Medication calculation skills of the medical-surgical nurse
Denise A. Ashby
Medication administration is one of the most time-consuming aspects of nursing practice. Expertise in medication calculation and administration is essential to the treatment of all patients; however, many nurses experience difficulty when calculating medications. In this study, 56.4% of nurses could not calculate medications correctly in 90% of the problems, suggesting the need for regular self-testing of medication calculation skills. Continuing education programs implemented for identified medication calculation errors influences nursing practice and patient care outcomes.
Purpose: The purpose of this exploratory study was to assess the medication calculation skills of practicing medical-surgical nurses. Significance: Medication administration is one of the most important and time-consuming aspects of nursing practice. Expertise in medication calculation and administration is essential to the treatment of all patients; however, many nurses experience difficulty when calculating medications. Methods: The study used Knowles Andragogical Theory of Adult Learning as a theoretical framework. A convenience sample of 62 practicing medical-surgical nurses was drawn from a large Midwestern hospital. Medication calculation skills were measured by the Bayne-Bindler Medication Calculation Test (odd-even split half reliability 0.82) (Bayne & Bindler, 1988). Results: The study found that 56.4% of the nurses could not calculate medications correctly in 90% of the problems using the Bayne-Bindler Medication Calculation Test. Medical-surgical nurses made significantly more errors calculating intravenous medications than when calculating oral, intramuscular, or subcutaneous medications (T = 3.01, p = .004). Educational preparation and number of years practiced by the nurses did not significantly influence their scores on the Bayne-Bindler Medication Calculation Test. Implications for Nursing Practice: The medication calculation skills of nurses should be routinely tested. When necessary, subsequent remediation or skills enhancement at regular intervals prior to mandatory medication calculation skill testing are potential ways to improve nurses’ performance. In addition, periodic continuing education programs are necessary for those nurses who need help calculating IV dosages. Medication errors are a major problem in many hospitals and are associated with excess cost to the entire health care industry and undesirable patient outcomes. Experts estimate the number of medication errors in the United States’ 7,000 hospitals could be as high as 15 billion a year (Medication Errors, 1992). Reports of drug errors on combination medical-surgical nursing floors have documented medication errors ranging from 1.6% to 38.5% (Girotti, Tierney, & Brown, 1988).
Cohen, Senders, and Davis (1994) contend that medication errors result from knowledge deficiencies, performance deficiencies, or both. Bindler and Bayne (1984) found that a substantial number of nursing students failed a basic mathematics proficiency test. Since medication administration is a major responsibility of the registered nurse, this raises concern that practicing medical-surgical nurses may lack the mathematical skills for safe medication administration and demonstrates the need to evaluate nurses’ mathematical competency skills for medication calculations. The purpose of this study was to assess the medication calculation skills of practicing medical-surgical nurses in one hospital.
As early as 1962, studies have been published addressing the problem of nurses committing medication error (Barker & McConnell, 1962). Barker and McConnell (1962) observed a small convenience sample of nine nurses as they prepared and administered medications. They noted no significant differences in medication errors made on different services or shifts and no relationship between the nurses’ medication workload and medication error rate. However, their study demonstrated that the average nurse made approximately one medication error for every six medications given (18%), suggesting that the incidence of medication errors was more significant than previously understood.
Several studies have focused on medication errors as they relate to math and dosage calculation problems (Bindler & Bayne, 1984, 1991; Blais & Bath, 1992; Perlstein, Callison, White, Barnes, & Edwards, 1979; Shockley, McCurn, Gunning, Gravely, & Tillotson, 1989; Trimpke & Janney, 1981).
Perlstein et al. (1979) studied nurses in a pediatric intensive care unit to determine computational errors that may result in harmful or fatal drug doses. Of the 95 nurses observed, 31 had more than 1 year professional experience. The computational skills of the experienced nurses were no better than those of the inexperienced nurses. Although the experienced nurses had a greater tendency to feel certain in making judgments about the appropriateness of ordered doses, these judgments were frequently wrong. Bayne and Bindler (1988) examined medication administration as an important nursing function requiring skillful performance. Their study focused on the ability of nurses to accurately calculate medication doses. They tested 62 nurses and found that only 35% of these nurses could achieve a score of 90% or higher on a 20-item medication exam. Only 75% of the nurses could attain a score of 70% or better. Questions assessing knowledge about the intravenous (IV) route of administration had the highest percentage of error.
Blais and Bath (1992) used a 20-item medication dosage calculation exam to study junior level nursing students and found that only 7 of the 66 students achieved 90% or higher on the exam. Each answer was analyzed for three types of errors: conceptual (set up the problem correctly and select the proper form of administration), mathematical, and measurement (metric and apothecary). Those problems which required the nurses to set up the problem in order to solve it had the highest error rate (77%). Thus, the problem of incorrect medication calculation by RNs and students is a serious concern.
