinic-based longitudinal study. Cephalalgia 1995; 15: 414-422;discussion 336.

    57. Gross ML, Dowson AJ, Deavy L, Duthie T. Impact of oral sumatriptan 50 mg on work productivity and quality of life in migraineurs. Br J Med Econ. 1996; 10: 231-246.

    58. Jhingran P, Cady RK, Rubino J, Miller D, Grice RB, Gutterman DL. Improvements in health-related quality of life with sumatriptan treatment for migraine. J Fam Pract. 1996; 42: 36-42.

    59. Larbig W, Bruggenjurgen B. Work productivity and resource consumption among migraineurs under current treatment and during treatment with sumatriptan: an economic evaluation of acute treatment in moderate to severe migraineurs. Headache Quart Curr Treat Res. 1997; 8: 237-246.

    60. Legg RF, Sclar DA, Nemec NL, Tarnai J, Mackowiak JI. Cost-effectiveness of sumatriptan in a managed care population. Am J Manage Care. 1997; 3: 117-122.

    61.Lofland JH, Johnson NE, Batenhorst AS, Nash DB. Changes in resource use and outcomes for patients with migraine treated with sumatriptan: a managed care perspective. Arch Intern Med. 1999; 159: 857-863.

    62. Birnbaum HG, Morley M, Greenberg PE, Colice GL. Economic burden of respiratory infections in an employed population. Chest. 2002; 122: 603-611.

    63. King DE, Williams WC, Bishop L, Shechter A. Effectiveness of erythromycin in the treatment of acute bronchitis. J Fam Pract. 1996; 42: 601-605.

    64. Little P, Williamson I, Warner G, Gould C, Gantley M, Kinmonth AL. Open randomised trial of prescribing strategies in managing sore throat. BMJ. 1997; 314: 722-727.

    65. Lidman C, Berglund O, Tynell E, Lindback S, Elvin K. Aerosolized pentamidine as primary prophylaxis for Pneumocystis carinii pneumonia: efficacy, mortality and morbidity. AIDS. 1994; 8: 935-939.

    66. Aoki FY, Fleming DM, Griffin AD, Lacey LA, Edmundson S. Impact of zanamivir treatment on productivity, health status and healthcare resource use in patients with influenza. Zanamivir Study Group. Pharmacoeconomics. 2000; 17: 187-195.

    67. Monto AS, Webster A, Keene O. Randomized, placebo-controlled studies of inhaled zanamivir in the treatment of influenza A and B: pooled efficacy analysis. J Antimicrob Chemother. 1999; 44 ( Suppl B): 23-29.

    68. Burls A, Clark W, Stewart T, et al. Zanamivir for the treatment of influenza in adults: a systematic review and economic evaluation. Health Technol Assess. 2002; 6: 1-87.

    69. Treanor JJ, Hayden FG, Vrooman PS, et al. Efficacy and safety of the oral neuraminidase inhibitor oseltamivir in treating acute influenza: a randomized controlled trial. US Oral Neuraminidase Study Group. JAMA. 2000; 283: 1016-1024.

    70. Nichol KL, Lind A, Margolis KL, et al. The effectiveness of vaccination against influenza in healthy, working adults. N Engl J Med. 1995; 333: 889-893.

    71. Nichol KL, Mendelman PM, Mallon KP, et al. Effectiveness of live, attenuated intranasal influenza virus vaccine in healthy, working adults: a randomized controlled trial. JAMA. 1999; 282: 137-144.

    72. Saxen H, Virtanen M. Randomized, placebo-controlled double blind study on the efficacy of influenza immunization on absenteeism of health care workers. Pediatr Infect Dis J. 1999; 18: 779-783.

    73. Wilde JA, McMillan JA, Serwint J, Butta J, O'Riordan MA, Steinhoff MC. Effectiveness of influenza vaccine in health care professionals: a randomized trial. JAMA. 1999; 281: 908-913.

    74. Bridges CB, Thompson WW, Meltzer MI, et al. Effectiveness and cost-benefit of influenza vaccination of healthy working adults: a randomized controlled trial. JAMA. 2000; 284: 1655-1663.

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

    76. Dille JH. A worksite influenza immunization program. Impact on lost work days, health care utilization, and health care spending. AAOHN J. 1999; 47: 301-309.

    77. Khan AS, Polezhaev F, Vasiljeva R, et al. Comparison of US inactivated split-virus and Russian live attenuated, cold-adapted trivalent influenza vaccines in Russian schoolchildren. J Infect Dis. 1996; 173: 453-456.

    78. Colombo C, Argiolas L, La Vecchia C, Negri E, Meloni G, Meloni T. Influenza vaccine in healthy preschool children. Rev Epidemiol Sante Publique. 2001; 49: 157-162.

    79. Measuring Employee Productivity: A Guide to Self-Assessment Tools. Glen Allen, VA: Institute for Health & Productivity Management; 2001.

    80. Koopmanschap MA, van Ineveld BM. Towards a new approach for estimating indirect costs of disease. Soc Sci Med. 1992; 34: 1005-1010.

    81. Lofland JH, Locklear JC, Frick KD. Different approaches to valuing the lost productivity of patients with migraine. Pharmacoeconomics. 2001; 19: 917-925.

    82. Koopmanschap MA, Rutten FFH, van Ineveld BM, van Roijen L. The friction cost method for measuring indirect costs of disease. J Health Econ. 1995; 14: 171-189.

    83. Elders MJ. The increasing impact of arthritis on public health. J Rheumatol Suppl. 2000; 60: 6-8.

    84. Birnbaum HG, Barton M, Greenberg PE, et al. Direct and indirect costs of rheumatoid arthritis to an employer. J Occup Environ Med. 2000; 42: 588-596.

    85. Welliver R, Monto AS, Carewicz O, et al. Effectiveness of oseltamivir in preventing influenza in household contacts: a randomized controlled trial. JAMA. 2001; 285: 748-754.

    86. Liljas B. How to calculate indirect costs in economic evaluations. Pharmacoeconomics. 1998; 13: 1-7.

    87. Brouwer WB, Koopmanschap MA. How to calculate indirect costs in economic evaluations. Pharmacoeconomics. 1998; 13: 563-569.

    Journal of Occupational and Environmental Medicine 2003; 45(6):610-621 Copyright © 2003 Lippincott Williams & Wilkins All rights reserved

    Go top