Health Economic Analysis of the Burden of Infusion Reactions on Patients, Caregivers, and Providers

Publication
Article
OncologyONCOLOGY Vol 23 No 2_Suppl_1
Volume 23
Issue 2_Suppl_1

In recent years, both the cost and efficiency of medical care have emerged as important considerations and areas of research. These considerations are of particular importance in the outpatient community oncology setting, where the demands for clinical productivity and evidence for quality and effectiveness are increasing amidst an evolving reimbursement system.

The cost and efficiency of medical care is an ongoing issue that has a particular effect on patients and caregivers within the community oncology setting. Monoclonal antibody (MoAb) therapy has proven to be effective in the treatment of cancer patients. Although MoAbs are associated with minimal toxicity, they can cause infusion reactions (IRs) in some patients. Managing these infusion reactions leads to an increased burden on patients, caregivers, and providers. Health economic models can quantify the burden that MoAb-induced IRs have on the patient and can help identify ways to maximize the efficiency of providing quality health care. This review identifies the specific burden, including tasks and associated costs, that IRs have on the patient and caregivers, and discusses the importance of using this information to help manage IRs. Nurses are often the first to respond when an IR occurs and the particular impact IRs have on nurses is reviewed. Patients, caregivers, and providers should be aware of the potential burden that MoAb-induced IRs can have on a patient and use this information to help guide clinical decisions.

In recent years, both the cost and efficiency of medical care have emerged as important considerations and areas of research. These considerations are of particular importance in the outpatient community oncology setting, where the demands for clinical productivity and evidence for quality and effectiveness are increasing amidst an evolving reimbursement system.

Early evidence suggests that the management of the side effects of outpatient cancer treatment is associated with an increased burden on patients, caregivers, clinical staff, and the overall practice.[1-3] Beyond the inherent negative experience of the toxicity itself, the additional burden of chemotherapy-induced toxicities on the patient includes direct expenses stemming from copays, medication, and travel, along with indirect expenses, such as increased time away from work.[1,4-6] Chemotherapy-induced side effects also impose a significant burden on health-care providers, particularly nurses, by increasing time spent attending to the patient.[7,8]

Such events ultimately have an impact on the overall pattern of human resources required to sustain highquality, efficient, and cost-effective care. Health economic models provide information to potentially improve management and the decision-making process of treating cancer to maximize effectiveness, while reducing or minimizing the economic impact on the patient and the health-care system.[9]

The use of monoclonal antibodies (MoAbs) in the treatment of cancer has increased in the past 10 years. These therapies are generally associated with relatively fewer and milder side effects than traditional cytotoxic agents. However, MoAbs have been shown to induce specific side effects associated with inhibition of the therapeutic target (cardiovascular events with vascular endothelial growth factor [VEGF] and skin toxicity with epidermal growth factor receptor [EGFR]) and immunologic recognition of foreign protein (Erbitux package insert, 2008; Avastin package insert, 2008; Vectibix package insert, 2008).

The immunologic response, generally characterized as an infusion reaction (IR) or hypersensitivity reaction, can range from mild (itching, rash) to severe (anaphylaxis).[10,11] The National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE) grades hypersensitivity reactions on a scale of 1 (mild) to 5 (death).[12] Grade 1/2 reactions are generally considered mild and are associated with rash, fl ushing, or fever. Grade 3 reactions are associated with bronchospasm, urticaria, or hypotension, while grade 4 IRs are associated with anaphylaxis.[12]

The overall incidence of severe IRs to monoclonal antibodies used in the treatment of solid tumors varies from 0.1% for the fully human monoclonal antibody panitumumab (Vectibix) to 3% for the chimeric antibody cetuximab (Erbitux).[10] The humanized monoclonal antibodies trastuzumab (Herceptin) and bevacizumab (Avastin) have reported an incidence of severe infusion reactions of less than 1% (Herceptin package insert, 2008; Avastin package insert, 2008). An increased incidence of severe IRs to cetuximab of 22% has been observed among patients from the midsoutheastern region of the United States.[13]

Despite the relative low rate of severe IRs observed in the clinic, the occurrence of these and of less severe adverse events can be burdensome to patients and health-care providers. Initial studies examining the impact of IRs on patients, caregivers, and providers illustrate the need for effective preparedness and management of IRs to decrease the overall burden of care.[14,15]

Impact on Human Resource Costs

In addition to the direct physical, functional, and psychological impact of adverse events such as infusion reactions on patients, there is also an increased burden on the health-care provider. The NCI CTC guidelines recommend therapeutic intervention and/or infusion interruption for grade 2–4 IRs.[12] However, even with grade 1/2 reactions, staff must engage in careful monitoring of the patient’s status to guard against the escalation of symptoms.