Given the increased demands in terms of roles and responsibilities in medical-surgical nursing, the author was interested in examining accuracy of medication calculation skills. By identifying medication calculation errors, continuing education (CE) programs would be initiated, which in turn would have a positive impact on nursing practice and the outcome of patient care. In addition, it was hoped the study would add to the body of existing nursing knowledge about the usefulness of the Bayne-Bindler Medication Calculation Test (1988).
The author conceptualized the study to answer the following 12 research questions:
1. What is the percentage of practicing medical-surgical nurses who lack the skills necessary to calculate medications correctly in 90% of the problems on a standardized test?
2. Do medical-surgical nurses make significantly more errors in calculating IV medications than those of other routes of administration?
3. Do medical-surgical nurses make significantly more errors in medication calculations when a common conversion formula is not provided?
4. Do medical-surgical nurses make significantly more errors in medication calculations when a problem requires more than one calculation?
5. Do medical-surgical nurses make significantly more errors in medication calculations when the calculation involves the use of fractions or decimals?
6. Does the number of years as a medical-surgical nurse significantly influence medication calculation scores?
7. Do medical-surgical nurses who have worked 1 to 3 years make significantly less medication errors than those who worked greater than 16 years?
8. Do baccalaureate nurses make significantly less medication errors than those prepared in associate or diploma programs?
9. Do medical-surgical nurses who have experience in giving IV medications within the last 3 years make significantly less errors when calculating IV dosages than those nurses without this experience?
10. Do medical-surgical nurses have accurate perceptions of their own medication calculation skills?
11. Is medication calculation and administration a stress-producing task for medical-surgical nurses?
12. Can a typology of medication errors be identified?
A sample of 100 practicing medical-surgical nurses from a 380 bed hospital in the Midwest were approached to participate in the study. Over a 6-week period, 62 (62%) of these nurses consented to participate in the study and became study subjects. All subjects signed a consent form to participate in the study.
The sample was overwhelmingly white, and all respondents were female. The respondents ranged in age from 20 to over 60 years, with the majority of the respondents being between 30 to 40 years of age. Thirty-one (50%) held baccalaureate degrees, 23 (37%) associate degrees, and 8 (13%) diploma degrees. The majority (37) were full-time employees; however, 19 (31%) were part-time employees, and 6 (10%) worked on an as-needed basis. Practice experience ranged from 1 to 16 years, with the majority of the nurses (18) having practiced 16 or more years. Forty-four (77%) of the nurses had experience in giving IV medications within the last 3 years. Fifty-eight (93.5%) of the nurses had practiced nursing within the last year and were responsible for medication administration.
Age, race, level of education, years of practice, and perception of the nurses’ medication skills were assessed by the subjects’ responses to items on a selfreported demographic questionnaire.
Medication calculation skills were measured by the Bayne-Bindler Medication Calculation Test (1988). The Bayne-Bindler Medication Calculation Test is a 20-item test with each answer weighted five points for a total possible score of 100. Bayne and Bindler reported a high level of reliability (.82) when an odd-even split half test of reliability was performed. Content validity was established by review of nursing and pharmacology text books and use of a group of nursing experts.
Subjects were approached in individual unit meetings for participation in the study. After the study was explained, each participant received a packet containing a demographic questionnaire and the medication calculation test. Half the packets contained a conversion table and were identified by a pink dot located in the upper right hand corner of the packet. The other half contained no conversion table. The subjects were told they did not have to complete any questions which made them feel uncomfortable, or ones they did not understand. Subjects were asked to complete the questionnaire and medication calculation test within 48 hours and return to the researcher unmarked in the envelope provided via interdepartmental mail. The return rate was 62%.
The primary question posed in this study was: How many practicing medical-surgical nurses lacked the skills necessary to calculate medications correctly in 90% of the problems? Test scores on the Bayne-Bindler Medication Calculation Test ranged from 45% to 100%. The median score on the test was 85% and the mean was 82%. Twenty-seven (43.5%) of the nurses attained a score of 90% or more on the medication calculation test. Thirty-five (56.4%) of the nurses lacked the skills necessary to calculate medications correctly in 90% of the problems.
T-tests for paired samples were calculated to determine if medical-surgical nurses make more errors when calculating IV medications than those of other routes. There was a statistically significant increase in calculating IV medications by the medical-surgical nurse than when calculating those medications by other routes (t = 3.01, df = 61, p = .004).
Fifty-percent of the respondents had access to a conversion table and 50% did not have access to a conversion table. There was no statistically significant difference noted between the two groups, and thus no statistically significant increase in medication errors when the conversion formula was not provided to the nurses. There were no significant differences in the number of medication errors made by the nurses when problems required one or more calculations or when the problems involved fractions or decimals.
