VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S3

AbstractAllergic rhinitis is a common con-

dition in managed care populations.The direct medical cost of rhinitisexceeded $3 billion in 1996, and anadditional cost of $4 billion resultedfrom the exacerbation of other con-comitant conditions, such as asthmaor otitis media. Costs continued toincrease in 1999; sales of prescriptionantihistamines and nasal steroidsexceeded $3 billion and $1 billion,respectively. The indirect costs ofallergic rhinitis include lost work pro-ductivity, reduced performance andlearning, and increased workplace andtraffic accidents. Rhinitis treatmentsinclude allergen avoidance, over-the-counter (OTC) sedating antihista-mines, nonsedating antihistamines,nasal steroids, and immunotherapy.Allergen avoidance strategies forpatients with asthma and rhinitis areineffective or are of very limited ben-efit. Allergists criticize the use of OTCsedating antihistamines, which are

associated with reduced learning andperformance even when sedationdoes not occur. Evidence-based litera-ture reviews of clinical trials haveshown that nasal steroids are moreeffective than nonsedating antihista-mines in the treatment of rhinitis. Themost commonly prescribed nasalsteroid, fluticasone, has been shownto be effective in treating rhinitis andin improving patients’ quality of life. Itis also more cost effective than themost commonly prescribed antihista-mine, loratadine. Clinical trials haveindicated that immunotherapy isexpensive and of limited benefit. Asthese evidence-based findings areused to develop managed care treat-ment guidelines, nasal steroids arelikely to be recommended as the first-line treatment for rhinitis, whichshould result in lower treatment costsand improved outcomes for patientswith rhinitis.

(Am J Manag Care 2000;6(suppl):S3-S15)

As competition increases in themanaged care arena, organiza-tions are looking for methods

of lowering treatment costs andimproving outcomes for those withdiseases other than asthma or dia-betes. One such condition is allergicrhinitis. The chronic nature ofrhinitis, its high prevalence, the

Optimizing Quality of Care and Cost Effectiveness in Treating Allergic Rhinitis in a Managed Care Setting

David B. Nash, MD, MBA, FACP; Sean D. Sullivan, PhD; and John Mackowiak, PhD

. . .REPORT . . .

From the Office of Health Policy and ClinicalOutcomes, Thomas Jefferson University Hospital,Jefferson Medical College, Philadelphia, PA(D.B.N.); Graduate and Post-Doctoral FellowshipPrograms, Department of Pharmacy, Universityof Washington, Seattle, WA (S.D.S.); Center forOutcomes Research, Chapel Hill, NC (J.M.).

Address correspondence to: JohnMackowiak, PhD, President, Center forOutcomes Research, 1025 Pinehurst Drive, Suite100, Chapel Hill, NC 27514.

rate of provider visits, and the use ofknown effective treatments makethis a likely target for qualityimprovement programs. Recently,such a quality initiative was imple-mented by the Lovelace HealthSystems in New Mexico. The interven-tion team identified variations inrhinitis treatment practices and dis-covered that none of the primary carephysicians interviewed recommendedthe same medical management strate-gy. Approximately 50% of theproviders altered their practice pat-terns based on Lovelace’s evidence-based treatment guidelines andprovider education programs.1 Thesechanges resulted in improved patientsatisfaction and in a reduction in theuse of medications, such as antibi-otics and oral steroids. Such managedcare interventions and the followingreview of evidence-based researchshould be useful as managed careorganizations strive to change rhinitispractices and improve outcomes.

The Pathologic and EpidemiologicFactors of Rhinitis

Rhinitis is a common conditiontreated in managed care populations.Seasonal allergic rhinitis is the mostcommon, but other forms of rhinitisare also seen in most practices. Thetypes of rhinitis are defined below2:

• allergic rhinitis: An IgE-mediatedinflammation of the nasal mucosa.Symptoms may be perennial, sea-sonal, or perennial with seasonalexacerbations.

• infectious rhinitis: Inflammationof the nasal mucosa caused by aviral, bacterial, or other infection.

• vasomotor rhinitis: An imbalanceof the parasympathetic and sym-pathetic nervous systems in thenasal mucosa, in which theparasympathetic system is stimu-lated by undue response to typicalenvironmental factors such assmoke, fumes, dust, chlorine inpools, or cold dry air, all of whichcause rhinorrhea.

• rhinitis medicamentosa: Asyndrome of rebound nasalcongestion caused by the overuseof intranasal alpha-adrenergicdecongestants.

Diagnosis is complicated by thefact that multiple forms of rhinitis canoccur in the same patient.

