Abstract
BACKGROUND:
Though specialty pharmacies collect and report a myriad of measures to internal and external stakeholders, these data are infrequently used for evaluating specialty pharmacy performance or clinical decision-making. Establishing standardized specialty pharmacy measures could enable benchmarking, performance-based contracting, and high-cost drug utilization management while improving patient care and specialty medication management.
OBJECTIVE:
To determine consensus among specialty pharmacy stakeholders on important and usable measures for specialty pharmacies managing patients with rheumatoid arthritis (RA).
METHODS:
A modified Delphi study was conducted with a multisite study group (n = 25 sites). Expert panelists from diverse RA-related backgrounds participated in up to 3 survey rounds. An environmental scan informed an initial list of 10 generalized measures. In rounds 1 and 2, panelists rated each measure’s importance and usability on an 11-point Likert scale. Measures were categorized to be included, excluded, or rescored based on mean scores. Panelists received group scores and feedback between rounds. In round 3, specialty pharmacy–affiliated panelists assessed the feasibility of measures that reached consensus. Final study group voting categorized measures as core (should be collected and reported by all specialty pharmacies without exception), reach (important and actionable but not yet essential for specialty pharmacies to collect and report), or neither.
RESULTS:
Of 315 recruited panelists, 118 participated; response rates were 83% (round 1), 76% (round 2), and 73% (round 3). In round 1, 4 measures met consensus; 3 measures moved forward to be rescored in round 2, and 2 general measures were delineated to more specific measures for feasibility scoring. Feasibility consensus varied. Final study group voting identified 7 specific core measures: adherence, serious adverse effects, patient response to therapy, medication discontinuation, medication switching, and tuberculosis and hepatitis B screening (if applicable). Six reach measures included immunization screening, drug-specific laboratory screening, common AEs, patient functional status, patient disease activity, and medication persistence.
CONCLUSIONS:
There was a high survey response rate within each round. Consensus was reached for 13 measures deemed important and usable by specialty pharmacy stakeholders, with most considered moderately/very feasible by specialty pharmacists.
Plain language summary
A multidisciplinary panel of specialty pharmacy professionals completed 3 rounds of surveys to evaluate measures for specialty pharmacies to use when managing patients receiving treatment for rheumatoid arthritis. The panel reached consensus on 13 important and usable measures. A subsequent study group classified 7 of these measures as essential for implementation, whereas the remaining 6 were considered important and applicable, though not critical.
Implications for managed care pharmacy
This study addresses an important gap for managed care pharmacists because better standardization and reporting of measures used in specialty pharmacy practice is needed to drive optimal use of these costly therapies. These 13 consensus measures establish a standardized framework for performance-based contracting, quality-of-care evaluation, and effective utilization management of high-cost RA therapies.
Rheumatoid arthritis (RA) is a chronic inflammatory autoimmune disease affecting approximately 1.5 million individuals in the United States.1 RA significantly reduces patients’ quality of life and contributes to long-term disability.2 Biologic disease-modifying antirheumatic drugs (DMARDs) are recommended for patients with moderate to severe disease activity to induce remission and improve clinical outcomes.3 Biologic DMARDs account for a large majority of pharmacy benefits spending, which has increased significantly over the last decade.4 Because of their potential high cost, complex administration requirements, and need for ongoing monitoring, biologic DMARDs are often dispensed and managed by accredited specialty pharmacies.5
Specialty pharmacies that dispense high-cost complex therapies are critical in facilitating affordable access to appropriate treatment and managing patients, including coordinating care and monitoring treatment safety and response. Services provided by specialty pharmacies vary widely among different specialty pharmacy models. A recent economic report found that 67% of specialty prescription revenues were captured by 3 pharmacy benefits manager–owned specialty pharmacies.6 However, growing research has demonstrated that the model and services provided by specialty pharmacies can impact overall patient costs and clinical and financial outcomes.7–9 In the absence of standardized metrics, it is challenging for payers, manufacturers, and patients to select pharmacies based on performance, outcomes, and patient or provider experience.
Specialty pharmacies currently collect and report a large amount of data, yet there is a lack of standardization in the data being collected and limited insight into its use for patient care or practice changes.10 Third-party specialty pharmacy accreditation through nationally recognized organizations requires demonstration of practice excellence through performance metrics, with specific calculation guidance available for only a limited number of metrics, which are primarily operational (eg, dispensing accuracy rates, call center answer time), and few therapy-specific clinical outcome measure recommendations.11,12 Similarly, data are frequently required by drug manufacturers for specialty pharmacies to have access to their medications and by third-party payers for network inclusion, but data requirements also primarily to focus on operational metrics.13 Though evidence supports that specialty pharmacies can impact patient outcomes, clinical outcome measures have not yet been a focus for reporting requirements, likely because of a lack of consensus within the industry on which metrics to collect for each therapeutic area.14,15 Specialty pharmacies are also reliant on data management infrastructure to evaluate and collect standardized outcomes and to present aggregated metrics to stakeholders.15 Although individual specialty pharmacies often collect clinical outcome measures for specialty disease states, measures are typically developed internally by specialty pharmacists and may lack a research-based approach or involvement of other key stakeholders when selecting measures, both of which this current study addresses.9,16
Establishing standardized measures, agreed on by all stakeholders, for use by specialty pharmacies serving patients with RA could accomplish the following: (1) enhance patient care and improve DMARD management to meet therapy targets; (2) guide specialty pharmacies that currently lack disease-specific outcome measures; (3) facilitate industry-wide benchmarking in specialty pharmacy; and (4) enable standardized reporting for accreditation, contracting, and research purposes.
The Delphi technique is a common research method that engages a group of experts in a flexible framework with project-specific goals to build, then reach consensus through a series of survey questionnaires.17–19 A national effort involving key stakeholders is needed to reach consensus on specialty pharmacy outcomes measures to be used when managing patients with RA. RA was selected as the clinical area focus because of its high volume of specialty medication use, complex monitoring needs, and high cost of treatment. This modified Delphi study aimed to identify important and usable measures that can be implemented in specialty pharmacy practice to ensure the safe, appropriate, and effective use of specialty medications in patients with RA.
