Home > RNJ > 2007 > November/December > Adherence to Therapy: Using an Evidence-Based Protocol

Adherence to Therapy: Using an Evidence-Based Protocol
Linda A. Moore, EdD APRN BC ANP/GNP MSCN Michael D. Kaufman, MD Robert Algozzine, PhD Nikki Irish, MSN APRN BC ANP Mary Martin, LPN Carol Rosser Posey, MSN APRN ANP

The number of patients receiving injectable medications has increased significantly during the past few years. Today, patients with hepatitis, rheumatoid arthritis, and multiple sclerosis are added to the list of those, namely diabetics, who have been instructed in self-administration of injectable medications. Currently, some of these medications create significant skin site reactions, and patients tend to discontinue the medications without informing the healthcare provider. Determining the problem and developing a research study that provides evidence to demonstrate methods to help patients adhere to agreed-upon treatment modalities can be accomplished within the clinical practice setting. This study provided a method to decrease skin reactions with interferon 1-b injections for multiple sclerosis patients and has been continued as a method with other like medications.

Patients with multiple sclerosis (MS) have numerous issues related to disability, medication management, and quality of life. In addition to the symptoms (i.e., fatigue, tremors, incontinence, weakness, pain) that patients with MS experience, they are also asked to take medications on a long-term basis that will not cure the disease and may actually make them feel worse (Compston et al., 1999). The medications are injectable, either subcutaneously or intramuscularly and, depending on the medication, are taken either daily, every other day, three times per week, or weekly.

Adhering to a medical regimen continues to be a major concern for healthcare providers of MS patients, particularly since research has shown that early and continuous therapy reduces the number of brain lesions and relapses by approximately 30% (Compston et al., 1999). It has been documented that an average of 50% of patients nationally are discontinuing the injections by the end of the first year (Namey, 2001). Adherence with therapy focuses on fostering self-care, in which persons with chronic disease are active participants in the collaborative plan of care (Namey, 2001). The concept of adherence requires education, mutually established realistic goals, and developing a mutual plan to accomplish these goals. Previous researchers (e.g., Mohr et al., 1996) supported this concept of a collaborative plan of adherence to treatment based on expectations of therapy. With an option of four medications that help slow the progression of the disease but do not cure it, patients must consider these with recommendations from the healthcare provider to determine which treatment option is going to provide the best option and can be managed for adherence to treatment (Namey, 2001).

To help patients adhere to therapy, the healthcare provider must actively listen to the patient, and then they both can jointly determine the barriers to therapy that the patient is experiencing. A major barrier for patients with MS is injection site reactions from the medications used to slow the progression of the disease process. Reactions include minor redness to necrotic tissue.

Betaseron® (interferon beta-1b) is one of three interferon therapies, administered every other day, used in the treatment of MS to slow the progression of the disease process. However, it has a significant side effect of injection site reactions (Halper, 2002). In our experience, patients discontinue or reduce the number of injections of Betaseron® due to injection site reactions, especially those associated with skin necrosis.

Previous research with interferon beta-1b injections suggested low body mass and female gender may correlate with injection site reactions (Gaines & Varricchio, 1998). Walther and Hohlfeld (1999) reported that a higher proportion of adverse reactions were associated with injections in areas having less subcutaneous fat, such as the arms or thighs, as opposed to the buttocks or abdomen. Earlier research found that risk factors for skin site reactions included incorrect injection techniques, cold solution, repeated use of the same injection site, exposure of recent injection sites to sunlight or ultraviolet rays, and insufficient needle length to reach the deep subcutaneous tissue (Knobler, Kelley, Trantas, Webster, & Lublin, 1994). Other researchers (e.g., Inafuku, Kasem, Nagata, & Nonaka, 2004; Samuel & Lowenstein, 2006) found that injection site reactions resulted from medication penetration into the dermis. Skin site reactions are very worrisome and must not be neglected (Halper, Harris, & Machler, 2005).

At the time of this study, existing recommendations by Berlex pharmaceuticals for injection of Betaseron® included warming the refrigerated medication at room temperature for at least an hour prior to injection, using a dry needle either by changing the needle or using the vial adapter when mixing the medication, rotating sites, changing needle length, and using ice before and after injections. Despite adherence to this protocol, patients continued to report skin site redness with injections of Betaseron®. The authors then proposed an alternative method to decrease injection site reactions based on the researchers’ past experience with other medications and from injection technique methods described by Kozier, Erb, and Blais (2000). The purpose of this study was to compare injection site reactions in a control and experimental group using a modified injection technique to decrease injection site skin reactions.

