The need to measure our health
It is a very current topic in health care: Value Based Health Care.
Value Based Health Care was introduced by Michael Porter and Elizabeth Teisberg in 2006 with their book Redefining Health Care . In this book they lay out a framework for organizing health care around the one thing that matters most: the health outcomes delivered to patients.
One of the key issues that Porter and Teisberg address is the absence of insight into these outcomes of care. Knowing what health benefits a certain treatment has or hasn’t had for patients is an essential first step in delivering good quality care. That’s why one of the six core components of Value Based Health Care is the measurement of health outcomes, as shown in Figure 1 (adapted from ):
Figure 1 The measurement of health outcomes is one of six components of the Value Agenda. Source: The Strategy That Will Fix Health Care, Michael E. Porter and Thomas H. Lee, MD, Harvard Business Review October 2013 .
And that's where PROMs, or Patient Reported Outcomes Measures, come in.
In a series of blog posts I will take you through the history, science, applications and future development of PROMs. This series aims to address the question: what is needed if you want to turn health outcomes into numbers?
This is the first part of this series. So let’s start with defining what we are talking about:
What are PROs and PROMs?
To define PROMs, we first need to define what is meant with a Patient Reported Outcome, or PRO? According to the FDA guidelines from 2009 , a PRO is the following:
PRO - Patient Reported Outcome
“A PRO is any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else. The outcome can be measured in absolute terms (e.g., severity of a symptom, sign, or state of a disease) or as a change from a previous measure.”
So, it is a status of a health condition. But what is a health condition? What encompasses health? It is quite complex to position a concept like status of a health condition in the overall context of health care and health related quality of life. What is included in this, and what is not? In 2000, the FDA had introduced PROs as an umbrella term including ‘Health related quality of life, satisfaction with treatment, performance measures for productivity assessment, and any other patient-based assessment of health status, well-being, or functional status’ . This encompasses a lot. One could argue it even contains a bit too much: it is widely debated whether or not patient satisfaction with treatment is a ‘real’ PRO.
In short, there is no general consensus on what are, and aren’t, PROs.
This is also illustrated by the many different models with which these concepts and their interrelation have been represented in the literature, a selection of which is shown in Figure 2:
Figure 2 A selection of the various models found with which the interrelations of various health concepts are conceptualized [5-7].
A model which clearly defines PROs can be found in the ‘Health related quality of life conceptual model’ as described in Wilson and Cleary’s 1995 JAMA publication . They developed this model to address the ‘lack of conceptual models that specify how different types of patient outcomes measures interrelate’. A (visually adapted) version of their model is shown in Figure 3:
Figure 3 Wilson and Cleary’s model for Patient Reported Outcomes (top row, visually adapted from original) and PRO examples (bottom row). Source: Wilson IB, Cleary PD. Linking Clinical Variables With Health-Related Quality of Life: A Conceptual Model of Patient Outcomes. JAMA. 1995;273(1):59–65 .
The top row illustrates the model and it starts on the far left with a patients’ biological and physiological status. This status is linked to the patients’ symptoms, and from there, (a combination of) symptoms lead to the patients’ functional status and subsequently his/her health-related quality of life (HRQL). In combination with non-medical factors, you then get the patients’ overall quality of life (QoL), as shown on the far right.
In this model, PROs are limited to the three green boxes in the middle: symptoms, functional status, and health related quality of life. The bottom row indicates examples of PROs for each of these three categories, such as ‘fatigue’ as a symptom example, and ‘dressing’ as an example of functional status.
Now that we have PROs defined, what are PROMs?
In short: PROMs are the tools or instruments with which you turn qualitative health outcomes into quantitative data.
Figure 4 PROs can be translated into numeric values using PROMs
The official definition of a PROM is as follows (again, according to the FDA guidelines, but note that the FDA does not use the word “PROM”, but uses “PRO instrument” instead):
PROM - Patient Reported Outcome Measure
“A means to capture data (i.e., a questionnaire) plus all the information and documentation that supports its use. Generally, that includes clearly defined methods and instructions for administration or responding, a standard format for data collection, and well-documented methods for scoring, analysis, and interpretation of results in the target patient population.”
It is important to keep in mind that a PROM can be any “means to capture data”. It can thus be any method with which the patient’s input is obtained, from diaries and event logs using free text input, to health tracker devices, and from single item categorical scales to multi-item-multi-domain questionnaires (<-- an ‘item’ is a single question, and a ‘domain’ is a certain outcome, such as fatigue.)
However, the field is currently most focused on questionnaires. For the remainder of this blog series, when I speak about PROMs, I mean PROM questionnaires, or surveys.
Different PROM types
Connecting PROMs back to Wilson and Cleary’s model, you can divide types of PROMs broadly into three categories: Domain specific, Disease specific, and Generic PROMs, as shown below:
Figure 6 PROM examples and PROM types added to Wilson and Cleary’s model of patient outcomes.
