Introduction

Variations in genetic material are inherent in all humans. While many of these variations do not change the protein product for which the gene codes, others may have a more deleterious effect (Adkison and Brown 2007). In autosomal or X linked recessive conditions, one normal copy of the gene is usually sufficient to ensure the protein product is not adversely affected; however if individuals are heterozygous, having one normal and one mutated copy of the gene, they are said to be a “carrier” of the condition. The offspring of carriers could be at risk of inheriting the disease. A woman carrying an X linked recessive condition, such as fragile X or Duchenne muscular dystrophy, could pass the condition to her children (usually her sons) but in the case of autosomal recessive conditions, such as cystic fibrosis or thalassaemia, both parents need to be carriers of the same disease for their children to be at risk. Fragile X does differ from most other X linked conditions because premutation carriers [individuals with 55 to 200 CGG repeats (Kronquist et al. 2008)] can be mildly affected by the condition, can present with late-onset fragile X-associated tremor/ataxia syndrome (FXTAS) (Reis et al. 2008), and have an increased risk of premature ovarian failure (Hunter et al. 2008).

The term “carrier” is also used to describe individuals who have a balanced chromosome translocation. The translocation does not affect their own health, but their children could inherit an unbalanced form of the translocation that could have an adverse phenotypic effect (Meza-Espinoza et al. 2008).

Because of the potential reproductive implications of carrier status, individuals who are aware they could be carriers may wish to be certain of their own status. Carrier testing for a range of genetic diseases has been offered to individuals with a family history of the condition, or through community based screening programs in particular ethnic groups, for example community based screening for Tay Sachs disease in the Ashkenazi Jewish population conducted since the 1970’s (Kaback et al. 1993). Population carrier testing has also been offered either regionally or nationally, for conditions including cystic fibrosis, thalassaemia and fragile X (Hartley et al. 1997; Kaufman et al. 2008; Metcalfe et al. 2008; Sangkitporn et al. 2004; Zlotogora et al. 2009).

Numerous studies have assessed the psychosocial impact of carrier testing and living as a carrier (Anido et al. 2005; Childs et al. 1976; Kenen and Schmidt 1978; McConkie-Rosell et al. 2001). As early as 1976, Childs et al. highlighted a number of emotional issues experienced by carriers including anxiety, self-stigmatization, and concern for offspring. Since then, other studies have assessed the impact of carrier testing in relation to a number of variables such as the particular condition (Anido et al. 2005), the mode of inheritance (James et al. 2006), the population being screened (Axworthy et al. 1996) and the impact of gender (Marteau et al. 1997). We conducted a systematic review to synthesize this body of knowledge. Carriers of fragile X were included in the review because a body of knowledge exists relating to the psychosocial effect of carrier testing on this group (Anido et al. 2007; Anido et al. 2005; McConkie-Rosell et al. 2000; McConkie-Rosell et al. 2001). Because female fragile X carriers can sometimes be affected by the condition, although in general more mildly, fragile X does differ from autosomal recessive and some other X linked conditions. However, because we were trying to ascertain similarities and differences across a range of different conditions, we felt it was relevant to include this group in the systematic review.

Specifically, the aim of this systematic review was to answer the following questions: (1) What are the factors affecting the impact of carrier testing results on individuals? (2) What is the methodological quality of the body of literature examining the psychosocial effects of carrier testing? (3) Can we make any assumptions about the psychosocial impact of living as a carrier, based on studies assessing the impact of carrier testing?

Methods

In conducting this systematic review the methods described by Pope et al. (2007) which involve using specific search parameters, defining inclusion and exclusion criteria, and undertaking quality appraisal of the studies that are included, were used as a guide. Due to the wide range of methods, conditions and samples in the studies reviewed, we did not conduct a meta-analysis of the data.

Search Methods

The following databases were searched: CINAHL, Embase, Ovid, Medline, PsychINFO, Pubmed and Web of Science, using the following search terms:

carrier testing or carrier test* or carrier screening or genetic screening or population screening or cascade testing or heterozygote testing AND genetic or DNA or chromosome or autosomal recessive or recessive or X-linked AND depression or emotion or guilt or anxiety or worry or stress or blame or psychological or psychosocial or social or effect or impact or psychological impact or social impact or personal or carrier status or distress or relief or burden or coping or coping strategy or communication or coping behavior* or emotion* or stigma or self concept or attitude* or psychology or social adaptation or reproductive uncertainty or risk perception or genetic counselling or genetic counseling or carrier couples or family planning or prospective risk AND NOT children NOT cancer NOT prenatal NOT predictive.