Methods
A multisite (n = 25) modified Delphi study was conducted by the Vanderbilt Health System Specialty Pharmacy (HSSP) Outcomes Research Consortium from October 2023 to August 2024. The Vanderbilt HSSP Outcomes Research Consortium is open to any health system specialty pharmacy in the United States, with no membership or fee requirements. Interested Consortium members were invited to participate in the study. Study methods are summarized in Figure 1. The study protocol was determined to be exempt by the Vanderbilt University Medical Center Institutional Review Board.
FIGURE 1.
Modified Delphi Method
Modified Delphi methodology is shown, including a pre-Delphi environmental scan and expert panelist recruitment followed by 3 rounds of measure scoring. Expert panelists scored measures on importance and usability in rounds 1 and 2. Specialty pharmacy–affiliated panelists scored measures on feasibility in round 3. Mean scores determined measure outcome (met consensus, rescored in subsequent round, or excluded). As a final step, the Consortium study group voted on measures that met consensus as core, reach, or neither.
aSpecialty pharmacy-affiliated expert panelists.
RA = rheumatoid arthritis; SP = specialty pharmacy.
INITIAL MEASURE DEVELOPMENT
Four workgroups conducted an environmental scan and drafted the initial list of outcome measures. The environmental scan included a comprehensive literature review using PubMed, national pharmacy and medical association measures, pharmacy accreditation standards, and clinical guidelines and trials. Workgroups compiled a list of potential measures that were voted on by workgroup members to be used in an initial draft measure list, which were then reviewed by the primary investigators. Primary investigators discussed and agreed on grouping several specific measures into more generalized measures (eg, RAPID3 into patient response to therapy) based on what the measure broadly evaluates, which were also reviewed and agreed on by the larger study group. Measure generalization was done to reduce the number of items to be voted on and promote broad implementation, as clinical monitoring practices vary widely. Relevant outcome measures were identified and refined through discussions, resulting in a draft list of 10 outcome measures for inclusion in the first survey round (Supplementary Table 1 (1.8MB, pdf) , available in online article). The more specific measures were retained to be scored in round 2 when evaluating delineated specific measures.
EXPERT PANEL
Consortium study group members submitted names and contact information of expert panelists representing relevant areas of RA-related care and management, including prescribing providers, clinic-based nurses, ambulatory care pharmacists, specialty pharmacy providers with some RA involvement, specialty pharmacy providers dedicated to RA, managed care stakeholders, pharmaceutical industry stakeholders, pharmacy analysts/specialty pharmacy leadership, specialty pharmacy technicians/liaisons, and accreditors. Submitted panelists were recruited via e-mail with the option to participate in up to 3 rounds of surveys by completing an enrollment survey. Selection criteria for each panelist type are defined in Supplementary Table 2 (1.8MB, pdf) . Recruitment targeted diverse geographic representation, with a minimum goal of 50 participants overall. Panelists remained anonymous throughout the study.
SURVEY DESIGN AND INSTRUMENT PILOT
A primary investigator developed each survey, which was reviewed and piloted by coprimary investigators and workgroup leads. Study team members had the opportunity to review each survey prior to dissemination. Feedback was incorporated prior to dissemination. Each survey, delivered via REDCap (a Health Insurance Portability and Accountability Act of 1996–secured data warehouse),20,21 included study details, completion instructions, and a 4-week deadline. Reminders were sent 2 weeks after the initial request.
DELPHI ROUNDS AND CONSENSUS
Three rounds of surveys were administered. The first 2 rounds were used to determine consensus and the third was used to evaluate feasibility. In rounds 1 and 2, panelists scored measures based on importance and usability using an 11-point Likert scale (0-10). Importance and usability were chosen to ensure relevance and impact of the consensus measures. Importance was selected to align with prior Delphi survey methodology; usability was included to prioritize measures that could directly influence patient care.22 Importance was defined as how meaningful the expert panel deemed the measure to be for patient care. Usability was defined using the National Quality Forum definition: “the extent to which potential audiences are using or could use performance results for both accountability and performance improvement to achieve the goal of high-quality, efficient healthcare for individuals or populations.”23
Consensus scoring and measure thresholds for determining a measure outcome are shown in Figure 1. In round 1, measures having mean importance and usability scores greater than 7.5 met consensus for inclusion and those with mean importance and usability scores less than 6.25 met consensus for exclusion. If the mean score was between 6.25 and 7.5, the measure was rescored in round 2. Where mean importance and usability scores placed the measure in differing categories (include, exclude vs rescore), the coprimary investigators determined their category assignment. Though no specific criteria were set, importance was prioritized over usability.
In round 2, general measures scoring at least a median of 7 met consensus for inclusion and all others were excluded. The round 2 survey also included questions about delineated specific measures and panelists’ preferences related to documentation, methods for measure evaluation, and monitoring frequency for measures that met consensus in round 1 and those being rescored in round 2. If a large majority of panelists recommended that a delineated specific measure be used during the round 2 measure specification scoring, that delineated specific measure was included in the final measure set. Therefore, the final consensus measure set included both initial general measures and specific, delineated measures.
After rounds 1 and 2, two study group members reviewed expert panelists’ comments to develop themes. Particularly poignant comments were retained verbatim. After each round, the Consortium study group analyzed and consolidated results to refine measures and shared anonymous feedback via Delphi reports, which included scoring results and thematic feedback, 1 week prior to the next round.