A study involving 416 health professionals and staff from 21 community oncology clinics systematically described the clinical tasks involved in the management of both mild and severe infusion reactions and enumerated the associated human resource costs.[14] Data were collected from a modified time and motion survey that identified tasks involved in the management of IRs and the associated human resource costs.

 

 

 

Findings suggested that human resource tasks required to manage patients experiencing IRs fell into three categories[14]:
1. Mild IR not requiring discontinuation of infusion therapy
2. Severe IR requiring discontinuation of infusion therapy but managed in the outpatient setting
3. Severe IR resulting in hospitalization

For mild IRs, a total of 13 tasks were required, an average of 4,320 seconds (72 minutes) of time was spent, and a mean increased cost of $51 was incurred per patient with a mild IR. For severe IRs managed in the outpatient clinic, a total of 20 tasks were needed to manage the IR, resulting in an average total human resource time expenditure of 8,340 seconds (139 min) and an increased human resource cost of $102 per patient. For severe IRs requiring hospitalization, 22 tasks were needed to manage the IRs, resulting in an average of 6,320 seconds (106 min) of human resource time and an increased human resource cost of $134 per patient. (See Table 1 and Figure 1.)

 

Overall, severe infusion reactions resulted in increased estimates of time spent by staff to manage the IR, which resulted in increased human resource costs. The findings were presented at the 2006 Multinational Association of Supportive Care in Cancer International Symposium.

Retrospective Chart Review of Severe Infusion Reactions

While the risk of developing an infusion reaction for patients administered MoAb therapy is well known by clinicians and nurses, there are few studies describing the prevention and management of IRs, particularly in the community oncology setting. A study conducted by Schwartzberg and colleagues reviewed 76 charts identified through electronic billing records, manual searches of medical records, and staff recall of patients experiencing severe IRs resulting from treatment with cetuximab, rituximab (Rituxan), or bevacizumab.[15] The definition of a severe IR was that described in the package insert for each agent (Rituxan package insert, 2008; Erbitux package insert, 2008; Avastin package insert, 2008). Chart data included patient demographics, IR management, and severity of IR. An oncologist and an oncology nurse reviewed all submitted chart data from the 19 community oncology centers.

Of the 76 patients who experienced a severe IR (grades 3–5), 47 were treated with rituximab, 24 with cetuximab, and 5 with bevacizumab.[15] The median age of patients in the study was 62. Among patients treated with cetuximab or bevacizumab, ≥ 80% had colorectal cancer, while the majority of rituximab patients (64%) were treated for non-Hodgkin’s lymphoma. Among the nearly 50% of patients receiving MoAb monotherapy, 68% of rituximab patients and 60% of bevacizumab patients received MoAb therapy as first-line treatment, while only 29% of cetuximab patients received this agent as first-line therapy.

Premedication of patients prior to therapy administration varied between agents. Approximately half of rituximab patients (55.3%) and cetuximab patients (54.2%) were premedicated with antihistamines.[15] Acetaminophen was the most prevalent premedication for patients treated with rituximab, administered to 61.7%, while 41.7% of patients treated with cetuximab received corticosteroid pretreatment. It should be noted that premedication is recommended for all patients receiving rituximab or cetuximab, but not for patients receiving bevacizumab (Rituxan package insert, 2008; Erbitux package insert, 2008; Avastin package insert, 2008). In this study, 20% of bevacizumab patients received pretreatment.

This study was retrospective and all patients included in the convenience sample were required to have a moderate or severe IR. The relative study incidence of severe IRs by grade and cycle in which the reaction occurred is shown in Figure 2.[15] The majority of rituximab (87%) and cetuximab patients (58%) and all (100%) bevacizumab patients had grade 3 IRs. Grade 4 IRs were identified in 33% of the sample patients treated with cetuximab and two (8%) of these patients experienced fatal grade 5 IRs (Figure 2A). All cetuximab and the majority of rituximab (66%) IRs occurred during the first administration of the agent (Figure 2B).

While post-IR management varied significantly, the most common treatments for grade 4 reactions included the use of oxygen and corticosteroids.[15] Most patients experiencing grade 4 reactions to cetuximab (63%) and rituximab (83%) required hospitalization, with a mean hospitalization stay of 5 to 6 days. More than 80% of patients experiencing grade 4 reactions discontinued MoAb therapy.