Of the forms of rhinitis, citedabove, only the epidemiology of aller-gic rhinitis is well studied.3 Theprevalence of allergic rhinitis, whichvaries according to the populationassessed, is approximately 8% duringany given year. More than 20% ofpatients report allergic rhinitis symp-toms sometime during their lifetime.4

According to a recent survey, rhinitissymptoms occurred in an estimated31.5% of 15,000 households in theUnited States. Of those with rhinitissymptoms, 8.2% reported physician-diagnosed hay fever (seasonal allergicrhinitis); 5.7% reported persistent“stuffy nose” or “stuffy head” (peren-nial allergic rhinitis); and 18.3%reported allergies of the eyes, nose, orthroat.5 The authors concluded thatthe prevalence of allergic rhinitis is14.2%. As improved constructionpractices lead to less air exchange inhomes and as humidification increas-es in the winter, the opportunity forthe growth of molds and infestation ofdust mites increases, which in turnincreases the potential for sufferingfrom perennial rhinitis. The highestrate of allergic rhinitis occurs fromthe age of 18 years to 49 years, andthe lowest incidence of rhinitis occursin the youngest and oldest age groups.Allergic rhinitis also occurs in 28% to50% of patients with asthma.6-8

Cost of RhinitisEstimates of the cost of rhinitis to

patients and employers vary widelyand seem to be related to the meth-ods used to conduct the analysis.Using National Medical ExpenditureSurvey data from 1987, a study byMalone and associates estimated the

. . . REPORT . . .

S4 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S5

cost of rhinitis in that year to be $800million.9 In 1990, it was estimatedthat the cost of the disease was $1.2billion; and in 1997, it was estimatedthat treating rhinitis cost $4.5 bil-lion.10,11 Another $3.4 billion wasattributed to the indirect costs ofrhinitis, primarily those from reducedwork productivity.11 The costs thatresult from rhinitis are high, consid-ering that little expense is devoted tohospitalizations. The major factorsthat influence the cost of rhinitis areprescription medication use and thelarge numbers of patients who seekambulatory care. Prescription expen-ditures in 1999 continued to increasein the United States: Prescriptionantihistamine and nasal steroid salesexceeded $3 billion and $1 billion,respectively.12 In estimating a sub-component of rhinitis disease costs, a1996 study predicted that the cost ofreduced productivity secondary tothe sedating effects of over-the-count-er (OTC) sedating antihistamines wasnearly $4 billion annually.13

A study by Ray and associatesmodeled both the cost of rhinitis andthe expenses associated with theexacerbation of concomitant illness-es. The authors estimated that rhini-tis without concomitant illness cost$1.9 billion in 1996, and that the costof treating other diseases increasedby $4 billion in the same year becauseof those patients who were sufferingrhinitis concomitantly. For example,it was estimated that the cost of otitismedia was increased by $1.4 billion,sinusitis by $1 billion, asthma by $1billion, acute upper respiratory infec-tion by $0.2 billion, pharyngitis and/ortonsillitis by $0.2 billion, and otherconditions by $0.2 billion.14

It is not clear whether this esti-mate includes the cost of endoscopicsinus surgery, which is often used intreating the allergic patient with nasalpolyps. Although the estimates of thestudy cited above can be debated, theconcept that one primary condition,such as rhinitis, can exacerbate or

contribute to the cost of other relateddiseases or sequelae should be con-sidered in comprehensive calcula-tions of the cost of illnesses.

Indirect Costs of RhinitisAccidents are mentioned as a cost

of rhinitis but often are not quanti-fied. The 1992 and 1993 data fromGroup Health Cooperative of PugetSound indicate that the risk ofwork-related injury was elevated by50% among users of sedating anti-histamines (odds ratio, 1.5; 95%confidence interval = 1.1 to 1.9). Noother medication class had a higherrisk of inducing injury, not even med-ications that affect the central nerv-ous system. Nonsedating antihista-mines had been available, but thosemedications had not been approvedin the formulary used during thestudy period. The authors concludedthat some medications or the condi-tion requiring those medicationsmight contribute to the risk of a work-related injury. It could not be deter-mined from the observational studydesign whether rhinitis, the treat-ment for rhinitis, or both hadincreased the risk of accidents.15

A variety of controlled studies havebeen conducted to investigate thenegative impact of rhinitis or sedatingantihistamines on performance, butnone has quantified the economiccost of the negative impact. In a studyby Marshall and Colon, it was demon-strated that rhinitis itself has a nega-tive effect on verbal learning, decisionmaking, psychomotor speed, and pos-itive affect.16 Other studies havedemonstrated the negative effect ofsedating antihistamines on drivingand on the performance of otherskills.17-20

School-age children with untreatedrhinitis symptoms or those undergo-ing treatment with diphenhydramine(an OTC sedating antihistamine) forrhinitis had lower learning scoresthan did those treated with a nonse-dating antihistamine.21 The economic

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .

S6 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

cost of this reduced learning was notestimated as an indirect cost of rhinitis.

Efficacy of TreatmentsThree general categories of med-

ications are commonly used for thetreatment of allergic rhinitis: sedatingantihistamines (usually nonprescrip-tion), nonsedating antihistamines,and nasal steroids. In part becausesedating antihistamines wereapproved years before the other treat-ments, they have not been comparedin clinical trials with nonsedatingantihistamines and nasal steroids.