Using the of 0.1 ml “air bubble” injection technique in the administration of Betaseron®, the hypotheses tested were the following:

  • There will be decrease skin site size of redness as compared to baseline and to the control group.
  • There will be no difference in injection site reactions based on anatomical location of injections.
  • There will be an increased temperature at injection site at 12 hours following injections.

A secondary aim of the study was to assess participant adherence to medication protocol when using the air bubble technique.

After following the recommendations from Berlex and after having the exact medication in the syringe, 0.1 ml of air is added to the syringe before injecting the medication. The method allows the medication to be delivered through a dry needle technique, which prevents leakage into the skin and surrounding tissues at the injection site. Since some of the air remains at the plunger during the injection process and is last to enter the needle, this allows for a dry needle during withdrawal of the needle and prevents a siphoning effect of liquid as the needle is withdrawn.


This study used a two-group crossover design. Forty subjects were recruited through the newsletter of the Mid-Atlantic Chapter of the National Multiple Sclerosis Society, local neurologists, and nurse practitioners. Inclusion criteria were participants who had skin site reactions from Betaseron® and who agreed not to use an auto-injector.

In a prepilot study, several subjects tried the addition of the air bubble. The reduction in skin reaction both in size and duration led the principal investigator (PI) to anticipate that the study would result in a large effect size. The estimated sample size was 40 participants. A total of 46 participants were recruited. Over the course of the study, two participants stopped for personal reasons, leaving a usable sample of 44. A post-hoc power analysis was conducted to determine the actual power of the study. Based on the calculated effect size (2.31) that assessed the magnitude of mean difference in skin redness between the experimental and control group, a one-tailed test and a significance level of .05, the actual power of the study was .98 (Glantz, 2005; Kraemer & Theimann, 1987), indicating the study was adequately powered.


After receiving institutional review board (IRB) approval, subjects who consented to participate in the study were randomized by flip of a coin into the experimental or control group until the control group reached 10 participants for the purpose of controlling for Hawthorne effect. Regardless of group assignment, once subjects were enrolled, they received weekly home visits from the nurses who provide care to MS patients. At the first visit, several activities occurred. The nurse asked participants to demonstrate how they prepared the medication for administration. Remedial instructions were then given by the nurse to match the recommended injection procedure as described in the Betaseron® package insert. This ensured that skin site reactions were not related to individual injection technique. Participants in the experimental group were instructed on the safety and skill of inserting 0.1 ml of air into the syringe and ensuring that the air bubble rose to the top of the syringe next to the plunger prior to the injection. The addition of the air bubble was the only modification made from the existing Berlex technique. To ensure equivalency at baseline, the nurse administered the first injection to participants in each group. In addition, the nurse measured the size of skin redness from the previous injection. Skin redness was measured using centimeter measures. Skin temperature was measured using the Mallenckrodt skin sensors.

During the initial visit, participants were interviewed for symptoms associated with injections, completed demographic information including a medication history, and provided information on height and weight. The nurse also calculated body mass index (BMI) using skin fold calipers.

The nurse continued to visit over the next 4 weeks to observe the individuals giving their own injections, measure the previous injection, and evaluate completion of diary that included the information described as potentially important to skin site reactions (Gaines & Varricchio, 1998; Walther & Hohlfeld, 1999). At each visit, ratings of skin redness by the nurse and the participants were compared.

Participants recorded injection location, measurement of redness, swelling at the site, any other symptoms noted with injections and temperature of the skin prior to injections, at 1–2, 12, and 24 hours after injection, using Mallenckrodt skin temperature sensors.

Prior to the beginning of the study, the PI met with each of the nurses to review the protocol for data collection and using the air bubble technique. Each nurse demonstrated measurement of redness and technique of injections using a model example. Interrater reliability was 96% agreement for measurement and data collection of skin temperatures. When percent agreement was evaluated between participants in either group and the nurse, there was a minimum of 90% agreement between them. Between visits, participants continued to administer the injections and record location of injection, redness size, symptoms, and skin temperature in the diary. After the initial 4 weeks, intervention participants ended their involvement. However, participants in the control group then crossed over to the experimental protocol. During this phase, the same protocol was followed as described above with the control, now experimental subjects. The data were then entered into SPSS statistical package for analyses.


