Evidence of Human Papillomavirus Vaccine Effectiveness in Reducing Genital Warts: An Analysis of California Public Family Planning Administrative Claims Data, 2007–2010

Abstract

Because of the rapid development of genital warts (GW) after infection, monitoring GW trends may provide early evidence of population-level human papillomavirus (HPV) vaccine effectiveness. Trends in GW diagnoses were assessed using public family planning administrative data. Between 2007 and 2010, among females younger than 21 years, these diagnoses decreased 35% from 0.94% to 0.61% (P_trend < .001). Decreases were also observed among males younger than 21 years (19%); and among females and males ages 21-25 (10% and 11%, respectively). The diagnoses stabilized or increased among older age groups. HPV vaccine may be preventing GW among young people.

Human papillomavirus (HPV) is a common sexually transmitted infection causing genital warts and anogenital and oropharyngeal cancers.¹ An estimated 1.4 million Americans have genital warts (GW).² The quadrivalent HPV vaccine, available in the US since June 2006, prevents HPV types 6 and 11, which cause 90% of GW.^3,4 The Advisory Committee on Immunization Practices (ACIP) recommends routine immunization of females and males aged 11 to 12 years with catch-up for females aged 13 to 26 and males aged 13 to 21 years.^5,6 As of 2010, 49% of adolescent females aged 13 to 17 years in the US had received at least 1 of the 3 doses and 32% had received all 3 doses; in California, 56% received at least 1 dose.⁷ Additionally, 21% of US females aged 19 to 26 years had received at least 1 dose.⁸

In the United States, several approaches are under way to measure the population-based impact of the HPV vaccine.^9,10 Because of the rapid development of GW after infection, monitoring GW trends may provide early evidence of vaccine effectiveness.^11,12 Further, analyses of administrative data enable the monitoring of vaccine-preventable disease outcomes.

METHODS

Using an ecological study design, trends in GW diagnoses were assessed using clinical encounter claims data from the California Family Planning Access Care and Treatment (Family PACT) program, which serves low-income individuals. Following implementation of more specific diagnostic coding requirements for billing, reliable data on International Classification of Diseases (ICD-9) codes were available 2007 through 2010. GW cases were defined as unduplicated clients with either an ICD-9 diagnostic code for viral warts (078.10) or condyloma (078.11) or a National Drug Code (NDC) for pharmacy-dispensed imiquimod or podofilox. To limit the analysis to incident cases, clients with these codes documented prior to 2007 were excluded. Inclusion of procedure codes for destruction of genital lesions was unnecessary because these claims closely correlated with appropriate ICD-9 codes; more than 95% of claims with destruction codes were included on the basis of ICD-9 code. The proportion of clients with GW was calculated based on denominators that included unduplicated clients served. The proportions of clients with GW were stratified by age (< 21, 21–25, 26–30, ≥ 31 years), gender, and year. No data on vaccination status were available. Statistical significance of gender- and age-specific trends was assessed using the Cochran-Armitage test for linear trend; 95% confidence intervals (CI) were calculated for the initial year and final year percent changes.

RESULTS

Between 2007 and 2010, an average of more than 1 754 000 female and 258 000 male clients were served annually. Total clients served increased each year for both female and male clients across nearly all age groups, with the greatest increases seen among older age groups (Table 1). Overall, 0.7% of female clients and 3.3% of male clients were diagnosed with GW. For both male and female clients, the highest rates of GW were among those aged 21 to 25 years, whereas the lowest rates were among those older than 30 years. The median age of clients younger than 21 years was 18 years (range = 10–20).