In a recent report, the Joint TaskForce of the American College ofAsthma, Allergy and Immunology andAmerican Academy of Asthma,Allergy and Immunology reviewed theliterature on sedating antihistaminesand stated, “In the majority of states,patients taking sedating antihista-mines are legally considered ‘underthe influence of drugs’.” The report

also questioned the logic of bedtimeuse of sedating antihistamines giventhe extended elimination half-life ofthose medications. It further statedthat, “Consequently an ‘AM/PM’ dosingregimen combining a second-genera-tion agent in the AM with a first-gen-eration agent in the PM is an ineffec-tive strategy for avoiding daytimesedation and performance impair-ment from antihistamine.”22

The Joint Task Force report alsostated the role of the nonsedatingantihistamine and nasal steroid cate-gories. The authors noted that oralantihistamines are effective as a first-line treatment in reducing many ofthe symptoms of rhinitis except forcongestion. In the same report, theyconclude that nasal steroids are themost effective medication class forthe treatment of rhinitis, and thatthey do not produce significant sys-temic side effects in adults. The mes-sage appears to give approval for both

. . . REPORT . . .

Figure 1. Comparison of the Effects of Intranasal Corticosteroids and Oral H1 Receptor Antagonists(Antihistamines) on Total Nasal Symptom Scores

χ2 = 26.82; df = 8; P < 0.001.Source: Weiner JM, Abramson MJ, Puy RM. Intranasal corticosteroids versus oral H1 receptor antagonists in allergic rhinitis: Systematicreview of randomised controlled trials. BMJ 1998;1624-1629. Reprinted with permission.

-1.5 -1.0 -0.5 0 0.5

Favorsantihistamine Weight (%)

Total nasal symptom score

8.115.44.3

19.711.015.42.9

17.85.4

100

-0.677 (-1.055 to -0.299)-0.645 (-0.919 to -0.370)-0.645 (-1.165 to -0.124)-0.606 (-0.848 to -0.364)-0.498 (-0.822 to -0.174)-0.427 (-0.701 to -0.152)-0.386 (-1.015 to 0.244)-0.062 (-0.317 to 0.193)-0.389 (-0.076 to 0.853)

-0.423 (-0.531 to -0.315)

Standardized meandifference (95% Cl)

Favorssteroid

Stu

dy

1.0

Géhanno

Bronsky

Munch

Schoenwetter

Van Bavel

Berrnstein

Beswick

Vervloet

Wood

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S7

categories to be first-line treatment.However, it would be logical that themost effective medication should beused first line unless safety or costfactors or other significant issues pre-vent first-line use. No significantissues preventing the first-line use ofnasal steroids were mentioned in thereport.22

Older nasal steroids, such as dex-amethasone, have high bioavailabilityand are capable of producing minorsystemic steroid side effects.23

However, studies of new steroid prepa-rations in adults demonstrate safetywith no systemic steroid effect, even inrelatively high doses, on the hypothal-amic pituitary-adrenal axis as assessedby the measurement of morning corti-sol concentrations, cosyntropinstimulation and 24-hour urinary-freecortisol excretion.24 Newer steroids,

such as fluticasone, have a lowbioavailability of 0.51%.25

A meta-analysis of studies pub-lished in the British MedicalJournal,26 which compared the effec-tiveness of nasal steroids to that of oralantihistamines, concluded that nasalsteroids should be used instead ofnonsedating antihistamines as thefirst-line treatment for allergic rhini-tis. The authors reviewed publishedtrials involving 2267 patients. The totalnasal symptom scores presented in 9 ofthe 16 studies show a clear advantageof using nasal steroids (Figure 1).Several of the studies compared theefficacy of these 2 categories in reduc-ing specific symptoms. For example, 14of the 16 studies evaluated how wellthese 2 treatments reduced nasalblockage (Figure 2). Nasal steroidswere found to be more effective.

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .

Figure 2. Comparison of the Effectiveness of Intranasal Corticosteroids and Oral H1 Receptor Antagonists(Antihistamines) on Total Nasal Blockage

χ2 = 11.76; df = 13; NS.Source: Weiner JM, Abramson MJ, Puy RM. Intranasal corticosteroids versus oral H1 receptor antagonists in allergic rhinitis: Systematicreview of randomised controlled trials. BMJ 1998;1624-1629. Reprinted with permission.

-1.5 -1.0 -0.5 0 0.5

Favorsantihistamine Weight (%)

Nasal blockage

4.0

3.7

2.4

12.7

16.8

14.8

13.5

3.8

7.3

3.9

8.0

2.6

4.7

1.7

100

-0.835 (-1.355 to -0.334)

-0.830 (-1.353 to -0.308)

-0.830 (-1.479 to -0.181)

-0.797 (-1.079 to -0.515)

-0.778 (-1.024 to -0.532)

-0.678 (-0.940 to -0.416)

-0.606 (-0.879 to 0.332)

-0.503 (-1.018 to 0.012)

-0.471 (-0.843 to -0.098)

-0.444 (-0.956 to 0.069)

-0.428 (-0.783 to -0.073)

-0.377 (-1.006 to 0.252)

-0.256 (-0.718 to 0.205)

-0.155 (-0.925 to 0.615)

-0.628 (-0.729 to -0.527)

Standardized meandifference (95% Cl)

Favorssteroid

Stu

dy

1.0

Bunnag

Simpson

Brooks

Bernstein

Schoenwetter

Vervloet

Bronsky

Munch

Géhanno

Juniper

Darnell

Beswick

Wood

Robinson

S8 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

Another article that reviewed stud-ies comparing nasal steroids withnonsedating antihistamines was pub-lished by Stempel and Thomas.27 Eightof the 13 studies that the authorsreviewed compared the effectiveness ofthe 2 categories according to a totalsymptom score. In all 8 studies, thenasal steroid produced superior results.The number of patients in those 8 stud-ies totaled 1745 patients and weredivided almost equally between thosetaking nasal steroids and those takingnonsedating antihistamines.