In round 3, specialty pharmacy–affiliated expert panelists, including specialty pharmacy providers, specialty pharmacy leaders, specialty pharmacy technicians/liaisons, and pharmacy analysts, assessed the feasibility of the final measure set (general measures that met consensus in rounds 1 and 2 and delineated specific measures that scored highly in round 2). Panelists scored feasibility using an 11-point Likert scale, provided feedback on barriers and facilitators for implementing measures, and reported what elements they currently collect and/or report for patients managed by their specialty pharmacies. Feasibility scores were not used to determine if a measure met consensus, rather they were evaluated to inform final study group voting and implementation consideration. Possible facilitators included need/motivation (buy-in from stakeholders, perceived importance/usability), capability (workload or workflow integration, technology capabilities), opportunity (data availability, anticipated high patient engagement), or none. Possible barriers included need/motivation (buy-in from stakeholders, perceived importance/usability), capability (workload or workflow, time constraints, technology limitations), opportunity (data availability, concern for low patient engagement), or none.
After round 3, as a final step, Consortium study group members voted to categorize the final measure set as “core” (should be collected and reported by all specialty pharmacies without exception; fundamental to providing specialty pharmacy services for patients with RA), “reach” (not yet essential for specialty pharmacies to collect and report; benefits are recognized, yet they may be less feasible or necessary), or “neither” (should not be collected or reported by specialty pharmacies). No a priori thresholds were established for categorization. Instead, the final designation for each measure was determined based on the majority vote of participants. Results were shared via e-mail with all expert panelists.
DATA ANALYSIS
Descriptive statistics, including means, frequencies, and percentages, were used to summarize the study sample and outcome measures. Data visualization was performed using graphs. All data analyses and major graphs were generated using R (version 4.4.0).
Results
SAMPLE
Of 315 recruited panelists, 118 expert panelists consented to participate and were sent surveys for rounds 1 and 2 of consensus voting. There were 98 respondents (83% response rate) in round 1 and 90 respondents (76% response rate) in round 2. Of the 118 panelists, the round 3 feasibility assessment survey was sent only to the 70 affiliated with specialty pharmacies, with 51 responses (73% rate). Supplementary Figure 1 (1.8MB, pdf) shows the number of responses for each panelist type by round, with the majority of responses being from pharmacy leaders, pharmacists or nurses dedicated to RA, pharmacists or nurses with some RA involvement, or prescribing providers. Reports between rounds are included as Supplementary Exhibits 1, 2, and 3 (1.8MB, pdf) . Ultimately 7 core and 6 reach measures were identified; the process for how this consensus was reached is described here chronologically by round.
DELPHI ROUNDS AND CONSENSUS RESULTS
Round 1: Of the 10 measures evaluated for importance and usability during survey round 1, 4 measures met consensus for inclusion, 3 met consensus for exclusion, and 3 measures moved forward to be rescored in round 2. Importance and usability scores for each measure are shown in Figure 2. In round 1, adherence, medication outcomes, patient response to therapy, and safety screening met consensus for inclusion. Panelists commented that these metrics can offer clear intervention opportunities for pharmacists.
FIGURE 2.
Rounds 1 and 2 Measure Scoring Results
Expert panelist scoring on whether measures are important and usable is shown. In round 1, there were 98 responses, and in round 2, 90 responses. Zero indicates not at all, 10 indicates very important/usable. To the right of the measure, the outcome of that measure for each round is shown, followed by the scoring breakdown and the mean score. Grey and lighter shading indicates a higher score (more important/usable).
Unplanned health care utilization, planned health care utilization, and productivity met consensus for exclusion in round 1. Panelist comments about planned and unplanned health care utilization and productivity included views that these metrics are difficult to collect, would require baseline and repeat measurements, and that these could be influenced by any number of issues that may be unrelated to the specialty medication or condition.
Disease activity, patient functional status, and patient quality of life did not meet consensus for either inclusion or exclusion and advanced to round 2 for review and rescoring. Panelist comments for disease activity, patient functional status, and quality of life indicated that although these metrics are important for accreditation and prior authorization requirements, they could be subjective, imprecise, and difficult for pharmacists to act on. Themes from expert panelist feedback for all measures can be found in Supplementary Exhibit 1 (1.8MB, pdf) .
Round 2: Importance and usability scores increased for disease activity, stayed similar for patient functional status, and decreased for patient quality of life in round 2 (Figure 2). Disease activity and patient functional status met criteria for inclusion, whereas patient quality of life met criteria for exclusion. Though disease activity and patient functional status met consensus in round 2, panelists’ views on these measures and quality of life shifted in round 2, with comments stating that these measures should not be evaluated by specialty pharmacies and instead should remain the responsibility of the prescriber to assess during clinic visits. Preferences related to documentation, evaluation method, and frequency for the measures that had met consensus or were being rescored in round 2 are summarized in Table 1 and detailed in Supplementary Exhibit 2 (1.8MB, pdf) .
TABLE 1.