TABLE 1—

Diagnosis of Genital Warts by Gender, Age Group, and Year: California Family Planning Access Care and Treatment Program, 2007–2010

	Aged < 21 Years		Aged 21–25 Years		Aged 26–30 Years		Aged ≥ 31 Years		Total
Year	Total No.	Warts, No. (%)	Total No.	Warts, No. (%)	Total No.	Warts, No. (%)	Total No.	Warts, No. (%)	Total No.	Warts, No. (%)
Female clients
2007	394 789	3723 (0.94)	475 698	4765 (1.00)	334 080	2066 (0.62)	475 117	1664 (0.35)	1 679 684	12 218 (0.73)
2008	404 379	3581 (0.89)	491 655	5263 (1.07)	343 959	2347 (0.68)	492 195	1853 (0.38)	1 732 188	13 044 (0.75)
2009	405 454	3080 (0.76)	508 465	4950 (0.97)	358 031	2408 (0.67)	521 884	2027 (0.39)	1 793 834	12 465 (0.69)
2010	391 386	2405 (0.61)	513 111	4628 (0.90)	365 126	2487 (0.68)	543 599	2083 (0.38)	1 813 222	11 603 (0.64)
Male clients
2007	52 183	1385 (2.65)	63 810	3231 (5.06)	43 701	1650 (3.78)	72 338	1276 (1.76)	232 032	7542 (3.25)
2008	55 192	1540 (2.79)	66 634	3514 (5.27)	44 398	1952 (4.40)	73 672	1441 (1.96)	239 896	8447 (3.52)
2009	58 492	1501 (2.57)	73 481	3568 (4.86)	50 605	2151 (4.25)	89 848	1766 (1.97)	272 426	8986 (3.30)
2010	58 758	1269 (2.16)	76 778	3452 (4.50)	54 159	2228 (4.11)	100 761	1798 (1.78)	290 456	8747 (3.01)

Between 2007 and 2010, GW diagnoses decreased 34.8% (95% CI = −38.2%, −31.5%) among female clients younger than 21 years from 0.94% to 0.61% (P_trend < .001). Decreases were also observed among female clients aged 21 to 25 years (10.0% decline; CI = –13.6%, –6.3%; P_trend < .001), male clients younger than 21 years (18.6% decline; 95% CI = –24.8, –12.5%; P_trend < .001), and male clients aged 21 to 25 years (11.2% decline; 95% CI = –15.4%, -7.1%; P_trend < .001). By contrast, GW diagnoses increased among female clients aged 26 to 30 and older than 30 years (10.1% increase; 95% CI = 3.7%, 16.6%; P_trend = .004 and 9.4% increase; 95% CI = 2.4%, 16.5%; P_trend = .005, respectively) and male clients aged 26 to 30 years (9.0% increase; 95% CI = 2.2%, 15.8%; P_trend = .05; Figure 1). No change was observed among male clients older than 30 years (1.2% increase; 95% CI = –6.0%, 8.4%; P_trend = .97).

FIGURE 1—

Trends in genital wart diagnosis among females (a) and males (b) by age group, California Family Planning Access Care and Treatment program, 2007–2010.

DISCUSSION

This is the first US report to our knowledge to provide preliminary data suggesting population-level effectiveness of the HPV vaccine. Although the HPV vaccine is not a covered benefit of the Family PACT program and vaccine status among clients was not determined in this analysis, this vaccine seems to prevent GW among young women. Family PACT clients may have alternative means to receive the vaccine, including patient assistance programs. Furthermore, because of herd immunity, it is possible that both genders are protected even if unvaccinated. In Australia, where HPV vaccine uptake exceeded 65% in 2009 among female residents in the target age group (12–26 years), significant decreases in GW were reported among female residents aged 12 to 26 years as well as among heterosexual male residents aged 12 to 26 years (59% and 39%, respectively).^13,14

Among older age groups, both absolute numbers and rates of GW diagnoses increased. National surveillance of GW trends demonstrated a 2-fold increase in rates since the mid-1990s.¹⁵ The reason for these increases is not clear, but could be related to increased health care–seeking behavior as a result of increased awareness, or actual increases in HPV transmission or development of GW.

The primary limitation of this analysis is that the observed trends are ecological and may be explained by factors other than vaccination. These decreases could be spurious or the result of biases in diagnostic coding across age groups; however, divergent trends by age along with observed increases in the number of patients receiving services annually are reassuring.

Using existing administrative claims data to assess trends in GW proved to be inexpensive and expeditious; however, the quality of the data depended on the accuracy of diagnostic coding. Prior to 2007, ICD-9 codes for secondary (noncontraceptive) diagnoses in the Family PACT administrative database were incomplete and unreliable. Although more challenging to monitor, the most important effect of HPV vaccination will be the reduction of HPV-associated cancers.