Each of the 13 randomized trialsreviewed compared 1 nasal steroidwith 1 second-generation antihista-mine. All possible comparisons havenot been studied. Beclomethasone,budesonide, fluticasone, mometa-sone, or triamcinolone (nasalsteroids) have been compared toastemizole, loratadine, or terfenadine(antihistamines) in various head-to-

head comparator trials. Forty-six per-cent of the nasal steroid data were col-lected from patients who were usingfluticasone. Patients using budesonideprovided only 6% of the nasal steroiddata. Studies with flunisolide and dex-amethasone were not reported. Withregard to the antihistamines, lorata-dine was the most commonly selectedcomparator; it was being used by 57%of the 2104 patients. Cetirizine andfexofenadine were not reported in anyof the studies.

When the comparison of nasalsteroids with antihistamines focusedon the 4 major nasal symptoms ofrhinitis—stuffy nose, itchy nose,runny nose, or sneezing—the nasalsteroid was superior in the majority ofcomparisons (Table 1). Nonsedatingantihistamines were not found to besuperior to nasal steroids in any ofthe studies for any nasal parametersinvestigated.

. . . REPORT . . .

Table 1. Results From Several Studies Comparing the Efficacy of Nasal Steroids Versus Antihistamines

Source: Stempel DA, Thomas M. Treatment of allergic rhinitis: An evidence-based evaluation of nasal corticosteroids versus nonsedatingantihistamines. Am J Manag Care 1998;4:89-96. Adapted with permission.

# of Studies Change With Efficacy Results

# of Studies Symptoms From (P <0.05)Reviewed Sample Size Treatment Evaluated Baseline Favoring Nasal Steroids

11 598 Nasal steroid Sneezing -1.1 8598 Antihistamine -0.7

11 598 Nasal steroid Runny nose -1.2 8598 Antihistamine -0.6

9 568 Nasal steroid Itchy nose -1.4 6568 Antihistamine -0.8

12 598 Nasal steroid Stuffy nose -1.1 12598 Antihistamine -0.6

8 741 Nasal steroid Total symptoms -5.4 8752 Antihistamine -3.3

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S9

Ocular symptoms were reportedin 10 of the 13 studies.27 Only 1 ofthe 10 studies—astemizole versusbeclomethasone28—showed statisti-cally improved eye symptoms favor-ing the antihistamine. The other 9showed no statistical difference in therelief of this symptom between the 2treatment categories. Eye symptomswere measured in this study by asummary score of responses to ques-tions about eye redness, itching, andtearing. The actual mechanism bywhich nasal steroids improve eyesymptoms is not known, however theimprovement is not due to systemicabsorption.25 Relief of ocular symp-toms is often proposed as an advan-tage of taking oral antihistamines,which are absorbed systemically, butthe evidence of a clinical advantagehas not been shown in the literature.

Four of the studies in the reviewcompared the effectiveness of usingnasal steroids alone with that of usinga combination of nasal steroids andnonsedating antihistamines, but noneshowed a significant differencebetween the 2 treatments on the totalsymptom score. The combinationtreatment resulted in significantlybetter scores in only 129 of the 4 stud-ies when sneezing was measured.

A potential criticism of these trialsis that patients who fail on a particu-lar drug in one class were not reas-signed to another drug within thesame class. However, there are trialsthat address the variations in efficacyof different drugs within the nonse-dating antihistamine class. In a 4-waycrossover study of 14 patients whowere taking either terfenadine, lorata-dine, chlorpheniramine, or astemi-zole, no clinically significant differ-ences in efficacy were reported.30

Same-patient variation of response tothe drugs was not evaluated. A similarcomparison of nasal steroids has notbeen conducted.

With regard to the differing effectsof drugs in the same class, the JointTask Force report recognizes that

there is a great interpatient variabili-ty associated with the sedation, per-formance impairment, and otheradverse events of antihistamines.However, some studies have shownthat overall clinical efficacy andpatient acceptance appear similaramong the different nonsedating orless sedating preparations.23

Therefore, trying several drugs fromthis class in the same patient does notseem to be supported by the litera-ture, even though some physicianscontinue the practice.

Results of the Ratner andAssociates Study

A study by Ratner and associatesdetails a fluticasone versus loratadinecomparator trial.31 The double-blindtrial compared 4 arms: patients usingfluticasone, 2 sprays in each nostrilonce daily; those taking loratadine 10mg, 1 tablet daily; those using thecombination of fluticasone andloratadine; and those taking placebo.The report includes both patient-rated and clinician-rated total nasalsymptom scores and disease-specificquality-of-life results, and it providesinformation on treatment failures andsuccesses.

