Ung, Lerma RN, BSc(MedSc), MHSc, PhD, FRCNA; Cook, Sara RN, Grad.Dip Comm Nurs Masters Mgt, MRCNA; Edwards, Ben BA(HonsPsych); Hocking, Lynda RN Dip App Sc (Nurs), CritCareNurs, MBA; Osmond, Felicity RN; Buttergieg, Heather RN, Cert.OncNurs
Peripheral intravenous cannulation is one of the most common invasive procedures that nurses perform, 1 and it carries with it a high risk of complication. For example, phlebitis rates reported for patients receiving intravenous therapy have been as high as 80%, 2,3 with the rates in most hospitals ranging between 20% and 80%. 4–6 Other complications resulting from intravenous cannulation include thrombophlebitis, 7–11 extravasation, 7–10 and infection resulting from bacteremia and septicemia. 8–10 Patients also experience unnecessary discomfort or pain related to resting. 12 Given the many complications that can arise from peripheral intravenous cannulation, ensuring that nurses performing the procedure are competent is paramount.
In Australia and Britain, 12 peripheral intravenous cannulation is increasingly becoming a nursing responsibility. This is evident in a variety of settings such as oncology and emergency departments. However, no published empirical studies have investigated nursing competence in peripheral intravenous cannulation. Anecdotal evidence from unit managers of three Australian hospitals and overseas institutions 13,14 indicates that nurses’ competence in peripheral intravenous cannulation has an impact on the quality of patient care.
In Australia, variations exist not only in nurses’ educational preparation, but also in the type and length of nurses’ experience in settings wherein peripheral intravenous cannulation is performed. In Australia, practicing nurses may have gained their undergraduate qualifications through a 3-year hospital program, a 3-year college diploma, or a 3-year university baccalaureate degree. Although undergraduate nursing preparation in Australia has been offered only in tertiary institutions (ie, universities) since 1993, many practicing nurses still do not hold tertiary nursing qualifications.
The levels of postgraduate preparation also are varied. Some nurses have completed graduate certificate programs (1 year part-time), whereas others have completed graduate diploma programs (2 years part-time). These programs prepare registered nurses for specialty practice in areas such as palliative care, oncology, critical care, perioperative nursing, and emergency nursing. In addition, some nurses have completed master’s degree programs in nursing. Thus, in any one clinical setting, there are practicing nurses prepared at different education levels who also have varied types and lengths of clinical experiences.
The variation in nurses’ education and experience affects their performance of peripheral intravenous cannulation. The characteristics of patients undergoing peripheral intravenous cannulation also may have an impact on the success rates. Determining the impact of education and experience on nurses’ performance of peripheral intravenous cannulation is considered to play a significant role in ensuring positive patient outcomes. As of this writing, no empirical studies have investigated this issue. This article reports the findings from a study using a standardized assessment tool to investigate the impact that nurses’ education and experience and patient characteristics have on nurses’ performance of peripheral intravenous cannulation.
This study aimed to identify any relation between registered nurses’ performance of peripheral intravenous cannulation and nurses’ type and length of nursing experience, level of educational preparation, or both. The impact of patient characteristics on peripheral intravenous cannulation performance also was examined.
A correlational design using clinically based observation was used in this study. The design was selected because it allowed identification of relations between variables while the performance of participants was observed using a standard measurement tool.
Instrument: Peripheral Intravenous Cannulation Assessment Tool
At the time of the study, no standardized tools had been developed and published to measure peripheral intravenous cannulation performance. The Infusion Nurses Society (INS) had developed standards of practice for infusion nursing. 15 Dugger 16 also had developed an assessment tool using INS standards to assess generic skills and knowledge of intravenous practice. The tool covered relevant general knowledge as well as behaviors and skills in intravenous cannulation, but did not specifically focus on peripheral intravenous cannulation. Consequently, only certain parts of Dugger’s 16 tool were useful for assessing educational outcomes and the development of competency in intravenous cannulation. As Jackson 17 suggested, cannulation is a skilled technique that is best performed as a staged process. Fuller and Winn 18 stated the importance of selecting equipment for peripheral intravenous cannulation that has both the materials and design to reduce the risk of infection and ensure patient comfort.
Taking the relevant information from Dugger, 16 Jackson, 17 and Fuller and Winn 18 into consideration, the researchers defined specific criteria for measuring peripheral intravenous cannulation. This led to the development of the peripheral intravenous cannulation assessment tool (PICAT) used specifically for this study.