Prospective Analysis of IR Burden on Patients, Caregivers, and Practice

To accurately assess the incidence and impact of IRs, a prospective time and motion study was conducted among 161 patients who received MoAb therapy in 27 community oncology centers.[16] Patients were assessed during the initial administration of MoAb therapy, 24 hours after therapy, and at 10 days follow-up.

During the initial assessment, the patients and staff were observed through a modified time sample time and motion method. Patient symptoms, visit time, and human resource time and tasks were recorded. The follow-up assessment periods involved remote and clinic interviews of patients and chart reviews regarding additional medical visits, services, and patient time and expense.

The mean age of the patients was 63 years and 44% of the patients were female.[16] Of the 161 patients enrolled, 90 were treated with rituximab and 71 were treated with cetuximab. The most common premedications were antihistamines (94%) and corticosteroids (53%). The incidence of IRs to cetuximab in this study was 32%, with 18 IRs classified as grade 1/2 reactions and 5 classifi ed as grade 3/4 reactions. IRs to rituximab occurred in 39% of patients, with 30 IRs classified as grade 1/2 and 5 as grade 3/4.

The assessment of patients treated in this study demonstrated that the incidence of an IR resulted in increased MoAb infusion times and staff time, leading to increased human resource costs.[16] The mean staff times during infusion and during the 10-day followup are shown in Figure 3A. Compared to patients not experiencing IRs, statistically significant increases in staff time during infusion were observed in patients experiencing IRs to either agent. There was also a trend for increased human resource costs associated with the additional staff time needed to manage the IRs (Figure 3B). In addition, for cetuximab-treated patients experiencing IRs, there was a significant increase in staff-associated time through the 10-day follow-up period (Figure 3A). The tasks were related to infusion time, visits, and hospitalization.

This study provided the first prospective analysis of the burden that IRs can have on patients, caregivers, and the clinical practice.[16] This study demonstrated that the incidence of IRs results in significant increases in staff time and a trend for increased human resource costs. Studies such as these can provide a model for community oncology centers to assess the tasks involved and the associated costs in treating IRs caused by therapies such as rituximab and cetuximab. These data can be used to help community oncology centers evaluate their prevention and management strategies for IRs, as well as the choice of therapy. These decisions may have a particularly significant impact on centers with already limited staff available.

Impact of IRs on Caregivers

Monoclonal antibody infusion reactions can have profound consequences on practice settings, often necessitating extra time and resources for management of the reactions. Many patients may require hospitalization for administration of medications and oxygen therapy while monitoring for further effects or biphasic reactions.[17] The consequences on the staff caring for these patients can be significant as well, particularly on nursing professionals, since they are on the front line caring for patients with infusion reactions.

Although most reactions are classified not as severe, but as mild to moderate, these reactions often require significant nursing care and management, with the potential to affect other staff as well. Treatment with monoclonal antibodies often, although not always, requires premedication with additional agents to reduce the risk of infusion reactions.[18] Patients who experience mild to moderate infusion reactions are often managed by the slowing of infusion rates, requiring increased staff intervention.[16] As discussed previously in this paper, the infusion reactions can create a burden on health-care resources as well as increased hospitalization requirements for patients. The prospective time and motion study revealed that staff time and costs were higher for patients who experienced infusion reactions and that these events are common.[15,16] Limitations of the time and motion study include the lack of assessment of the value of both patient and family caregiver time, and not specifically asking nurses about the impact of infusion reactions on the staff.

Although there is a paucity of research regarding how staff reacts to infusion reactions, one study of oncology nurses attending a national meeting did reveal specific information regarding the nurses and their feelings toward infusion reactions.[19] The impact on the patients was also discussed in this survey. These nurses were asked to complete an interviewer- administered survey of 31 items regarding the frequency and severity of infusion reactions, management of reactions, and how disruptive the reactions were to patients, nurses, and other staff. The nurses were also queried regarding the impact of reactions on the patients themselves as well as their caretakers and the staff and practice. Other questions concerned the risks of specific therapies.

As might be expected, infusion reactions were somewhat common, with one-third of the nurses indicating that any grade reaction occurred “often.”[ 19] The most frequent reactions were seen with rituximab or paclitaxel, and more nurses in the outpatient setting reported that paclitaxel caused infusion reactions than did nurses in the inpatient setting. Management of infusion reactions generally consisted of temporarily stopping the infusion, hydration of the patient, and administration of steroids, regardless of the severity of the infusion reactions. Permanent cessation of therapy with grade 3 or 4 reactions was reported by 69% of the nurses.