A total of 11 control subjects and 33 experimental subjects, including 2 men (the ratio of MS in women to men is approximately 4 to 1), completed the protocol. Participants ranged in age from 24 to 57 with a mean age of 41 years. The BMI was calculated based on height and weight of the participants. The BMI ranged from 18.8% to 51.7% with a mean of 27.5%. There were 59% of the participants with a BMI that exceeded guidelines for upper limit of ideal BMI (27.8 for men and 27.3 for women) (Seidel, Ball, Dains, & Benedict, 1999). There were no statistically significant differences in the injection site reactions when Betaseron® was administered by the nurse at baseline as compared to when administered by the participants.

As previously described, during the initial visit by the nurse and prior to beginning the experimental technique, participants were measured for actual redness from the previous injection. The size of the redness prior to beginning the experimental technique ranged from 2 to 6 cm with a mean of 3.3 cm. Most participants complained that the redness was “very angry looking,” indicating that the redness was deep red, but without pain. In their description of previous symptoms with injection, they reported that the redness lasted for weeks, making it difficult to find a site for the next injection (injections are given every other day). Seven subjects reported previous necrotic sites.

Independent t tests compared the control group with the experimental group in terms of measurement of redness in centimeters (cm). There was a statistically significant difference between the groups (t = 4.35, df = 10.68, p = .001) as shown in Table 1. The experimental group included two persons who said they forgot to add the bubble on several occasions. These were not deleted from the experimental results because they had not recorded the exact date of forgetting to use the air bubble.

When the two groups were compared for differences in skin temperature, they were no statistically significant differences. When skin temperatures were taken at the same place on both sides of the body simultaneously, temperatures varied from one side of the body to the other side, as do many other MS symptoms. It is noted that a rise in temperatures did not follow the expected trend except for a slight increase at the 12-hour interval; however, this trend was not statistically significant, as previously reported (Moore, 2003).

The crossover participants were analyzed separately, comparing their mean redness during their participation as control and experimental participants. The injection size redness was greater during the control phase as compared to the experimental phase (t = 4.59, df = 8, p = .002), as shown in Table 2.

Site redness was then compared with location of injection, because previous research suggested that specific locations would have a higher incidence of skin site reactions (Walther & Hohlfeld, 1999). The injection site redness was descriptively compared across injection sites (Figure 1). The variation in redness across sites for the participants in the experimental group was minimal. However, there was greater variation in redness among the participants in the control group. Anecdotally, one participant in the control group reported the abdomen had the worst redness, and for several other others, it was the arm or leg. After the intervention, the redness in sites became more consistent across all areas instead of being specific to any one area. It is interesting to note that the right side of the body sites, except for the right arm, tended to have increased redness compared to the left side of the body sites. Information regarding right- versus left-handedness was not recorded to determine if there was a correlation between handedness and site reaction. However, recall of data from the nurses indicated that all were right-handed; therefore no data could be analyzed regarding right- versus left-handedness.

To determine whether body mass had any effect on injection site reaction, skin folds were measured three times, then averaged. There was no statistically significant difference between the control and experimental group in reactions based on skin fold analysis and BMI.

Participant Comments

To assess whether participants reported greater adherence to the medication regimen, participant comments were elicited. Soon after the trial ended, one participant wanted to test if the air bubble could be added to the syringe when using the auto-injector. After the study was completed, the participant added the air bubble and then used the auto-injector. It was determined that all medication was injected and the air bubble did not create any problem for the MS patient or the auto-injector delivering the medication. This participant noted to the researcher that she has continued to use this technique for more than 6 months with excellent results. Others have also reported that they used the air bubble after the study with the auto-injector and found the results to be very satisfactory.

The PI telephoned the participants approximately 6 months after the intervention and found that only one person had discontinued using the air bubble. She had difficulty managing syringes and could not administer her own injections. The remaining participants continued with the air bubble and indicated how satisfied they were at the change in the redness from their injections. As noted in the findings, not all redness was eliminated. The redness was decreased, and no necrotic areas occurred during the study or were reported later at 6 months.