Expert Panel Preferences for Documentation and Measurement of Consensus Measures (Round 2, n = 87)
| Adherence | |
| Specification | Preference |
| Documentation | Actions taken by the pharmacy to address adherence (90%) |
| Adherence scores (87%) | |
| Documentation that adherence has been assessed (71%) | |
| Methods | Patient-reported missed doses (83%) captured monthly (60%) |
| PDC (77%) captured quarterly (51%) | |
| Medication outcomes | |
| Specification | Preference |
| Elements to capture | Serious AEs (90%) measured/aggregated monthly (43%) |
| Medication discontinuation (82%) measured/aggregated quarterly (35%) | |
| Medication switching (75%) measured/aggregated quarterly (28%) | |
| Common AEs (66%) measured/aggregated monthly (44%) | |
| Specific medication persistence (64%) measured/aggregated either quarterly (32%) or every 6 months (30%) | |
| Response to therapy | |
| Specification | Preference |
| Documentation | Outcomes of patient response assessment (82%) |
| Actions taken by the pharmacy to address patient response (70%) | |
| Documentation that patient response has been assessed (70%) | |
| Methods | Disease activity measure (72%) |
| Flare occurrence/frequency (69%) | |
| Treat to target progression based on patient goals (59%) | |
| Functional status measure (53%) | |
| Monitoring after initiation | Within 3 months (58%) |
| Monitoring frequency | Quarterly (37%) |
| Every 6 months (31%) | |
| Safety screening measures | |
| Specification | Preference |
| Elements to capture | Documentation that safety screening assessed (87%-96% for all elements) |
| Screening to be captured | Tuberculosis screening (83%) before initiation only (33%) or based on package insert (35%) |
| Drug-specific laboratory monitoring (77%) based on package insert (57%) | |
| Hepatitis B virus screening (76%) prior to initiation only (39%) or based on package insert (35%) | |
| Immunization screening (76%) annually (49%) | |
| Functional status | |
| Specification | Preference |
| Documentation | Documentation that functional status has been assessed (68%) |
| Outcomes of functional status assessment (63%) | |
| Methods and frequency | Pain (51%) measured using a validated assessment using patient-reported outcomes (61%) within 3 months of initiation (59%), then quarterly (36%) |
| Morning joint stiffness (46%) measured using a validated assessment using patient-reported outcomes (50%) within 3 months of initiation (67%), then quarterly (32.5%) or every 6 months (38%) | |
| Free-text options | Pain: respondents recommended the VAS or 0-10 pain scale for validated assessment using patient-reported outcomes (50%) |
| Morning joint stiffness: respondents recommended the RAPID3 (67%) or PGA for validated assessment using patient-reported outcomes (30%) | |
| Monitoring frequency | Quarterly (37%) |
| Every 6 months (31%) | |
| Disease activity measure | |
| Specification | Preference |
| Documentation | Outcomes of disease activity assessment (75%) |
| Documentation that disease activity has been assessed (72%) | |
| Methods and frequency | Validated assessment using patient-reported outcomes (68%) measured within 3 months of initiation (58%), then quarterly (39%) |
| Flare occurrence/frequency (54%) measured within 3 months (57%), then quarterly (36%) | |
| Free-text options | Validated assessment using patient-reported outcomes: RAPID3 commonly recommended (31%) |
| Flare occurrence/frequency: patient reported in specified time frame (56%) | |
AE = adverse event; PDC = proportion of days covered; PGA = Patient Global Assessment; RAPID = Routine Assessment of Patient Index Data 3; VAS = Visual Analog Scale.
Two general measures that met consensus in round 1 were delineated to specific measures for the final measure set based on measure specification voting in round 2. The general measure “medication outcomes” was delineated to specific measures, including medication discontinuations, medication switches, medication persistence, common adverse effects (AEs), and serious AEs. Safety screening was delineated into 4 specific screening measures (tuberculosis [TB], hepatitis B virus [HBV], immunization, and drug-specific).
Round 3: Expert panelists affiliated with specialty pharmacies scored the remaining original and delineated measures (n = 13) that met consensus on feasibility.
FEASIBILITY OF COLLECTING AND REPORTING MEASURES
Measure feasibility scoring, current elements reported, and barriers/facilitators are summarized in Supplementary Table 3 (1.8MB, pdf) and detailed in Supplementary Exhibit 3 (1.8MB, pdf) . The final 13 measures were scored moderately/very feasible by over 70% of specialty pharmacy–affiliated stakeholders. (Figure 3) Across all consensus measures, adherence and medication outcomes (primarily discontinuations and serious AEs) were scored highest for collecting and reporting feasibility. In contrast, disease activity and patient functional status measures were scored lowest in feasibility for collecting and reporting. Many respondents seemed reliant on providers for these measures and as such, there were comments regarding the lack of availability and standardization of these measures in practice as well as questions on their clinical utility. Currently reported measure elements varied (Supplementary Table 3 (1.8MB, pdf) ).
FIGURE 3.
Feasibility Scoring Results
Results from specialty pharmacy–affiliated panelists (n = 51) scoring the feasibility of measures that reached consensus in rounds 1 and 2 are shown. The percentage of respondents who scored the measure as moderately/very feasible is represented by the bar/dot. Delineated medication outcome measures are shown in grey, delineated safety screening measures are shown in purple, and all other measures that met consensus are shown in navy.
HBV = hepatitis B virus; TB = tuberculosis.
COMMON COLLECTION AND REPORTING BARRIERS AND FACILITATORS ACROSS ALL MEASURES
Capability and opportunity consistently emerged as the most cited barriers to collecting and reporting data across all measure types, with respondents noting the lack of infrastructure or data access to capture certain measures effectively, as well as lack of provider buy-in. Conversely, need/motivation and opportunity were the most common facilitators, indicating that strong clinical or organizational incentives can help overcome existing barriers. Notably, opportunity emerged consistently as both a barrier and facilitator. These patterns held true across the consensus measures, with medication outcomes being an exception, in which respondents indicated capability as a stronger facilitator than opportunity.
FINAL CONSORTIUM STUDY GROUP VOTING RESULTS
Of the 13 measures that met consensus for inclusion and were deemed feasible by specialty pharmacy–affiliated stakeholders, 7 were voted to be core measures by over half of the Consortium study group: adherence (using any measure), serious AEs, patient response to therapy (using any measure), discontinuations, TB screening, HBV screening, and medication switches. The remaining 6 were deemed reach measures: immunization screening, drug-specific laboratory screening, common AEs, patient functional status (using any measure), patient disease activity (using any measure), and medication persistence (Figure 4).
FIGURE 4.
Study Group Categorization of Consensus Measures
Final voting on measures that met consensus from 18 of the Consortium study group members is shown. Members could vote for the measure to be considered core, reach, or neither. Darker cells indicate higher scores for that category. Final consensus results are shown on the right.
AE = adverse event; HBV = hepatitis B virus; TB = tuberculosis.
