A spectrum of problems with using color

In our July 24, 2003 issue, we provided an example of how color-coding can contribute to medication errors. Specifically, we mentioned how a color-coding system for the pharmacologic class of ophthalmic medications, along with similar corporate logos, fonts, and package sizes, have led to numerous errors with these products. Antiinfectives are tan, steroids are pink, mydriatics and cycloplegics are red, beta-blockers are yellow, and so on. But individual products and dosage strengths within the class are not color-coded, so mix-ups have been widely reported within each class.

Since publication of that article, we have received several questions that reflect unresolved issues about the use of color to reduce the risk of errors. To fully understand the issues, the distinct differences between color-coding, color differentiation, and color matching need to be explored, as well as how each may be applied by end-users and manufacturers.

Color-coding. Color-coding is the systematic, standard application of color to aid in classification and identification. A color-coding system allows people to memorize a color and match it to its function. The black cap used in the US for all vials of potassium chloride concentrate injection is one example. No other products can use black caps. Thus, practitioners can assume that a manufacturer’s vial with a black cap contains potassium chloride concentrate, as long as the color black has been memorized to mean “potassium chloride.”

Color differentiation. Color differentiation, on the other hand, entails the use of color to make certain features stand out, or to help distinguish one item from another. To cite an example, a vaccine product for adults may be packaged in an orange box while the same company’s pediatric formulation may be packaged in a light blue box. However, the color itself is not a standard code that is applied systematically to classify and identify the product, as it is with color-coding.

Color matching. Color matching is also used sometimes to reduce the risk of errors. For instance, a medical device may have a blue plug that attaches to a blue receptacle and yellow plug that attaches to a yellow receptacle. However, the colors have no special meaning beyond matching one item with another.

Frequent problems with color-coding. Color-coding for pharmaceutical products should be used with extreme caution, as there are several problems associated with its widespread adoption. For one, there’s a limit to the variety of discernible colors available for commercial use. Well demonstrated in color-coding research in other industries (1,2) subtle distinctions in color are poorly discernible unless products are adjacent to each other. Contrast with background or surrounding colors also can be problematic if a certain color must be used for product identification. And of course, clinicians might be color-blind, resulting in possible misidentification of color-coded products. This could be the reason that FDA and the pharmaceutical industry have frowned on color-coding for the most part.

Next, color-coding has not been scientifically tested as a way to prevent medication errors. But, like any change, there’s evidence that it may actually contribute to some errors that would not happen without it. As mentioned above, color-coding ophthalmic products by pharmacologic category has led to numerous errors. Another example is the recent change in the colors used to identify needle gauges on disposable syringes. Several 10-fold insulin overdoses have been reported because the 25 gauge needles on some tuberculin syringes are now coded in orange, a color previously associated with insulin syringes (see our March 6, 2003 issue).

Color-coding can also be error-prone if it’s not applied consistently across the industry, or within a single manufacturer’s product line. For example, one company used product-specific color-coding schemes for the labels on some of their products, and yellow and blue labeling for the remainder of their products. On occasion, practitioners have mixed up some of the products with blue and yellow labels, failing to recognize that the blue and yellow labels were not a product-specific color code.

We’ve also observed occasional problems with user-applied color-code schemes. For example, a color code from the American Society for Testing and Materials (ASTM) is employed as the basis for user-applied labels on syringes. The label color identifies a drug category, but it does not necessarily identify a specific drug, strength, or dose contained in a syringe. So anyone other than the person who prepared the syringe may not be aware of its actual contents, and thus administer the wrong drug or dose. Of course, if the system is used as intended to quickly distinguish the different drug classes, the ASTM color scheme has merit. However, the user should add the actual drug name (if it’s not already printed on the label) and strength.

Less frequent problems with color differentiation. Color differentiation also has not been scientifically proven to prevent medication errors. However, practitioners favor it for commercial use to help reduce the risk of product-selection errors within a manufacturer’s product line. The idea is for users to efficiently find and select medications from storage areas using color differentiation, then carefully read the label for verification. But users don’t always read the labels, or they may suffer from confirmation bias and fail to see that they do not have the correct product in their hand.

However, color differentiation has been used successfully in commercial applications to enhance recognition of various label elements by making the information stand out. For example, a product warning using the color red is more likely to be seen if the product’s primary label is black and white. On occasion, we’ve also suggested using a yellow highlighter to draw attention to important label information, such as the concentration of the drug, or the total volume of product in the vial.

Occasional problems with color matching. Likewise, little scientific evidence exists to prove the value of color matching related to medication use or pharmacological products. In fact, it’s rarely, if ever used for such purposes. However, its benefits appear sound when applied to the use of medical devices, given that the user or manufacturer applies the proper colors to products.

Black and white labels. Of course, if all products had identical black and white labels and were packaged in the same size and shape, careful reading of the labels would be the only way to differentiate them. Perhaps this would reduce error potential, but it’s not likely to happen in our free enterprise system for marketing pharmaceuticals. We look forward to the time when all products are bar-coded or identified through radio-frequency methods. Then, we’ll be better able to store them all together but tell them all apart, even if we misread or misunderstand the label.

References: 1. Christ RE. Review and analysis of color-coding research for visual displays. Human Factors 1975;17:542-570 (web.engr.oregonstate.edu/~pancake/cs552/guidelines/color.coding.html). 2. American National Standards Institute, Association for the Advancement of Medical Instrumentation. Human factors engineering guidelines and preferred practices for the design of medical devices (ANSI/AAMI HE-48). Arlington, VA: Association for the Advancement of Medical Instrumentation, 1993; pg. 41.