Limits were set on publication dates (January 1990 to May 2010), language (English), and population (Human; Age: Adult.).

An author search and an ancestral search (reference search) were also carried out after relevant studies were identified. Keywords from relevant studies found this way were fed back into the search terms to ensure the search was thorough.

Inclusion and exclusion factors

Studies were included if they were:

  • systematic reviews, literature reviews randomized controlled trials, quasi-experiments, observational studies, surveys or qualitative studies

  • published between January 1990 to May 2010. We included studies published from 1990 onwards as around this time DNA carrier testing became feasible clinically for patients with a family history of recessive and X-linked conditions (Broide et al. 1993; Kerem et al. 1989). At the same time, studies that assessed the impact of the test on the patient began to appear in the literature

  • focused on the psychological and social impact of the test result on the patient

  • focused on either autosomal recessive and X linked conditions, or carriers of chromosomal changes such as translocations.

Studies were excluded if they were:

  • about cancer, adult onset conditions or other dominantly inherited conditions, because the nature of the information derived from these tests will be different from receiving carrier information for recessive, X-linked or chromosomal conditions

  • ones in which there was potential for participants to find out that they were homozygotes for a particular gene mutation where the age of onset of the disease was in adulthood (e.g., hemochromatosis)

  • focused on pregnant women because their feelings may be influenced by worry for their offspring, and also because their decision to seek testing would be influenced by the immediate needs of a current pregnancy (Cheuvront et al. 1998)

  • ones that included children or adolescents because it is likely that they will have very different psychosocial reactions and information needs than adults

  • focused only on recall of information about risk

  • focused only on motivation for taking/not taking the test.

Search Outcome

The literature search generated 1694 articles for consideration. Following exclusion based on title and abstract, the full text of 41 articles was retrieved. An ancestral and an author search identified 10 further studies. After reading the papers in full, 31 studies were excluded because they did not meet the inclusion criteria, leaving 20 relevant studies to be included in the systematic review. There were substantial differences in construct, design, measures, population and outcomes across the studies. In this section we cite one example of each particular design, measure or outcome studied. Table 1 contains a more detailed report of the characteristics of each study. Thirteen studies were quantitative, three were qualitative and four were mixed methods. Study designs comprised longitudinal studies (Bekker et al. 1994), randomised controlled trials (Callanan et al. 1999), and cross sectional studies (Dunn et al. 2008). Samples from different populations including the general population (Henneman et al. 2002), high risk groups (McConkie-Rosell et al. 2001) of Jewish decent (Marteau et al. 1992) and women only (Anido et al. 2005) were included. Sample size varied from eight participants (Anido et al. 2007) to 2220 participants (Honnor et al. 2000). Data collection methods varied from questionnaires (Bekker et al. 1994), to focus groups (Anido et al. 2005) and in-depth interviews (Williams and Schutte 1997). Various measures were used, including the Spielberger State-Trait Anxiety Inventory (STAI) (Spielberger et al. 1970), the Health Orientation Scale (HOS) (Wooldridge and Murray 1988) and the Tennessee Self Concept Scale (TSCS) (Fitts and Warren 1996). In two cases the same cohort of participants was involved in two studies, however because different findings were presented in the different papers, both studies were included in the review (Cheuvront et al. 1998; Newman et al. 2002).

Table 1 Summary of studies
Full size table

Quality Appraisal

Each study was assessed using a quality appraisal tool developed by Kmet et al. (2004). The Kmet scale enables assessment of both qualitative and quantitative studies. This tool has proven internal validity and provides a systematic, reproducible and quantitative means of simultaneously assessing the quality of research encompassing a broad range of study designs. Using this tool, the first author scored each paper based on quality criteria including the description of the research question, appropriateness of design, justification of sampling strategy, appropriate data collection and analysis and estimates of variance (for quantitative studies) to produce a score phrased as a percentage. Five papers across the range of scores were selected and a blind appraisal was made by the second author to verify the results. The papers were ranked in the same order by both appraisers.