The PICAT was developed through consultation with a panel of peer-nominated intravenous cannulation nursing experts. Available literature also was examined, and relevant information from the INS standards 15 was taken into consideration, as well as contemporary nursing practice in Australia.
The tool consisted of behavioral rating scales that assessed nurses’ performance in
* provision of patient education
* correct choice and use of a catheter
* appropriate site selection and preparation before catheter insertion
* correct insertion technique and attachment of an IV line
Nurses were examined to determine whether they had educated the patient on the need for infusion therapy and assessed the patient’s need for the provision of holistic patient care. Nurses also were assessed on their explanation of the cannulation process.
In the area of equipment selection, nurses were assessed on their ability to select the most appropriate catheter for the intended task and the type of substance or medication to be infused.
The nurses’ site selection was assessed for appropriateness in relation to the volume of the infusion and the type of substance or medication to be infused. The method used by nurses to assess vein characteristics also was examined.
The insertion angle of the catheter was assessed, considering the type of vein to be accessed (surface, medium, or deep vein), to ensure that the angle of insertion never exceeded 30°. Secured attachment of the IV line to the patient also was assessed.
Each of these areas was rated, and an aggregate score was obtained to determine each participant’s overall rating of peripheral intravenous cannulation performance, which could vary from 0 (lowest possible score) to 18 (highest possible score).
A pilot study was conducted to ensure face and content validity of the PICAT. Two investigators conducted the pilot study, with 10 nurses participating. Among the participants, 100% agreement was reached on the appropriateness and relevance of the content for measuring performance of peripheral intravenous cannulation.
Sample and Sampling Process
The study was conducted in day oncology units and oncology wards at three Melbourne metropolitan acute care hospitals. Of all the registered nurses working in these units, 54 were able to cannulate, and therefore met the inclusion criteria. Given the small number of nurses who met the criteria, convenience sampling was used instead of random sampling. Of the 54 registered nurses, 38 consented to participate in the study, giving a recruitment rate of 70.37%. Eight eligible nurses could not be recruited because of variation in shift times during the data collection period.
A critical care nurse specialist was employed as a research assistant to observe the participants’ performance and collect the data using the PICAT. This research assistant was oriented to the research aims and given specific information relating to chemotherapy and site selection.
Together with the unit manager in each of the three hospitals, the research assistant briefed the nursing staff about the project, recruited nurses, and for those consenting, arranged appointment times for observations and collection of data. Because patients also were being observed while the nurse performed intravenous cannulation, the research assistant and consenting nurses sought patients’ permission before the observation. All the patients who were approached gave written consent.
In the participating nurses’ absence, the research assistant assessed the patient’s veins for the appropriate site or sites and the most appropriate type of catheter. On completion of patient assessment, the nurse was invited back, and observation for data collection using the PICAT was completed. In addition, data were collected on patients’ and nurses’ demographic variables. As part of debriefing, the research assistant offered feedback to each nurse participant on his or her performance.
Statistical analysis was performed using SPSS version 9.0 (SPSS Inc., Chicago, Ill). Because the ages of the patients and the number of nurses with no university degree or graduate diplomas were the only variables that differed significantly among the hospitals, analyses were pooled across hospitals. The pooling of data across hospitals also accounted for the small sample from each individual hospital: 14 from hospital A, 12 from hospital B, and 12 from hospital.
Patients’ demographics and nurses’ education and experience were used as predictors of overall intravenous cannulation ratings in a hierarchical multiple regression analysis. To control for patient characteristics, patient demographic variables were entered at step 1, with variables assessing nurses’ education and experience entered at step 2.
For further exploration of the impact that nurses’ education and experience had on peripheral intravenous cannulation performance, an analysis of variance (ANOVA) was conducted using overall ratings of peripheral intravenous cannulation performance as the dependent variable. The two independent variables were 20 or more years of general nursing experience and a graduate diploma.
The mean overall rating of nurses’ cannulation across the three hospital sites was 16.57 ± 1.08. This result suggests that there was a high standard of intravenous cannulation because the maximum possible overall rating of intravenous cannulation was 18. Means, standard deviations, numbers, and frequencies (where meaningful) for patient demographics and nurses’ education and experience summed across hospitals are presented in Table 1.
Hierarchical multiple regression was used to investigate the amount of variance in overall ratings of intravenous cannulation accounted for by nurses’ education and experience exceeding that of patient demographic characteristics. During step 1, the age, gender, first language, and day or inpatient status of the patients were entered. During step 2, the nurses’ grade (level of employment), years of experience in general and oncology nursing, and years of cannulating were entered. Also during step 2, the nurses’ levels of educational preparation were entered as follows: no tertiary qualifications, undergraduate degree, graduate diploma, or master’s degree.
