Objective: Fears of potentially costly use of topical retinoids for cosmetic treatment of photodamaged skin have resulted in many managed care organizations placing prior authorization requirements on this class of medications. The purpose of this investigation was to examine whether prescribing patterns of a nationally representative sample of US physicians shed light on the incidence of use of topical retinoids for indications other than acne.

Methods: A retrospective, cross-sectional study of data from the National Ambulatory Medical Care Survey (1996-2000) was used to examine the impact of physician specialty as well as patient diagnosis of acne on the probability of retinoid prescription in weighted multivariate logistic regression models.

Results: Topical retinoids were prescribed in 0.4% (14.7 million out of 3.67 billion) physician visits for any diagnosis in the 5-year period from 1996 to 2000, and in nearly 31% (12.0 million out of 38.7 million) of physician visits for a diagnosis of acne. Topical retinoids were prescribed for acne in 77.1% of the cases. This finding held when individual retinoids (tretinoin and adapalene) were examined separately. Clear age-related prescription trends are observed, with a significant decrease in prescriptions beyond the teen years. In older patients, tretinoin prescribing did not decrease as much as adapalene prescribing.

Conclusions: These data suggest that managed care organizations may want to examine their own data to determine the optimum criteria for operation of prior authorization (PA) programs for retinoids. PA requirements for these medications appear unnecessary in young patients, given the very small probability of non-acne related use. PA in older patients might be targeted to those patients on topical retinoids (such as tretinoin) for which there is evidence of efficacy in treatment of cosmetic photoaging.

**********

Introduction

The term "retinoids" encompasses any synthetic or natural compound that has activity like that of vitamin A. Many of these actions have anti-acne effects, making topical retinoids an excellent therapeutic choice in the management of acne vulgaris. Some of these agents have multiple therapeutic indications, including photodamaged skin treatment. (1) The Global Alliance to Improve Outcomes in Acne strongly recommends that topical retinoids should be used as first-line therapy for mild to moderate inflammatory acne and comedonal acne, excluding very severe disease. (2) In an effort to minimize antibiotic use and development of antibiotic-resistant organisms, topical retinoids are also the preferred agents for the maintenance therapy of acne. (2) The efficacy of these agents in comedonal and inflammatory acne lesions has been demonstrated in randomized controlled studies. (3) In the United States, the currently available topical retinoids for acne include tretinoin, adapalene, and tazarotene. Some of these agents have multiple indications, one of which is photodamage. Insurance companies generally cover most of the costs associated with medications to treat acne vulgaris and psoriasis, but not for medications used to treat photodamaged skin, often defined as cosmetic use. A list of the approved and non-approved indications of topical retinoids (4) is shown in Table 1.

There is increasing concern about the spiraling expenditures for prescription medications in the United States. In 2001, national health expenditures exceeded $1.4 trillion, and prescription drugs accounted for 10% of the expenses. (5) Several pharmacy benefit management tools have been developed with the intent to deliver prescription drug benefits while controlling pharmaceutical costs. Prior authorization (PA) is one method commonly used to control utilization of, and expenditures for, medications. This process limits the dispensing of certain medications by requiring approval by the health plan of attestations from the prescriber, usually based upon explicit criteria for appropriate use. (6) The philosophy of the PA method is to target new, expensive, or potentially unnecessary or dangerous medications, while encouraging the delivery of less expensive and/or safer alternatives. (7) Growth hormones, drugs used for infertility, drugs for treatment of lifestyle conditions or with a potential cosmetic use are frequently controlled by PAs. Although potentially beneficial in avoiding possible unsafe or unnecessary drugs, PA programs incur administrative costs and may require significant time requirements for managed care organizations and pharmacy benefit management companies. (8) The total cost of administering PAs may be more than the savings in avoided medical spending, thereby defeating one of the goals, ie, cutting expenditures. Since the safety of topical retinoids has been well established, (9) we adopt a cost perspective in considering the implications of PAs required by third party payers.

We suggest that analyses of prescriber patterns and the demographics of patients may aid in the development of guidelines to encourage cost-effective PA programs. One method of analysis is investigation of the diagnoses associated with a specific type of medication, stratified by age group. If the target diagnoses are those associated with non-reimbursable costs in a certain age group, then PA in that particular age group may represent a reasonable expenditure of health plan resources. In a prior study, we demonstrated that topical tretinoin use for non-acne patients was uncommon in young adults (up to approximately age 35) and, therefore, PA requirements in this age group were not an efficient tool for cost management. (10,11) Another approach is to study the differences among diagnoses associated with prescriptions for specific agents within a pharmaceutical category (indication) to determine if PAs are cost-effective for certain agents, but not for others.

