Cases of human papillomavirus (HPV)–associated head and neck cancers are rapidly increasing in the United States. Little is known about the economic burden of these cancers. A literature review identified 7 studies that characterized aspects of the overall economic burden of HPV-associated head and neck cancers in the United States. Other cost studies are detailed to highlight the clinical reality in treating these patients. As the clinical awareness of the role of HPV in head and neck cancers continues, the economic impact of cancers caused by this virus will have implications for the role of various preventive measures.
- •
A literature review shows that a complete cost-of-illness study has not been conducted on head and neck cancer (HNC) in the United States.
- •
An International Classification of Diseases (ICD) code indicating the human papillomavirus (HPV) status of patients with HNC would facilitate the characterization of the economic burden of HPV-associated HNC.
- •
Establishing which HPV tumor-detection method in HNC is the most cost-effective would be a useful addition to clinical practice.
- •
The cost of treatment-related complications and patients’ health-related quality of life should be factored into any cost-consequence analysis of HPV-associated HNC treatment options. Morbidity rather than mortality is the main concern in treating HPV-associated HNC.
- •
The increasing incidence and subsequent burden of HPV-associated HNC will be an important consideration in the HPV vaccination debate.
| ASCO | American Society of Clinical Oncology |
| BIA | Budget impact analysis |
| CDC | Centers for Disease Control and Prevention |
| CEA | Cost-effectiveness analysis |
| COI | Cost of illness |
| CRD | Center for Reviews and Dissemination |
| EGFR | Epidermal growth factor receptor |
| HCUP | Healthcare Cost and Utilization Project |
| HNC | Head and neck cancer |
| IMRT | Intensity-modulated radiation therapy |
| ISPOR | International Society for Pharmacoeconomics and Outcomes Research |
| ISRCTN | International Standard Randomized Controlled Trial Number |
| LOHRAN | Longitudinal Oncology Registry of Head and Neck Carcinoma |
| MEPS | Medical Expenditure Panel Survey |
| MeSH | Medical Subject Heading |
| NCCN | National Comprehensive Cancer Network |
| NCI | National Cancer Institute |
| NHS EED | National Health Service Economic Evaluation Database (UK) |
| PBT | Proton beam therapy |
| QALY | Quality adjusted life year |
| QOL | Quality of life |
| RT | Radiation therapy |
| SEER | Surveillance, Epidemiology, and End Results |
| SMDM | Society for Medical Decision Making |
| TORS | Trans-oral robotic surgery |
Introduction
The rising incidence of human papillomavirus (HPV) as the causative agent of a subset of head and neck cancers (HNC) has recently been described as an epidemic. The reported proportion of oropharyngeal cancers attributable to HPV in the United States has increased from 16.3% during the 1980s to 72.7% during the 2000s. More careful anatomic site stratification has made it apparent that the age-adjusted incidence of oropharyngeal cancer is rising dramatically (estimated to be a 5% annual increase). In comparison with HNC not associated with HPV, incident cases of HPV-associated HNC occur primarily among younger (aged 40-59 years), nonsmoking, white men. Hence, these virus-related cancers have been characterized in the clinical literature as being “ a distinct epidemiologic, clinical and molecular entity .”
The increasing awareness of the role of HPV in HNC in both sexes has amplified the profile of this virus even further in the public health and mass media arena. Therefore, understanding the costs of the condition is useful for making an economic argument about efforts to reduce the burden of the virus. Moreover, various decision makers (providers, payers, and policy makers) will be concerned with the resulting financial impact on clinical management issues in treating such patients.
The most common way to characterize the economic burden of a disease is to perform a cost-of-illness (COI) study. In this article, the components of a COI study are described. The authors then report on a literature review that was undertaken to look at the economic burden of HNC. The focus of the review was to ascertain what inferences (if any) were made about HPV. The authors elaborate on the current medical costs involved in the diagnosis, treatment, and management of these patients. Subsequently, the challenges facing the use of economic data and the controversies associated with the economic data are described, and suggestions for future research are made. The overarching aim of this article is to summarize, critique, and elaborate on the published studies that are pertinent to characterizing the economic burden of this emerging disease entity.
What is included in a COI study?
COI studies are descriptive analyses assessing the economic burden of health problems on the population overall. The traditional approach considers:
- •
Direct medical costs: Associated with emergency department and hospital services, physician services, diagnostic procedures, laboratory tests, medications, treatments, ancillary therapies, and other health care services.
- •
Productivity costs: Result from lost work productivity, disability, and premature death caused by a disease or condition.
- •
Intangible costs: Primarily related to losses in quality of life.