Quality of Life. Although otherplacebo-controlled trials haveassessed the quality-of-life impact ofusing a nasal steroid or a nonsedatingantihistamine to treat rhinitis, noother head-to-head active comparatortrial reporting quality-of-life differ-ences beside the Ratner and associ-ates study has been identified. Thestudy used the RhinoconjunctivitisQuality of Life Questionnaire(RQLQ), which measures quality oflife globally and across differentparameters often affected byrhinoconjunctivitis, such as nasal andeye symptoms, fatigue, irritability,and tiredness.32,33

Quality of life was measured atbaseline and at the end of the trial.The patients taking fluticasone and

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .

S10 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

those who took the combination hadsuperior quality-of-life scores whencompared with the patients takingplacebo and loratadine. The combina-tion drug regimen was not superior tonasal steroids in any RQLQ category,nor was loratadine superior to place-bo in any of the RQLQ categories.

Cost-Effectiveness AnalysisPerformed Using Trial Results

Using the data from the Ratner andassociates comparator trial, we wereable to make treatment success/fail-ure assumptions to be used in a cost-effectiveness analysis model (Table2). The comparison was performed onan intent-to-treat basis. Each armconsisted of 150 patients. The place-bo arm of the trial, however, was notincluded; it is not a relevant com-parator because such treatment is notused in general practice.

Cost in the model was defined asthe expense of 14 days of drug treat-ment, and average wholesale prices

were used to determine costs. Becausethere was no difference in the cost ofphysician visits and in sequelae oradverse events reported in the trialbetween the different treatments,these factors were not included in theanalysis. [The length of the trial wasprobably not sufficient for thesesequelae to develop.] Treatment suc-cess in the model was defined aseither a significant or a moderateimprovement in rhinitis symptoms atthe end of the trial as reported by thestudy patients. A mild improvement,no change, or any level of worseningwas treated as a treatment failure.

As shown in the model, fluticasoneproved to be the most cost effective ofthe 3 treatment options. The combi-nation arm has the highest cost oftreatment, yet the number of treat-ment successes is not significantlygreater than those obtained with fluti-casone monotherapy. Consequently,a health plan will be paying $4704more for combination therapy than

. . . REPORT . . .

Table 2. A Cost-Effectiveness Analysis Model Comparing the 3 Arms of the Ratner and Associates Trial31

Model:

(treatment cost per day) x (14 days) x (150 patients per arm) = cost per treatment success

# of treatment successes

Fluticasone Arm

($1.66 / day) x (14 days) x (150 patients) $3486= $48.42

72 successes 72

Loratadine Arm

($2.24 / day) x (14 days) x ( 150 patients) $4704= $109.40

43 successes 43

Combination (fluticasone plus loratadine) Arm

($3.90 / day) x (14 days) x ( 150 patients) $8190= $109.20

75 successes 75

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S11

for fluticasone monotherapy ($2.24per day per patient more) without sig-nificant improvement in outcomes.

In a comprehensive economicanalysis, sensitivity analyses would beperformed using these calculations.

Quality Improvement InitiativesImplications for Treatment

Guidelines. A number of treatmentguidelines have been distributed bymanaged care plans to their pre-scribers. The Lovelace guidelines,mentioned earlier, are an example ofa comprehensive undertaking inmanaged care.1 Guidelines for man-aged care are often thoroughlyreviewed by the organization’s inter-nal staff and by an external pharmacyand therapeutics committee ofexperts. Publication is usually notsought.

The first element in a rhinitis treat-ment guideline is usually allergenavoidance. Although prevention isalways logical, the evidence pertainingto that strategy is very limited. In astudy by Reisman and associates,34 itwas reported that High-EfficiencyParticulate Air (HEPA) filters reduceparticulate matter but do not signifi-cantly improve symptoms of rhinitis.The National Cooperative Inner-CityAsthma Study35 showed that afterextensive cockroach allergen avoid-ance, there was no clinical differencein asthma symptoms between theintervention and control groups. In astudy of cat allergen, the combinationof a HEPA room air cleaner, mattressand pillow covers, and cat exclusionfrom the bedroom did reduce airbornecat-allergen levels but had no effect onasthma or rhinitis disease activity forany parameter studied.36 In a study ofasthma patients allergic to house dustmites, an extensive dust mite inter-vention program had no significanteffect on symptom scores, peak flow,or lung function.37 It is not knownwhether the lack of effectiveness ofallergen reduction in asthma patientswould also be observed in rhinitis

patients. The results of the above-men-tioned studies indicate that a reductionin allergen load does not result in aclinical reduction of symptoms. Totalallergen elimination may be necessaryto achieve a clinical improvement.Thus, the utility of allergen reductionas a strategy for rhinitis treatmentmust be questioned.

