SEARCH

SEARCH BY CITATION

Keywords:

  • carrier screening;
  • genetic nurse specialist;
  • genetic testing;
  • physician

Abstract

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

barnoy s., levy o. & bar-tal y. (2010) Nurse or physician: whose recommendation influences the decision to take genetic tests more? Journal of Advanced Nursing66(4), 806–813.

Abstract

Title. Nurse or physician: whose recommendation influences the decision to take genetic tests more?

Aim.  This paper is a report of a study conducted to explore factors influencing the decision to take genetic carrier testing.

Background.  Genetic testing has become a popular means of elucidating the risk of giving birth to a sick/disabled child but what influences prospective parents to take genetic tests or not is unclear.

Methods.  The study was based on a factorial 2 × 2 × 2 within-between subjects design. The questionnaire presented four scenarios, each illustrating a healthcare worker presenting information on a devastating genetic disease. The variables manipulated were: the health care worker’s expertise (novice vs. expert), the recommender’s role (nurse vs. physician) and type of recommendation (recommendation vs. no recommendation). After each scenario participants were asked their own intention to take the test and their intention to recommend a friend to take the test. Data were collected in 2007.

Results.  Altogether 156 questionnaires were collected. Expert nurses and expert physicians were accorded the same level of compliance. However, a novice physician’s recommendation to take a genetic test was complied with as much as an expert physician’s, whereas a novice nurse’s recommendation was less accepted than that of an expert nurse. When no recommendation was made, more compliance with the nurse was found (i.e. less intention to take the test).

Conclusion.  Genetic information should be presented to the public with caution to avoid over-testing. For the public to accept a nurse’s recommendation, the nurse has to be perceived as an expert; therefore, expert nurses’ proficiency should be made clear.


What is already known about this topic

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References
  •  Genetic carrier testing is prevalent among couples before/during pregnancy.
  •  The specialist genetic nurse is an established role in some countries.
  •  Recommendations from perceived experts are accepted more easily than those of perceived novices.

What this paper adds

  •  The recommendations of expert nurses and expert physicians lead to the same level of compliance.
  •  While the recommendations of novice and expert physicians are both complied with, a novice nurse’s recommendations are complied with much less than those of an expert nurse.
  •  Prospective parents respond to being informed of a potential disease with a high intention to take a genetic carrier test for the disease, even when professionals make no recommendation to take the test.

Implications for practice and/or policy

  •  Since much of primary care is managed by nurses, the proficiency of expert nurses should be made more evident to the public.
  •  Genetic information should be presented to the public with caution to avoid over-use of tests.

Introduction

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

Genetic testing has become a popular means for elucidating the risk of giving birth to a sick/disabled child. One type of testing is carrier testing, in which healthy people with no family history of a given disease are tested to find out if they are carriers of a disease gene.

Background

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

The Israeli context

The number of such tests (which are expensive and mostly paid for privately) offered to the public is increasing rapidly. In Israel, over 20 are available but only the cystic fibrosis, Tay-Sachs and familial dysautonimia tests are funded by the Ministry of Health, while the test for beta thalassaemia is funded by health maintenance organizations (HMOs – sick funds) (Ministry of Health 2008). Typically in Israel, tests are offered prior to or at the onset of pregnancy. Although the Israeli Association of Medical Geneticists has issued recommendations as to which genetic carrier tests should be suggested to which ethnic groups (Ministry of Health 2009), the decision to take the test or not remains with prospective parents. Nevertheless, a study of 281 Jewish College and University students in the south of Israel showed that the students preferred non-directive genetic counselling (Raz & Atar 2003). These results point to the notion that the cultural expectation of the Jewish population in Israel is to receive relevant information that can help their personal decision-making regarding genetic testing.

The role of specialist genetic nurses, who refer people for genetic testing and give genetic counselling, is already established in some countries (Skirton et al. 1997, Greco & Mahon 2003, Lea et al. 2006). In Israel, however, recommendations to take genetic carrier tests are usually given by physicians or genetic counsellors, but recently nurses have begun to take an expanded role and HMO’s have trained nurses to advise the public about appropriate genetic carrier testing. No study has yet compared compliance with a nurse’s recommendation to take genetic tests against compliance with a physician’s recommendation, and the present study is designed to fill this gap.

