Numbers Game
by Peter Fretty
April 2007
Prescription data can be a powerful tool for pharmaceutical sales representatives looking to tailor sales to specific audiences. But due to the sensitive nature of the data, it also can be an ethical minefield.
In the pharmaceutical industry, diligently collecting prescriber data is a popular way to tap into market knowledge and guide sales force activities. One of the biggest concerns with this practice, however, is the potential for overstepping ethical boundaries with such highly sensitive personal information. These concerns have grown to such an extent that various U.S. state legislatures have created statutes, guidelines and notifications to get physicians’ attention.
Although these concerns are valid, the current approach is not necessarily the best course of action, says Jody Fisher, vice president of product management at Verispan LLC, in Yardley, Pa., USA. “There have been arguments made that patient privacy is affected when [prescriber data] is allowed into the market, which is untrue, because all of the data we receive is well-governed and guarded [in part through the Health Insurance Portability and Accountability Act],” he says. “Because of these measures, we have no idea about the [identity of] patients. If I personally wanted to find my data, I could never find it.”
Physician and Patient Privacy
Prescriber data helps guide not only sales activities, but also research and development, and there is no reason for the industry to suffer as a result of the unethical behavior of a few sales representatives, Fisher says. “What some people have trouble getting past is the fact that in some instances, [prescriber data] has been used fairly aggressively, and some unscrupulous individuals have used it as an aid to convince doctors to use their medication,” he says.
To protect patients’ identities, there should bemeasures in place to ensure that pharmaceutical companies are unable to get to the actual consumer level, says Dr. Robert George Wilcox, head of cardiovascular medicine at University Hospital, Queen’s Medical Centre, in Nottingham, U.K. “When you are talking about an individual’s health, there needs to be confidentiality, or at least the ability of a patient or physician who wishes for confidentiality to be able to opt out,” he says.
A large part of the problem is that most physicians don’t know the extent to which their prescribing behaviors are used, says Graham Leask, Ph.D., lecturer at the economics and strategy group at Aston University, in Birmingham, U.K. “The ethical issue is that the doctor is often unaware of this data and has not given informed consent,” he says. “Knowledge of what competitor’s product the doctor is using makes it easier for the sales representative to tailor the detail and effectively displace the competition. Data is often purchased from organizations who employ the doctors, and in cases where the doctors themselves give access to the data, it is generally financially advantageous for them to do so.”
Defining Differences
There are two types of data under discussion at the prescriber level: standard prescription dispensing data and de-identified longitudinal patient data. Prescription dispensing data tracks prescription information through pharmacies or insurance companies that show a doctor has prescribed a certain medication a precise number of times and at specific dosage levels. Longitudinal data is the actual patient record, without individually identifiable information, from which specific prescriptions may be ascertained for a given disease area.
Although its practical application is quite similar to that of prescription dispensing data, longitudinal data is more valuable in certain applications. “Simply put, the value rests with the fact that longitudinal data is based on actual patients, and includes valuable information about their medical backgrounds and history,” says Martin Fagan, founding partner of Infozyme Consulting International, a marketing and management consultancy in Stathern, U.K. “[Longitudinal data collection] is an effective means of tracking when a product is prescribed to treat specific ailments. While these data were originally intended for pharmacovigilance purposes, this type of information also can allow sales representatives to make more educated decisions based strictly on the prescribing habits of physicians within their region or in some cases, within their specific territory.” Confidentiality regulations do not allow the physician prescriber data to be individually identified, Fagan says.
According to Fagan, longitudinal data picks up very detailed prescription activity from only a nominal sampling of actual practitioners across the country (between 100 and 400 general practitioners in the U.K., depending on the source of data), whereas prescription data routinely picks up 50 to 60 percent, and sometimes more, of the entire physician population through retail pharmacy sales. This can represent a significant difference in numbers, especially since very few physicians opt out of inclusion in the instance of prescription data, or even know it is being collected. Opt outs are more common for longitudinal data, the collection of which requires specific cooperation from the physician, Fagan says. “Undoubtedly, the biggest challenge about longitudinal data is that the information collection company needs to have more of an intimate relationship with the physician or the care network,” Fagan says.