Domain specific PROMs cover a specific symptom and/or functional status. For example, the Neuro-QOL Fatigue Short Form. This questionnaire consists of 8 items (questions) covering a single domain (fatigue). Another example is NOSE, or Nasal Obstruction and Septoplasty Effectiveness Scale, which has 5 items and assesses the functional domains sleeping, breathing, and nasal function.
Disease specific PROMs are developed to measure health related quality of life in patients with a certain condition. A commonly used example is the EORTC-QLQ-C30, a 30 item questionnaire for cancer patients, which assesses their overall global health status as well as 5 functional domains, and 9 symptom domains. And the HOOS, or Hip Disability and Osteoarthritis Outcome Score, which assesses 5 domains (pain, symptoms, activity of daily living, sport and recreation function and hip related quality of life) using 40 items.
And Generic PROMs assess the general impact of health on everyday life. For example, the Whodas, a 12 or 36 item questionnaire which can be used across all diseases and assesses the following 6 domains: cognition, mobility, self-care, getting along, life activities and participation.
It becomes evident that the type of PROM you use depends on what you want to measure. You cannot study planets with an electron microscope, nor use a regular ruler to measure the size of cells. The exact same thing applies to PROMs, and this becomes clear when you think of PROMs as having a certain 'resolution' (a level of depth, or precision), as well as a certain 'field of view' (breadth, or comprehensiveness).
Broadly speaking: the 'resolution' of a PROM is determined by the number of items covering a single domain. And its 'field of view' is determined by the total number of domains that are interrogated. By organizing PROMs on a 'resolution' and 'field of view' axis, the three (overlapping) PROM categories become apparent, as is illustrated in the next figure:
Figure 7 When organized along a Depth (precision of the tool) and Breadth (comprehensiveness) axis, the three PROM types (Domain, Disease, and Generic) can be coarsely divided as shown above. From a single item pain scale (top left) to the very precise 42 item Obsessive-Compulsive Inventory (OCI) (bottom left) to a very broad, extensive generic health related quality of life assessment with the SF-36 (far right).
And with that, we've reached the end of this first post about the world of PROMs. I hope it has been able to give you some context and clarity.
To further understand these questionnaires well, it is worth diving into their history. That will be addressed in the next post. See you there!
- Michael E. Porter and Elizabeth O. Teisberg. Redefining Health Care: Creating Value-Based Competition on Results. Boston: Harvard Business School Press, 2006
- Michael E. Porter and Thomas H. Lee, MD. The Strategy That Will Fix Health Care. Harvard Business Review October 2013
- Guidance for Industry. Patient-Reported Outcome Measures: Use in Medical Product Development to Support Labeling Claims. U.S. Department of Health and Human Services, Food and Drug Administration, December 2009
- Chassany, Sagnier, Marquis, Fullerton, and Aaronson for the ERIQA Group. Patient-Reported Outcomes: the example of health-related quality of life - a European guidance document for the improved integration of Health-Related Quality of Life assessment in the drug regulatory process. Drug Information Journal, Vol. 36, pp. 209–238, 2002
- Cohen RM, Greenberg JM, IsHak WW. Incorporating Multidimensional Patient-Reported Outcomes of Symptom Severity, Functioning, and Quality of Life in the Individual Burden of Illness Index for Depression to Measure Treatment Impact and Recovery in MDD. JAMA Psychiatry. 2013;70(3):343–350.
- Kluetz PG et al. Focusing on Core Patient-Reported Outcomes in Cancer Clinical Trials: Symptomatic Adverse Events, Physical Function, and Disease-Related Symptoms. Clin Cancer Res 2016; 22:1553-1558.
- McKenna, S.P. Measuring patient-reported outcomes: moving beyond misplaced common sense to hard science. BMC Med 9, 86 (2011).
- Wilson IB, Cleary PD. Linking Clinical Variables With Health-Related Quality of Life: A Conceptual Model of Patient Outcomes. JAMA. 1995;273(1):59–65.
About the author
Sara Maria Sprinkhuizen, PhD
I am a physicist who fell in love with MRI scanners, which launched my path into health care. After finalizing my PhD in MRI physics (Utrecht University, the Netherlands) I moved to Boston for a post-doc (Harvard Medical School, USA). I then decided to explore the medical field from a system-level perspective and joined ICHOM (Boston, USA). At ICHOM I guided working groups of physicians, patient representatives and registry leaders through the process of defining health outcomes of most importance for patients. The last years I have worked as a data analyst and visualizer, getting a close look at the daily inner workings of hospitals and the health care industry. With my health care, data analytics, and visualization experience, I provide human centered and tailor-made data support for health care organizations and scientists.