Patients view benefits in terms of whether their symptoms improve, by how much, and for how long. For drugs that prevent or suppress a disease or problem, benefits are likely to be viewed based on how well the drugs accomplish those goals according to reports they are given by their healthcare providers.

Risks are viewed as the likelihood of having adverse events, exacerbations or new episodes of their underlying disease, or other drug-related problems. Risks that a patient voluntarily accepts (e.g., smoking cigarettes while using oral contraceptives) are often differentiated from those where little or no choice is possible (e.g., cosmic radiation while walking down the street by one’s home). Some risks that are accepted involve elements that are both voluntary and involuntary (e.g., driving a car for one’s business).

Patients also assess their own treatment in terms of quality of life and compliance factors in order to decide whether their current treatment is better or worse than previous treatments. Their assessments are usually discussed with their physician and usually influence the physician’s benefit-to-risk judgment about how to continue to treat that specific patient. Two patients experiencing the same benefits and risks may react quite differently to the risks. For example, one patient may decide to forego the benefits rather than to risk the adverse events, whereas another patient may reach the opposite conclusion. The benefit-to-risk ratio of the drug would, therefore, be different for these two patients. This trade-off is illustrated in Figure 65.6.

Physicians view benefits of a drug in relative terms. They assess or guess whether other doses, other drugs, or a nondrug treatment will enhance or reduce the degree or type of benefit their patient is experiencing or potentially could experience. Risks of a drug may sum in a complex way with concomitant drugs being used, and the resulting risk is often difficult to predict. In addition, a drug may increase the risk of serious adverse events resulting from the drug, but at the same time, the drug might decrease the risks attributable to the disease being treated. This is a frequent trade-off experienced by physicians in emergency care medicine.

Overall, a physician must determine the benefit-to-risk ratio for a patient by considering all benefits and risks and then comparing the present treatment with previous treatments, other treatments, or no treatment at all. The risks a physician should assess are both the real risks and the perceived risks that may or may not be real. To make this assessment, the physician needs to have access to current literature. The physician must then decide whether the current treatment is better than, the same as, or worse than one or more of the other treatments available (Fig. 77.1). If the current treatment is better or about the same, the physician must evaluate whether it is worth the cost in money, convenience, and other factors such as any paperwork or administration required (e.g., if an Investigator Investigational New Drug Application is required). If the current treatment is worse than others, the physician must evaluate the convenience and cost of each of the other alternative treatments that may be preferable.

The most objective way for a physician to evaluate benefits and to compare the benefits of different treatments is in terms of the number of patients needed to be treated for a specified period to obtain one beneficial response. This concept has been described as the number needed to treat (NNT). The NNT is the reciprocal of the difference in the rate of the benefit between the treated group and control group (i.e., the reciprocal of the absolute risk reduction). This concept was described by Laupacis, Sackett, and Roberts (1988) and has been widely used not only to describe benefits but also risks, which are described as the number of patients needed to receive a treatment for a specified period that will result in one patient being harmed [i.e., number needed to harm (NNH)].

Pharmaceutical companies view a drug’s beneficial properties in terms of its ability to demonstrate sufficient efficacy so that:

Risks are usually viewed as adverse events (including abnormal vital signs, symptoms, laboratory abnormalities, physical abnormalities, and other clinical signs) relative to existing treatments. Companies may accept the presence of increased risks until they reach the point where regulatory authorities will not approve the drug or physicians will not prescribe it. Within a company, some groups traditionally tend to focus to a greater degree on these risks (e.g., attorneys, research and development staff), while others tend to focus primarily on benefits (e.g., marketers, public relations, advertising staff). Obviously, these are generalizations with many exceptions.

A pharmaceutical company must focus on the interests of all groups concerned with benefits and risks. To obtain a more complete concept, a company must determine the benefit-to-risk ratio for the following:

In evaluating this concept over time, it is obvious that the benefits expected at the start of a drug’s development often vary greatly from those found as additional data are collected. Similarly, projected benefit-to-risk ratios at the start of a drug’s development often have little resemblance to those observed after the drug reaches the market. Thus, the concept of benefit to risk is a very dynamic one that is subject to great changes over time.

During the marketing period, the benefit-to-risk ratio sometimes changes as new information is gathered. Such changes may be gradual or sudden. Thus, it makes little sense to believe that the benefit-to-risk ratio on the first day of marketing when it is based on data from only a few thousand patients at best will be the same when several hundred thousand or millions of patients have used the drug. Of course, the ratio may improve (as it did for many drugs of the statin class), or it may remain the same (as it does for the majority of drugs) or decrease to the point where the drug’s use is restricted or it is removed from the market.

Also, the pharmaceutical company considers that each drug has a technical risk of failure in development, expressed as the probability of its reaching the market. This, in turn, depends in part on the phase of development (i.e., how close it is to the market). Generally, the more advanced a drug is, the greater the likelihood of its being approved and launched. The technical risk also depends on whether the scientific rationale for using the drug is based on strong or weak preclinical data and on how well those data can be extrapolated to humans.

Intellectual property (patents), licensing, and competitive issues are also risks for the pharmaceutical manufacturer throughout the life of a drug.

Institutional Review Boards (IRBs)/ECs view benefits and risks in terms of the acceptability for patients to enter a clinical trial to evaluate new or marketed drugs. These committees also must consider the benefits and risks of using a placebo or active treatment if one is included in the trial design.

IRBs/ECs usually focus on the benefits and risks of the activities described in the protocol. They do not generally evaluate a trial to determine the sins of omission, that is, whether procedures or other activities not included should be or whether the data to be obtained justify the benefits and risks of the trial. For example, are there too few patients to be enrolled to provide statistically and clinically meaningful results when the trial is completed? Are the endpoints measured going to provide the most meaningful results that are possible in this particular trial? Is the trial open-label or double-blind? Many other such questions could be posed but rarely are, particularly if there is no statistician or clinical trial methodologist on the IRB/EC. The author believes that these and also other related methodological questions should be posed by every IRB of every clinical trial protocol because patients may be at risk because of what is not included in the trial (e.g., sufficient power to show an effect if it is there, adequate trial design that minimizes biases and confounding factors, using a double-blind procedure instead of an open-label one) as well as what is included. This is because a trial that will not yield meaningful data is unethical to conduct and should not place patients at risk.

Data Safety Monitoring Boards (DSMBs; sometimes referred to as independent data monitoring committees) view benefit to risk in terms of the rules that have been established in advance for terminating a trial on futility grounds (i.e., inability to demonstrate efficacy if the trial continues to completion) or for reasons of safety that show increased levels of clinically significant adverse events in the treatment group compared with the control. Trials are also able to be stopped for overwhelmingly positive efficacy that achieves a preset “p” value demonstrating a major positive response. It should be noted that a DSMB may only be constituted to assess safety, and the efficacy issues mentioned may not be in their remit. DSMBs are discussed in Chapter 81.