Discussion
This study represents the first systematic effort to establish consensus among specialty pharmacy stakeholders on measures that are important, usable, and feasible for specialty pharmacies to collect and report when managing patients with RA. Using a diverse panel of specialty pharmacy stakeholders, the modified Delphi process culminated in 13 measures, 7 deemed to be core and 6 reach. Specialty pharmacies that adopt the core measures could elevate clinical care provided to patients with RA and standardize outcome reporting for the industry. Additionally, results provide guidance to specialty pharmacy payers and manufacturers regarding important and usable practice measures that may be considered for reporting. The established core measures facilitate performance-based contracting, benchmarking across specialty pharmacies, and value assessment for high-cost therapies. The structured approach ensured that each measure was thoroughly evaluated, refined, and prioritized to achieve a robust set of consensus-based outcome measures for RA management in specialty pharmacy. Notably, the majority of consensus measures were rated as moderately/very feasible for specialty pharmacy implementation.
ADHERENCE AND MEDICATION OUTCOMES
Adherence and medication outcomes are fundamental metrics in specialty pharmacy, and both measures met consensus for inclusion in round 1 and scored as highly feasible. Specialty pharmacy accrediting organizations require specialty pharmacies to measure and trend adherence to ensure patients are taking medications appropriately and to promote pharmacy intervention if needed. Because most specialty pharmacies seek accreditation, accrediting organizations help shape which measures a pharmacy builds capabilities to collect and report on a semiregular basis. Accrediting organizations have not selected a single standard for measuring adherence but instead provide several options, including proportion of days covered, medication possession ratio, and patient-reported adherence.24 Claims-based calculations (proportion of days covered and medication possession ratio) evaluate medication-filling behavior but do not account for actual administration. Patient-reported adherence can be subjective or affected by recall bias. Because of the limitations to patient-reported and claims-calculated adherence measures, it is likely beneficial to measure adherence using both patient-reported and pharmacy claims calculations.
The general measure “medication outcome” met consensus for inclusion in the first round of voting and was further delineated based on measure preference voting in round 2. Accrediting bodies require specialty pharmacies to have a process to enroll and discharge patients in patient management programs. Medication discontinuations, which may be a reason for discharge, are required to be captured within the discharge summary. Other medication outcomes, including switching, common AEs, and medication persistence, are not noted in accreditation standards, which may have led to panelists scoring these as less feasible. Therefore, for external stakeholders (eg, payers/manufacturers), to gain these valuable reach metrics, it is likely necessary to incorporate them into reporting or contracting requirements to incentivize specialty pharmacy capability development.
PATIENT RESPONSE TO THERAPY
Measuring patient response to therapy also met consensus in the first round. Patient response to RA therapy may range from total remission to no response at all, and the time to achieve response is variable.3 Additionally, patients receiving antiTNF agents may have a secondary loss of response if they develop antibodies to the medication.25 Therefore, measuring patient response to RA treatment is important and usable information to optimize therapy and support patients to attain or regain remission. For managed care specifically, patient response to therapy can inform treatment value and avoid costly therapy switches or failed courses. One expert discussed that response to therapy data helps justify (to patients, prescribers, and payers) continuation of costly medications, reinforces medication adherence by demonstrating to patients the effectiveness of their treatments, and addresses pharmacy accreditation requirements.
In a recent survey, almost all (94%) health system specialty pharmacies reported tracking patient response to therapy and using disease state–specific activity measures.10 Because response to therapy can be measured in many appropriate ways, the final core measure was to evaluate any measure of response to therapy. Currently, no single patient-reported outcome disease activity measure can be recommended for RA.26 Overall, respondents indicated high feasibility of collecting and reporting any measure of patient response to therapy, though specific metrics to be used to evaluate response to therapy scored lower (Table 1). Although validated, standardized measures are ideal, they may not be feasible or clinically appropriate to be collected by specialty pharmacies. Therefore, the multidisciplinary team may consider implementing the disease activity measure that provides the most valuable information to their practice, is easiest to complete, and is patient-reported, prioritizing measures that have been validated.
SAFETY MONITORING
The survey identified barriers faced in documentation and distinguished the feasibility of collecting and reporting between serious and common AEs. Although both were considered moderate/very feasible to collect by most, the ability to report these events was lower, particularly for common AEs. Qualitative responses provided insight into this disparity, with respondents noting that reliance on patient self-reporting and difficulty in attributing AEs to medications posed significant challenges. Additionally, integrating the collection of common AEs into existing workflows was cited as a barrier. Serious AEs often require additional reporting to manufacturers or the US Food and Drug Administration (FDA) and are more likely attributable to the medication. Limited feasibility likely led the study group to vote for common AEs as a reach measure and serious AEs as a core measure. The distinction between serious vs common AEs provides a practical solution for payers by focusing mandatory reporting on high-risk, high-impact events that directly affect health care costs and utilization.
AE reporting is a key requirement for specialty pharmacy accreditation. Accreditation Commission for Health Care, Inc, standards mandate AE collection as a performance improvement metric, with established policies and procedures for documentation. URAC also requires 24/7 pharmacist availability via on-call systems and emphasizes AE identification and reporting to appropriate external databases. However, neither accreditation body differentiates between serious and common AEs. Although reporting AEs to the FDA’s MedWatch program remains voluntary, it is strongly encouraged as part of national safety surveillance efforts.27
SAFETY SCREENING
The expert panel reached consensus in round 1 on the importance and usability of safety screening measures. An expert panelist commented that the usability of safety screenings is an “important and common intervention point that can reduce complications and future care costs.” In round 3, collecting and reporting TB, HBV, and drug-specific laboratory monitoring scored higher than immunization screening. Many sites reported currently collecting TB and HBV screenings, but drug-specific laboratory monitoring and immunization screenings were lower (Supplementary Exhibit 2 (1.8MB, pdf) ). Feasibility likely drove the study group members to score only TB and HBV as core measures, and immunization screening and drug-specific laboratory screening as reach measures.