The next line of therapy in thetreatment guideline is often the use ofOTC sedating antihistamines.Although those medications may costmanaged care plans less than nonse-dating antihistamines, reductions inproductivity, performance, and learn-ing caused by sedating antihistaminesas demonstrated in research studiesmust be considered. The recent JointTask Force report is the most stronglyworded document to date that warnsagainst sedating antihistamine use. Italso states that reductions in perform-ance, productivity, and learning canoccur with antihistamine therapy,even when sedation does not occur.This report reflects a summary of thebest evidence for changing the stan-dard of care to exclude the use ofsedating antihistamines in patientswho must maintain their functionalstatus while their allergic rhinitissymptoms are being treated.

Current prescribing patterns forthe treatment of allergic rhinitis indi-cate nonsedating antihistamines areoften used before nasal steroids.27

Evidence from clinical trials indicatesthat this is not the most effective orthe most efficient treatment for rhini-tis.26,27 As this type of evidenceincreases as well as the move to evi-dence-based medicine grows, man-aged care plans may begin recom-mending the use of nasal steroids as afirst-line therapy to be used before atrial of a nonsedating antihistamine.The use of cost-ineffective combina-tion therapy may be safely avoided bymeans of appropriate pharmacy ben-efit edits.

Treatment guidelines recommendthe use of immunotherapy or allergy

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .

S12 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

shots to treat rhinitis after pharmaco-logic treatment has failed.38 In a ran-domized, open-label trial in Japan ofpatients who suffered from perennialrhinitis and were allergic to der-matophagoides farinae, the grouptreated with pharmacotherapy hadsuperior outcomes to those of thegroup treated with immunotherapyfor the first 2 years.39 During years 3

and 4 of the trial, the 2 groups hadsimilar outcomes. At 5 years, theimmunotherapy group demonstratedbetter outcomes. In a 2-year US-basedblinded study of children with asth-ma, no incremental advantage ofimmunotherapy was seen at anytime.40

The cost of and patient adherenceto immunotherapy was studied in a

. . . REPORT . . .

Figure 3. Evidence-Based Allergic Rhinitis Treatment Guideline for the Primary Care Physician

Make allergic rhinitis diagnosis

If the treatment is not effective

1. Conduct medication-use history. Warn patients about the consequences of using sedating antihistamines. Note delivery of the warning in each patient's chart to limit liability.2. Educate the patient about allergen elimination. (Allergen load reduction is usually not clinically effective.)3. Prescribe a nasal steroid and provide information about initial use and maintenance therapy.

1. Reassess diagnosis, technique, and patient compliance.2. If the patient prefers oral tablets, discontinue the nasal steroid and prescribe a nonsedating antihistamine. (Adding antihistamines to nasal steroid use provides very little incremental benefit.)

If revised treatment is not effective

Refer the patient to an allergist or an otolaryngologist for diagnosis and treatment.

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S13

population of 122,196 Harvard PilgrimHealth Maintenance Organizationasthma and rhinitis patients, of whom2667 were treated with immunother-apy.41 Only 33% of patients completedtherapy. The cost of immunotherapyfrom 1988 to 1992 was $416 per yearfor the treatment of those with rhini-tis and $496 per year for those withrhinitis and asthma. The intendedduration of therapy was 3.5 years.The actual treatment duration onaverage was 2.7 years, for a meantotal cost of $1123 to $1339 perpatient. Clinical trial data lead us toquestion the value of that investment.

As a result of the evidence favoringthe use of nasal steroids over otherrhinitis treatments, managed careorganizations might begin proposingguidelines for the treatment of rhini-tis that demonstrate an evidence-based protocol, similar to that shownin Figure 3.

Implications for Cost Savings.The daily cost of loratadine therapy(1 tablet b.i.d.) is 35% higher than ini-tial fluticasone therapy (2 spraysb.i.d.). If the patient can be managedwith the approved maintenance doseof fluticasone (1 spray b.i.d.), the costsavings could be significantly greater.Converting patients from loratadineto fluticasone results in cost savings ifeach drug were to be used for thesame duration. More important to ahealthcare provider is improvedpatient outcomes at a lower cost.Because fluticasone has been shownto be more clinically effective thanloratadine in the treatment of rhini-tis, the 35% cost difference citedabove could become increased to a126% difference in cost effectivenessbetween the 2 drugs. It should benoted that inputting different nonse-dating antihistamines and nasalsteroids into the analysis may alterthe results because of differences indrug prices and effectiveness.Without similar definitions of treat-ment success and failure as noted in

the Ratner and associates trial, theextrapolation of these results to otherapplications of nasal steroids andantihistamines cannot be made.

Methods of Influencing PrescribingPatterns. A wide variety of methodshave been used by managed careorganizations to influence change inprescribing patterns. These methodsare not compared in a scientific man-ner to determine which techniqueswork best in different situations.Methods can range from low-level orsoft interventions to high-level orhard interventions (Table 3). Internet-based interventions have not yet beenfully explored. The intervention mustbe appropriate to the disease and tothe organizational culture and mustbe matched to the magnitude of thechange required without instigating anegative counteraction.

Implications of Improving RhinitisOutcomes. The cost of the rhinitis

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .

