New Instruments: Patient-Reported Health Outcomes
Understanding a patient's perception of his quality of life may play an important role in evaluating AMD treatment options.
BY SANJAY SHARMA, MD, FRCS, MSC (EPID), MBA

Clinicians have a broad array of methods for defining outcomes for any given ocular disease treatment. We're all familiar with the traditional methods used to quantify clinical outcomes in age-related macular degeneration (AMD) treatment: Visual acuity on a Snellen or an ETDRS chart, fluid leakage on fluorescein and ICG angiography and retinal thickening on optical coherence tomography (OCT).

Increasingly, researchers and clinicians are adding a new dimension to understanding the impact of AMD. Patient-reported health outcomes (PRO) is a discipline aimed at exploring the effect of disease on quality of life, taken from the patient's perspective. PRO research typically employs both health-related quality-of-life (HRQoL) questionnaires and utilities. These techniques can also be used to measure any potential changes that treatments may have on health-related quality of life. In this article, I'll describe PRO instruments and how their use may impact future treatment decisions.

HRQoL QUESTIONNAIRES

Health-related quality-of-life questionnaires can be generic in nature, function-specific or disease-specific. Examples of questionnaires that have been used in ophthalmology include the SF-36, a generic index, the Visual Function Index (VF-14), a vision specific instrument, and the RetDQoL, a condition-specific questionnaire.

A vision-specific questionnaire such as the National Eye Institute Visual Function Questionnaire 25 (NEI-VFQ-25) or the VF-14 includes a series of questions that can be grouped into different domains. The NEI-VFQ, for instance, asks questions that are pertinent not only to how patients are functioning with respect to standard vision-dependent tasks, such as reading and driving, but also regarding their emotional well-being and social functioning.

TRADE-OFFS AND GAMBLES

While HRQoL questionnaires provide valuable information regarding the ability of AMD patients to perform specific tasks, there are other techniques that can be used to assess how patients value their vision. To understand patients' preferences and how they view their disease, we can use a quantitative method, such as the time trade-off technique (TTO) or the standard reference gamble (SRG).¹

With the TTO, we ask patients to assume they'll live a hypothetical but finite period of time, say another 10 years. The patient then is asked how many years of life he'd be willing to trade off to eliminate his condition.

The standard reference gamble is derived from economic literature and was developed to value the outcomes of potential decisions and to ultimately guide decision-making when significant uncertainty exists.² Today, it is used across many disciplines, including business, defense and healthcare. To quantify how patients value their vision using the SRG, a framework similar to the TTO is employed. In this scenario, patients are offered a gamble; either they can accept their AMD condition as it is, or they can gamble for a chance to eliminate it. To do so, they are offered a hypothetical treatment that, when it works, works perfectly and eliminates their disease. But when it doesn't work, the patient dies. What changes in the various iterations of this test is the probability of death; it can range from 0% to 100%. Patients are questioned to the point that they are indifferent between their present state and opting for the treatment that has a specified complication rate.

My colleagues and I have talked to hundreds of people using these assessment tools. Essentially, we've found the average AMD patient would take a 30% reduction in length of life to be cured using the TTO. Those who were bilaterally blind were willing to trade off more than 50% of their remaining life. AMD patients responding to the SRG would accept a 15% to 25% risk of immediate death for a procedure that could effect a cure.³

CLINICIANS' PERCEPTIONS

In another interesting study, we applied the same sort of framework to interview a group of ophthalmologists.⁴ We found a significant disconcordance between what patients were telling us and what the clinicians were telling us. The clinicians dramatically underestimated how macular degeneration reduces a patient's quality of life.

I believe this disconcordance will decrease as more papers are published and clinicians become more educated as to what outcomes research really is. In addition, and perhaps more importantly, I believe that quality-of-life information will serve as another filter for clinicians to understand emerging treatments for AMD.

For example, we may have a treatment that reduces the risk of visual loss in terms of three lines of Snellen acuity or doubling the visual angle, but how does that really impact the patient? That's what clinicians are starting to ask themselves. Will the treatment I'm offering -- not just for macular degeneration but for anything -- really improve my patient's quality of life?

IMPACT ON TREATMENT DECISIONS

The primary statistic driving most treatment decisions in clinical practice is a change in visual acuity. While most clinical trials are powered around detecting changes in visual acuity, more and more, we are seeing investigators adding HRQoL measures into their protocols. This is true not only of NEI-sponsored trials but for the vast majority of industry-sponsored ones, too.

This will influence treatment in several ways. Let's say we have four or five new medicines or treatments for macular degeneration that will emerge over the coming years. If one has a specific impact on quality of life, that certainly will be a differentiating factor in the eyes of clinicians and patients.