The results of the hierarchical multiple regression indicated that the patient demographic variables did not significantly predict overall ratings of intravenous cannulation:F (4,26) = 1.20;P > .05 (step 1). The nurses’ years of experience and levels of educational preparation significantly predicted overall ratings of cannulation:F (12,18) = 3.97;P < .01 (step 2). These data also accounted for 54% of the variance in overall ratings of cannulation.
In step 2 of the analysis, preparation at a graduate diploma level was a significant positive predictor of overall ratings, indicating that nurses with graduate diplomas had higher overall ratings for peripheral intravenous cannulation. In contrast, experience in general nursing was a significant negative predictor of overall ratings, indicating that nurses with more general nursing experience had lower overall peripheral intravenous cannulation ratings (Table 2).
Although the years of experience in oncology nursing were significant, they acted as a suppressor variable. Suppressor variables enhance the importance of the other independent variables by suppressing irrelevant variance, and hence increasing the overall variance explained by the multiple regression. 19 Years of oncology nursing experience were identified as a suppressor variable because the bivariate correlation with peripheral intravenous cannulation performance (r = −.01;P > .05) and its beta weight (.63) in the regression equation had opposite signs. 20,21 Another indicator of a suppressor variable is its correlation with another independent variable that has a substantial correlation with the dependent variable. 22 In this case, years of oncology nursing correlated with years of general nursing (r = .65;P < .001), which had a substantial correlation with the dependent variable (r = −.42;P < .05). Consequently, the positive beta weight of years of oncology nursing in the multiple regression was an artifact of the high correlation between years of oncology nursing and years of general nursing rather than an indication that a relation existed between years of oncology nursing and peripheral intravenous cannulation.
For further exploration of the impact that nursing education and experience has on peripheral intravenous cannulation, a 2-by-2 factorial ANOVA was conducted with 20 or more years of general nursing experience and a graduate diploma as the independent variables. The dependent variable was overall ratings of peripheral intravenous cannulation performance.
As expected from the results of the hierarchical multiple regression, a significant main effect was associated with possession of a graduate diploma (F [1,33] = 8.15;P < .01) and 20 or more years of general nursing experience (F [1,33] = 7.98;P < .01). Moreover, there was significant interaction between possession of a graduate diploma and 20 or more years of general nursing (F [1,33] = 6.27;P < .05). Inspection of means indicated that nurses with 20 or more years of general nursing experience and no graduate diploma had lower overall ratings of intravenous cannulation (Χ = 14.80 ± .67) than nurses with fewer than 20 years of general nursing experience and no graduate diploma (Χ = 16.82 ± .91), nurses with 20 or more years of general nursing experience and a graduate diploma (Χ = 16.83 ± 1.04), and nurses with fewer than 20 years of general nursing experience and a graduate diploma (Χ = 16.96 ± .96).
In this study, patient characteristics were not associated with overall ratings of peripheral intravenous cannulation performance, whereas nurses’ education and experience were significant predictors of overall performance. Nurses with graduate diploma qualifications had higher performance ratings. Moreover, nurses with more experience in general nursing had lower performance ratings.
These findings can be discussed best in the context of nursing in Australia. The graduate diploma course in Australia aims to prepare nurses for clinical practice in specialized areas such as critical care, operating theater, emergency nursing, oncology nursing, and palliative care. In these courses, components of clinical practice in conjunction with theory are related to the specialty stream of study, and consequently are more tailored toward skills development for clinical practice. Therefore, graduates who have completed a graduate diploma are expected to have specialist clinical skills, advanced theoretical knowledge in their specialty practice, and research skills. A graduate diploma qualification is one of the pathways for entry into a master’s program in nursing.
The findings highlight the importance of postgraduate education with a focus on clinical practice development, and in this context, nurses with graduate diplomas. The higher overall ratings of this group suggest strong support for courses that offer both the theoretical and clinical skills development necessary for specialized practice in specific areas of nursing.
As the findings show, nursing experience in an oncology setting acted as a suppressor variable, indicating that it was not an important predictor of performance in peripheral intravenous cannulation. Despite this finding, future studies should include this variable in their analysis because suppressor variables suppress error variance in other independent variables, and hence provide a more accurate reflection of the relation between other independent variables and the dependent variable. 19 Further studies are needed to determine whether the suppressive effect of years in oncology nursing is robust in other samples of oncology nurses.