Interestingly, 96% of the nurses indicated that grade 3 or 4 reactions were “very” or “extremely” disruptive for patients, and disruptive to nurses 80% of the time, and to the entire staff 79% of the time. Among the nurses surveyed, 87% stated that patients and their caregivers experienced fear or apprehension related to infusion reactions, with 20% of the nurses indicating that patients felt afraid and anxious about experiencing another infusion reaction and some patients worried about the appropriateness of their ordered therapy. A higher percentage of outpatient nurses (88%) reported that “other patients are frightened when infusion reactions occur” than did inpatient nurses (62%).

This study shows that reactions are common, and that the surveyed nurses felt that even reactions considered not severe or life-threatening could be emotionally difficult and disruptive to patients, nurses, and other staff. The reactions take up the staff’s clinical time, limiting the ability of staff to care for other patients or tend to other duties. Although this study was limited by a convenience sample of nurses at a major conference, which might not represent the experience of all oncology nurses administering parenteral treatments for cancer, it nonetheless represents the first effort to measure the emotional impact of infusion reactions on staff and patients.

Discussion

Chemotherapy-induced side effects such as neutropenia and anemia have been shown to result in an increased burden on the patient and provider and an increase in the cost of providing care.[2,20] The data reviewed here demonstrate that infusion reactions constitute another class of side effects that lead to this increased burden. Given the life-threatening nature of severe IRs and the overall burden of all IRs, identifying clinical guidelines for the prevention and management is an important component of improving the quality and efficiency of the delivery of cancer treatment.[13,21] It may also be beneficial to identify best practice guidelines in community oncology centers, where staff are working at maximum capacity.[22]

Nurses are integral to the management of hypersensitivity reactions and participate in early identification of signs and symptoms of infusion reactions and in the comprehensive management of these events. Standing orders or protocols, as well as increased staff familiarity with infusion reactions, can assist staff with management of infusion reactions, facilitating immediate intervention by the nursing staff without waiting for a physician to implement orders.[10] It should be recognized, however, that these reactions do have an emotional impact, as well as a financial impact involving the cost of resources for staff and institutions.

In addition to the burden of IRs on the health-care provider, the incidence of IRs may have a significant impact on the patient and family. This includes out-of-pocket expenses associated with extended hospital stays and indirect expenses including time off work and travel time to the outpatient clinic and hospital.[16] Prevention and proper management of IRs may minimize these expenses for patients and families, and proper education of patients and caregivers about the associated risks of IRs with specific therapies should be discussed before therapy is administered.

The protocols for premedication of patients receiving MoAb therapy appear to differ among community oncology centers, although trends have been observed.[ 15,16] Differences in premedication may have effects on the incidence of infusion reactions, and more research may be needed to determine the most effective prevention strategies. This topic is covered in more detail in other articles within this supplement.

The presence of MoAb-induced IRs appears to lead to an increase in both tasks and staff time, which translates into increased human resource costs.[14] Interestingly increases in human resource costs and staff time are observed with mild and moderate IRs. Although a severe IR can be much more traumatic to the patient than mild IRs, the human resource times needed to manage mild and severe IRs are comparable. Monitoring of patients experiencing a mild IR, particularly by nursing staff, leads to increased allocation of time and resources to attend to the patient, regardless of the ultimate severity of the reaction.

Conclusion

The efficiency, effectiveness, and quality of delivering cancer care is an important and present challenge as patient populations increase and available specialty trained and experienced physicians and physician extenders are increasingly in short supply. These concerns and the overall cost of health care have led to the need for health economic models that quantify the costs associated with chemotherapy-induced adverse events, such as IRs. Health economic models have been described for other cancerrelated symptoms, such as breakthrough pain, to enumerate the burden placed on the patients, caregivers, and providers within a decision-making framework to better determine the most effective and efficient management strategies.[9] Future research should focus on advancing preliminary current data on the burden of IRs and further developing comprehensive health economic models that will guide daily clinical decision-making and health policy.

Financial Disclosure: Pamela H. Viale has acted as a speaker for Amgen, Bristol-Myers Squibb, Merck, and Novartis. She has served on advisory boards and acted as a consultant for Bristol-Myers Squibb, IMER, and Meniscus.

This supplement and associated publication costs were funded by Amgen.

Acknowledgment: The authors wish to acknowledge William Fazzone, PhD, from MediTech-Media, Ltd., supported by Amgen, for editorial assistance including: preparation oof the initial draft of the manuscript, collating the authors' comments, and formatting of the mansucript for submission.
 

