Braille Monitor                                                 July 2010

(back) (contents) (next)

Special Report:
A Study on the Accuracy of Dosing with Insulin Pens by Blind and Sighted People

by Ann S. Williams, PhD, RN, CDE

Ann S. WilliamsFrom the Editor:  Ann Williams has worked in the field of diabetes and blindness for some time. She has been a frequent contributor to our former publication, Voice of the Diabetic.  Here is a report on a research project she undertook, in consultation with the National Federation of the Blind, during last year's convention.

Last summer, through the Diabetes Action Network (DAN) and the Jernigan Institute, the National Federation of the Blind collaborated with me in a small but significant research project. Its purpose was to compare the accuracy of blind and sighted people who used insulin pens to deliver doses of insulin into a small rubber ball.


To understand the importance of this project and why the NFB was involved, it helps to know a little about insulin pens and the history of their use. Insulin pens were first invented in the late 1980s as a convenient, easy, and accurate way to deliver precise amounts of insulin. They are the size and shape of an old-fashioned fountain pen except that instead of ink they are filled with insulin, and instead of pen tips they have disposable, screw-on needles. To set the dose, the person using the pen turns a knob on the end opposite the needle, which makes a small click that can be heard and felt for each unit of insulin dialed. (A few pens make a click for each 1/2 unit.) People who see well can confirm the dose set by reading a number in a small window near the knob.

Although insulin pens were not specifically designed for use by blind people, they are particularly useful for blind diabetics because a person merely needs to count the number of clicks to determine the amount of insulin dialed. Consequently, blind people with diabetes began using them almost as soon as they appeared on the market. Many blind people have used them safely for years. For a more detailed discussion of how to use an insulin pen, see an article by Tom Ley that first appeared in the Voice of the Diabetic, archived on the NFB Website at <>

Shortly after the pens were first introduced, the major manufacturers began including a disclaimer in the instructions saying that insulin pens should not be used by people with visual impairment. This disclaimer has remained in the packaging for many years, even though thousands of blind people all over the world have used insulin pens safely. Unfortunately, the disclaimer causes some healthcare professionals to doubt that blind people can in fact safely use insulin pens. Some doctors and other prescribers therefore refuse to write prescriptions for insulin pens for people who do not see well.

The Diabetes Action Network has recognized this as a problem for blind diabetics. In 2008 the NFB passed a resolution calling for the removal of this disclaimer. You can read the resolution at <> by following links from the Publications page.

My experience with insulin pens, both as a nurse diabetes educator and as a person with diabetes, has been very positive. I have used insulin pens myself and have taught hundreds of blind and sighted people how to use them. I have seen in clinical experience that most blind people are able to use insulin pens accurately unless they have additional problems such as numb hands or impaired memory. Sighted people who have such problems often have trouble using insulin pens. Both because of this experience with insulin pens and because I know some people who were denied the benefits of using insulin pens because of vision loss, I decided to do a research project to develop the rigorous evidence that could encourage manufacturers to remove the disclaimer from the insulin pen instructions. 1

Design of the Study

The study was designed to compare the accuracy of dosing between thirty sighted and thirty blind people. Each person either read the instructions included with the pen or listened to a recording of those instructions, depending on their eyesight. The recordings contained the same information included in the printed instructions, with verbal descriptions of all pictures and nonvisual techniques for steps in the procedures that assumed good vision. While people read or listened to the instructions, they handled the pen and the supplies that go with it, trying it out until they felt sure they knew how to use it. Each person injected ten doses of insulin into a rubber ball. The ball was weighed on a very precise scale immediately before and after each injection, giving the exact weight of the insulin delivered into the ball. This allowed us to calculate exactly how much insulin was delivered in each dose.

The reason for designing the project as a comparison is that any group of people delivering a lot of insulin doses would make at least some mistakes. If only blind people had participated in the study and had made mistakes, people reading the study report might debate what level of accuracy was acceptable. However, if blind people made some mistakes with their dosing but overall were about as accurate as sighted people, manufacturers would have no reason to keep the disclaimer on their insulin pens.