Furthermore, the findings also show that nurses with more experience in general nursing had lower overall ratings of peripheral intravenous cannulation performance. The results from the 2-by-2 factorial ANOVA also provided information about the interaction of years in general nursing and graduate diplomas on performance of peripheral intravenous cannulation. The interaction effect between a graduate diploma and 20 or more years of experience indicated that only nurses with 20 or more years of nursing experience but no graduate diploma had lower ratings of peripheral intravenous performance. These results suggest that a low performance rating in peripheral intravenous cannulation was influenced by the lack of contemporary specialist skills, and not general nursing experience in particular. Therefore, ongoing education and training to develop and upgrade skills in peripheral intravenous cannulation are important.
The nonsignificant relation between years of intravenous cannulation and higher overall ratings of cannulation performance may be a function of the frequency of intravenous cannulations performed over the years by nurses. Some of the participants may have performed intravenous cannulation infrequently over a long period. Future studies should include this variable to determine whether it has an impact on cannulation performance.
Conclusions from the current study are preliminary because these findings need to be replicated in studies on a larger scale. Future studies should include larger sample sizes of nurses with undergraduate and postgraduate educational preparation. Similarly, consideration should be given to including nurses who perform cannulation in different clinical settings such as emergency units. Future research also could include other patient characteristics that have been suggested as having an impact on intravenous cannulation such as diabetes, severe debilitation, and disease that compromises vascular integrity. 12
Despite the limited sample size, this is the first study to explore the impact of patient and nurse characteristics on peripheral intravenous cannulation performance. Very few studies examining intravenous cannulation performance have used a standardized assessment procedure. Overall, the importance of applied education, such as that indicated by graduate diplomas, in determining the performance of advanced clinical skills for specialist practice also has important implications for nursing career structures, nursing education, models of care, and hospital management.
1. Davies S. The role of nurses in intravenous cannulation. Nurs Stand. 1998; 12:43–46.
2. Peters JL, Frame JD, Dawson SM. Peripheral venous cannulation: reducing the risks. Br J Parenter Ther. 1984; 5:56–58.
3. Feldstein A. Detect phlebitis and infiltration before they affect your patient. Nursing. 1986; 16:44–47.
4. Larson E, Hargiss C. A decentralized approach to maintenance of intravenous therapy. Am J Infect Control. 1984; 12:177–186.
5. Angeles T, Barbone M. Infiltration and phlebitis: assessment, management, and documentation. J Home Health Care Pract. 1994; 7:16–21.
6. Maki DG, Ringer M. Evaluation of dressing regimens for prevention of infection with peripheral intravenous catheters. JAMA. 1987; 258:2396–2403.
7. Collignon P. Intravascular catheter associated sepsis: a common problem. The Australian study on intravascular catheter-associated sepsis. Med J Aust. 1994; 161:374–378.
8. Campbell L. IV-related phlebitis, complications, and length of hospital stay: 1. Br J Nurs. 1998; 7:1304–1312.
9. Campbell L. Intravenous cannulation: potential complications. Prof Nurse. 1997; 12:s10–s13.
10. Dougherty L. The management of intravenous therapy in palliative care. Int J Palliat Nurs. 1999; 5:177–184.
11. Todd J. Peripherally inserted central catheters and their use in IV therapy. Br J Nurs. 1999; 8:140–147.
12. Castledine G. Nurses’ role in peripheral venous cannulation. Br J Nurs. 1996; 5:1274.
13. Dougherty L. Intravenous cannulation. Nurs Stand. 1996; 11:47–51.
14. Campbell J. Intravenous cannulation: potential complications. Prof Nurse. 1997; 8(suppl):S10–S13.
15. Infusion Nurses’ Society. Infusion nursing standards of practice. J Intraven Nurs. 2000;23(6S):S14.
16. Dugger B. Competency for intravenous nursing practice. J Intraven Nurs. 1993; 16:293–298.
17. Jackson A. Performing peripheral intravenous cannulation. Prof Nurse. 1997; 13:21–25.
18. Fuller A, Winn C. Selecting equipment for peripheral intravenous cannulation. Prof Nurse. 1999; 14:233–236.
19. Smith JW. Suppressor variables in multiple regression/correlation. Educ Psychol Measure. 1992; 52:17–29.
20. Cohen J, Cohen P. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. New York: Erlbaum; 1975.
21. Darlington PB. Multiple regression in psychological research and practice. Psychol Bull. 1968; 69:161–182.
22. Collins JM, Schmidt FL. Can suppressor variables enhance criterion-related validity in the personality domain? Educ Psychol Measure. 1997; 57:924–936.