Because of their immunosuppressive nature, DMARDs used to treat RA have preventative safety requirements identified in the package insert, which are also required for some insurance prior authorizations. Despite the safety screening requirements and rating of high importance from our expert panel, there are still gaps in safety screening.28 Results from the current study demonstrate that safety screening should be a core measure for specialty pharmacies.
PATIENT FUNCTIONAL STATUS AND DISEASE ACTIVITY
Measuring patient functional status and disease activity met consensus for inclusion in round 2. Alhough these measures were valued, they scored lower than other included measures on feasibility, with less than 75% of respondents reporting them as moderately or very feasible to collect. Although pain and morning joint stiffness were the most recommended elements of functional status to be measured, only about half of the respondents agreed they should be used (Table 1). Low feasibility of collecting and reporting these clinical measures is likely multifactorial. The variety of tools available makes standardization challenging, and administering and documenting many of the tools, which often require clinician assessment, can be resource-intensive. Low feasibility of many tools and reliance on provider data present a critical barrier to specialty pharmacy accountability (particularly for non-HSSP pharmacies who have limited access to clinical information) and make cross-pharmacy benchmarking of these clinical outcomes unreliable for payers.
The American College of Rheumatology recommends collecting patient-reported functional status and disease activity assessment measures.29,30 The variety of validated measures may be a benefit for practitioners who can determine what works best in their specific practice yet creates challenges with standardization. In this study, the method most agreed on to measure functional status was pain (51%), whereas for disease activity, it was a validated assessment using patient-reported outcomes (68%) or flare occurrence (54%). The low agreement suggests no clear consensus on a single usable measure.
Specialty pharmacies may be reliant on providers to document these measures during routine clinic appointments. Rates of current elements collected or reported in this study ranged from 26% to 55% for functional status and from 12% to 53% for disease activity, highlighting significant gaps and variations in standardized data collection and reporting. Therefore, the final measure recommended for use from the current study is any measure of functional status and any measure of disease activity. This provides opportunity for practices/pharmacies to use what is best while also ensuring a measure is being collected.
IMPLEMENTATION CONSIDERATIONS
Because of competing health information technology priorities and feasibility challenges, successful implementation of these consensus measures likely requires accreditation, payer, or manufacturer uptake to secure the necessary electronic health record (EHR)/workflow resources and leadership support at the specialty pharmacy level. Successful implementation of prioritized measures into the EHR and standard workflows hinges on a strategic, multipronged approach that integrates standardized documentation to capture and share quality data, coupled with strong stakeholder and leadership support. Understanding available data sources is essential: some data points may be automated for reporting, whereas others require standardized forms within the EHR for discrete reporting. Ideally, outcomes should be tracked dynamically using a dashboard to visualize success over time. A previous study details how one site pioneered clinic-based pharmacist outcome identification through the modified Delphi methodology.22 Authors highlighted the importance of providing tailored approaches for each clinic, using a multidisciplinary stakeholder meeting, establishing clear measurement parameters, and capitalizing on existing dashboard metrics to facilitate successful implementation. Leadership support is essential to secure resources, modify workflows, and sustain practice changes when new outcomes are prioritized for collection. Expanding implementation strategies from individual clinic-based pharmacist practices to multiple sites introduces distinct challenges, particularly related to EHR variability and workflow differences. Proactively addressing these challenges is crucial to enabling consistent data tracking over time.
FUTURE DIRECTIONS
Advancing implementation efforts will require strategic collaboration of specialty pharmacy with accreditors, manufacturers, and payers to align outcomes data with accreditation and reimbursement objectives. Leveraging performance improvement teams and implementation science frameworks can promote sustainability through iterative cycles such as plan-do-study-act.31 Specialty pharmacy stakeholders should collaborate to implement standardized, consensus-based measures that generate meaningful, actionable data. This ensures data are used to drive improvements rather than collected for its own sake. Additionally, accreditors, payers, and manufacturers should use the 7 core consensus measures from this study as the baseline for all specialty pharmacy performance evaluation and reimbursement programs.
LIMITATIONS
The current study was limited by a small sample size of certain expert panelist types, though all panelist types were represented. The largest response rates being from specialty pharmacies is appropriate, but additional expert panelists from other areas would have been preferable. The low response rate from payers was likely due to study group members having fewer contacts from this stakeholder group and is a limitation as payers are primary consumers of pharmacy performance data. Lack of or varying familiarity with the assessments or proposed measures by some stakeholders could have contributed to variability in measure scoring. Nonspecialty pharmacy–affiliated stakeholders may not completely understand the role or scope of specialty pharmacists. Consensus scoring thresholds were defined in rounds 1 and 2, but feasibility scores and study group voting thresholds for the final measure set were not clearly defined a priori. This is an aspect of the modified Delphi method, but it did lead to some subjectivity when determining how to proceed with certain measures. After round 1, it was noted that expert panelists were often using the terminology of usability and feasibility interchangeably. This was corrected by adding usability and feasibility to the survey definitions and attaching a copy within the round 2 survey e-mail. Finally, the length of the survey could have been a deterrent from participating or sustaining participation.
Conclusions
This modified Delphi study successfully identified 13 measures considered to be important, usable, and feasible for specialty pharmacies to collect and report when managing patients with RA. The 7 core measures identified in this study provide a standardized, actionable framework for specialty pharmacies to promote optimal patient care and for payers, accreditors, and manufacturers to evaluate specialty pharmacy performance and accountability for high-cost RA therapies.
Disclosures
Dr Zuckerman reports research support paid to the institution unrelated to this work from the following companies in the past 36 months: AstraZeneca, Beigene, Pfizer, Sanofi, and UCB.
The publication described was supported by CTSA awards No. UL1 TR002243 and UL1 TR000445 from the National Center for Advancing Translational Sciences. Its contents are solely the responsibility of the authors and do not necessarily represent official views of the National Center for Advancing Translational Sciences or the National Institutes of Health.