Address all correspondence to:
Barry Fortner, PhD
4523 Ravenwood Oak Dr
Lakeland, TX 38002
e-mail: bfortner@p4healthcare.com

References:

References
1. Houts AC, Loh GA, Fortner BV, et al: Patient and caregiver time burden associated with anaemia treatment in different patient populations. Support Care Cancer 14:1195-1204, 2006.
2. Fortner BV, Okon T, Zhu L, et al: Costs of human resources in delivering cancer chemotherapy and managing chemotherapy-induced neutropenia in community practice. Commun Oncol 1:23-28, 2004.
3. Carelle N, Piotto E, Bellanger A, et al: Changing patient perceptions of the side effects of cancer chemotherapy. Cancer 95:155-163, 2002.
4. Moore KA: Breast cancer patients’ out-ofpocket expenses. Cancer Nurs 22:389-396, 1999.
5. Cella D, Peterman A, Passik S, et al: Progress toward guidelines for the management of fatigue. Oncology (Williston Park) 12:369-377, 1998.
6. Cremieux PY, Finkelstein SN, Berndt ER, et al: Cost effectiveness, quality-adjusted life-years and supportive care. Recombinant human erythropoietin as a treatment of cancer-associated anaemia. Pharmacoeconomics 16:459-472, 1999.
7. Moore K, Johnson G, Fortner BV, et al: The AIM Higher Initiative: New procedures implemented for assessment, information, and management of chemotherapy toxicities in community oncology clinics. Clin J Oncol Nurs 12:229-238, 2008.
8. Given B, Given CW, McCorkle R, et al: Pain and fatigue management: Results of a nursing randomized clinical trial. Oncol Nurs Forum 29:949-956, 2002.
9. Abernethy AP, Wheeler JL, Fortner BV: A health economic model of breakthrough pain. Am J Manag Care 14:S129-S140, 2008.
10. Lenz HJ: Management and preparedness for infusion and hypersensitivity reactions. Oncologist 12:601-609, 2007.
11. Kang SP,Saif MW: Infusion-related and hypersensitivity reactions of monoclonal antibodies used to treat colorectal cancer-Identifi - cation, prevention, and management. J Support Oncol 5:451-457, 2007.
12. National Cancer Institute. Common Terminology Criteria for Adverse Events v3.0 (CTCAE). Available at http://ctep.cancer.gov/forms/CTCAEv3. pdf: Accessed September 18, 2008.
13. O’Neil BH, Allen R, Spigel DR, et al: High incidence of cetuximab-related infusion reactions in Tennessee and North Carolina and the association with atopic history. J Clin Oncol 25:3644-3648, 2007.
14. Houts AC, Fortner BV, Moore K, et al: Human resource (HR) costs of chemotherapy infusion reactions (IR) in community oncology (abstract 14- 084). Support Care Cancer 14:626, 2006.
15. Schwartzberg LS, Stepanski EJ, Fortner BV, et al: Retrospective chart review of severe infusion reactions with rituximab, cetuximab, and bevacizumab in community oncology practices: Assessment of clinical consequences. Support Care Cancer 16:393-398, 2008.
16. Schwartzberg LS, Stepanski EJ, Walker MS, et al: Implications of IV monoclonal antibody infusion reaction for the patient, caregiver, and practice: Results of a multicenter study. Support Care Cancer 17:91-98, 2008.
17. Gobel BH: Hypersensitivity reactions to biological drugs. Semin Oncol Nurs 23:191- 200, 2007.
18. Chung CH: Managing premedications and the risk for reactions to infusional monoclonal antibody therapy. Oncologist 13:725-732, 2008.
19. Colwell HH, Mathias SD, Ngo NH, et al: The impact of infusion reactions on oncology patients and clinicians in the inpatient and outpatient practice settings: Oncology nurses’ perspectives. J Infus Nurs 30:153-160, 2007.
20. Fortner BV, Schwartzberg L, Tauer K, et al: Impact of chemotherapy-induced neutropenia on quality of life: a prospective pilot investigation. Support Care Cancer 13:522-528, 2005.
21. Chung CH, Mirakhur B, Chan E, et al: Cetuximab-induced anaphylaxis and IgE specifi c for galactose-alpha-1,3-galactose. N Engl J Med. 358:1109-1117, 2008.
22. Elting LS, Fortner BV, Bosserman L, et al: If at fi rst you don’t succeed, don’t quit, try again: The Centers for Medicare and Medicaid Services demonstration project on quality of care-2005 and 2006. J Support Oncol 4:147- 151, 2006.
 

Recent Videos