Funding for the Study

I was able to obtain funding for this study through a small grant from the American Association of Diabetes Educators and Sigma Theta Tau, an international nursing honors society. The Jernigan Institute of the NFB also contributed to the project by making the pilot study possible, by helping recruitment of blind participants for the larger study, and by allowing me to do part of the study at the 2009 NFB Convention.

Doing the Study

The first part of the study was a small pilot conducted with twelve NFB staff members at the National Center. The purpose was to try all the procedures and work out the problems before doing the larger study. I went to Baltimore with a research assistant, and the NFB provided a room to conduct the study while we were there. The NFB staff members who tried out the study procedures gave me valuable feedback that we used to fine tune the instructions and procedures. This feedback also helped us understand how to ensure accurate measurements of the doses that the participants in our main study actually delivered.

After the pilot was complete, we did the main study. The Jernigan Institute helped by publicizing the study and asking NFB members to participate. To do the study itself, I went to the 2009 NFB convention in Detroit along with two research assistants. There we met with blind diabetics so they could listen to the recordings, deliver insulin doses into the ball, and have the doses weighed. Being able to meet with so many blind diabetics in one place was an important part of making this study possible.

After I returned home, I recruited thirty sighted people with diabetes who also received instruction in insulin pens, gave ten doses into a rubber ball, and had the doses weighed.


After the study was completed, we found no significant differences in the accuracy of doses between those given by blind people and those given by sighted people. Both groups delivered many accurate doses, and both also made some mistakes. The level of mistakes was about the same for both groups.

The results of this study have been published in the May issue of the Journal of Diabetes Science and Technology. I chose to publish in this journal because many people who design diabetes equipment and make decisions about manufacturing it read it.

The first major lesson of this study is that the evidence does not support inclusion of a disclaimer warning against use of insulin pens by visually impaired people. The study was small enough that the insulin pen companies and government regulatory agencies may not accept the evidence as conclusive. They will probably want to see a study that includes many more people before they actually remove the disclaimer. However, this study represents an important first step, one that will make it easy for interested researchers to justify doing a larger study to confirm these results.

Another lesson of this study for manufacturers of health care technology is that it is important to include people with disabilities in research about the usability of their products. For example, according to the U.S. Centers for Disease Control, about 3.6 million people have diabetes and visual impairment. This is about 20 percent of the people with diagnosed diabetes. However, many manufacturers ignore this 20 percent when they conduct their studies. They should, however, be interested in knowing whether their product can be used properly by this very large minority group. Including people with disabilities in diabetes technology studies is not only ethically right, but also financially sound. The 20 percent of people with diabetes who do not see well do buy products for treating diabetes. Any company that has tested their products in this population and found them useful to blind people will have a marketing advantage. With proper product testing among people with disabilities of all kinds, a company could market to all groups that can use their product--not just to people who happen to have no current disability.

Finally, a third important lesson concerns the importance of intelligent collaboration between the NFB and researchers. We need each other and can benefit each other greatly. As a researcher I was very grateful for the help I received from the Jernigan Institute--from the use of a room for the pilot study, to the expert advice from NFB staff members, through the help with recruitment of participants and access to people at last year’s national NFB convention. Without this valuable assistance I would have had a much harder time conducting sound research with blind participants, and overall it would have been much more difficult to complete the project. NFB’s help benefited me greatly. In return this project will ultimately benefit NFB members and all blind diabetics by providing real evidence of the true abilities of blind people and ultimately by enhancing the access of blind people to a useful product. Our collaboration in this project has been a win-win situation for us all and is a good model for further work together.


1. Patrick Schnarrenberger, a graduate student, was the research assistant for both the pilot project and the full research project. He reviewed this article and gave many valuable suggestions to make it clear and easy to understand.

(back) (contents) (next)