Participants’ comments about the new technique include the following:

  • The deep redness became very light pink.
  • The site reactions lasted only 2–3 days instead of 1–2 weeks.
  • African Americans reported more bruising than Whites. However, the limited number of African Americans in the study prevented analysis.
  • The ice recommended by the Betaseron® nurses, before and after, was too much trouble. More than one third of the participants discontinued the ice with no change in the redness while still using the air bubble technique.
  • There were no necrotic areas after beginning the air bubble technique.
  • Several participants noted severe reactions when they forgot the air bubble during the study and afterward during follow up conversations. They commented on how this reminded them of the difference in the reactions.
  • Several people had tried just changing needles before injection with little success.
  • At least two people reported their intention to discontinue the injections because of skin site reactions, but because of the new injection technique, they planned to adhere to their medication schedule. One noted that her husband had her stop her medication for almost a year due to the skin site reactions before trying the new technique. Even he was pleased with the improvement in skin site reactions with the air bubble technique.
  • Prior to the study, several patients were taking their injections inconsistently because of skin site reactions. They would delete as many as 4–6 injections per month without informing their healthcare provider. At follow-up, they reported consistence in adhering to their treatment regime.


Adherence to medication therapies is particularly challenging for persons with MS. Even though they followed the recommended guidelines for administration of Betaseron® including the use of the auto-injector, they continued to have injection site reactions. In this study most participants had abandoned the use of the auto-injector even though it had helped decrease the reactions slightly. The auto-injector allows patients a method of injecting without seeing the needle at the time of injection and to maintain the same depth of injections, especially for those who find it difficult to “stick” themselves with a needle. During the study, participants were asked to not use the auto-injector because this had already been noted to not eliminate injection site reactions in these participants, and there was no evidence that the auto-injector could accommodate the additional air bubble.

This study found important, both statistically and clinically, that injection of 0.1 ml of air decreased skin site reactions. However, unlike previous reports (Gaines & Varricchio, 1998; Walther & Hohlfeld, 1999), this study found that neither BMI nor injection site location influenced the size of skin site reactions.

Evidence from this study indicates that participants continued their injections following the method for decreasing skin site reactions by injecting 0.1 ml of air into the syringe. Even participants who had some slight injection site reactions found these more tolerable than the previous large “angry, red” areas. Since adherence to therapy is a collaborative process between the healthcare system and the patient, determining methods to help patients solve issues such as injection site reactions is essential to success of therapy. In this particular disease process, the adherence to therapy is more likely to decrease progression of the disease process of this incurable, chronic, progressive disease.

Recommendations and Nursing Implications

The nurse working with MS patients or any patients receiving injectable medications should assess patients for site reactions. Creating the dry needle with the air bubble technique of using 0.1 ml of air is a simple technique to decrease skin site reactions. Anyone who has had necrotic areas should be evaluated for injection technique and then correctly evaluate the use of the air bubble.

In-service programs are needed for nurses working with patients receiving injectable medication to teach them the importance of follow-up with patients on this medication and the step-by-step approach to managing site redness and necrosis.

Patients should be asked to describe how they are doing injections to determine correctness of injection. As noted by Halper (2002), more than one visit is necessary to provide understanding and consistent technique. Healthcare providers may believe that the initial instruction in technique is all that is needed. Managed care does not pay for the extra time required to follow all patients and their injection techniques. Follow-up would be a major asset for increased adherence, and several of the pharmaceutical companies have recognized this need for adherence and follow-up and now provide calling service and, if needed, additional visits for training.

The pharmaceutical companies have helped address issues of persons with MS by either having prefilled syringes with correct diluents with new adapters for mixing or prefilled ready-mixed syringes. After noting that 25% of the participants in both the control and experimental groups were mixing the medication incorrectly, this is a major improvement in dosing correctly. This new prefilled diluents syringe is especially important for those patients who have tremors or visual disturbances. Having to measure accurately more than one thing to prepare the medication increased the risk of error. The authors found more than one individual with visual disturbances could not read the directions and figure out what to do with the new techniques. Other injectables for MS have now changed their packaging instructions to state that persons should keep the air bubble that is present in the prepackaged syringe (2007 Physicians’ Desk Reference, 2006). Careful instruction is still needed to be sure the air bubble is at the plunger end of the syringe during injection.

Pharmaceutical companies who change their packaging and method of administration need to contact all healthcare providers prescribing these medications prior to the release of the new techniques. They need to provide instruction with demonstration so that the right questions can be asked of the patient to determine possible errors in administration and to enhance patient adherence.