Table 3. Health Plan Interventions Used to Influence PrescribingBehavior

■ National Drug Code lockout

■ Shifting drug cost risk to prescribers

■ Tiered co-pay structure

■ Prescription coupons

■ Physician prescribing report cards with penalties, with rewards, not confidential, confidential

■ Internet-based interventions

■ Academic detailing

■ HMO direct-to-patient communications (patient specific)

■ Required attendance at meetings

■ Required attendance at continuing education meetings

■ Managed care organization coordination of industry representatives

■ Generalized direct-to-patient communications

■ Preprinted prescription pads

■ Voluntary continuing education meetings

■ Written general communications

S14 THE AMERICAN JOURNAL OF MANAGED CARE JANUARY 2000

treatment strategy proposed by Rayand associates14 indicates that therecould be additional long-term cost sav-ings because of reduced expenses andimproved outcomes in the treatmentof otitis media, asthma, sinusitis, andother diseases that occur with rhinitis.This theory has not been demonstrat-ed in long-term clinical trials but canbe evaluated by examining claimsdatabases from managed care organi-zations. Improved patient satisfactionfollows more effective treatment,which results in improved outcomes.42

ConclusionHealth plans can take an active

role in improving outcomes in dis-eases by taking steps to improve pre-scribing practices, in particular forallergic rhinitis, which is a ubiquitousdisease and a likely target for managedcare intervention. Comprehensive lit-erature reviews have demonstratedthat treatment of allergic rhinitis withnasal steroids produces outcomessuperior to those in which nonsedat-ing antihistamines are used. The cost-effectiveness analysis model describedearlier comparing 2 rhinitis treat-ments showed that the nasal steroidfluticasone had a lower cost per treat-ment success than did the nonsedat-ing antihistamine loratadine. Giventhis dominance (an economic termfor this type of cost-effectiveness rela-tionship with both lower cost andimproved effectiveness), managedcare organizations and clinicians maybegin to approve nasal steroids as first-line treatment for allergic rhinitis.

. . . REFERENCES . . .

1. Gregory C, Cifaldi M, Tanner LA. Targetedintervention programs: Creating a customizedpractice model to improve the treatment ofallergic rhinitis in a managed care popula-tion. Am J Manag Care 1999;5:485-496.2. Mygind N. Essential Allergy. Oxford:Blackwell Scientific Publications; 1986:279-350.

3. Sibbald B, Strachan DP. Epidemiology ofrhinitis. In: Busse WW, Holgate ST, eds.Asthma and Rhinitis Boston: BlackwellScientific Publications: 1997;32-43.4. Druce HM. Allergic and nonallergic rhini-tis. In: Middleton E Jr, Reed CE, Ellis EF,Adkinson NF Jr, Yuninger JW, Busse WE, eds.Allergy: Principles and Practice, 4th ed. St. Louis, MO: Mosby; 1993:1433-1453.5. Nathan RA, Meltzer EO, Sener JC, Storms W. Prevalence of allergic rhinitis inthe United States. J Allergy Clin Immunol1997;99:S808-S814.6. Edfors-Lubs ML. Allergy in 7000 twinpairs. Acta Allergol 1971;26:249-285. 7. Weeke ER. Epidemiology of hay fever andperennial allergic rhinitis. Monogr Allergy1987;21:1-20.8. Peckham C, Butler N. A national study ofasthma in childhood. J EpidemiolCommunity Health 1978;32:79-85.9. Malone DC, Lawson KA, Smith DH, et al.A cost of illness study of allergic rhinitis inthe United States. J Allergy Clin Immunol1997;99:22-27.10. McMenamin P. Costs of hay fever in theUnited States in 1990. Ann Allergy1994;73:35-39.11. Mackowiak JI. The health and economicimpact of rhinitis. Am J Manag Care1997;3:S8-S18.12. IMS RPP Sales Audit. Data on file.GlaxoWellcome Inc. 1999.13. Ross RN. The costs of allergic rhinitis.Am J Manag Care 1996;2:285-290.14. Ray NF, Baraniuk JN, Thamer M, et al.Healthcare expenditures for sinusitis in1996: Contributions of asthma, rhinitis, andother airway disorders. J Allergy ClinImmunol 1999;103:408-414.15. Gilmore TM, Alexander BH, Mueller BA,Rivara FP. Occupational injuries and med-ication use. Am J Ind Med 1996;30:234-239. 16. Marshall PS, Colon EA. Effects of allergyseason on mood and cognitive function. AnnAllergy 1993;71:251-258.17. Moskowitz HM, Burns M. Effects of ter-fenadine, diphenhydramine, and placebo onskills performance. CuTIS 1988;42:14-18.18. Ramaekers JG, Uiterwijk MMC,O’Hanlon JF. Effects of loratadine and ceti-rizine on actual driving and psychometrictest performance, and EEG during driving.Eur J Clin Pharmacol 1992;42:363-369. 19. O’Hanlon JF, Ramaekers JG.Antihistamine effects on actual driving per-formance in a standard test: A summary of

. . . REPORT . . .