Together with prevalence and incidence, morbidity and mortality help portray the overall burden of disease in society. This raises methodology concerns, specifically the adding-up constraint: it is not always entirely clear what costs are associated with each disease and how to ensure that all medical spending is allocated to one and only one disease. For analysts using retrospective datasets, the attribution of costs to a particular disease can be difficult, especially if patients have several other medical conditions. Moreover, despite the popularity of COI studies, it is surprising that there is little published guidance to support the choice of methodological approach to be used. Therefore, many COI studies in the United States are not comparable because they differ in terms of the valuation approaches used, the perspective adopted, and the components of care analyzed.
What is included in a COI study?
COI studies are descriptive analyses assessing the economic burden of health problems on the population overall. The traditional approach considers:
- •
Direct medical costs: Associated with emergency department and hospital services, physician services, diagnostic procedures, laboratory tests, medications, treatments, ancillary therapies, and other health care services.
- •
Productivity costs: Result from lost work productivity, disability, and premature death caused by a disease or condition.
- •
Intangible costs: Primarily related to losses in quality of life.
Together with prevalence and incidence, morbidity and mortality help portray the overall burden of disease in society. This raises methodology concerns, specifically the adding-up constraint: it is not always entirely clear what costs are associated with each disease and how to ensure that all medical spending is allocated to one and only one disease. For analysts using retrospective datasets, the attribution of costs to a particular disease can be difficult, especially if patients have several other medical conditions. Moreover, despite the popularity of COI studies, it is surprising that there is little published guidance to support the choice of methodological approach to be used. Therefore, many COI studies in the United States are not comparable because they differ in terms of the valuation approaches used, the perspective adopted, and the components of care analyzed.
What is the reported economic burden of HNC in the United States?
A literature review was conducted to ascertain the published data regarding the economic burden of HNC. This review builds on previous economic reviews of HNC. The initial search strategy was performed in PubMed. Multiple searches using various terms pertaining to costs and the site of disease were used. Supplemental databases searches were also performed (see Appendix for search strategy). As a review, it is a summary of the literature specifically relevant to the United States but more tellingly it critiques the HPV dimension of economic studies on HNC. Seven studies were identified that reported estimates of direct medical costs and productivity losses.
Of these, 2 of the 7 studies ( Table 1 ) estimated the economic burden of HPV-associated HNC over a patient’s lifetime. Hu and Goldie base their estimates on a previous study by Lang and colleagues that looked at Medicare patients. From the epidemiologic evidence, we know that HPV-associated HNC are predominately in patients aged 40 to 59 years, whereas Medicare claims are based on patients aged older than 65 years. This is the main shortcoming in using this data source. Also, it should be noted that the evaluated patients were from 1991 through 1993, when combined modality therapy had not come into play.
| Author, Year, and Type of Patients | Cost Methodology | Data Sources | Cost Estimates | Main Conclusion and Limitation | HPV Perspective |
|---|---|---|---|---|---|
| Hu & Goldie 2008 Report looked at noncervical HPV-related conditions: oropharyngeal and mouth cancer | Discounted lifetime cost per case expressed in present value Incidence-based approach applied to costs to estimate economic burden | US-linked SEER-Medicare data (Lang et al 2005) British & Dutch studies used for plausible range American Cancer Society (2003) incidence rates | Average cost per case of HNC in 2003 is $33,020 (range: [min] $15,340– $46,800 [max]) | Total lifetime costs for new cases in 2003: $38.1 million (range: $17.7 million–$54.1 million) Uses SEER-Medicare claims data | Underestimated HPV prevalence (10.7% of all oropharyngeal cancer caused by HPV-16, 18) |
| Ekwuene et al, 2008 HPV-associated cancers: cancers of the tonsil, tongue, and other oral cavity/pharyngeal cancers | Societal burden of mortality: Mortality, YPLL, value of productivity loss from premature death a | SEER US census National mortality data: CDC’s NCHS National Vital Statistics system US life tables ICD-10 | Year (2003): Number of deaths: 3379 YPLL: 63,587 YPLL per death = 18.8 PVFLE: $406,061,000 | Total mortality costs = $1.37 billion Productivity loss per death = $406,061 Human capital approach used for productivity loss | Used subsites as proxy for HPV-associated cancers |
a Productivity costs of premature mortality were estimated by multiplying the number of deaths in 2003 (stratified by age, sex, and race/ethnicity) by the present value of future lifetime earnings (PVFLE) stratified by age and sex. The PVFLE estimates that were applied took into account factors like life expectancy, the labor force participation rate and future growth rate in productivity, and the imputed value of housekeeping services (eg, cooking, cleaning, childcare).