Persons with chronic illnesses continue to have difficulty with treatment adherence, either leading to frequent missed medications or stopping the medication altogether, especially those requiring injectables. Nurses have a unique position of being the first and last line of communication in the healthcare system. Using evidence-based practice, nurses can contribute greatly to patient adherence for long-term therapies using injectable medications for other chronic diseases (i.e., rheumatoid arthritis, hepatitis) as has worked with the MS patients.


The researchers wish to thank Berlex for making this research possible and to the Nellcore corporation for providing the skin temperature monitors. Thanks also go to Mary Anthony, PhD, Kay Hough, RN BSN, and Jackie Dienemann, PhD RN, for editorial assistance in the development of this manuscript.

About the Authors

Linda A. Moore, EdD APRN BC ANP/GNP MSCN, is an associate professor and nurse practitioner at UNC Charlotte, College of Health and Human Services & the Multiple Sclerosis Center, Carolinas HealthCare System in Charlotte, NC. Address correspondence to her at lmoore@uncc.edu.

Michael D. Kaufman, MD, is a medical director in the MS Center, Carolina Healthcare System, Charlotte, NC.

Robert Algozzine, PhD, is a professor at UNC Charlotte School of Education.

Nikki Irish, MSN APRN BC ANP, is a nurse practitioner and professional educator.

Mary Martin is an LPN at Raleigh University.

Carol Rosser Posey, MSN APRN ANP, is a nurse practitioner at Advanced Occupational Health Services.

Funded by Berlex Grant and support with temperature monitors from Nellcore.


Compston, A., Ebers, G., Lassmann, H., McDonald, I., Matthews, B., & Wekerle, H. (1999). McAlpine’s multiple sclerosis (3rd ed.). New York: Churchill Livingstone.

Gaines, A. R., & Varricchio, F. (1998). Interferon beta-1b injection site reactions and necroses. Multiple Sclerosis, 4, 70–73.

Glantz, S.A. (2005). Primer of biostatistics (6th ed.). New York: McGraw-Hill.

Halper, J. (2002). Pharmacotherapeutics in multiple sclerosis and nursing implications. In J. Halper & N. Holland (Eds.) Comprehensive nursing care in multiple sclerosis. (pp. 175-182). New York: Demos Medical.

Halper, J., Harris, C., & Machler, B. (2005). Managing injection-site reactions in patients with MS. Multiple Sclerosis Counseling Points, 1, 1–10.

Inafuku, H., Kasem, K., Nagata, T., & Nonaka, S. (2004). Cutaneous ulcerations following subcutaneous interferon beta injection to a patient with multiple sclerosis. Journal of Dermatology, 31, 671–677.

Knobler, R. L., Kelley, C. L., Trantas, F., Webster, G. F., & Lublin, F. D. (1994). Interferon beta 1-b induced ulcerative skin lesions in MS. Journal of Neuroimmunology, 54, 173.

Kozier, B., Erb, G., & Blais, K. (2000). Fundamentals of nursing: Concepts, process and practice (6th ed.). Upper Saddle River, NJ: Prentice Hall Health.

Kraemer, H. C., & Theimann, S. (1987). How many subjects: Statistical power-analysis in research. Newbury Park, CA: Sage.

Mohr, D. C., Goodkin, D. E., Likosky, W., Gatto, N., Neilley, L. K., Griffin, C., et al. (1996). Therapeutic expectations of patients with multiple sclerosis upon initiating interferon beta-1b: Relationship to adherence to treatment. Multiple Sclerosis, 2, 222–226.

Moore, L. A. (2003). Correlations of injection reactions to BMI and to anthropometric measures. [Abstract]. International Journal of MS Care, 5, 114.

Namey, M. A. (2001). Promoting adherence to complex protocols. In J. Halper (Ed.), Advanced concepts in multiple sclerosis nursing care. New York: Demos Medical.

2007 Physicians’ Desk Reference. (61st ed.). (2006). Montvale, NJ: Thomson.

Samuel, L., & Lowenstein, E. J. (2006). Recurrent injection site reactions from interferon beta 1-b. Journal of Drugs in Dermatology, 5, 366–367.

Seidel, H. M., Ball, J. W., Dains, J. E., & Benedict, G. W. (1999). Mosby’s guide to physical examination (4th ed.). St. Louis, MO: Mosby.

Walther, E. U., & Hohlfeld, R. (1999). Multiple sclerosis: Side effects of interferon beta therapy and their management. Neurology, 53(8), 1622–1627.