Factors influencing test take-up

Most studies that have examined influences on the uptake of genetic tests have focused on predictive tests. Several factors of influence were found, for example: attitudes, risk perception and the desire to reduce uncertainty (for a review, see Etchegary 2004). Studies that focused on the general population’s interest in genetic testing for hypothetical diseases used scenarios describing devastating diseases; these studies have shown several factors that explain the interest. For example, research by Barnoy (2007), Shaw and Bassi (2001) and Shiloh et al. (1999) (= 121, = 226, = 209 respectively) shows that control over the disease (Shiloh et al. 1999, Shaw & Bassi 2001, Barnoy 2007), likelihood of contracting the disease (Shiloh et al. 1999), test predictivity (Barnoy 2007) and cure availability (Shaw & Bassi 2001, Barnoy 2007) influenced the interest in genetic testing.

As for carrier screening, a systematic review that included 40 studies identified several major factors related to interest in cystic fibrosis testing. Among them were perceived benefits of undergoing the test, fewer perceived barriers to taking the test, higher risk perception of being a carrier, more knowledge about the disease, more severe perception of the consequences of being a carrier, and having higher internal locus of control. Other factors identified were socio-demographic factors (being female, having higher education and income) (Chen & Goodson 2007). Two retrospective studies in Israel examined the uptake of genetic carrier testing by pregnant women; in one all Jewish women in maternity wards on the same day in 2003 (= 377) were interviewed (Sher et al. 2003). In the second study, 361 pregnant women visiting Maternal Child Clinics were interviewed (Barnoy et al. 2006). Both studies showed that educational level, religiosity and socio-economic level were associated with more uptake of carrier tests. A qualitative study conducted among 46 women who undertook genetic carrier tests and 23 women who declined testing in Israel showed that all those who chose to take the testing cited fear of having a disabled child as a reason. Compliance with physicians’ advice was another theme: this was mentioned by 85% of the women who chose to be tested but by only 30% of women who declined testing (Remennick 2006).

Compliance with recommendations

Compliance or not with health behaviour recommendations may also be connected to whether the recommender is regarded as an expert, as advice from experts is reportedly accepted more easily than that from a novice, even when both give the same advice (Harvey & Fischer 1997). In every profession there are novices and experts, so that the perceived status of the recommender is not necessarily determined by their profession: a nurse can be an expert and a physician a novice, and vice versa.

On the other hand, receiving accurate information from a credible source is reported to influence the decision to take presymptomatic genetic tests (Cassidy & Bove 1998), and the perceived epistemic authority of the information provider is also reportedly a factor influencing the acceptance of information (Kruglanski et al. 2005). Ascribed epistemic authority is influenced by role (Kruglanski et al. 2005), so that it is likely that a physician per se will be perceived as a greater expert than a nurse, and a physician’s recommendations can be expected to carry greater weight than a nurse’s, regardless of the level of their expertise. Even a novice physician may be perceived as more expert than an expert and experienced nurse.

The study

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

Aim

The aim of the study was to explore factors influencing the decision to take genetic carrier testing.

The hypotheses tested were: (i) there will be more compliance with experts’ than novices’ recommendations, (ii) the recommender’s role (nurse/physician) will be associated with differential compliance and (iii) the type of recommendation will affect the intention to take the test.

Design

The study was based on a factorial 2 × 2 × 2 within-between subjects design, where the level of expertise of the recommender (expert vs. novice) and the type of recommendation (recommendation vs. no recommendation) were within-subject variables (that is, all participants responded to four scenarios in which these variables were manipulated). The recommender’s role (nurse vs. physician) was a between-subject variable (that is, each participant responded to scenarios which all featured either a nurse or a physician). The assignment of participants to questionnaires (featuring nurse or physician) was random. The response rate was 94%.

