Abstract
Doctors and nurses play a crucial role in sexual health education. They are the primary contacts for patient information about sexual health and help prevent the spread of sexually transmittable diseases such as HIV infection by screening patients for, and educating them about, the consequences of high-risk sexual behaviours. Three aptitudes are critical to taking meaningful sexual histories and providing effective sex counselling: knowledge, skills and attitude [1]. Doctors and nurses must not only know what to ask patients, and how to do so in a sensitive, caring manner, they must also convey to patients a supportive, non-judgemental attitude [2]. However, there is evidence to suggest that many doctors and nurses, as well as medical and nursing students, have deficient sex knowledge, lack adequate sexual history-taking and sex counselling skills, and have negative attitudes toward the sexual behaviours of some of their patients [2, [3], [4], [5], [6], [7], [8], [9], [10], [11]]. While the relationship between negative attitudes and actual medical and nursing practice is uncertain, it is clear that those with a lack of sex knowledge will not make effective sex counsellors or sexual history-takers.
The present study was designed to assess sex knowledge and attitudes toward controversial aspects of human sexuality among medical and nursing students. It is based on an unpublished preliminary study carried out among medical students at a medical school in the southern United States. The stimulus for the preliminary study was the experience of one of the present study's authors at the school, where students would sometimes absent themselves from lectures in which sexual behaviours they considered ‘immoral’ were being discussed. The results of the preliminary study indicated that medical students attending religious services of any religious denomination three or more times per month, identifying themselves as politically right-wing and with no experience of sexual intercourse expressed the most negative attitudes toward controversial aspects of human sexuality. Negative attitudes, however, were not correlated with deficient sex knowledge. Our hypothesis, derived from the preliminary study, was that Australian medical and nursing students with background and sociodemographic characteristics similar to students in the US study would also be more likely to express negative attitudes, and would not have lower levels of sex knowledge.
Methods
Questionnaire
A self-administered, anonymous questionnaire, based on the preliminary study and modified to reflect an Australian context, was employed. The questionnaire consisted of three parts. Part 1 elicited background information about subjects' sociodemo-graphic characteristics, political philosophy, religious affiliation and practices, means of acquiring sex information and sexual experience. Part 2 determined students' attitudes toward certain controversial forms of sexual behaviour, including pre- and extramarital sex, masturbation, sex between male and female homosexuals and bisexuality. It also asked about their attitudes toward contraception, therapeutic and non-therapeutic abortion and child-bearing for unmarried women. Subjects indicated their level of approval or disapproval of each item along a five-point Likert scale on which 5 indicated strong disapproval and 1 strong approval. Part 3 consisted of a version of the Kinsey Institute/Roper Organization National Sex Knowledge Test [12], modified for use in Australia.
Subjects
The sample consisted of first-through fifth-year medical students at the University of Western Australia and first-through third-year undergraduate nursing students at Edith Cowan University. Ethics committee approval for the study was received from each university, and subject anonymity was assured. Questionnaires were distributed to all students present during university campus lectures. A specific lecture was selected for each class year and discipline based on the anticipation that all or most members of the class would be present. This was done to offer participation in the study to as many students as possible and to ensure a representative sample. Members of the research team explained the purpose of the study, distributed the questionnaires and supervised their administration and collection. Before completing the questionnaires, students were asked to read an enclosed information sheet and, if they chose to participate, to sign an informed consent document. Approximately 90% of students receiving questionnaires returned them. The 288 medical students returning questionnaires represented 47.0% of first-through fifth-year medical students at the University of Western Australia, while the 288 nursing students returning questionnaires represented 39.8% of undergraduate nursing students at Edith Cowan University.
Data analysis
For the analyses, some of the attitude measures were combined. This was done because some questions were designed to assess the same construct with reference to different target populations, such as males and females. Such questions tended to be highly correlated with one another. Attitudes toward gay/lesbian/bisexual behaviour were assessed by combining questions 5, 6 and 7, which referred, respectively, to gay, lesbian and bisexual behaviour. The correlation between questions 5 and 6 was 0.93, that between questions 5 and 7 was 0.76 and that between questions 6 and 7 was 0.77. Attitude toward masturbation was assessed by combining questions 8 and 9 (r == 0.99). Attitude toward extramarital sex was assessed by combining questions 3 and 4 (r == 0.98). All other attitude questions were included in the analysis as single items.
