Patient Outcomes Research Moves Drug Development Beyond Clinical Results

In the ever-escalating battle for medicinal value and competitive positioning, drug developers and marketers are relying on outcomes research By Suzanne Shelley

What makes a drug worth commercializing, worth prescribing, and worth paying for? Years ago, the answer was simple: a drug cured a disease or corrected a medical condition. But today, as drug research ventures farther into specialized areas like better mental or emotional health, diffuse conditions like obesity or chronic conditions where there’s a fine line between using and not using a drug, the answer is more complicated. As this complexity increases, and as everyone in healthcare focuses on costs, the analysis of “patient outcomes” becomes more critical.

“Economists, pharmacists, physicians and payors are all looking for rational ways to justify drug expenditures, so increasingly, they’re embracing all types of patient outcomes studies to gain greater insight and strengthen their approach,” says Mick Kolassa, PhD, chairman of Medical Marketing Economics LLC (MME; Oxford, MS).

MME, along with a host of consulting organizations, contract research organizations (CROs), market-research and data-analysis houses, and even insurers, PBMs and healthcare providers themselves, are consulting with biopharma manufacturers to arrive at useful patient-outcomes information. For providers and payors, the work can uncover both cost savings and better service to patients. For manufacturers, the objectives are manifold: getting drugs past the FDA approval process; winning market share against competing treatments (drug and non-drug); and gaining acceptance and better compliance among patients.

Clinical, financial and emotional

Patient outcomes is a broadly encompassing term, which is perhaps more accurately classified by its main subcategories: health economics and outcomes research (HEOR) and patient-reported outcomes. Comparative effectiveness (Pharmaceutical Commerce, April, 2009, p. 1) is a related specialty, addressing head-to-head comparisons of medical benefit directly.

In general, three types of patient outcomes data can be gathered and analyzed — clinical, economic and humanistic. All are essential to paint a complete picture.

Physicians look to clinical outcomes to assess the impact a specific medication may have on one or more particular clinical endpoints, to weigh the advantages and disadvantages of competing products in the same therapeutic class

Payors look at both clinical and economic outcomes to understand the direct and indirect costs associated with using a particular drug (Will this change in blood pressure be sustained over time long enough to directly reduce the patient’s risk of stroke or heart attack?), and to identify the financial implications of not using the drug (for instance, when a less-compelling competitor product is selected). Such findings can help to ensure that pricing and reimbursement strategies for a given product are appropriate — or justify a renegotiation.

Meanwhile, patients want to know whether the medications they are taking will work as promised, and whether or not taking them can help them feel better so they can get on with their lives. Toward that end, humanistic outcomes research aims to produce the data that verify how a drug can improve the patient’s overall health and quality of life (QoL).

While interest in outcomes research is growing, its use is far from universal in the US, which lags behind many other countries in terms of payors requiring pharmacoeconomic data. By contrast, many countries where single-payor models prevail, and health technology appraisal (HTA) bodies are in place — such as the National Institute for Health and Clinical Excellence (NICE) in the UK, Canada’s Common Drug Review (CDR), Germany’s Institute for Quality and Efficiency (IQWiG), and Australia’s Pharmaceutical Benefits Advisory Committee (PBAC) — already routinely require extensive economic dossiers to accompany the clinical data required for regulatory approval and reimbursement decisions. However, many large U.S. insurance companies are starting to request this type of information “in order to approve drugs for reimbursement because they want to know how the inclusion of a new drug to their formulary will affect their budgets,” says William Lenderking, PhD, senior research leader, United BioSource Corp. (UBC; Lexington, MA).

Pricing and reimbursement implications

The right types of patient outcomes data are needed when it comes to negotiating the most favorable formulary tier designations, pricing and reimbursement terms for individual medications (not just for products that are about to launch, but for existing products, product line extensions and product acquisitions, as well). “Formulary and reimbursement decisions both rest squarely on how well the payor understands the overall impact of, and value associated with, using a given product,” says Bob Mauch, PhD, PharmD, president of Xcenda (Palm Harbor, FL), a subsidiary of AmerisourceBergen Specialty Group. “The goal for all drug makers and for the healthcare industry in general is to better understand where the costs are accumulating, and then develop and use outcomes data to help find ways to address these cost issues,” adds Rebecca Hyde, EVP, global commercialization, for inVentiv Advance Insights (Somerset, NJ).

