Comparison of Specialty Injection and Infusion Adverse Events Among Hospital Outpatient Settings vs Non-Hospital Outpatient Settings

J Clin Pathways. 2025;11(1):34-38. doi:10.25270/jcp.2025.11.01

Specialty drugs are used to treat rare, chronic, and difficult-to-manage conditions, such as cancer, rheumatoid arthritis, multiple sclerosis, and Crohn’s disease.¹ These drugs can be injected or infused and often require special handling and administration by a health care professional, making them extremely complex and expensive. In addition, specialty drugs may require monitoring for clinical outcomes and effective cost control measures when appropriate.^1,2 They can be covered under either the medical or the pharmacy benefit and are often administered in a hospital outpatient setting, doctor’s office, infusion suite, or even the patient’s home with a home health nurse.¹

Whereas specialty drugs account for over 60% of total drug spend, less than 5% of patients use specialty medications (medical plus pharmacy).^3,4 From 2016 to 2021, the cost of specialty drugs increased by 43%, totaling $301 billion in 2021.4 Specialty drug spend was projected to increase by more than $100 billion between 2020 and 2025.⁵ Globally, between 2019 and 2023, more than two-thirds of new drug launches were projected to be specialty drugs, contributing to substantial increase in cost trends.⁶ As a class, specialty drugs continue to be expensive, however, some new specialty drugs are priced at unprecedented levels. For instance, the gene therapy Lenmeldy, approved by the US Food and Drug Administration (FDA) in March 2024 is priced at $4.25 million, making it the world’s most expensive drug.⁷

The total cost of a specialty drug administration can be two to three times higher in the hospital outpatient setting compared with non-hospital outpatient settings, based on the medication charge structure and the inclusion of a facility fee.^8-12 In addition, administration costs associated with the top 10 specialty drugs, which account for approximately half of overall specialty medical spend, can be reduced by 50% if these drugs were to be administered in a physician’s office or non-hospital affiliated ambulatory infusion suite (AIS) rather than in a hospital outpatient setting.^9,10,13,14 Therefore, optimizing site-of-care by administering expensive specialty drugs at clinically appropriate, convenient and low-intensity care settings can be associated with substantially lower costs.^11,12 Cigna Healthcare’s proactive site-of-care optimization programs promote clinically appropriate, safe, cost-effective, and highest value site of care with positive health outcomes and highlight the significance of value-based care.^1,15 These innovative solutions can drive access and affordability of specialty drugs while offering meaningful value for patients, employers, and the health care ecosystem.

Although injected and infused specialty medications are generally well tolerated with a good safety profile, they are associated with the potential risk of infections, acute infusion reactions, and escalation of care.^16-18 Acute infusion reactions, usually occurring within 24 hours of infusion, can be mild to moderate and associated with chills, fever, mild hypotension, dyspnea, and rash. Severe reactions are associated with severe hypotension, anaphylaxis, and cardiac dysfunction.¹⁷ Further, adverse events can also include potential escalation of care through an emergency department visit or hospital admission the same or next day after the administration of an infusion.¹⁸

To our knowledge, there are no documented studies in the US analyzing adverse events associated with injected and infused specialty medications among hospital outpatient setting vs non-hospital outpatient settings. This study analyzed whether there are significant differences in adverse events associated with specialty medications among hospital outpatient setting vs non-hospital outpatient settings such as physician’s office, patient’s home, and non-hospital affiliated AIS. The study hypothesized that adverse outcomes were not unfavorably impacted by shifting specialty medications to low-intensity non-hospital outpatient settings from high-intensity hospital outpatient setting.

