A major challenge for nursing faculty is to teach and evaluate all aspects of the HIV/AIDS curriculum, including knowledge of and compliance with universal precautions. To address this challenge, an instrument was developed and administered in 1991 and revised and administered again in 1993, The purpose and intended use of this instrument is to survey students' attitudes toward nursing HIV-seropositive patients and document their self-reported understanding of and compliance with universal precautions.
Several researchers have reported studies of health care professionals or students regarding attitudes and behaviors related to care of AIDS/HIV-infected patients (Goldenberg & Laschinger, 1991; Kelly, St. Lawrence, Hood, Smith, & Cook, 1988; Lawrence & Lawrence, 1989). However, few researchers have published brief instruments which were validated using nursing students.
One such attitude instrument (Bliwise, Grade, Irish, & Ficarrotto, 1991) was developed for use in research on students in a variety of health care professions, and it was validated using two different cohorts of medical and nursing students. That instrument contains 15 items and has a Cronbach alpha coefficient of .86. The subscales and Cronbach's coefficients for the total scale are Fear of Contagion (.65), Negative Emotions (.70), and Professional Resistance (.75). These coefficients suggest good internal consistency. However, the Bliwise, Grade, Irish, & Ficarrotto instrument deals only with attitudes and not with universal precautions.
The only instrument of known measurement properties that addresses both attitudes and compliance with universal precautions is the HTV-Impact Questionnaire (Wiley, Heath, Acklin, Earl, & Barnard, 1990). That 15-item scale was used to survey registered nurses on three subscales: Precautions, Worry and Policies, and Job Change. Cronbach's coefficient for the total scale was .61, and for the subscales, .51, .76, and .76, respectively. This instrument has the advantages of satisfactory reliability coefficients as well as the dual focus on attitudes and universal precautions.
However, Wiley and colleagues (1990) tested their instrument with a sample of registered nurses, not students. The need remained, then, for an instrument to assess the impact of AIDS/HIV infection instruction on student nurses' attitudes as well as their understanding of and compliance with universal precautions. This dual focus is important in order to explore the possibility that the primarily late adolescent student population would consider themselves invulnerable (Weinstein, 1984) and therefore be less likely to report compliance with universal precautions. The present report describes the development, measurement properties and potential uses of the second revision of the HTV-Impact Questionnaire: Student Form for measuring nursing students' attitudes about, and precautions in, HIV/AIDS patient care.
Participants - The sophomore, junior, and senior nursing classes at a mid-sized, urban Catholic university participated in this study after they had completed clinical courses for that year. About a third of the junior and senior classes were placed, during their sophomore year, on an AIDS unit for their initial clinical course. Because of the high mortality rate of relatively young persons, faculty selected students whom they believed were more paychosocially mature for placement on the AIDS unit.
Demographic data for this school indicate that 75% of the students report that they are Catholic; and 55% Caucasian. About 90% of these students are age 20 to 23. Approximately 230 students received the instrument, and 190 (83%) completed it.
Instrument - The instrument presented to the students was a 28-item scale called the HIV-Impact Questionnaire: Student Form. It was a third revision of an instrument (but the second revision of the student form) used to survey this same population in 1987 (Wiley, Heath, & Acklin, 1988). The original instrument addressed the concept of personal susceptibility (Becker, 1974), and it queried students about their perceptions of risk of becoming HIV-infected while giving patient care. That 66-item pilot instrument used several different types of response options, Students said they found it confusing and that they spent about 45 minutes completing it. Only 36% returned completed questionnaires.
The first revision of the 1987 instrument was developed to survey a registered nurse population (Wiley, Heath, Acklin, Earl, & Barnard, 1990). For the RN survey, 15 statements were selected from the original instrument, and respondents rated the items using a Likert-type response scale. The response rate to the survey was 54%; 323 nurses responded. Cronbach's coefficient for the 15-item scale was .61; three subscales were identified.
In 1991, when the 15-item registered nurse questionnaire was adapted for the student population, some items were revised and others added to improve the reliability of the three subscales. To address content validity of the Student Form, one of the non-nursing authors discussed the instrument with a group of clinical nursing students. These students' comments were used to add new items and revise other items in the 1991 and 1993 instruments.
The 1991 edition of the HTV-Impact Questionnaire: Student Form was administered to sophomores and juniors during the last class of the year. The response rate was 99%; 118 students responded. Recognizing that the sample size was small for factor analyses, the authors examined with caution several analyses. The best fit was a seven-factor solution of Worry (4 items), Satisfaction (3 items), Refusal (4 items), Prevention (2 items). Precautions (3 items), Job Change (2 items) and Test Results (1 item). One item did not load on a factor. Cronbach's coefficient for the entire scale was .78.
In Spring 1993, the instrument was revised using the above data. Eight items were added, including ones about students' confidence in universal precautions. The questionnaire presented to students had 28 items.
At the top of the instrument, students were informed that HTV-seropositive body fluids "excluded those body fluids considered to be lowly infectious, such as bloodfree saliva, tears and urine." For the 28 statements, students were instructed to mark one of four responses: strongly agree, agree, disagree, and strongly disagree. In the statements, the phrase "HTVseropositive patients" was used; there was no distinction made between HIV-infected patients and AIDS patients. And, although students have been taught "Body Substance Isolation," the term "Universal Precautions" was used throughout because the latter term is more widely used.