VOL. 6, NO. 1, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S15

Dutch experience, 1989-1994. Allergy1995;50:234-242.20. O’Hanlon JF. Antihistamines and drivingsafety. CuTIS 1988;142:10-13.21. Vuurman EFPM, van Veggel LMA,Uiterwijk MMC, et al. Seasonal allergic rhini-tis and antihistamine effects on children’slearning. Ann Allergy 1993;71:121-126.22. Dykewicz MS, Fineman S. Diagnosisand management of rhinitis: Parameter docu-ments of the Joint Task Force on practiceparameters in allergy, asthma, and immunol-ogy. Ann Allergy Asthma Immunol1998;81:463-468.23. Dykewicz MS, Fineman S, eds.Diagnosis and management of rhinitis:Complete guidelines of the Joint Task Forceon practice parameters in allergy, asthma,and immunology. Ann Allergy 1998;81:506. 24. Van As A, Bronsky E, Grossman J, et al.Dose tolerance study of fluticasone propi-onate aqueous nasal spray in patients withseasonal allergic rhinitis. Ann Allergy1991;67:156-162. 25. Daley-Yates PT, McAllister T. Systemicbioavailability of fluticasone propionateadministered as nasal drops (FP-drops) andaqueous nasal spray formulations (FPANS)Allergy 1998;53:158. [Abstract]26. Weiner JM, Abramson MJ, Puy RM.Intranasal corticosteroids versus oral H1receptor antagonists in allergic rhinitis:Systematic review of randomised controlledtrials. BMJ 1998;317:1624-1629.27. Stempel DA, Thomas M. Treatment ofallergic rhinitis: An evidence-based evaluationof nasal corticosteroids versus nonsedatingantihistamines. Am J Manag Care 1998;4:89-96.28. Frolund L. Efficacy of oral antihistamine,loratadine, as compared with a nasal steroidspray, beclomethasone dipropionate, in sea-sonal allergic rhinitis. Clin Otolaryngol1991;16:527-531. 29. Simpson RJ. Budesonide and terfena-dine, separately and in combination, in thetreatment of hay fever. Ann Allergy 1994;73:497-502.30. Crawford WW, Klaustermeyer WB, LeePM, Placik IM. Comparative efficacy of terfe-nadine, loratadine, and astemizole in peren-nial allergic rhinitis. Otolaryngol Head NeckSurg 1998;118:668-673.31. Ratner PH, van Bevel JH, Martin BG, etal. A comparison of the efficacy of fluticas-one propionate aqueous nasal spray andloratadine, alone or in combination, for the

treatment of seasonal allergic rhinitis. J FamPract 1998;47:118-125.32. Juniper EF, Guyatt DH. Development ofa new measure of health status for clinicaltrials in rhinoconjunctivitis. Clin Exp Allergy1991;21:77-83.33. Juniper EF, Guyatt GH, Dolovich J.Assessment of quality of life in adolescentswith allergic rhinoconjunctivitis:Development and testing of a questionnairefor clinical trials. J Allergy Clin Immunol1994;93:413-423.34. Reisman RE, Mauiello PM, Davis GB, etal. A double-blind study of effectiveness of ahigh-efficiency particulate air (HEPA) filter inthe treatment of patients with perennialallergic rhinitis and asthma. J Allergy ClinImmunol 1990;85(6):1050-1057.35. Gergen PJ, Mortimer PM, Eggleston PA,et al. Results of the National CooperativeInner-City Asthma Study (NCICAS) environ-mental intervention to reduce cockroachallergen exposure in inner-city homes. JAllergy Clin Immunol 1999;103:501-506. 36. Wood RA, Johnson EF, Van Natta ML, etal. A placebo-controlled trial of a HEPA aircleaner in the treatment of cat allergy. Am JRespir Crit Care Med 1998;158:115-120.37. Marks GB, Tovey ER, Green W, et al.House dust mite allergen avoidance: A ran-domized controlled trial of surface chemicaltreatment and encasement of bedding. ClinExp Allergy 1994;24:1078-1083. 38. Hadley JA. Evaluation and managementof allergic rhinitis. Med Clin North Am1999;83:13-25. 39. Ohashi Y, Nakai Y, Tanaka A, et al. Acomparative study of the clinical efficacy ofimmunotherapy and conventional pharma-cological treatment for patients with peren-nial allergic rhinitis. Acta Otolaryngol Suppl(Stockh) 1998;538:102-112.40. Adkinson NF Jr, Eggleston PA, Eney D, etal. A controlled trial of immunotherapy forasthma in allergic children. N Engl J Med1997;336:324-331. 41. Donahue JG, Greineder DK, Connor-Lacke L, et al. Utilization and cost ofimmunotherapy for allergic asthma andrhinitis. Ann Allergy Asthma Immunol1999;82:339-347. 42. Mackowiak JI. Fluticasone’s rapid reliefof allergy symptoms leads to greater patientsatisfaction. Poster presented at: AAAAI’s54th Annual meeting, March 13-18, 1998,Washington, DC.

. . . TREATING ALLERGIC RHINITIS IN A MANAGED CARE SETTING . . .