Program Background

Cigna Healthcare has various strategies that clients can choose to manage the cost of medical benefit specialty drugs, and these strategies manage the setting in which the specialty drugs are administered. Specialty Care Options leverages the review of clinical appropriateness of high-intensity hospital outpatient settings during the medical necessity review. An individualized approach allows for providers and customers to choose less-intensive options (eg, patient’s home or non-hospital affiliated AIS) that best meet the clinical needs of the customer.¹ To qualify, the patient must be stable on their infusions, and there must be clinically equivalent sites available in the patient’s area of residence.¹ Pathwell Specialty is Cigna Healthcare’s leading solution to manage medical specialty drug spend, with a specific specialty drug network made up of more than 28 000 quality providers, including home infusion, physician offices, non-hospital affiliated AIS, hospital outpatient providers, and specialty pharmacies.¹⁵

Data

The source of the retrospective data used in this study is Cigna Healthcare’s administrative and commercial medical claims data. These standard data are routinely collected by the company to fulfill its operational purposes. Claims and administrative data were used only for the purposes of the study analyses post facto. A separate internal team deidentified these data, in accordance with security measures, to protect member confidentiality before any secondary data analysis. These data cannot be shared externally since the information contained therein are sensitive and proprietary in nature. The study methods follow the ethical guidelines of the 1975 Declaration of Helsinki.

The study sample comprised 122 448 patients aged 18 years or older receiving specialty medications between January 1, 2021, and October 31, 2023. These patients received one or more of the 72 specialty drugs that were commonly administered across the study’s four focus sites of care: hospital outpatient department, physician’s office, patient’s home with home health nurses, and non-hospital affiliated AIS. The study analyzed 72 specialty drugs comprising 22 oncology and 50 nononcology specialty drugs (Table 1). These 72 specialty drugs are managed by Cigna Healthcare’s site-of-care optimization programs and Cigna Pathwell Specialty.^1,15 The top 10 utilized specialty drugs were Prolia, Inflectra, Entyvio, Xolair, Neulasta, Keytruda, Remicade, Eligard, Kanjinti, and Gamunex-C; these drugs comprised approximately 50% of the injection/ infusion instances.

The patients included in this study received one or more specialty drugs through one or more injections or infusion instances. Overall, 995 333 injection/infusion instances of the specialty drugs were used in the study model.

Measures

Healthcare Common Procedure Coding System codes and drug names were used to identify specialty medication claims. Patients who received these specialty medications were identified for the data period. All instances of specialty administration that any given patient was on for the data period were included in the study. Specialty drugs were identified by names and categorized as either oncology or non-oncology.

A categorical variable was created to classify patients based on the site of care where they received specialty injection/infusion services: “HOPD” (hospital outpatient department), “office” (physician’s office), “home” (patient’s home), or “AIS” (non-hospital-affiliated ambulatory infusion suite).

Injection/infusion episode severity was measured using the Symmetry Episode Treatment Groups (ETG) severity score.^19,20 The ETG severity score of each injection/infusion episode was assessed by considering demographic, condition status (eg, condition specificity, disease progression), and comorbidity markers associated with the episode and then summing their weights.^19,20 The patient characteristics included in the study were age and sex. The analysis included medical costs, encompassing all health care service expenses incurred while patients were managed by Cigna Healthcare during the study period.

The following adverse outcome measures were studied: medication-related adverse event 1-day post-injection/infusion (MRA1; Table 2), injection/infusion-related infection within 7 days post-injection/infusion (INF7; Table 3), emergency department visit within 1, 7, and 30 days post-injection/infusion (ED1, ED7, and ED30).

International Classification of Diseases (ICD) codes, Current Procedural Terminology (CPT) codes, and revenue codes were used to identify adverse outcome measures. Dichotomous variables were created grouping injection/infusion instances into adverse outcome “yes” vs adverse outcome “no” categories for any given outcome measure.

Statistical Analysis

Descriptive and chi-square (χ2) statistics were performed to assess any overall associations between specialty injection/infusion site-of-service and adverse clinical outcome measures,

such as MRA1, INF7, ED1, ED7, ED30, HA1, HA7, and HA30. Logistic regression models were used to compute odds ratios (ORs) and their 95% CIs to determine the magnitude and direction of potential associations between site-of-service and adverse outcomes after adjusting for explanatory variables, such as type of infusion, type of drug, age, sex, medical cost, and episode severity. SAS Enterprise Guide, version 8.2 (SAS Institute Inc) was used to conduct analyses.