Years of experience as a medical-surgical nurse were coded into six categories (1-3 years; 4-6 years; 7-9 years; 10-12 years; 13-15 years; [is greater than] 16 years). It is interesting to note that the medical-surgical nurses with 13 to 15 years of practice had the highest test scores, while medical-surgical nurses with 1 to 3 years had the lowest test scores. However, an analysis of variance indicated that there were no statistically significant differences among the mean scores on the Bayne-Bindler Medication Calculation Test by years of practice as a medical-surgical nurse (F = .473, df = 5, p = .795). As years of practice as a medical-surgical nurse increased, the number of medication errors did not decrease as might be expected.
Educational preparation was not significantly related to the number of medication errors. Furthermore, nurses who had experience in giving IV medications within the last 3 years did not make significantly fewer errors when calculating IV doses than those medical-surgical nurses without this experience.
Respondents were asked to rate their perception of their medication skills as above average, average, or below average. Almost 17% of the nurses rated their perceived skills as above average; however, 43.5% scored 90% or more on the medication calculation test. The findings suggest that some of the nurses may have lacked self-confidence in their ability to perform well on a written medication calculation test, when in actuality they can perform well. There was no statistically significant difference noted between the medical-surgical nurses’ perception of their medication calculation skills and their actual test scores. Fifty-seven percent of the medical-surgical nurses reported that medication calculation and administration was a stress-producing task.
In this study, 56.4% of the nurses could not calculate medications correctly in 90% of the problems as contrasted to Bayne’s and Bindler’s findings of a 35% failure rate. These results provide further support for previous findings that many nurses lack the basic skills necessary to calculate medication problems correctly an acceptable percentage of time.
The study found the majority of practicing medical-surgical nurses made significantly more errors when calculating IV medications (M =.0822). Again, this finding supports Bayne’s and Bindler’s studies (1988, 1991), which also found a significantly higher mean error rate by route of administration for IV medications (M = 4788). This finding is not surprising because the nurses in this particular institution were not required to calculate IV medications or drip rates on a daily basis. All medications were prepared, calculated, and sent to the unit via the pharmacy and administered via IV pumps.
The study indicated there was no statistically significant difference in error rate between the group of nurses who received a conversion table and the group which did not receive a conversion table. These findings support Bayne’s and Bindler’s (1988) earlier work and are in contrast to their later study (1991), which found questions that required nurses to know conversions that resulted in significantly more errors (43.9%) than those which did not. The findings from this study suggest that practicing medical-surgical nurses lack the knowledge and skills necessary to accurately use correct formulas to solve medication calculation problems, with or without the provision of conversion tables. It appears that most medication calculation errors are related to nurses’ lack of knowledge and skills to formulate and solve problems. The need for institutions to provide testing and inservice programs to teach nurses how to use formulas and solve problems with or without the availability of a conversion table is substantiated.
The findings demonstrated that the scores on the medication examination were not significantly related to the number of years the medical-surgical nurse had practiced. This supports previous research (Bayne & Bindler, 1988; Perlstein et al., 1979). Fuqua and Stevens (1988) found that medication errors decreased with nursing experience and increased with the number of distractions when calculating and administering medications. Thus, other variables related to the conditions surrounding medication calculation and administration may play an even more important role in medication errors. With the numerous changes occurring on medical-surgical units currently, the effects of situational and environmental factors on nurses’ job performance, especially in the area of medication calculation and administration, warrant further study.
A surprising finding was that as years of nursing practice increased the number of medication errors did not decrease. Experienced nurses did not make less errors, but had a greater tendency to feel certain about their calculations, even though these were wrong. The judgmental uncertainty of the inexperienced nurse may, therefore, serve as a protection against the incorrect computation of medication doses assuming the uncertain nurse seeks consultation before drug administration (Perlstein et al., 1979). Knowles (1970) noted that adults tend to be more closed to changing ways of doing things which have been working for them in the past. The fact that this institution used a unit dose system and the availability of a pharmacist to answer or double-check medication computations may have limited the nurses’ ability to calculate their own medications, thereby, reducing these skills. By the same token, the availability of these resources should provide nurses who question the accuracy of this calculation with a comfortable way to identify their uncertainty and obtain accurate information. Thus, the thinking that unit-dosing and unit-based pharmacy services protect against medication errors may be flawed. The role of multidisciplinary services in medication ordering, calculation, preparation, and administration obviously needs further investigation. This is particularly apparent as new, nonprofessional workers take on some of these responsibilities.