Participants

As this was a study involving hypothetical situations, we wanted the questions to be as relevant as possible to the participants; therefore we chose respondents of reproductive age by approaching parents collecting their children from a large elementary school. The convenience sample consisted of 156 participants. The study design was a between-subjects design [two types of questionnaires] and we wanted a fairly equal representation of men and women to guarantee variability in each sub-group of about 40. All the 76 men and 80 women were aged from 20 to 40 years, with an average of 33·26 ± 4·68. All were fluent in Hebrew. With respect to religious practice, most (= 127) were secular, 23 were traditional and 6 devout. Slightly more than half had had university education (= 86), 37 had had some form of higher education, and 33 high-school education, a spread which is fairly representative of the educational level in Israel since, according to the Central Bureau of Statistics (2008), 48% of the Jewish Israeli population has more than 13 years of schooling.

Data collection

Data were collected in 2007 using a questionnaire developed by the researchers and then pilot-tested with 30 volunteers. The research goals were explained to these volunteers and, once they had completed the questionnaire, they were asked to comment on its clarity and the relevance of the items to the research questions. The questionnaire was modified accordingly.

The questionnaire is based on four scenarios, each illustrating a health professional (nurse or physician, both presented as females) who presents information about a devastating genetic disease. In two scenarios, the health professional is a novice and in two is an expert. In two scenarios, the health professional recommends taking a genetic test and in two she gives the information but makes no recommendation. To avoid bias between the different levels of expertise attributed to nurses and physicians, the effect of the role of recommender variable was tested by being randomly assigned among the study participants. Thus, all participants completed an identical questionnaire, with the one exception that it featured either a nurse or a physician.

With respect to each scenario, participants were requested to answer two items on a scale from 0 to 100: (i) their intention to take the test themselves and (ii) their intention to recommend a friend to take the test. To achieve higher reliability in the dependent measure, intention to take genetic testing was calculated as the average of the answers to both questions. Using the average of the two answers was legitimate as both answers represented the dependent variable and so replicated each other (Cronbach alpha reliabilities for each of the four pairs of questions was between 0·93 and 0·95), with the scale transformed to range from −50 to +50, when −50 represents certainty of not taking the test, 0 represents uncertainty and +50 represents certainty of taking the test).

The four scenarios

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

The four scenarios are described below:

  •  Imagine that before your marriage you decide to get information on genetic tests. You call the clinic and make an appointment to visit a nurse/physician specialist in genetics. At the clinic the nurse/physician asks you about your and your partner’s ethnic origin and takes your family history. On the basis of this information the nurse/physician informs you of a genetic disease which is common in your ethnic group. The disease is characterized by an enzyme deficiency which causes the accumulation of fatty acids. Affected children develop normally for 3 to 4 months, then deteriorate and typically die before the age of 3 years. The nurse/physician recommends you to take a genetic carrier test to see if you are a carrier of the disease.
  •  Imagine that you and your partner arrive at a clinic to perform a routine blood test during pregnancy. The nurse/physician who sees you does not work in antenatal care. After drawing blood, the nurse/physician asks you about your ethnic origin and genetic carrier tests. She tells you about a genetic disease that is common in persons of your ethnic origin, explaining that the disease causes proliferation of blood vessels, leading to brain haemorrhage, which can cause handicap or death. She adds that in most cases children who suffer from the disease start to develop normally but then deteriorate and death occurs within 3 years. The nurse/physician recommends to you to take the genetic carrier test.
  •  Imagine that you and your partner come to a Women’s Health Centre for pregnancy monitoring by an unfamiliar nurse/physician. You are told that the nurse/physician has 15 years’ experience in antenatal care, is a specialist in genetics and teaches the subject in a nursing/medical school. During the appointment she discusses issues concerning the pregnancy and asks you about your ethnic origin. She tells you about a genetic disease common in your ethnic group that causes an imbalance which leads to inability to walk. The immune system is also impaired, which causes severe lung infections and, typically, affected children die by the age of 3. The nurse/physician does not make a recommendation regarding taking a genetic carrier test.
  •  Imagine that you and your partner visit a Mother and Child health clinic for the first time for pregnancy monitoring. At the clinic a nurse/physician who has just graduated university sees you. She takes vital signs and asks you about your ethnic origin. She also inquires about genetic tests and gives details of a genetic disease prevalent in your ethnic group. Affected children lack an enzyme and their brain degenerates. The symptoms start at the age of a few months and death occurs by the age of three. The nurse/physician makes no recommendation as to whether you should take a genetic carrier test.