Gauging Effectiveness
Paradoxically, while the pharmaceutical industry is awash with data, it is also devoid of information, Leask says. “A considerable asset of a pharmaceutical company is the data that we collect,” he says. “We benchmark everything, but unfortunately we often fail to make use of it, or choose to discard useful categories in favor of a simpler approach. This is a mistake that may lead to a loss of competitive advantage.”
But since pharmaceutical companies are continually looking for ways to cut costs, “there is usually little consideration given to the nature of the data or what the correct analysis tools to use are,” Leask says. Prescriber data could yield reliable results that are both accurate and reproducible if pharmaceutical companies used proper statistical methodologies, he says.
However, because the data comes from collection systems integrated into the prescribing physician’s network or extended network, the data is high-quality and, at least in theory, provides unparalleled insight into physician prescribing habits, Fagan says. “Even though the data is extruded in a manner that the buying pharmaceutical company desires to suit its needs, it is extremely valuable and can serve as an effective means of guiding sales efforts,” he says.
“Companies value seeing actual usage, because it tells them if their sales staff are working in an effective manner. However, the whole complexity of the market is changing how things are working,” he says. He cites generic substitution and prescribing nurses, among other factors, as changing the dynamic of prescription data. Prescription data can show what is being prescribed, but not the rationale behind the decision, he says.
From a salesperson’s perspective, knowing how often particular brands are prescribed is valuable information. This type of information could serve a key role in driving prescription costs down, Wilcox says. For example, if a pharmaceutical representative learns that the majority of prescriptions being written for a certain drug are for a competitor’s product, “this may spark dialogue to undercut the pricing of the competition [in order] to gain more of the market share.”
Prescriber data also could help health departments keep an eye on practitioners, Wilcox says. “Knowing that the prescribing habits of practitioners in the U.K. are scrutinized heavily, the use of such data could ensure that general practitioners avoid prescribing unnecessarily high levels of an expensive drug,” he says.
Cultural Considerations
The acceptance of using prescriber data as a sales tool differs by country, due to cultural variances and the stringency of laws. “Exactly how much value the country’s health care system places on each provider makes a difference on the national perspective,” Fagan says. “For instance, there is a big move in the U.K. to get more patient treatment in the general practitioners’ hands, a move that would surely weigh heavily on the use of prescriber data. Add to this a move to have the ownership of the patient’s data away from the hospital and general practitioner [and] into the hands of the actual patient, [and the effect] will mean collection and use of these data will have added complexity in the future.”
Prescriber-level data is available in the United States, but current laws do not allow the same level of detail in the U.K. “Longitudinal prescriber data is routinely used in the United States, but its effectiveness is limited in Europe, due to the European Union’s stricter data protection, culture and confidentiality issues,” Fagan says. “The same is true with the lack of direct-to-consumer advertising in EU countries. While its practice is commonly deployed in the United States, and arguably makes a difference in what physicians prescribe, its practice is prohibited in most of the EU.”
Despite the varying international regulations, sales representatives around the world share the common goal of trying to influence physician prescribing behavior. When gathered ethically and used responsibly,longitudinal prescription data can be a powerful tool to gain valuable market insight and formulate more effective sales strategies.
Send us Your Comments
Web Exclusives
- Statistical Advantage
-
Gary Shorter, director of biostatistics for Quintiles, talks to Envisage editor Adam Istas about the growing importance of incorporating biostatistical analysis into all stages of drug development.
- Measure for Measure
-
A proposed Health Care Comparative Effectiveness Research Institute in the United States may force the biopharmaceutical industry to emphasize patient outcomes and quality of life issues when developing a new medicinal product.