Procedure - For the sophomore and junior classes, the survey was conducted during the last non -exam class of the year; for the senior class, it was conducted at the end of a final exam. Students were informed that participation in the research was voluntary and that their responses were anonymous. They were also informed that no nursing faculty would see the students' responses, and that a psychology graduate student would tabulate the data and keep the raw data in his possession. The consent form was "signed" by having the student choose two digits and writing those digits in the two blank spaces provided.
Of the approximately 230 students who were asked to participate, 190 completed the survey. The sophomore class had the lowest response rate (80%); the response rate for the three classes together was 83%.
Results and Discussion
After tabulating the percentages of agreement with each item, responses to approximately half of the 28 items were reverse scored so that higher scores indicate a more favorable orientation toward HIV/AIDS patient care. The Table lists the reverse scored items as well as the percento of agreement before the items were reverse scored.
Using Factor Analysis with Maximum Likelihood Extraction and Varimax Rotation, the 28-item instrument was analyzed, and items with low factor loadings were deleted and the resulting 22-item scale was analyzed. The following discussion reports the findings of the 22-item scale analysis.
The seven factors of the 22-item scale and the percent of variance accounted for are found in the Table. Although the 1993 analysis included the same number of factors as were found in the 1991 analysis, the best fit solution was slightly different than the solution found in the 1991 study. Both the 1991 and 1993 factor analyses suggest these five factors: Satisfaction (items 1, 12, 13, 21, 22), HTV-testing (items 8, 9, 15, 16, 19, 20), Refusal of Assignment (items 17, 18), Precautions (items 4, 5), and Prevention (items 3, 10, 11). However, for the purpose of clarity, the Prevention factor was retitled, Confidence in Universal Precautions. In addition, the 1993 analysis suggested that the Worry Factor be divided into two separate factors: I worry (items 2, 7), and Others worry items (6, 14). Lastly, the Job Change factor found in the 1991 study was deleted and one of the items formerly found in Job Change loaded on the satisfaction factor.
Cronbach's coefficients for the seven factors are: Satisfaction (.80), HTV-testing (.72), Knowledge/Use of Universal Precautions (.75), Others Worry (.79), I Worry (.71), Confidence in Universal Precautions (.61), and Refusal of Assignment (.63). Cronbach's coefficient for the 22-item scale was ,82.
Also included in the Table are the percents of agreement with the statements. For the purpose of reporting the results concisely, the authors combined the responses of strongly agree and agree into a "percent of agreement" total. When reading these percento, the reader should keep in mind that the students may have responded in the manner they believed was wanted by faculty. No measure of "social desirability" (or lying) was used.
There was 100% agreement with item 4 (I understand the guidelines for universal precautions), and 99% agreement with item 5 (I routinely follow universal precautions for all patients). Also, 96% agreed with item 3 (On-the-job exposure to HTVinfection can be prevented if nurses use blood and body fluid precautions for all patients) and 64% agreed with item 11 (If I always use universal precautions, there is virtually no chance of me becoming HIVinfected during my nursing practice). These percento of agreement suggest that students have confidence in universal precautions but also know that accidental exposure can happen.
Generally speaking, the students expressed satisfaction with nursing HTVseropositive patients. Unfortunately, because students were not randomly assigned to the AIDS unit, conclusions cannot be drawn about the causal role of the ATDS unit experience in influencing their attitudes.
In future research, larger and diverse populations will be surveyed to enable the researchers to analyze further the validity of this instrument. These replication studies also will enable faculty elsewhere to evaluate their teaching of HTV-related content.
Although brief enough to be completed in less than 15 minutes, the HIVImpact Questionnaire: Student Form yields both a reliable overall score as well as subscale scores for several aspects of AIDS/HIV-related teaching of nursing students. Included in the subscales is documentation about studente' selfreported understanding of and compliance with universal precautions.
Results of Factor Analyses Using Maximum Likelihood Extraction and Varimax Rotation and Percents of Agreement with Statements
- Becker, M. (Ed.). (1974). The health belief model and personal health behavior. Health Education Monograph, 2(4), 409-419.
- Bliwise, N., Grade, M., Irish, T., & Ficarrotto, T. (1991). Measuring medical and nursing students' attitudes towards AIDS. Health Psychology, 10(4), 289-295.
- Goldenberg, D., & Laschinger, H. (1991). Attitudes and normative beliefs of nursing students as predictors of intended care behaviors with AIDS patients: A test of the AjzenFishbein theory of reasoned action. Journal of Nursing Education, 30, 119-126.
- Kelly, J., St. Lawrence, J., Hood, H., Smith Jr., S., & Cook, D. (1988). Nurses' attitudes toward AIDS. The Journal of Continuing Education in Nursing, 19, 78-83.
- Lawrence, S., & Lawrence, R. (1989). Knowledge and attitudes about acquired immunodeficiency syndrome in nursing and non-nursing groups. Journal of Professional Nursing, 5(2), 92-101.
- Weinstein, N. (1984). Why it won't happen to me: Perceptions of risk factors and susceptibility. Health Psychology, 3(5), 431-457.
- Wiley, K, Heath, L., Sc Acklin, M. (1988). Care of AIDS patients: Student attitudes. Nursing Outlook, 36(5), 244-245.
- Wiley, K., Heath, L., Acklin, M., Earl, A, & Barnard, B. (1990). Care of HIV-infected patients: Nurses' concerns, opinions, and precautions. Applied Nursing Research, 3(1), 27-33.
Results of Factor Analyses Using Maximum Likelihood Extraction and Varimax Rotation and Percents of Agreement with Statements