The concept of one treatment for everyone in a specific disease state is now becoming a thing of the past. We are seeing it all across medicine. Different treatments work better for different stages of diseases. In terms of AMD, if we have different types of treatments that affect different strata of the disease or if they affect quality of life in different ways, clinicians may be able to make a more highly individualized treatment decision for a particular patient.

For instance, let's say reading is very valuable to your patient and let's say one particular treatment is proved much more beneficial in terms of ability to improve reading as measured through either a HRQoL questionnaire or reading speed. That one treatment may be the best modality available for that patient.

However, another treatment may impact another form of quality of life, which may be more appropriate for yet another patient. Thus, we may be able to offer a more targeted solution based on a patient's quality-of-life preferences.

BUILDING CONFIDENCE

Any time there's something new, there's always skepticism, which I think is completely appropriate. A lot of the quality-of-life methodology was derived from the psychology and education literature. But is it reliable and scientifically valid? Absolutely.

Gary Brown, MD, MBA, Melissa Brown, MD, MN, MBA, and I have done most of the legwork on utilities. When we brought the concept to ophthalmology, we wondered: Is it really valid? Is it reliable? So initially, we concentrated on validating our results and making sure the tests were reliable in terms of test/
re-test. Our results have demonstrated that the intraclass correlation coefficient between the initial and follow-up visual utilities obtained 28 days later was 0.7634 (95% confidence interval 0.6655-0.8355), indicating excellent reliability.⁵

Utilities are also valid measures of HRQoL; they are strongly correlated to vision in the better-seeing eye, and VF-14 scores.⁶

The more these tests are demonstrated to be methodologically sound, the more likely they are to be believed and incorporated in further studies. Over the past 5 years, patient preferences and utilities as well as both disease-specific and generic questionnaires have been proven to be valid. They're really here to stay.

WHAT WILL DRIVE ACCEPTANCE?

As with any new technology, there are the early adopters followed by the masses and finally the laggards. The early adopters already are using quality-of-life questionnaires and utilities to help guide their treatment decisions.

At what point will the masses adopt this assessment method? That answer may be related to managed care. Here in Canada, a provincial government may decide to pick up the tab for a given treatment based on demonstrable improvement in patient-reported quality of life. So if a treatment program that's funded through managed care, Medicare or a provincial government starts asking for formal documentation of quality-of-life data, then the masses will certainly adopt it.

As formulary managers become more sophisticated, they're looking for ways to understand who will benefit from treatment. I think it's very reasonable that they look at quality of life.

PROFOUND CONSEQUENCES

AMD is a disease that profoundly affects people's quality of life. That's borne out in patient interviews and questionnaires that demonstrate the activities they can no longer do -- from driving to reading. It affects patients on an emotional level, too. There's a much higher depression rate among people with macular degeneration who are socially isolated.⁷

In terms of utilities, people are willing to trade off a significant amount of time. They're willing to accept a significant theoretical risk in terms of immediate death to eliminate their problem.

Ultimately, we're looking for cost-effective treatments that significantly improve quality of life and eliminate the economic burden associated with the disease.

Dr. Sharma is the director of the Cost-Effective Ocular Health Policy Unit. He's an associate professor of ophthalmology and an assistant professor of epidemiology at Queen's University in Kingston, Ontario.

REFERENCES

1. Sharma S, Brown GC, Brown MM, et al. Converting visual acuity to utilities. Can J Ophthalmol. 2000;35:267­272.

2. von Neuman J, Morgenstern O. Theory of Games and Economic Behaviour, 3rd ed. New York, NY: Wiley; 1953.

3. Brown GC, Sharma S, Brown MM, Kistler J. Utility values and age-related macular degeneration. Arch Ophthalmol. 2000;118:47­51.

4. Brown GC, Brown MM, Sharma S. Difference between ophthalmologists' and patients' perceptions of quality of life associated with age-related macular degeneration. Can J Ophthalmol. 2000;35:127­133.

5. Hollands H, Lam M, Pater J, et al. Reliability of the time trade-off technique of utility assessment in patients with retinal disease. Can J Ophthalmol. 2001;36:202­209.

6. Sharma S, Brown GC, Brown MM, et al. Validity of the time trade-off and standard gamble methods of utility assessment in retinal patients. Br J Ophthalmol. 2002;86:493­496.

7. Rovner BW, Casten RJ, Tasman WS. Effect of depression on vision function in age-related macular degeneration. Arch Ophthalmol. 2002;120:1041­1044.