More than half of specialty injections/infusions were administered in the office setting (55.0%) and under one-third were administered in the HOPD setting. Home and AIS settings represented approximately 10.2% and 3.4% specialty injections/ infusions, respectively. Most injections/infusions (63.4%) were administered to patients aged 40 to 64 years. Approximately two-thirds (63.3%) of injections/infusions were administered to female patients. The ETG severity score was less than or equal to 2 for most of the injection/infusion episodes (70.8%). Incidence of MRA1 occurred in 0.3% of injections/infusions, and incidence of INF7 occurred in 294 injections/infusions. ED1 and ED7 occurred in 0.4% and 1.7% injections/infusions, respectively. ED30 incidence occurred in 5.5% of injections/infusions compared with HA30 incidence in 3.3% of injections/infusions (Table 4).

The results of χ2 analyses showed that HOPD injections/ infusions had the highest likelihood of developing MRA1 (0.4%; P < .05) compared with home and AIS injections/infusions, which had the lowest incidence of MRA1 (0.1%; P < .05) (Table 5). Among the various sites of care, AIS had the lowest incidence of ED1 (0.2%; P < .05), ED7 (1.0%; P < .05), ED30 (3.4%; P < .05), HA1 (0.1%; P < .05), HA7 (0.4%; P < .05), and HA30 (1.3%; P < .05) (Tables 6 and 7).

Age and sex were significantly associated with all eight adverse outcomes analyzed in the study (P < .05). Female patients had higher likelihood of MRA1 compared with male patients, and male patients had higher likelihood of ED7, ED30, HA1, HA7, HA30 compared with female patients. Among the various age groups, patients aged 65 years or older had significantly higher incidence of ED7 (2.0%; P < .05), ED30 (6.8%; P < .05), HA7 (1.3%; P < .05), and HA30 (4.6%; P < .05), compared with (0.8%; P < .05), ED7 (3.2%; P < .05), ED30 (9.1%; P < .05), HA1 (0.8%; P < .05), HA7, (2.6%; P < .05), and HA30 (7.8%; P < .05) was observed in patients with medical costs of more than $1 million. Oncology injections/infusions had higher likelihood of MRA1, INF7, ED7, ED30, HA1, HA7, and HA30 incidence compared with non-oncology injections/infusions (Tables 5, 6, and 7).

The results of logistic regression analyses showed that the incidence of MRA1 was 84%, 80%, and 49% less likely in AIS (OR, 0.16; 95% CI, 0.10-0.25), home (OR, 0.20; 95% CI, 0.16-0.25), and office (OR, 0.51; 95% CI, 0.47-0.56) injections/ infusions, respectively, compared with HOPD injections/ infusions (Table 8). The odds of ED1 (OR, 0.70; 95% CI, 0.61-0.79), ED7 (OR, 0.76; 95% CI, 0.71-0.81), and ED30 (OR, 0.86; 95% CI, 0.83-0.89) incidence with in-home injections/infusions were 30%, 24%, and 14% lower, respectively, compared with HOPD injections/infusions (Table 9). There were 23% and 25% lower odds of HA7 (OR, 0.77; 95% CI, 0.71-0.84), and HA30 (OR, 0.75; 95% CI, 0.71-0.79) incidence, respectively, with in-home injections/infusions compared with HOPD injections/infusions. There were 62% and 26% lower odds of HA1 (OR, 0.38; 95% CI, 0.25-0.59), and HA7 (OR, 0.74; 95% CI, 0.62-0.90) incidence, respectively, in AIS injections/infusions compared with HOPD injections/infusions (Table 10).