Ethical considerations

The study was approved by a university ethics committee.

Data analysis

The data were analysed using a 2 × 2 × 2 within-between anova with a mixed design (repeated measures and between subjects) followed by a posteriori tests. The independent variables were type of recommendation (to take the test vs. no recommendation) and expertise (expert vs. novice) as within-subjects factors, and professional role (physician vs. nurse) as a between-subjects factor. The dependent variable was the intention to take the test. < 0·05 was considered as statistically significant.

Results

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

Table 1 shows that the analysis yielded two statistically significant main effects for the within-group factors, two simple interactions, and a three-way interaction. Since the lower-level effects were constrained by the higher-order interaction only the latter will be discussed. Figure 1 presents the means of the cells formed by this interaction.

Table 1. anova of the intention to take the test: the independent variables were the type of recommendation, expertise and role of the healthcare provider
 SSd.f.MSFΔη2
  1. * = < 0·01.

Expertise16,952·81116,952·8132·82*0·18
Recommendation13,049·16113,049·1647·03*0·23
Role3952·3313952·333·000·02
Expert × Recommendation6283·2016283·2021·09*0·12
Expert × Role7014·7417014·7413·58*0·08
Recommendation × Role16·55116·550·060·00
Expert × Recommendation × Role2256·0512256·057·57*0·05
Error45880·05154297·92  
image

Figure 1.  Participants’ rating of their decision as a function of the role and level of clinician expertise and type of recommendation.

Download figure to PowerPoint

A posteriori Bonferroni tests showed that the statistically significant three-way interaction may stem from the different effect of the type of recommendation made by novice physicians and nurses. Under almost all conditions there was no difference between responses to a physician and a nurse. The only exception was the response to a novice’s recommendation to take the test. That is, while participants gave a statistically significantly lower compliance score to a novice physician’s recommendation to take a genetic test compared to her non-recommendation, their reactions to the two types of recommendation from a novice nurse did not differ statistically significantly. Participants declared statistically significantly more intention take the test when a novice physician recommended it than when the recommendation came from a novice nurse.

Finally, we examined whether participants’ sex or level of education affected the results. Participants’ sex could have affected the results because the nurses and physicians in the scenarios were described as women. Women and men may react differently to recommendations made by nurses, in particular because nurses have lower status than physicians. Participants’ level of education could have affected their perception of the nurse as an expert and therefore their intention to comply with her recommendation. To test the possible effect of each demographic variable, we performed two-four-way anova’s, adding one of the two levels of the demographic variables (men vs. women in the case of sex, and below-college education vs. college education and above, for education). The results of both analyses showed that no main effect and no interaction with the demographic variable achieved statistical significance. That is, neither of the demographic variables had any effect on the results.

Discussion

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

Study limitations

The study had a number of limitations. First, it was based on scenarios which might not reflect the real level of interest in genetic testing. However, studies using scenarios to examine the intention to take genetic tests have shown this to be a suitable means for predicting genetic test uptake (for a review, see Persky et al. 2007). Second, the scenarios were not completely consistent: for example, in the first scenario the respondent was not pregnant but in the other three she was. This discrepancy might have influenced respondents’ intentions to take the test, for it has been found that when women are offered carrier testing for cystic fibrosis, pregnant women are more interested in being tested than non-pregnant women (Chen & Goodson 2007). However, it was the first scenario that elicited the highest interest in testing. A second inconsistency is that the third scenario gives a more detailed description of the nurse/physician than the other three. However, a review assessing scenario use in predicting genetic test uptake showed that number of words was not associated with estimates of test uptake (Persky et al. 2007). Also, interest in being tested when an expert made no recommendation was of a similar level to when a novice made no recommendation; this points to the fact that the detailed description in scenario 3 had little, if any effect, on the level of interest in being tested. In addition, in the present study, all health professionals were presented as women (nurses and physicians). It is possible that if the recommenders had been presented as women vs. men (both nurses and physicians) we would have achieved different results. This issue was beyond the scope of the present study and is an interesting question for future research.