All 20 papers scored greater than 60% on the Kmet scale (range between 63% to 95%; median 81%) and therefore none were excluded on the basis of quality (Kmet et al. do not provide a ‘cut-scale’ at which studies should be discarded on the grounds of poor quality). The studies included in the review had notable strengths. Validated questionnaires were used in many of the studies [e.g., Honnor et al. (2000)]. The use of a longitudinal study design enabled changes in psychosocial wellbeing to be measured over time [e.g., Lakeman et al. (2008)], and the use of in-depth qualitative interviews enabled rich and complex data to be acquired by the researchers [e.g., Anido et al. (2007)]. There were a number of study limitations that should be considered when interpreting the findings from these research studies. First, potential confounding variables, such as whether participants had completed their families at the time of testing (e.g., Bekker et al. 1994) or whether they knew anyone with the condition for which they were a carrier (e.g., Watson et al. 1992) were not addressed by some of the authors. In addition, authors of many studies did not justify their sample size (e.g., Pastore et al. 2008) while some studies lacked clear conceptual definitions and were not based on particular theoretical models (e.g., Henneman et al. 2002).

Data Synthesis

To enable comparison across the studies, a matrix (Miles and Huberman 1994) of studies was drawn up including aspects of each relevant study considered to be most important (study design, sample and size, methods, quality issues and findings related to the psychosocial impact of carrier testing). This matrix is summarized and displayed as Table 1. Due to the range of different quantitative measures and quantitative philosophical approaches used, a meta-analysis was not performed. Instead we used an approach based on Grounded Theory (Strauss and Corbin 1998), inductively deriving codes and themes from the data. We undertook a thematic analysis to elicit general overarching themes from the papers studied. The findings were discussed by the present authors and compared to identify areas of agreement and disagreement across studies; this resulted in a set of overarching themes. These themes were labeled using the terminology commonly contained in many of the studies, such as guilt, anxiety and stigmatization. Once this process was complete, a “spider diagram” or “mind map” (Burgess-Allen and Owen-Smith 2010) was drawn so that the key themes could be visualized and the relationships between them identified (Fig. 1). Finally, after reviewing the studies, we compared our results with well established psychological models relating to self concept (Shavelson et al. 1976) and coping (Lazarus and Folkman 1984). These two models have been used in other health-related studies looking at psychosocial adaptation (McConkie-Rosell et al. 2000; Street et al. 2010).

Fig. 1
figure 1

Mindmap showing relationship between key themes

Full size image

Results

The impact of carrier testing for cystic fibrosis was the condition most commonly investigated, with ten studies focusing on this condition, followed by carrier testing for fragile X (five studies), Tay Sachs (one study) and hemophilia A and B (one study). The focus of the remaining three studies was the effects of carrier testing for a number of conditions. No studies which assessed the impact of carrier testing on carriers of chromosomal abnormalities were identified from the literature review. Carrier testing for people who had a family history of a genetic condition (and were therefore at an increased risk) was assessed in 11 studies, risk in the general population was assessed in seven studies, and in two studies people in both groups were assessed. Only three of the studies included in the systematic review were intervention studies (Callanan et al. 1999; Cheuvront et al. 1998; Newman et al. 2002). All three compared levels of anxiety related to home education and testing with clinic education and testing. A number of overarching themes were identified. The most prominent were anxiety, guilt, relief, effect on self image, active coping mechanisms, impact on reproductive issues and disclosure of test results (Table 2).

Table 2 Themes by mode of inheritance
Full size table

Anxiety

Two categories of anxiety emerged; one related to testing and the other related to child health. In relation to testing, all longitudinal studies investigating patient anxiety over time either found no significant difference in anxiety between carriers and non carriers (Honnor et al. 2000), or found that any anxiety experienced by carriers upon first receiving their test result had, for the vast majority, dissipated by six months as assessed by the state STAI (Bekker et al. 1994; Callanan et al. 1999; Cheuvront et al. 1998; Lakeman et al. 2008; Watson et al. 1992), the Fragile X Visual Analog Scale (VAS) (McConkie-Rosell et al. 2001), or qualitative interviews (Anido et al. 2007; Anido et al. 2005).