In male patients, the incidence of MRA1 (OR, 0.84; 95% CI, 0.77-0.91), INF7 (OR, 0.67; 95% CI, 0.50-0.89), ED7 (OR, 0.95; 95% CI, 0.92-0.98), and ED30 (OR, 0.95; 95% CI, 0.93-0.97) was 16%, 33%, 5%, and 5% less likely, respectively, compared with female patients (Tables 8 and 9). The odds of MRA1 (OR, 0.66; 95% CI, 0.58-0.75), and HA1 (OR, 0.75; 95% CI, 0.68-0.83) were 34% and 25% lower, respectively, in patients aged 65 or older than those aged 40 to 64 years. Additionally, patients aged 18 to 39 years were 1.17, 1.21, and 1.19 times more likely to develop MRA1, ED30, and HA30, respectively, than those aged 40 to 64 years (Tables 8, 9, and 10).

Patients with medical cost between $500 000 and $1 million were 1.18, 1.31, 1.87 times more likely to have MRA1 (OR, 1.18; 95% CI, 1.05-1.33), ED1 (OR, 1.31; 95% CI, 1.19-1.44), and HA1 (OR, 1.87; 95% CI, 1.72-2.07), respectively, compared with those with medical cost between $100 000 and $500 000 (Tables 8, 9, and 10). For every unit increase in severity score, the odds of ED1 (OR, 1.04; 95% CI, 1.03-1.05) and HA1 (OR, 1.06; 95% CI, 1.05-1.07) increased by 4% and 6%, respectively (Tables 9 and 10).

To our knowledge and based on a literature review, this is the first study in the US analyzing adverse events associated with injected and infused specialty medications among hospital outpatient setting vs non-hospital outpatient settings, such as physician’s office, patient’s home, and non-hospital affiliated AIS. This study demonstrated statistically significant clinical outcomes associated with injected and infused specialty medications administered across various sites of care, such as hospital outpatient setting vs non-hospital outpatient settings.

The results showed that the odds of incidence of MRA1, ED1, ED7, ED30, HA7, and HA30 were lower in non-hospital outpatient settings compared with hospital outpatient setting. Further, the results demonstrate that the AIS setting had the lowest incidence of MRA1, HA1, and HA7, and the home setting had the lowest incidence of ED1, ED7, ED30, and HA30. Also, the odds of MRA1, ED1, ED7, and HA30 were lower in the office setting compared with the HOPD setting. These favorable clinical outcomes associated with specialty medications administered in the non-hospital outpatient settings observed in this real-world clinical practice among a large sample of patients show that these medications can be safely administered in the non-hospital outpatient settings when clinically appropriate.

Site-of-care optimization strategies that move patients from high-cost hospital outpatient settings to cost-effective non-hospital outpatient settings, such as physician’s offices, patient’s home, and non-hospital affiliated AIS, can lead to substantial reduction in specialty medication costs.^11,12,21 Even though site-of-care optimization strategies aimed at managing expensive injected and infused specialty medications are gaining adherents due to the substantial cost savings, some clinicians have expressed reservations about the implications for patient safety of routinely administering these drugs in non-hospital outpatient settings.²² A 2023 study conducted by Baker and colleagues¹⁸ showed that compared with biologic infusions administered at physician’s office, those administered at home were associated with increased adverse events that required escalation of care. The specialty medications analyzed in the Baker et al study included a small list of biologic infusions used to treat immune-mediated diseases.¹⁸ Our study analyzed 73 specialty drugs used to treat a broad spectrum of oncology and non-oncology conditions, such as cancer, immune-mediated diseases, hematologic disorders, bone disorders, and ophthalmologic conditions.¹⁸ The Baker et al study analyzed two sites of care, namely the physician’s office and patient’s home, whereas our study analyzed four sites of care.¹⁸ Our study examined substantially larger patient sample and infusion instances compared with the Baker et al study.¹⁸ Our findings present evidence that administration of expensive specialty medications in clinically appropriate, cost-effective, and highest value sites of care can be achieved without compromising safety outcomes.