Acknowledgments
The authors would like to acknowledge the nonauthor Vanderbilt Health System Specialty Pharmacy Outcomes Research Consortium RA Delphi study group members that assisted with the environmental scan and final measure voting: Amanuel Kehasse, Kristel Geyer, Jessica Mourani, Alicia Zagel, Kama Thomas, Heather Dalton, Jennifer Loucks, Paul Groehler, Jayson Verdick, Sarah McGill, Leonard Petrik, Karen Kelley, Jamie Pond, Agnes Cha, Stefanie Cisek, Alexander Philbrick, Amy Wenker, Logan Murray, Sheena Babin, Lauren Laforge, Katharine McCarthy, Jennifer Donovan, Melissa Ortega, Erica Diamantides, Meagan Fowler, Christian Rhudy, Courtney Queen, Wendy Ramey, Rushabh Shah, Chelsea Maier, and Julia Pate.
The authors would like to acknowledge the experts and/or their companies that participated in this Delphi process (only those that have agreed to share their name/company are represented): Adenrele Olajide, MBBS, Allison Trawinski, PharmD, MBA, Amber Oakes Hudgins, PharmD, CSP, CPP, Amber Rollo, PharmD, CSP, AAHIVP, PharmD, Ann McNamara, PharmD, Bernice Y. Man, PharmD, BCPS, CPHQ, CSP, DPLA, Elizabeth Ratz, CPhT, Brandon Hardin, PharmD, MBA, CSP, Breanna Radel, BSN, RN, Brett Glasheen, PharmD, Bridget Lynch, PharmD, MS, Cassandra Ramel, PharmD, Chelsey McPheeters, PharmD, BCPS, BCACP, Christen Tatum, PharmD, MHA, MSL, CSP, AAHIVP, Christopher Palma, MD, ScM, Clay Hayes, PharmD, BCPS, Courtney Wakefield, CPhT, Danielle Gatti-Palumbo, PharmD, BCACP, David Mitchell, PharmD, MBA, CSP, FCPhA, Diane Lewis Horowitz, MD, Drew Paszotta, PharmD, Elizabeth Costanzo, PharmD, BCPS, Elizabeth Kirchner, DNP, Elizabeth Rightmier, PharmD, BCPS, Erin Hamai Tom, PharmD, APh, MBA, BCACP, CSP, Fiona Bush, PharmD, Isabelle To, PharmD, Jacqueline Varnyan, PharmD, CSP, Jessica B. Michaud, PharmD, BCPS, Jessica A. Walsh, MD, Jessica Wedel, PharmD, CSP, Johnson Ching, PharmD, CSP, Jonna Zelie, PA-C, Julie Snyder, CPhT, Karen Maillet, RN, Katie Jones, PharmD, Kaitlin Ciaramitaro, PharmD, MHA, CSP, Kelsey Hennig, PharmD, BCPS, Kevin Byram, MD, Kristine Keaton, PharmD, BCACP, Laura Jester, PharmD, CSP, Lindsey MacQueen, PharmD, Marsha S. Jackson, RN, BS, Michelle Hoagland, CPhT, Michelle Schlief, CPhT, Narender Annapureddy, MD, MS, Nina Chhabra, PharmD, BCPS, CSP, Norman Madsen, MD, MSc, FACR, RhMSUS, Olivia Brown, PharmD, Orgesa Cepo, PharmD, Drew Paszotta, PharmD, Rebecca S. Overbury, MD, MS, Rebecca Freedman, PharmD, Robert W. Corty, MD, PhD, Russel Burge, PhD, S. Bobo Tanner, MD, Steven C. Vlad, MD, PhD, Thuyvan Phan, PharmD, BCPS, CPP, Timothy M. Hinson, PharmD, Trisha A. Elkins, PharmD, Olutuminiun Osunsanmi, PharmD, Tyler Reese, MD, Uche Ndefo, PharmD, Wendy Ramey, BSPharm, RPh, CSP, Amy Wenker, PharmD, MBA, Matt Nguyen, PharmD, Eleni Theodoropoulos, MBA, CPHIMS, Aaron Woodman, PharmD, RPH, Alicia Verret, PharmD, CSP, and Rachael Smith, PharmD, BCACP.
References
- 1.Foundation A . Rheumatoid arthritis: Causes, symptoms, treatments and more. 2021; https://www.arthritis.org/diseases/rheumatoid-arthritis. Accessed March 25, 2025.
- 2.Krati C, Jandu JS, Brent LH, Al-Dhahir MA. Rheumatoid Arthritis. StatPearls Publishing; 2025. [PubMed]
- 3.Fraenkel L, Bathon JM, England BR, et al. 2021 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol. 2021;73(7):1108-23. doi: 10.1002/art.41752 [DOI] [PubMed] [Google Scholar]
- 4.Alqahtani ZA, Yue X, Guo JJ. Utilization, expenditure, and price trends of nonbiologic and biologic disease-modifying antirheumatic drugs in the US Medicaid programs: An empirical data analysis of over three decades. J Pharm BioTech Ind. 2025;2(2):7. doi: 10.3390/jpbi2020007 [DOI] [Google Scholar]
- 5.NASP . NASP Definitions of specialty pharmacy and specialty medications. 2016. Accessed November 15, 2019. https://naspnet.org/wp-content/uploads/2017/02/NASP-Defintions-final-2.16.pdf
- 6.Fein AJ. The top 15 specialty pharmacies of 2024. 2025. Accessed November 7, 2025. https://www.drugchannels.net/2025/04/the-top-15-specialty-pharmacies-of-2024.html
- 7.Looney B, Crumb J, White S, et al. Financial impact of integrated specialty pharmacy efforts to avoid oral anticancer medication waste. J Manag Care Spec Pharm. 2024;30(5):465-74. doi: 10.18553/jmcp.2024.30.5.465 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zuckerman AD, Carver A, Cooper K, et al. An integrated Health-System Specialty Pharmacy model for coordinating transitions of care: Specialty medication challenges and specialty pharmacist opportunities. Pharmacy (Basel). 2019;7(4):163. doi: 10.3390/pharmacy7040163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.O’Reilly E, Foushee S, Nolan L, Maier C, McPheeters C. Evaluation of the impact of a health-system specialty pharmacy patient management program on achieving predefined clinical outcome measures. Am J Health Syst Pharm. 2024;81(9):e226-33. doi: 10.1093/ajhp/zxad341 [DOI] [PubMed] [Google Scholar]
- 10.Zuckerman AD, Mourani J, Smith A, et al. 2022 ASHP Survey of Health-System Specialty Pharmacy practice: Clinical services. Am J Health Syst Pharm. 2023;80(13):827-41. doi: 10.1093/ajhp/zxad064 [DOI] [PubMed] [Google Scholar]
- 11.URAC . Specialty pharmacy services accreditation. 2025. Accessed March 5, 2025. http://blog.urac.org/how-pharmacy-accredita-tion-can-boost-patient-outcomes
- 12.ACHC . Pharmacy: Supporting your success. 2025. Accessed November 24, 2025. https://achc.org/specialty/
- 13.ASHP . ASHP Specialty Pharmacy Resource Guide. 2015. Accessed November 24, 2025.