Three of the background/sociodemographic variables were dichotomous and were dummy coded as zero or one. These variables were: male/female, had experienced intercourse/had not experienced intercourse, and medical student/nursing student. Ethnicity was coded using three categories: Anglo-Saxon Australian (those listing Australian, Irish, English, or any combination of these), Asian (any Asian country, plus those listing both Australian and Asian), and ‘Other’ (those listing Europe, the Americas, or the Middle East). There were 320 Anglo-Saxon Australians, 109 Asians and 48 ‘Other’, proportions reflective of the ethnic break-down of the medical and nursing schools surveyed. To code these three categories, two vectors were created in which each of the three groups was coded as 1 or 0. The category of Anglo-Saxon Australian was used as the reference group and hence was coded as zero in both vectors. In the first vector, Asians were coded as one. In the second, the ‘Other’ category was coded as one.
The data were analysed using canonical analysis. The purpose of this analysis was to determine the dimensions on which background/sociodemographic characteristics, attitudes toward various aspects of sexual behaviour and sex knowledge were related.
Multiple regression analyses were done for each of the eight attitude measures and the sex knowledge score. For each of these analyses, the attitude or knowledge score was the dependent measure and the background and sociodemographic measures were the independent variables. Because the eight dependent measures were not independent of one another, Bonferroni's procedure (in which alpha is divided by the number of comparisons) was used to protect against type I errors.
Methods
A comparison of the relationship between attitudes, sociodemographic variables and sex knowledge, loading matrices for pairs of canonical variates in two dimensions
Examination of the second pair of canonical variates indicated that those with negative attitudes toward masturbation, extramarital sex, child-bearing by unmarried women and non-therapeutic abortion tended to be non-Asian and non-Anglo-Saxon Australian (i.e. those classified as ‘Other’ in terms of ethnicity), female and medical students. The sexual attitudes in this pair of canonical variates (with the exception of masturbation) seemed to reflect more controversial sexual practices, and were related to demographic factors such as gender and ethnicity.
Examination of the results of the separate regression equations for each of the eight attitude and knowledge measures indicated that the background factor most consistently related to attitude and knowledge was the frequency of attendance at religious services of any religious denomination in the past month. This measure was significantly related to each of the attitude and knowledge measures. Sex knowledge, as assessed by the Kinsey Institute/Roper Organization National Sex Knowledge Test, was also significantly related to attitudes toward gay/lesbian/bisexual behaviour, masturbation, premarital sex and contraception. For each of these attitudes, those with lower sex knowledge scores expressed more negative attitudes. Whether one had experienced sexual intercourse was related to attitudes toward premarital sex and contraception, with those who had not experienced sexual intercourse expressing more negative attitudes. Thus, low sex knowledge and lack of personal sexual experience were independently related to negative attitudes toward premarital sex and contraception, while only low sex knowledge was related to negative attitudes toward gay/lesbian/bisexual behaviour and masturbation. Political orientation was related to attitudes toward gay/lesbian/bisexual behaviour, with those who were more right wing politically more likely to express negative attitudes in this area. Family income was related to attitudes toward premarital sex and childbearing by unmarried women, with those with lower incomes expressing more negative attitudes. Gender was related to attitudes toward gay/lesbian/bisexual behaviour, with males expressing greater disapproval. Conversely, females expressed stronger disapproval of extramarital sex than males. Ethnicity was related to attitudes toward premarital sex and contraception, with Asians expressing more negative attitudes than the other two ethnic groups. Student status (medical or nursing) and number of sex partners were related only to attitudes toward non-therapeutic abortion. Medical students' attitudes were slightly more positive, and those who had fewer sex partners had more negative attitudes.