Site-of-care optimization strategies not only offer avenues for significant reduction of specialty medication costs but also improve access and affordability of specialty medications. Cigna’s site-of-care optimization and Pathwell Specialty programs proactively manage specialty drugs by administering them at clinically appropriate, safe, cost-effective, and highest value sites of care with improved patient outcomes.^1,15 Further, site-of-care optimization strategies highlight the importance of value-based care by promoting cost-effective care with optimal clinical outcomes and best patient experience.²³ Also, the home setting provides additional benefits to patients, such as convenience of in-home administration, avoidance of costs and logistics associated with traveling, and minimization of institutional visits for patients with compromised immune function who have an increased risk of infection.

This study has several strengths. First, the study data comprised large real-world commercial claims data across 3 years. We analyzed approximately 1.1 million injection/infusion instances of specialty drugs administered to approximately 135 000 patients aged 18 years or older. Second, we analyzed adverse events associated with specialty drugs across comprehensive sites of care, such as a hospital outpatient setting, physician’s office, patient’s home, and non-hospital affiliated AIS. Third, by studying 73 specialty drugs that were commonly administered across all four sites of care, we analyzed the same group of drugs administered for similar conditions across all four settings rather than site-specific drugs that could vary by any given site of care. Fourth, we analyzed 73 specialty drugs used to treat a broad spectrum of oncology and non-oncology conditions, such as cancer, immune-mediated diseases, hematologic disorders, bone disorders, and ophthalmologic conditions. Fifth, to account for heterogeneity in patient disease conditions, we adjusted for specialty drug names and drug categories (oncology vs non-oncology).

We acknowledge certain limitations of our study. First, the claims sample is not representative of the Medicare and uninsured US population. Second, the study did not adjust for potential predictors, such as race, education, and socioeconomic status. Third, the results of the study are from claims data associated with site-of-care optimization programs of a single payer. Fourth, data signifying primary indication for emergency department visit or hospital admission that occurred 1, 7, and 30 days post-injection/infusion were not available in the database. However, this same limitation was observed with data associated with patients receiving specialty drugs across all four sites of care. Despite these limitations, the strengths outlined above provide significant clinical value to our study.

This study results elucidate that Cigna Healthcare’s site-of-care management strategies of injected and infused specialty medications is associated with favorable clinical outcomes and quality in the non-hospital outpatient settings compared with the hospital outpatient setting. A variety of clinical endpoints were measured confirming significantly lower incidence of medication-related adverse event, injection/infusion infection, emergency department visit, and hospital admission in the non-hospital outpatient settings compared with the hospital outpatient setting. These findings are consistent across all three less-intensive sites studied (ie, physician’s office, patient’s home, non-hospital affiliated AIS). The study findings demonstrate that using alternative sites instead of hospital outpatient sites is both safe and effective, with the potential to improve clinical outcomes.

The findings of this large real-world, retrospective study confirm that Cigna Healthcare’s site-of-care optimization strategy of shifting injected and infused specialty medications from high-cost hospital outpatient settings to more affordable, clinically appropriate alternative settings is associated with favorable clinical outcomes and quality in the non-hospital outpatient settings compared with hospital outpatient settings. Our study presents significant evidence that specialty medications can be safely administered in the non-hospital outpatient settings when clinically appropriate. Widespread adoption of site-of-care management strategies that offer alternatives to the hospital outpatient setting might reduce the burden of rising health care costs, increase affordability, enhance patient convenience, and improve patient choice. These findings underscore the importance of offering similar alternatives as an option to patients receiving specialty medications as part of their treatment. Further, randomized controlled studies should be conducted to confirm the improved outcomes identified in many of the assessed metrics.

Affiliation:

The Cigna Group, Bloomfield, CT

Correspondence:

Lavanya Raj, MBBS, MS

900 Cottage Grove Road

Bloomfield CT, 06002

Email: lavanya.raj@evernorth.com

Disclosures:

The authors disclose no financial or other conflicts of interest.