- 14.Patel K, Chim YL, Grant J, Wascher M, Nathanson A, Canfield S. Development and implementation of clinical outcome measures for automated collection within specialty pharmacy practice. J Manag Care Spec Pharm. 2020;26(7):901-9. doi: 10.18553/jmcp.2020.26.7.901 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jolly JA, Pierson TJ, Pulvermacher A. Turning differentiation into value: How health-system specialty pharmacies can align to an outcomes-focused mission. Am J Health Syst Pharm. 2021;78(11):1009-14. doi: 10.1093/ajhp/zxab087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zuckerman AD, Whelchel K, Kozlicki M, et al. Health-system specialty pharmacy role and outcomes: A review of current literature. Am J Health Syst Pharm. 2022;79(21):1906-1918. doi: 10.1093/ajhp/zxac212 [DOI] [PubMed] [Google Scholar]
- 17.McMillan SS, King M, Tully MP. How to use the nominal group and Delphi techniques. Int J Clin Pharm. 2016;38(3):655-62. doi: 10.1007/s11096-016-0257-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Olsen AA, Wolcott MD, Haines ST, Janke KK, McLaughlin JE. How to use the Delphi method to aid in decision making and build consensus in pharmacy education. Curr Pharm Teach Learn. 2021;13(10):1376-85. doi: 10.1016/j.cptl.2021.07.018 [DOI] [PubMed] [Google Scholar]
- 19.Schifano J, Niederberger M. How Delphi studies in the health sciences find consensus: A scoping review. Syst Rev. 2025;14(1):14. doi: 10.1186/s13643-024-02738-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research Electronic Data Capture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Harris PA, Taylor R, Minor BL, et al. ; REDCap Consortium. The REDCap consortium: Building an international community of software platform partners. J Biomed Inform. 2019;95:103208. doi: 10.1016/j.jbi.2019.103208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Summerlin CM, Vest MH, Valgus JM, Deyo ZM, Alexander MD, Waldron KM. Development and implementation of a standardized process for identifying ambulatory pharmacy clinical outcomes measures. Am J Health Syst Pharm. 2023;80(3):148-58. doi: 10.1093/ajhp/zxac301 [DOI] [PubMed] [Google Scholar]
- 23.National Quality Forum . Measure evaluation criteria and guidance for evaluating measures for endorsement. Washington, DC: National Quality Forum; 2021. [Google Scholar]
- 24.Canfield SL, Zuckerman A, Anguiano RH, et al. Navigating the Wild West of Medication Adherence Reporting in Specialty Pharmacy. J Manag Care Spec Pharm. 2019;25(10):1073-1077. doi: 10.18553/jmcp.2019.25.10.1073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Aletaha D, Smolen JS. Diagnosis and management of rheumatoid arthritis: A review. JAMA. 2018;320(13):1360-72. doi: 10.1001/jama.2018.13103 [DOI] [PubMed] [Google Scholar]
- 26.Pickles T, Macefield R, Aiyegbusi OL, et al. Patient reported outcome measures for rheumatoid arthritis disease activity: A systematic review following COSMIN guidelines. RMD Open. 2022;8(1):e002093. doi: 10.1136/rmdopen-2021-002093 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.US Food and Drug Administration . Reporting serious problems to FDA. 2024. Accessed March 12, 2025. https://www.fda.gov/safety/medwatch-fda-safety-information-and-adverse-event-reporting-program/reporting-serious-problems-fda
- 28.Patterson S, Schmajuk G, Evans M, et al. Gaps in ambulatory patient safety for immunosuppressive specialty medications. Jt Comm J Qual Patient Saf. 2019;45(5):348-57. doi: 10.1016/j.jcjq.2018.12.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Barber CEH, Zell J, Yazdany J, et al. 2019 American College of Rheumatology recommended patient-reported functional status assessment measures in rheumatoid arthritis. Arthritis Care Res (Hoboken). 2019;71(12):1531-9. doi: 10.1002/acr.24040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.England BR, Tiong BK, Bergman MJ, et al. 2019 Update of the American College of Rheumatology recommended rheumatoid arthritis disease activity measures. Arthritis Care Res (Hoboken). 2019;71(12):1540-55. doi: 10.1002/acr.24042 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Leis JA, Shojania KG. A primer on PDSA: Executing plan-do-study-act cycles in practice, not just in name. BMJ Qual Saf. 2017;26(7):572-7. doi: 10.1136/bmjqs-2016-006245 [DOI] [PubMed] [Google Scholar]