Intercorrelations among attitudes toward sexual behaviours and background/sociodemographic factors
Means and standard deviations for group differences on attitudes toward sexual behaviours (higher scores represent more critical attitudes)
Discussion
As expected, the study demonstrated a significant relationship between background and sociodemographic variables and attitudes toward controversial aspects of human sexuality among medical and nursing students. Of concern, and contrary to expectations based on the preliminary study, students with negative attitudes were also more likely to have lower sex knowledge scores. Students who identified themselves as Asian, attended religious services of any religious denomination three or more times per month, had fewer sex partners and had not experienced sexual intercourse were significantly more likely to express negative attitudes and to have lower sex knowledge scores. Among background variables, frequent attendance at religious services was most powerfully linked to negative attitudes. It is important to note in this context that, while ethnicity and frequency of religious attendance were significantly related to attitudes, religious affiliation was not. Theoretically, religious affiliation might have overlapped with ethnicity and frequency of attendance (for example, Asians might be more likely to be Muslim and thus more likely to attend religious services frequently). Lower sex knowledge scores were directly related to more negative attitudes toward gay/lesbian/bisexual behaviour, masturbation, premarital sex and contraception. Other background and sociodemographic variables significantly related to more than one negative attitude were right-wing political orientation, lower family income and ethnicity.
Of the three critical aptitudes (knowledge, skills and attitude) required for effective sexual history-taking and counselling, the present study assessed only two, attitude and knowledge. The clear linkage between negative attitudes and lower sex knowledge scores suggests that those medical and nursing students predisposed by background and sociodemographic variables to be critical of controversial aspects of human sexuality are also less likely to acquire sex knowledge. While negative attitudes may or may not be directly related to sexual history-taking and sex counselling skills, it is unlikely that doctors and nurses poorly informed about sex will make effective sex counsellors or sexual history-takers.
It is unclear how attitudes expressed on a questionnaire translate into attitudes conveyed to patients. However, our classroom observations of some less tolerant students, who not only expressed negative attitudes, but also absented themselves from lectures in which controversial material was discussed, suggest that their attitudes may well result in unhelpful interactions with patients.
What can be done to modify negative attitudes and increase sex knowledge among medical and nursing students? The connection between deficient sex knowledge and negative attitudes [7, [11], [13], [14]] suggests that increased time devoted to sex education in the medical and nursing curricula is indicated. Didactic instruction could be combined with simulated patient interviews in which students practice taking sexual histories under supervision. T h i s approach has been shown to be quite effective in improving the rate at which practising doctors take sexual histories [15]. It would also offer an opportunity to discuss with students how their attitudes toward sexual behaviour affect patient care.
It may also be helpful to expose students to a broader cross-cultural framework. Cross-cultural research has revealed few, if any, true ‘norms’ for human sexual behaviour [16]. What we experience as normal reflects only the standards prevailing during a brief moment in time and space. Mixing more and less tolerant students in ongoing discussion groups about the impact of cultural beliefs on sexual practices may expand students' horizons beyond judgement to empathy for other ways of being human. While we have emphasised the difficulties students with a more conservative sexual outlook may experience in dealing with patients whose sexual practices they regard as controversial, the same difficulties may be experienced by sexually liberated students in dealing with patients whose sexuality is more traditional. Similarly, from the perspective of ethnicity, it is also possible that the openness and directness of students from western backgrounds might offend patients from more traditional Asian backgrounds. Thus, the emphasis of cross-cultural training must be on increasing tolerance and sensitivity for practices and views different from one's own, whatever they may be.
Future research should investigate the relationship between negative attitudes expressed on a questionnaire and behaviour with patients during sexual history-taking and sex counselling. This might be assessed by observing and evaluating medical and nursing students taking sexual histories from actors simulating patients who engage in controversial sexual behaviours [15]. It will also be important to assess whether educational approaches are effective in broadening students' perspectives on human sexual practices different from their own.
Footnotes
Acknowledgements
This research was supported by grants from Edith Cowan University and the Western Australian Health Promotion Foundation.