Outcomes data are also figuring in a number of “pay for performance” arrangements between drugmakers and payors, in which the manufacturer in effect provides a financial guarantee of some agreed-on level of benefit for using a drug.

Last spring, Merck and Cigna (the health insurance company) agreed on a program to both peg the cost of Merck’s Januvia and Janumet diabetes treatments to how well blood sugar is controlled, while Cigna was also committing to helping raise adherence to the drug regimen.* Outcomes data are a key part of administering many comparable pay-for-performance arrangements in Europe, and in some cases in the US (see p. 1).

Similarly, SDI has carried out a study for a client that was about to launch a new drug it felt would be able to provide distinct advantages in terms of both efficacy and effectiveness. “We first analyzed the economic and clinical outcomes for existing products in the same therapeutic class to create a set of baseline expectations,” says Gregory Hess, MD, MBA, MSc, VP of HEOR for SDI Health (Plymouth Meeting, PA). “During formulary negotiations, we were then able to present our findings to a large insurer and propose that if our client’s drug were to provide outcomes that exceeded those baseline outcomes values, it would maintain a favored position on formulary, and the drug manufacturer would benefit from increased sales resulting from that favored status. If the outcomes didn’t meet expectations, then the drug company would provide some financial giveback to the insurer.”

‘Show me the evidence’

“Historically, the goal of comparative effectiveness studies was to pit one drug against another in order to beat down the unit cost from vendors,” says Sam Malik, EVP, managed markets, for inVentiv Advance Insights. “Now, people are starting to really appreciate that solid comparative effectiveness studies that can link specific outcomes to medication use are useful for numerous other purposes.”

When it comes to formulary tier designation, “the prevailing model is that certain medications enjoy lower co-pays largely because of more-favorable prices that were negotiated between the payer and the drugmaker,” says Hyde of inVentiv. “But in the future, it’s conceivable that we’ll see lower-co-payments assigned to drugs that have better proven outcomes data.”

However, she adds: “The conditioned response among healthcare plans is generally ‘generics first’ so this paradigm shift will require a major turnaround in philosophy — and of course access to valid, reliable outcomes data to make the case.”

“Given the cost-containment pressures that are at work today, post-marketing, comparative-effectiveness studies could become the most important marketing tool, as they are able to establish value (in terms of clinical, economic and QoL outcomes) of a product and compare the value of its nearest competitor(s) or standard of care,” predicts Samir Bhattacharyya, PhD, VP, Covance Market Access Services (Gaithersburg, MD).

And the execution of healthcare-related comparative effectiveness studies — carried out not just by the pharma industry, but by any number of industry, academic and government researchers (Box, p. 30) — is expected to grow in the years to come, thanks to the $1 billion in funding that has been earmarked carrying out such studies as part of the Obama Administration’s recent stimulus funding and the ongoing $600,000/year contained in the current health reform bill. “That’s going to be a game changer in the years to come,” says Jonothan Tierce, C.Phil., GM of HEOR at IMS Consulting (Falls Church, VA).

Get started early

“The critical factor for success in terms of marketing and access to formularies is planning the overall strategy early on in the drug development, certainly before the start of Phase II,” says Lenderking of UBC. “Too many sponsors come to companies like ours when their products are already in Phase III and that is so late in the planning process that the likelihood of generating the right kinds of outcomes research evidence to support successful launch and market access is low. It takes time to develop and implement a successful strategy.”

As such, “Health economics research should be an early and ongoing part of developing any brand strategy, and should be carried out as early as possible in the product-development process,” adds Tierce of IMS. “To develop the most valuable breakthroughs, outcomes research needs to be more of a core function — not just an afterthought.”

“The clinical trial team’s focus is really on getting the drug through the FDA-approval process, and — sad but true — efforts to gather data that could help support the reimbursement question becomes almost an afterthought,” says Robert Epstein, MD, MS, SVP and chief medical officer at Medco Health Solutions (Franklin Lakes, NJ), who is also past-President of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). “So often, we see pharma companies wake up to the reimbursement issues in Phase III and by that point, it’s really too late to change the studies.”

