How to build a clinical outcome assessment (COA)
In the modern healthcare environment, data is necessary for assessing efficacy of a new treatment, facilitating market access and reimbursement, and ultimately, improving patient care. Regardless of care model, therapy area, or patient makeup, healthcare professionals rely on data to eliminate uncertainty and generate more consistent (and desirable) care outcomes.
Clinical outcome assessments (COAs) are essential tools in this process, providing a structured way to evaluate how a patient feels, functions, or survives as a result of treatment. In this how-to guide, we’ll explore what a clinical outcome assessment is, the methods used in these assessments, and the factors that determine whether you should build a new COA or adapt an existing one to most effectively assess the efficacy of your drug or treatment. Finally, we’ll walk through how to properly build a COA.
What is a clinical outcome assessment?
A clinical outcome assessment is a measure used in clinical trials and healthcare settings to evaluate the effects of a medical treatment or intervention on patients. These assessments help providers understand the impact of a treatment on various aspects of a patient’s health, including symptoms, overall well-being, and daily functioning.
COAs deliver critical insights into the efficacy of interventions and are often used as a data source to support regulatory approval, guide clinical decision-making, and inform patient care.
There are four primary types of COAs:
- Patient-reported outcomes (PROs) are reports directly from patients about how they feel or function in relation to a health condition or its treatment, without interpretation by clinicians or any other healthcare professional.
- Clinician-reported outcomes (ClinROs) are assessments made by healthcare professionals based on their observations of a patient’s condition.
- Observer-reported outcomes (ObsROs) are measures based on observations made by someone other than the patient or healthcare professional, such as a caregiver or family member. These are usually used when the patient is unable to self-report, such as in pediatric or cognitively impaired patients.
- Performance outcomes (PerfOs) are objective measures of a patient’s ability to perform tasks or activities, such as timed walking tests or cognitive function tests.
What data-collection methods are used for COAs?
Every patient is unique, and likewise, every COA will vary somewhat in its methods of data collection depending on its specific goals.
Questionnaires and surveys are the most common methods for collecting PROs, for instance, as they give patients the opportunity to share specific aspects of their experience, such as pain, physical function, or mental health. They’re also easily administered via traditional pen and paper, electronic/digital form, or mobile apps.
ClinROs, however, are more likely to involve clinical evaluations, in which a healthcare professional assesses a patient based on predefined criteria. These evaluations can include physician examinations, lab tests, and imaging studies.
Behavioral observations from caregivers or family members are the primary source of data for ObsROs, while performance assays like the 6-minute walk test, grip strength test, or cognitive function test are usually used for PerfO reports.
Should I build a new COA or adapt an existing one?
There are advantages and disadvantages to both creating a COA from scratch and adapting an existing assessment. The right choice for you will depend on your needs, available resources, and ultimate goals.
Building a new COA allows you to design an assessment that is perfectly aligned with the specific goals of your trial or treatment plan. A custom-built COA can incorporate questions and metrics most relevant to your unique patient population or treatment, while also including the latest scientific insights, methodologies, and technologies.
However, developing a COA from scratch is time-consuming and requires extensive research, validation, and testing. The process will also take significant resources and expertise around psychometrics, clinical research, and the regulatory environment—expertise you may need to outsource or hire for. Navigating the regulatory process can be especially challenging for less experienced healthcare professionals and can quickly add to the time and cost associated with COA development.
Adapting an existing COA, on the other hand, will likely save time and resources: The assessment groundwork is already laid, regulatory hurdles have already been cleared, and a COA that has already been used in clinical trials will almost certainly be accepted as valid by other members of the medical community.
Of course, existing COAs may not align perfectly with the needs of your study or patient base, and customization could be time-consuming or impossible, depending on the COA and your application. Some adaptations could even render the COA invalid. Certain existing COAs could be built on outdated methodologies or technologies and fail to capture the full scope of a treatment or patient population.
A step-by-step guide to building a clinical outcome assessment
Whether you’re building a totally new COA or adapting an existing one, you’ll want to take a systematic approach to ensure your assessment is reliable, valid, and suited to your needs. Here’s a high-level walkthrough of the process:
1. Define the objectives
As with any assessment, you’ll need to define the objectives of your COA. What outcomes are you trying to measure? Who is your target patient population and what are the symptoms or disfunctions they want relieved? Will your assessment be used in clinical trials, routine inpatient care, or by patients and caregivers at home? Answering these questions will guide the rest of the process.
2. Review existing literature
You need to understand the current state of COAs related to your objectives. A literature review will help you identify possible assessments for adaptation, determine best practices for your own COA development, and find gaps in datasets and other information sources that you might address. After this step, you should have a good idea of whether a new or existing COA is best suited to support your needs.
3. Engage stakeholders
To encourage participation and ensure best outcomes, you should engage stakeholders in the COA development process as early as possible. Include patients, fellow clinicians, researchers, and regulatory contacts in the development process and seek their insights early and often. Their input will keep your COA relevant, patient-centered, and compliant with regulatory concerns. You might consider forming a steering committee or advisory board to guide the process forward.
4. Develop the conceptual framework
At this point, you should outline the key concepts that the COA will measure and identify the relationships between those concepts. This framework will shape the assessment and keep it on track according to your goals. For example, if you’re working on a COA for a new treatment for lung disease, your framework might include concepts like level of breathing difficulty, the relationship between shortness of breath and daily activities, and the impact of breathing difficulties on a patients mental and emotional health.
5. Define questions and measures for inclusion
Using your conceptual framework as a guide, begin to develop questions and measures (or select from existing options) to include in your COA. Pay close attention to details like response scales, question wording and clarity, potential bias in your language, and any other factor that could impact how a subject responds. Use the process of linguistic validation to ensure translations are accurate. Consider conducting interviews with patients or other subjects to refine your measures and make them as clear as possible.
6. Begin pilot testing
With your conceptual framework established and questions and measures written out, it’s time to test your COA with a small sample of your target population. This should reveal potential issues with the measures, response scales, and methods of administration. Use preliminary data and subject feedback to make adjustments and align the COA to your objectives and patient needs.
7. Expand testing and obtain regulatory approval
After making adjustments, you should test your COA with a larger sample to confirm its reliability, validity, and responsiveness. Important factors for assessment include internal consistency, test-retest reliability, and criterion validity. If the COA will be used in clinical trials or for other regulatory purposes, you may need to submit it for review with the U.S. Food and Drug Administration or another regulatory agency.
8. Implement the COA and keep monitoring
Once the COA is developed, validated, and approved for use, you can start to implement it in your practice or study. Continuously monitor its usage and collect feedback from users to refine and improve the assessment over time.
Learn more
Clinical outcome assessments are critical components of modern healthcare, enabling healthcare professionals and treatment developers to make data-driven decisions about medical treatments that can improve care outcomes and change patients’ lives.
If you’re building a new COA or adapting an existing one, you’ll want to leverage a range of healthcare intelligence to plan the assessment, engage stakeholders, find patients, and validate outcomes.
The Definitive Healthcare platform delivers powerful insights and analytics built on procedure and diagnosis claims, reference and affiliation data, population intelligence, and more. Sign up for a demo today to see how our platform can support you throughout the COA development process.