In 2008, a National Cancer Institute (NCI) State of the Science meeting used these studies to estimate the annual cost of HPV-associated oropharyngeal cancers. The cost of treatment and disease management was calculated to be in the order of $151 million. It is likely that this figure is a conservative estimate of the treatment cost burden because the cost of the treatment has increased substantially with multiple modality regimens coupled with the increased incidence of HPV-associated cancers.
An analysis by the Centers for Disease Control and Prevention (CDC) reported on the societal burden of mortality due to HPV-associated cancer sites for 2003. The investigators used a human capital approach to estimate the mortality burden in terms of years of potential life lost and mortality-related productivity costs. Specific to oral cavity/pharynx, they estimated the present future value of lifetime lost productivity of cancer to be $1.37 billion, with men accounting for $1.1.billion. The investigators do not take into account the attributed fraction caused by HPV, which would significantly lower the burden estimate. However, this report does provide an upper bound estimate on what the productivity losses would be if all the cancers were HPV-associated.
What direct medical cost studies have been conducted in the United States?
Five pertinent direct medical cost studies that looked at clinically diverse populations of patients with HNC are given in Table 2 . The range of patients evaluated, the cost methodology adopted, the length of data collection, and the data sources used differ among the studies. All of these studies used the International Classification of Diseases , Ninth Revision ( ICD-9 ) and Tenth Revision ( ICD-10 ) codes as the basis of their disease diagnosis. However, none of these studies considered HPV-association with HNC. Perhaps only an ICD code indicating the HPV status of patients with HNC would facilitate an accurate characterization of the economic burden of HPV-associated HNC. It should be noted that Ekwuene and colleagues (see Table 1 ) used cancers of the tonsil, tongue, and “other oral cavity/pharyngeal” cancers as a proxy for HPV-associated HNC. These sites are the specific sites where HPV-associated cancers develop, and it is the cost-of-care estimates associated with these sites that are of particular interest. Fortunately, the Longitudinal Oncology Registry of Head and Neck Carcinoma (LORHAN) are tracking whether HPV testing is being performed. Hopefully, future cost studies may avail of stratification by HPV status.
| Author, Year, and Type of Patients | Cost Methodology | Data Sources | Cost Estimates | Main Conclusion and Limitation | HPV Perspective |
|---|---|---|---|---|---|
| Amonkar et al, 2011 Resected SCCHN (N = 1104) | Retrospective claims-based analysis of commercially insured patients (2004–2007) | Medical, pharmacy, and laboratory data & enrollment information from a large US database of commercially insured patients | Patients incurred ∼$94 million in costs following index surgery (average: $85,000 per patient [2008 USD]). Mean total health care cost was $34,450 per patient per year (2008 USD). | Patients with resected SCCHN incur substantial health care costs and have high use rates Managed care setting, not generalizable | Not mentioned in report ICD-9 codes used to identify patients Possible to separate by HPV-associated subsites |
| Le et al, 2011 Metastatic (N = 1042) & recurrent, locally advanced (N = 324) HNC | Retrospective payer-based analysis (2004–2008) Compared rate frequency and costs of health care use during the 6 mo after index period | Thomson MarketScan databases (Medicare data & private-sector health data from ∼100 payers) |
| “Advanced HNC patients, pose a significant health economic burden on the payer” Based on patients receiving employer-sponsored health insurance, not generalizable | Not mentioned in poster ICD-9 codes used to identify patients Possible to separate by HPV-associated subsites |
| Choi et al, 2009 HNC diagnosis (N = 6570) | First -year expenditures associated with HNC diagnosis in the US managed care population | US commercial managed care claims database | Projected average Medicare payment per individual in 1 y after HNC diagnosis = $18,000 (2007 USD) Average heath care cost per patient 1 y after HNC diagnosis = $29,608 (±$77,500) (2007 USD) | Annual cost associated with HNC is higher in the managed care population than reported on Medicare population | Not mentioned in abstract ICD-9 codes used to identify patients Possible to separate by HPV-associated subsites |
| Epstein et al, 2007 OSCC and pharyngeal squamous cell carcinoma (N = 3422) | Direct medical costs of patients were defined as being treated for early or late- stage disease based on treatment modality | Retrospective analysis of California Medicaid claims data CPT-4 coding in claims data | Median year-1 cost of care following initial diagnosis = $25,319 (n = 229) (2002 USD) Estimated range of year-1 cost of care in a commercial PPO in California = $42,198–$72,340 (2002 USD) | Costs for patients treated as having early stage OSCC were approximately 36% less than those treated with late-stage disease ( P = .002). Did not include patients that died within 1 y of diagnosis | Not mentioned in report ICD-9 codes used to identify patients Possible to separate by HPV-associated subsites |