Discussion of results

We examined how people’s intentions to take genetic carrier tests are affected by the recommender’s role (nurse or physician) and expertise level (expert vs. novice). The main finding was that intention to comply was identical when both nurse and physician were experts. This finding was surprising, in that we expected that compliance with a physician’s recommendation would be higher, regardless of their perceived expertise. We expected this finding in Israel in particular, where the role of specialist nurse is not established as it is in the United States of America (Lea et al. 2006). This result, however, does conform to that in a recent study in Israel in which the public was receptive to the idea of getting genetic information from nurses (Barnoy et al. 2006).

Other results show that the recommendation of a novice physician to take a genetic test gains the same level of compliance as that of an expert physician. By contrast, the recommendation of a novice nurse gains less compliance than that of an expert nurse. It seems that the title ‘physician’ makes the difference and that a physician is considered an epistemic authority regardless of level of expertise. Not so a nurse. We found, therefore, that in the case of genetic tests, the public is open and willing to recognize a nurse as an expert in genetics and willing to comply with her recommendations as much as with those of an expert physician, provided that the client knows the nurse to be an expert. A physician’s recommendations, by contrast, are followed regardless of expertise level. Holding the title ‘physician’ apparently makes the difference.

The results show a high level of intention to take the genetic test in all scenarios, which confirms previous studies in Israel (Barnoy 2007). The case of non-recommendation by a novice nurse served as the baseline for taking the tests. That is, intention to take the test after receiving information from a novice nurse, but having received no recommendation, shows that merely knowing that a disease exists is enough to convince prospective parents to take a carrier test. A clue to this finding comes from a previous study in Israel showing that the main reason for not performing carrier testing for fragile-X and Tay-Sachs was not being referred for testing by a healthcare worker (Sher et al. 2003).

In our study, when the physician or nurse made no recommendation there was more compliance with the nurse’s non-recommendation than with the physician’s non-recommendation (i.e. less intention to take the test). It seems that when a potential problem is shown to exist by a physician, even though no recommendation for action is made, the physician’s presentation of the problem is perceived as more relevant to health and, hence, there is more intention to take the test.

We compared the case of a recommendation to take a test against making no recommendation. The latter is not the same as an explicit recommendation not to take the test. It would be interesting to examine interest in taking genetic tests after receiving an explicit recommendation not to be tested. In genetics, a recommendation not to take a test is made when the test is inappropriate given ethnic origin and family history. Taking unnecessary tests does not cause harm but does cause expense and overload of the healthcare system. The response to explicit recommendations not to perform a given health behaviour may also be relevant beyond the genetic testing context, for example, in the case of influenza vaccination managed by nurses, when the vaccination might cause harm to the patient.

Conclusion

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

It is important to be cautious in the way genetic tests are presented to the public as, even when no recommendation to take the tests is made, prospective parents are still interested in taking them. It seems that merely mentioning the test is enough to persuade people to take it. Emphasizing a nurse’s level of expertise elicits responses similar to those accorded a physician. However, in Israel and many other countries, nurses’ credentials are much less salient to the public than those of physicians. Since much of primary care is managed by nurses, to increase compliance we recommend that the proficiency of expert nurses be made clearer to the public. This can be achieved by giving expert nurses formal titles (i.e. genetic nurse) and having this title displayed on their name badges and door signs. Another option is to publicize nurses’ expertise by means of public education programs.

Author contributions

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References

SB, OL and YBT were responsible for the study conception and design. OL performed the data collection. SB, OL and YBT performed the data analysis. SB was responsible for the drafting of the manuscript. SB and YBT made critical revisions to the paper for important intellectual content. YBT provided statistical expertise.

References

  1. Top of page
  2. Abstract
  3. What is already known about this topic
  4. Introduction
  5. Background
  6. The study
  7. The four scenarios
  8. Results
  9. Discussion
  10. Conclusion
  11. Funding
  12. Conflict of interest
  13. Author contributions
  14. References