Educational preparation was not significantly related to medication errors. Nurses with experience in giving IV medications within the last 3 years did not make significantly fewer errors than nurses without this experience. It was a surprise to learn that as years of experience in giving IV medications increased, the number of errors did not decrease. The study suggests the need to periodically test nurses’ ability to calculate IV medications and to offer continuing education (CE) programs to those nurses who need help with IV dosage calculations. CE programs should also be offered to those nurses who do not perceive their medication calculation abilities to be acceptable in order to increase and promote self-confidence and appropriate double-checking.
Nurses are keenly aware that they serve “as the final checkpoint in the system” for administering medications (Adams & Burleson, 1992). Nurses rely on professional and personal rituals, hospital policies and procedures, action plans, as well as advice from their colleagues to prevent them from making errors (Keill & Johnson, 1993). At times these preventive strategies fail, leaving nurses distressed by their mistakes and searching for additional tactics to prevent future errors.
It should be of major concern to nurse educators, nurse managers, and nurses in general that 56.4% of the nurses studied in this sample were unable to calculate medications at a 90% level of proficiency and that 19.4% of the test scores were below the 70% level. There is an urgent need for the continual evaluation of the basic mathematical skills of all practicing medical-surgical nurses. Based on nurses’ self-assessment of needed review, opportunities for self-testing of medication calculation skills with subsequent remediation or enhancement should be offered at regular intervals. These opportunities should be afforded to the nurses prior to mandatory medication calculation skill testing.
Practicing medical-surgical nurses require expertise in both drug knowledge, as well as the ability to calculate medication doses. It is important for nurses to understand the formulas for “setting up” calculation problems to assure the correct answer, especially when these skills are not used on a daily basis. Educators should provide yearly inservices on drugs and medication calculations, and health care institutions should provide readily accessible conversion tables, reference materials, as well as charts of common drugs and drug interactions. Nurses should be encouraged to double-check their computations with a colleague, especially those which may result in high error rate, such as IV medications. How nurses use pharmacy services and other resources to prevent medication errors is poorly understood and warrants further study. A potential check in the drug administration system is possibly poorly utilized.
This research project identified the need for the medication calculation skills of practicing medical-surgical nurse to be evaluated similarly to other skills, such as cardiopulmonary resuscitation (CPR). By improving nurses’ medication calculation skills, an important source of unnecessary cost and patient suffering may be substantially reduced.
Adams, T.D., & Burleson, K.W. (1992, June). Continuous quality improvement in medication error reporting system. P & T, 943-951. Barker, K., & McConnell, W. (1962).
Barker, K., & McConnell, W. (1962). Detecting errors in hospitals. American Journal of Hospital Pharmacy, 19, 361-369.
Bayne, T., & Bindler, R. (1988). Medication calculation skills of registered nurses. The Journal of Continuing Education in Nursing, 19(6), 258-262.
Bindler, R., & Bayne, T. (1991). Medication calculation ability of registered nurses. Image, 23(4), 221-224.
Bindler, R., & Bayne, T. (1984). Do baccalaureate students possess basic mathematic proficiency? Journal of Nursing Education, 23(5), 192-197.
Blais, K., & Bath, J. (1992). Drug calculation errors of baccalaureate nursing students. Nurse Educator, 17(1), 12-15.
Cohen, M.R., Senders, J., & Davis, N.M. (1994, February). 12 ways to prevent medication errors. Nursing 94, 34-42.
Fuqua, R., & Stevens, K. (1988). What we know about medication errors: A literature review. Journal of Nursing Quality Assurance, 3, 1-17.
Girotti, M.J.T, Tierney, M.G., & Brown, S.J. (1988). Medication administration errors in an adult intensive care unit. Heart and Lung, 16(4), 449-453.
Keill, P., & Johnson, T. (1993). Shifting gears: Improving delivery of medications. Journal of Nursing Quality Care, 7(2), 24-33.
Knowles, M.S. (1970). Gearing adult education for the seventies. The Journal of Continuing Education in Nursing, 1(1), 11-16.
Medication errors: High liability and price for the hospital. (1992). Hospital Risk Management, 14(10), 129-134.
Perlstein, P., Callison, C., White, M., Barnes, B., & Edwards, N. (1979). Errors in drug computations during newborn intensive care. American Journal of Diseases in Children, 133, 376-379.
Shockley, S.J., McCurn, W.C., Gunning, C., Gravely, E., & Tillotson. (1989). Effects of calculator use on arithmetic and conceptual skills of nursing students. Journal of Nursing Education, 28, 402-405.
Trimpke, J., & Janney, C. (1981). Teaching drug dosages by computer. Nursing Outlook, 29, 376-377.
Denise A. Ashby, MSN, RN,C, is Director of Medical-Surgical Nursing, Columbia Suburban Hospital, Louisville, KY.
COPYRIGHT 1997 Jannetti Publications, Inc.
COPYRIGHT 2007 Gale Group