| Lang et al, 2004 Retrospective cohort analysis of newly diagnosed elderly (>65 y) SCCHN (N = 4536) | Linked clinical data to Medicare claims | SEER and Medicare claims Selected diagnosis-related groups, ICD-9-CM diagnosis and procedure codes, and Healthcare Common Procedure Coding System codes in the Medicare claims data | Total mean Medicare payments = $48,847 IQ range: $16,314–$65,682 (1998 USD) Average Medicare payments among patients with SCCHN were $25,542 higher than those of the matched comparison group ( P <.001) (1998 USD) | Patients with advanced SCCHN had shorter survival and higher costs than patients diagnosed as having distant, regional, local, and in situ cancer Medicare looks at patients aged >65 y; data 1991–1993 | Not mentioned in report ICD-9 codes used to identify patients Possible to separate by HPV-associated subsites |
Two studies (Lang and Epstein ) used exclusively publicly funded cost sources. Such cost data refer to payments for services paid for by the state/federal government for a subset of the general population. These studies use public payment rates that may underestimate the economic burden of the disease. As noted in a study (Choi and colleagues ) using a large US commercial managed care claims database (n = 6570), the average first-year expenditures associated with HNC diagnosis in this population ($29,608 ± 77,500) is higher than the projected average Medicare payment ($18,000). The other two studies also used private-sector health cost data (Amonkar and colleagues and Le and colleagues ). They highlight the direct medical costs associated with specific types of patients with HNC: treated with surgery or diagnosed with metastatic or recurrent locally advanced cancer. For purposes of collecting data on HPV-associated HNC, the essential first step is to obtain an accurate, standardized, HPV status so that future analysis can refer to confirmed HPV-positive HNC.
What are the costs involved in diagnosis, treatment, and management of HNC?
Various choices exist in how to diagnose, treat, and manage patients with HNC. In the 2011 National Comprehensive Cancer Network (NCCN) guidelines, testing for tumor HPV is suggested and immunohistochemical staining for the surrogate biomarker p16 is recommended. It is likely that HPV/p16 testing will become common practice, but a standardized method has not emerged thus far. A recent survey performed in the United Kingdom reported that the associated laboratory resources (ie, to determine HPV status) cost between £45 and £60. The cost is borne by the publically funded National Health Service (NHS) and depends on the HPV detection technique. In the United States, a more nuanced payment structure for diagnostic services exists. For a general summary of Medicare coverage, coding, and payment for therapeutic and diagnostic devices, refer to Ackerman and colleagues’s (2011) edited book: Therapeutic and Diagnostic Device Outcomes Research . The charge to Medicare is likely to be part of a single bundled payment for the facility’s services furnished to a Medicare beneficiary coupled with the physician fees.
It should be noted that the identification of a novel biomarker, such as HPV or p16, would never make economic sense if it were not clinically useful. Currently, the NCCN and others note that the results of HPV testing should not change management decisions except in the context of a clinical trial. Historically, HNC, whether or not associated with HPV, have been treated in the same manner. However, it has been suggested from multiple retrospective case series that patients with HPV-positive HNC have an improved overall prognosis. Moreover, the literature calls for less-intense treatment strategies that do not compromise survival outcomes but lower the risk of debilitating side effects in HPV-positive HNC.
Currently, clinicians and patients are faced with a variety of treatment modalities with huge uncertainties regarding the best sequence of treatment. In 2010, the Cochrane Library conducted systematic reviews on the 3 broad treatment modalities in HNC (surgery, radiation, and chemotherapy), although there was little reference to HPV status. An optimal treatment algorithm based on HPV and smoking status will likely occur once the results from clinical trials are known.
Treatment: Surgery
A 2007 review identified 6 costing studies on various treatment strategies for HNC of which 5 involved some form of surgery in the treatment algorithm. Since that review, a study using data from the Maryland Health Service Cost Review Commission database identified attributes to the cost of surgery for oropharyngeal cancer surgical cases (1990–2009). These cost drivers were:
- •
Postoperative wound complications
- •
Length of hospital stay
- •
In-hospital death.
For patients aged younger than 60 years (n = 735), the mean cost of hospital care was $24,537 (median $19,655; range: minimum $1493 and maximum $298,032 in 2009). These figures did not include physician-related costs and that hospital-related charges for each index admission were converted to the organizational cost of providing care using cost-to-charge ratios for individual hospitals.