The purpose of this investigation was to analyze more recent data since earlier studies utilized the NAMCS data from the years 1990 to 1994. (10,11) Using data from the 1990 to 1994 NAMCS survey, the earlier set of studies showed that the distribution of outpatient visits for acne treatment is skewed toward younger patients and does not persist beyond age 40. (10) In addition, analyses of these data showed that as the PA age decreases, the cost of requiring PA increases. (11) Therefore, this paper examined whether prescribing patterns of a nationally representative sample of US physicians shed light on potential cosmetic use of topical retinoids. Identification of these prescription patterns and related factors may help to better direct rational use of PAs for topical retinoids.

Methods

Data for this study were obtained from the National Ambulatory Medical Care Survey (NAMCS) for the years 1996 to 2001. For 20 years, the National Ambulatory Medical Care Survey has collected nationwide outpatient data from physicians in the United States not employed by the federal government. The NAMCS is conducted by the National Center for Health Statistics (NCHS) and provides data on the demographics, diagnoses (designated by International Classification of Disease, Ninth Revision, Clinical Modification [ICD-9-CM] codes), and medications prescribed in physician encounters. (12) A one-page patient log is completed for every patient visit detailing the reason for the visit, diagnoses, services provided, medications prescribed, referral practices, and demographic description. The ICD-9-CM code 706.1 identifies visits at which acne vulgaris was diagnosed. (13) The demographic data were used to quantify the frequency of each race, gender, and age group seeking acne management. Sampling was limited to non-federally employed physicians engaged principally in the delivery of outpatient care. The multistage probability sampling design was stratified by primary sampling unit (contiguous counties, or standard metropolitan statistical area in the United States), then by physician practices within the sampling unit, and, finally, by patient visits within 52 weekly randomized periods. Within small physician practices, a 100% sample of 1 week's visits was possible. For very large practices, 20% of patient visits were randomly sampled.

[FIGURE 1 OMITTED]

For normalization with national estimates (ie, to make national inferences from sample estimations), each individual record in the NAMCS is assigned an inflation factor called the patient visit weight that is used to predict the total number of office visits made in the United States, and analyses are weighted using statistical software. The resulting national estimates describe the use of ambulatory care services in the United States. (12) The study interval of 1996 to 2000 was chosen because these were the most recent data available at the time of conduct of the study. The topical retinoids that were included in the search were tretinoin, adapalene, tazarotene, and all of the brand names associated with these medications. We analyzed the primary, secondary, and tertiary diagnoses for these physician visits. ICD-9-CM code 706.1 pertains to a diagnosis of acne vulgaris. The demographic data were used to analyze patient and physician characteristics as related to receipt of topical retinoid prescriptions, in general and individually for tretinoin (eg, Renova[R] and Retin-A[R]), adapalene (Differin[R]), and tazarotene (Tazorac[R] and Avage[R]). Figure 1 shows the data subsetting procedure used for the study.

Certain categorical variables were created from the original data set to simplify the interpretation of the analyses. These included 3 dummy variables for age: < 18 years, 18 to 39 years and [greater than or equal to] 65 years. The 40 to 64 age group served as the comparison group. To facilitate trend analyses, we created dummy (1/0) variables for each year of the NAMCS data utilized in the study. The comparison group was the 2000 NAMCS data. To determine topical retinoid usage by physician specialty, we classified each patient visit based on type of physician (dermatologist or non-dermatologist, primary care physician) seen. We defined primary care physicians included in the following NAMCS-defined specialties: family/general practice, internal medicine, and pediatrics. We have found in our previous research (14) that internists and pediatricians serve as primary care providers and, therefore, to compare dermatologist with primary care providers, we combined the pediatricians, family physicians, and internists into one group. We examined predictors of both receipt of any topical retinoid prescription as well as specific topical retinoid in separate multivariate logistic regression models. All analyses were adjusted using the NAMCS sampling weights. All analyses were conducted using the STATA statistical software. (15)

[FIGURE 2 OMITTED]

Results

From 1996 to 2000 for patients 10 years and older, there were approximately 3.67 billion outpatient patient visits as estimated by NAMCS. There were 38.7 million visits (1.1%) with a diagnosis of acne vulgaris, and 31% (12.0 million) of these visits were associated with a prescription for a topical retinoid. An estimated total of 15.6 million patients (0.4%; each patient has one visit in the database) were prescribed a topical retinoid.