Epstein says this trend is not just the result of ignorance. “I think there’s long been a fear factor at work here — if the initial clinical trials end up developing too much information around the drug, the product could end up getting niched or pigeonholed, rather than being launched as a potential blockbuster.” However, he notes that “some companies are starting to realize that it’s better to determine earlier in the process which specific patient sub-populations will truly benefit the most from the product, and then be able to directly promote the product to those sub-groups and get them priced and reimbursed appropriately.”

“Marketing people and clinical people need to work more closely to identify all critical endpoints to investigate,” adds Kolassa of MME. “Clinical teams need to think ‘If I can capture key lifestyle outcomes during clinical trials and post-marketing studies, that will give the marketing folks something to talk about.’”

Post-marketing studies

“What we are able to learn from clinical trials is that — under the best of circumstances — the drug will work compared to placebo, and that’s good news, but these trials rarely tell us how the drug will behave in the real world,” says Epstein of Medco. By comparison, retrospective database and registry studies are not subject to the same strict inclusion/exclusion criteria as clinical trials or prospective patient studies, “so they will — for better or worse — capture data about how the drug performs among real people behaving naturally with all their real-life co-morbidities, poor adherence practices, tobacco and alcohol use, antipathy toward exercise and so on. In this way, they will paint a more realistic picture of what can be expected,” he adds.

“As such, these studies help us to gain a better understanding of how the medication performs in actual care settings under non-ideal conditions,” says Hess of SDI. “When it comes to clinical practice, we don’t have the luxury of saying ‘I’m not going to give this patient this medication because they don’t meet the trial’s testing criteria,’ so these more naturalistic or ‘real-world’ studies ensure greater external validity, and that’s very useful in terms of the effectiveness results they reveal.”

“When devising post-marketing outcomes studies, we try as hard as possible to design studies that will establish a link between the use of the medication with specific outcomes and endpoints that have some clear clinical or economic value,” says Malik of inVentiv. “For instance, you might design a post-marketing study for a given cardiovascular drug to look for trends in usage data that could establish what number of patients would need to be treated to avoid one death or one heart attack or other desired outcome.” Such a result “would have a clear clinical value to both physician and patient, and a clear economic value to the healthcare system in terms of the avoided costs associated with avoidable deaths or hospitalizations,” he adds.

“Within the last two years or so, studies like these — those that dig a little deeper to link more disparate data sources to look at patients longitudinally — have begun to gain currency,” says Malik. He credits ongoing improvements in data-collection, warehousing and data-mining capabilities, coupled with improved analytics and a greater willingness of healthcare organizations and payors to provide such data for the purpose of patient outcomes studies, have helped to make this more possible.

Post-marketing outcomes studies to examine various trends that can be associated with, or attributed to, the use of the drug can be either retrospective or prospective. Retrospective studies are, by design, mainly as database-driven studies, which rely on extensive third-party databases from insurers, PBMs and pharmacies, but from data aggregators such as SDI Health, IMS Health and others.

In general, retrospective analyses “can often be done quickly and at relatively low cost, and they are more reflective of real-world settings,” says Mauch of Xcenda. “However, you don’t always have access to all of the key clinical indicators or lab results from a single data source, so you may have to seek multiple data sources and work to integrate them. This will require you to make more assumptions about the relationship between surrogate markers and the outcomes.”

Alternatively, prospective post-marketing studies look for specific endpoint trends among a cohort of actual patients using the drug under real-world conditions. “Because prospective studies reflect outcomes related to real world experience, the results may look significantly different from the ideal results developed in the earlier clinical trials, so they are perhaps more accurate in terms of what patients and physicians can expect during actual usage,” says Mauch.

In recent years, there has been a gradual shift in emphasis from retrospective to prospective studies, says Hyde of inVentiv, mainly because retrospective pharmacoeconomic studies sponsored by drug manufacturers are sometimes met with skepticism by health plans that question whether the studies are bias-free. “As a result, we’re seeing much greater interest in, and receptivity to, prospective studies that involve actual patients and are carried ‘from scratch.’”

Enticing employers

Employers — who are the biggest payers of healthcare costs for their employees —should not be overlooked by drug companies when it comes to structuring various types of patient outcomes studies and sharing their results.