As noted in a study using case reviews (n = 100) from one US medical facility, postoperative medical complications were statistically far more important in negatively affecting the outcomes and true costs of microsurgical reconstruction for patients with HNC than microsurgical complications. Perhaps the cost of surgery in HNC depends more on the physical condition of patients and not just the progression of the disease. As noted, patients who are HPV positive are generally younger and healthier than patients who are HPV negative.
A promising development in surgery is the use of minimally invasive transoral robotic surgery (TORS) for difficult-to-access cancers. TORS represents the surgical equivalent of delivering targeted therapy for HNC.
Treatment: Radiation Therapy
A nonoperative approach is favored for patients with HNC for whom surgery followed by either radiation therapy (RT) alone or radio chemotherapy may lead to severe functional impairment. In an expert review piece, David Sher highlighted that no cost-effectiveness-analysis (CEA) studies have been performed evaluating the use of radiation therapy for HNC. CEA is a form of full economic evaluation whereby both the costs and consequences of alternative health programs or treatments are examined and compared between treatment options with consequences most often measured in natural units (eg, cost per millimeters of mercury decrease in diastolic blood pressure). In light of the better prognosis of patients with HPV-positive HNC, the focus is to reduce the treatment morbidity. Hence, the Eastern Cooperative Oncology Group and the Radiation Therapy Oncology Group are planning a complex treatment regimen with lower dose radiation (Total dose 54 Gy, conventional dose 70 Gy).
For advanced cancers, the RT is usually delivered with more expensive intensity-modulated RT (IMRT) and there is also interest in using proton beam therapy (PBT). The increase in the use of IMRT and PBT is expected to further add to national expenditures on RT services. Again, Sher states that there is a clear need for a CEA comparing IMRT and PBT with 3-dimensional conformal RT in head and neck squamous cell carcinomas.
Treatment: Chemotherapy
Chemotherapy is often used in combination with RT for treating patients with HNC. In a recent review (2011), chemotherapy in HNC includes one or a combination of the following: cisplatin, carboplatin, 5-fluorouracil, paclitaxel, docetaxel, leucovorin, and cetuximab. In a managed-care population (2004–2007), the mean 2008 US dollar figure for total chemotherapy cost per patient per year was $2004 for pharyngeal cancer (n = 185) and $1177 for lip/tongue cancer (n = 367). This cost is considerably less than the total mean radiation cost of $11,833 and $7264 for pharyngeal and lip/tongue cancer respectively.
The LORHAN group reported that for 1144 patients, inexpensive (∼$40/100 mg vial) cisplatin-based chemotherapy was the most frequently used regimen (51%) and that the vastly more expensive ($10,000+ per treatment cycle) cetuximab was the next most commonly used regimen (21%) in the United States. One of the secondary aims of Clinical trial NCT01302834 is to explore differences in the cost-effectiveness of cetuximab as compared with cisplatin in HPV-associated oropharynx cancer. Cetuximab, a monoclonal antibody inhibitor, works by antagonizing the epidermal growth factor receptor (EGFR). HPV and p16 status is important to determine the prognostic value; however, EGFR status has a predictive treatment value. An Australian study showed that only 2 of 126 patients (1.6%) with oropharyngeal cancer were found to be p16+/EGFR fluorescence in situ hybridization positive. This finding may suggest that cetuximab may not be appropriate for all HPV-associated HNC.
Cost of Posttreatment Management
One of the significant features about all 3 modalities is the risk of complications associated with each treatment option. From an economic perspective, other direct medical costs related to rehabilitation with auxiliary health care professionals and the cost of treatment complications ought to be factored into any cost analysis. Using a retrospective cohort study (PharMetrics Patient-Centric Database 2000–2006), Lang and colleagues found significantly higher rates of treatment-related complications among patients receiving chemoradiotherapy (86%) than among patients receiving radiotherapy alone (51%) ( P <.001). The mean per-patient costs associated with treatment-related complications were approximately $10,000 higher among patients who received chemoradiotherapy than those treated with radiotherapy alone ( P <.001). These costs represented 17% of the total costs during follow-up for patients who received chemoradiotherapy and 11% of costs for those who received radiotherapy.
Other US studies have also examined the cost of radiation-induced oral mucositis on managing dysphagia and xerostomia in patients with HNC. Treatment options have an array of severe acute toxicities and long-term morbidities that have not been fully documented and are particularly concerning in younger patients with highly curable HPV-associated oropharyngeal cancers. Therefore, it is likely that treatment-related morbidities impose substantial unnecessary costs on patients and payers.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