The descriptive characteristics of this study population are outlined in Table 2. Females comprised 62% of the population. Mean age of the entire population was approximately 49 years, and the mean age of the population that was prescribed retinoids was 28 years. Racial distribution was similar across retinoid users and non-retinoid users. Topical retinoid prescriptions were more frequently associated with dermatologist visits (80%) as compared to non-dermatologists. Among groups that were prescribed different retinoids, we found several differences in the distributions of patient characteristics. Nearly 77% of the patients receiving a topical retinoids prescription had a diagnosis of acne. Of patients receiving a prescription for adapalene, 91% had a diagnosis of acne vulgaris, whereas only 45% of patients receiving tazarotene had a diagnosis of acne vulgaris. Nearly 74% of patients receiving a prescription for tretinoin had an acne vulgaris diagnosis. The mean age of the tazarotene group (38.9 years) was higher than adapalene group (23.3 years) and tretinoin group (29.8 years). Patients who were prescribed tazarotene were also more likely to be established (rather than new) patients compared to the other two medications. Those who were prescribed adapalene had comparatively shorter visits than those who were prescribed tretinoin or tazarotene. Number of people insured and geographical distribution was similar across all three retinoid groups. The fraction of visits that were to a dermatologist (as opposed to non-dermatologists) associated with retinoid prescription was much lower in the tretinoin group (74%) compared to the adapalene group (92%) and the tazarotene group (97%). The average total number of medications prescribed at the time of each visit was comparable for all three drugs.

Table 3 displays the results of the multivariate logistic regression analyses that examined the predictors of topical retinoid prescriptions. There was negligible prescription of topical retinoids for non-acne related conditions (OR for topical retinoid prescription with acne diagnosis: 56.58, 95% CI: 32.33, 99.04). This association was the strongest for adapalene (OR: 82.31, 95% CI: 27.51, 147.06) versus tretinoin (OR: 66.37, 95% CI: 29.96, 246.32) and lowest for tazarotene (OR: 3.61, 95% CI: 1.84, 7.09). Patients aged 65 years and over were less likely to be prescribed a topical retinoid (OR: 0.34, 95% CI: 0.21-0.55). Patients aged 40 to 64 years were 51% less likely to be prescribed adapalene than patients aged 18 to 39 years (OR: 0.49, 95% CI: 0.25-0.93). Patients at a visit to a dermatologist were more than 12 times more likely to be prescribed topical retinoid as compared to a non-dermatologist visit (OR: 12.10, 95% CI: 6.80, 21.56). One point to note is that the estimates in the tazarotene only group seem to be unreliable because the standard error estimates were higher than allowed NAMCS norms.

Discussion

Retinoids are integral therapeutic alternatives in dermatologic care. (16) The primary FDA-approved indications for topical retinoids are in the treatment of acne vulgaris and psoriasis and as an adjunctive treatment of photodamaged skin. Fears of potentially costly use of topical retinoids for cosmetic treatment of photodamaged skin have resulted in many managed care organizations placing prior authorization requirements on this class of medications. Our study finds that in 77.1% of the cases, topical retinoids are prescribed for acne only. This finding held when individual retinoids (tretinoin and adapalene) were examined separately. Clear age-related prescription trends were observed, with a significant decrease in prescriptions beyond the teen years, paralleling the percentage of patients with acne (Figure 1, Table 2). In older patients (age [greater than or equal to] 65 years), tretinoin prescribing did not decrease as much as adapalene prescribing. Given that more than 90% of patients given a prescription for adapalene had a diagnosis of acne vulgaris, prior authorization of adapalene does not seem warranted at any age.

We also explored the issue of whether there were increases in prescriptions for certain age groups and for non-acne related conditions (detailed results not shown) over time through inclusion of interaction terms for each year or age group and acne related use. For topical retinoids overall, as well as the individual topical retinoids, we only found increased use over time, all of which was related to acne. There was one exception with the decreased use of tretinoin in 1999, but there was no indication that non-acne related prescriptions increased. There were no age-group related relationships noted in non-acne related use of topical retinoids, adding confidence to the findings that there was very little non-acne related use of topical retinoids, and this non-acne related use did not increase with increase in age, which would have suggested potential cosmetic use.

In a previous series of studies, we demonstrated that topical tretinoin use for non-acne patients was sufficiently uncommon in young adults (up to approximately age 35) and that prior authorization requirements in this age group were not a cost efficient management tool. That work, however, assumed that the low frequency of non-acne prescribing was not due to a deterrent effect of the existing prior authorization requirements. Subsequent to our previous publication on topical tretinoin prior authorization--and the increased prior authorization age that followed, restricting prior authorization for women aged 40 years and older only--there has been no significant increase in prescribing topical tretinoin for non-acne conditions in young patients.

Limitations

We relied on the ICD-9-CM coding of the clinical diagnoses of the patients and, therefore, there is a possibility that cosmetic uses were coded as acne for reimbursement purposes. The study findings especially related to tazarotene could be unreliable because of small cell sizes. Two of the study medications were introduced in the US market at the end of the study period; adapalene was introduced in the third quarter of 1996 (the end of the first year of the study period) and tazarotene was introduced in the third quarter of 1997 (the end of the second year of the study period). We did not find any evidence of higher use in older age groups however, and it is highly unlikely the age of the patient was miscoded.