“At the employer level, there’s great value in developing and disseminating outcomes study results that can confirm the value of a given therapeutic agent in real, quantifiable terms, such as reduced hospitalizations, improved QoL for the patient, reduced absenteeism and greater overall productivity,” says Hyde of inVentiv.

For instance, when it comes to chronic pain, employers shoulder significant direct costs (in terms of the cost of the drug therapies and other interventions the patient is using) and indirect costs (in terms of productivity losses and increased absenteeism that result from chronic pain issues among their employees). “In this case, the ability to demonstrate that one migraine pain medication over another can improve QoL in a way that reduces absenteeism and increases productivity would speak to an employer in ways that would never be of interest to the payor,” says Hyde of inVentiv, adding: “Studies should always be designed and communicated around who is going to be the ultimate consumer of the outcome results.”

Drug companies already work with many major employers to carry out overall wellness programs for employers, to help employees quit smoking and to assist with the control of diabetes and management of other chronic conditions. “In the future, we expect to see drug companies working more closely with employers with respect to educating them about compelling outcomes findings related to specific drugs,” says Hyde. “If a study is able to link medication use with direct cost savings, it will be impossible for employers to ignore that conversation.”

Taking that connection further, employers could then be pressed into service to help influence formulary tier designation and reimbursement strategies with their payors and PBMs. “If the universal goal is to manage therapies in a way that improves outcomes and reduces healthcare-related costs, then pressure from well-informed employers on their health plans and PBMs can help to make all parties more receptive to discussions related to formulary tier designations and reimbursement decisions,” says Malik of inVentiv. “Similarly, as these parties become more aware of the useful connections that can be made, they will likely begin to exert more pressure on drug companies to carry out specific studies that may be able to help connect the dots.”

Defining the patient response

Perhaps the hardest aspect of patient outcomes research is to put numbers to the qualitative satisfaction that patients have with a med. But this is also the most valuable in the long run. Higher ratings for “quality of life” (QoL) should translate into both better sales performance of a product, and higher reimbursement, all other things being equal.

“There are two dimensions of QoL,” says Tierce of IMS. “These include patients’ subjective feelings about their health and its impact on their sense of well-being, and more objective measures of their ability to carry on the “activities of daily living” that are essential to independence, lack of custodial care (whether by family members or through paid caregivers) and the ability to conduct a robust lifestyle.” He adds: “This isn’t all ‘touchy-feely;’ there can be significant societal and even additional medical costs for people with low QoL, who cannot do for themselves and who may be discouraged or depressed.”

“Historically, pharmacoeconomic studies have typically focused on clinical performance and economic aspects — not really on patient outcomes — but that’s starting to change,” says Kolassa of MME. “It’s one thing to tout the benefits of receptor sites and mechanisms of action, but if the patient doesn’t ultimately feel better — something that can only come out with QoL studies — the dollars-and-cents arguments don’t necessarily make the case.”

Specifically, findings related to patient satisfaction — a measure that reflects for instance, the drug’s ability to alleviate symptoms and make the patients feel better and to verify such appealing features as minimal side effects and adverse impacts and relatively easy dosing and administration requirements compared to similar products — can give brand managers invaluable insight into factors that may influence physicians’ willingness to prescribe a given medication, and patients’ willingness to take it. Armed with the right outcomes information, brand teams can structure their DTP and DTC outreach more effectively.

“Recent interest by FDA and industry to capture subjective information in a scientific way is a recent development that is being driven by several factors,” says Lenderking of UBC. “Today’s patients are more active participants in their own healthcare so it is natural that their voice should be incorporated into the evaluation of medicines when appropriate. And, with the proliferation of competing medications for the same condition, brand teams realize that the patient’s perspective might provide some ability to differentiate between products.”

“Whatever the drivers, from the perspective of overall brand management and reimbursement strategy, pharma companies should embrace this trend,” adds Xcenda’s Mauch. “Being able to measure and articulate the full value of your product to all potential stakeholders is more important than ever.” PC

* See http://www.pharmaceuticalcommerce.com/frontEnd/1191-Performance_based_pharma_pricing_comes_to_US_payers.html