There is a potential for some Hawthorne effect since physicians participating in the survey had knowledge of being observed by the survey collecting organization (National Center for Health Statistics) and, therefore, may have altered their prescribing behaviors for the week of the year the data were gathered for the NAMCS survey. Another limitation is that because of small sample sizes in the tazarotene group (being a relatively new therapy), the standard errors of the estimates regarding its prescribing were higher than the allowable norms of the NAMCS, thus rendering the estimates with regard to this medication potentially unreliable. Our study was cross-sectional in nature and did not capture previous treatment history of patients. Our focus in this study was to examine if there was cross-sectional national evidence of significant non-medical uses of topical retinoids that might justify potentially expensive PA requirements of these medications. Although we did not find such evidence, further investigation using a longitudinal study design would be useful to confirm our findings.

Conclusion

These data suggest that managed care organizations may want to examine their own data to determine the optimum criteria for operation of PA programs for retinoids, as there seems to be evidence from national data to suggest that the majority of the prescribing for these medications occurs for acne only. PA requirements for these medications are not necessary in young patients (aged 40 years and younger) given the very small probability of non-acne related use. In older patients (age [greater than or equal to] 65) PA, if needed at all, should focus on those topical retinoids for which there is evidence of efficacy in treatment of cosmetic photoaging, such as tretinoin. (4)

Disclosure: No funding was obtained for this study. The Center for Dermatology Research is funded by a grant from Galderma Laboratories, L.P.

References

1. Webster G et al. Acne vulgaris. BMJ. 2002;325:475-479.

2. Gollnick H et al. Management of acne: a report from a global alliance to improve outcomes in acne. J Am Acad Dermatol. 2003;49:S1-38.

3. Kakita L. Tazarotene versus tretinoin or adapalene in the treatment of acne vulgaris. J Am Acad Dermatol. 2000;43(2 Pt 3):S51-4.

4. Rolewski SL. Clinical review: topical retinoids. Dermatol Nurs. 2003;15(5):447-50, 459-65.

5. Heffler S, Smith S, Won G, et al. Health spending projections for 2001-2011: The latest outlook. Faster health spending growth and a slowing economy drive the health spending projection for 2001 up sharply. Health Aff. 2002;21:207-218.

6. Kreling D, Mucha R. Drug product management in maintenance organizations. Am J Hosp Pharm. 1992;49:374-81.

7. MacKinnon NJ, Kumar R. Prior authorization programs: a critical review of the literature. J Manag Care Pharm. 2001;7:297-302.

8. Burton SL, Randel L, Titlow K, Emanuel EJ. The ethics of pharmaceutical benefit management. Health Aff (Millwood). 2001; 20(5):150-63.

9. Cunliffe WJ, Caputo R, Dreno B, et al. Clinical efficacy and safety comparison of adapalene gel and tretinoin gel in the treatment of acne vulgaris: Europe and U.S. multicenter trials. J Am Acad Dermatol. 1997;36(6 Pt 2):S126-34.

10. McConnell RC, Fleischer AB, Williford PM, Feldman SR. Most topical tretinoin treatment is for acne vulgaris through the age of 44 years: an analysis of the National Ambulatory Medical Survey, 1990-1994. J Am Acad Dermatol. 1998;38(2 Pt 1):221-226.

11. Feldman SR, Fleischer AB, Chen GJ. Is prior authorization of topical tretinoin for acne cost effective? Am J of Manag Care. 1999;5:457-63.

12. Anonymous. 1996-2000 NAMCS micro data file documentation. Available at: http://www.cdc.gov/nchs/about/major/ahcd/ahcd1.htm. Accessed September 12, 2003.

13. Physicians ICD-9-CM; vol 1. Salt Lake City: Medicode Publications; 1997.

14. Balkrishnan R, Dugan E, Camacho FT, Hall MA. Trust and satisfaction with physicians, insurers, and the medical profession. Med Care. 2003;41(9):1058-64.

15. STATA Statistical Software. Release 6.0. College Station, TX: 2000

16. Sardana K, Sehgal VN. Retinoids: Fascinating up-and-coming scenario. J Dermatol. 2003;30:355-80.

Rajesh Balkrishnan PhD, (a,b) Julia C. Sansbury MD, (a) Rahul A. Shenolikar MS, (b) Alan B. Fleischer Jr. MD, (a) Steven R. Feldman MD PhD (a)

a. Department of Dermatology and Center for Dermatology Research, Wake Forest University School of Medicine

b. Department of Pharmacy Practice and Administration, Ohio State University College of Pharmacy

Address for Correspondence

Rajesh Balkrishnan, PhD

Merrell Dow Professor

Ohio State University College of Pharmacy and School of Public Health

500 W. 12th Avenue

Columbus, OH 43210

Phone: 614-292-6415

Fax: 614-292-1335

Email: balkrishnan.1@osu.edu

Return to Differin

Home

Contact

Resources

Exchange Links

ebay