Carrier anxiety dissipated for a number of reasons. Watson et al. (1992) found that the provision of written information and genetic counseling was helpful for most participants (92% and 97%, respectively). Bekker et al. (1994) found that the passage of time appeared to dissipate anxiety. Gender was also an issue discussed in relation to anxiety, in a number of studies. Newman et al. (2002) and Henneman et al. (2002) found that women reported higher anxiety than men while waiting for their test results (mean = 16.5 and 14.6, respectively on the STAI in the Newman study, p < 0.001; and 24% versus 13%, p < 0.001 measured on a five-point Likert scale in the Henneman study); however there was no significant difference between the genders once the test results had been received. Lakeman et al. (2008) found that Western participants generally reported lower levels of anxiety compared with non-Western participants (General Linear Model analysis at 4 time points, p < 0.001).

Anxiety did however appear to be an issue for both carrier and non-carrier siblings of people with cystic fibrosis, in the interview-based study conducted by Fanos and Johnson (1995b). Identified carriers and non-carriers were equally likely to have moderate or severe anxiety around their child’s health. Forty-one percent had had their children sweat-tested to rule out the condition, nine percent had had their child tested for carrier status, and 55% planned to do so before their child reached 18 years of age. Siblings who had had their children sweat-tested or tested for carrier status were equally divided between those who knew their own carrier status and those who did not.

Guilt

Guilt was a prominent theme in the data. Feelings of guilt associated with carrier status were cited as findings in five studies. These results were identified through interviews (Anido et al. 2005; McConkie-Rosell et al. 1997; Williams and Schutte 1997), an open-ended questionnaire (Dunn et al. 2008), a VAS (James et al. 2006; McConkie-Rosell et al. 1997) and the guilt subscale of the Multiscore Depression Inventory (sMDI) (James et al. 2006). Guilt is also an issue found to be closely associated with gender, mode of inheritance, and whether the participants had affected children. In the study conducted by Dunn et al. (2008) in which 81% of respondents had a son with hemophilia, 18 of 48 (38%) female carriers reported the timing of testing as negative. Reasons cited for the timing being negative included feeling blamed by their partner and a prolonged sense of guilt. James et al. (2006) found that mothers who were carriers of X-linked conditions felt substantial guilt and self-blame related to their child’s condition. When measured on the VAS, mothers of children with X linked conditions had significantly higher levels of guilt than mothers of children with recessive conditions (p < 0.01) and were more likely to blame themselves (p < 0.001). A similar finding was identified in the Williams and Schutte study, in which it was found that many of the participants who expressed feelings of grief and guilt were women who were carriers of fragile X or Duchenne muscular dystrophy. Anido et al. (2005) also found that in families affected by fragile X, even those women without affected children experienced guilt to some extent, by virtue of the condition being in the family.

In one study (Gordon et al. 2003) there was no significant difference evident between carriers and non-carriers on the “guilt” scale (as measured by the HOS). The participants were from the general population, were screened for cystic fibrosis, and did not have a family history of the condition.

Relief

Anido et al. (2005), McConkie-Rosell et al. (1997) and Lakeman et al. (2008) all identified that relief was an emotion experienced by carriers. In the study conducted by Anido et al. reactions of relief were expressed equally as strongly as reactions of guilt, with nearly all carriers expressing this emotion during interviews. For these individuals, finding out their carrier status was an inevitable result of finding a diagnosis for their child. Similarly, in the study by McConkie-Rosell et al. (1997), participants’ responses indicated that while they felt angry or depressed about their carrier status, there was an “emotional relief in finding out the cause of the mental retardation in the family” (p. 65). Lakeman et al. (2008) found that 68% of participants, including seven out of ten carriers, felt relief one week after receiving their test results, as measured on a structured questionnaire assessing emotional outcomes.

Effect on Self-image

Three main issues arose within this theme: perception of health, self-stigmatization and social-stigmatization.

Perception of Health

Of the seven studies in which perception of health was measured, findings from three studies indicate that some carriers believed their current or future health to be significantly poorer after learning their carrier status (Fanos and Johnson 1995b; Henneman et al. 2002; Marteau et al. 1992). Seven out of 17 carriers (41%) in the study conducted by Henneman et al. (2002) felt less healthy (measured on a multiple-choice questionnaire) due to their test results, despite being informed both verbally and by letter that their carrier status would have no effect on their own health. Marteau et al. (1992), also using a multiple-choice questionnaire which measured perceived health from three time perspectives, identified that carriers of Tay Sachs held the least optimistic view of future health compared with non-carriers and the control group (p < 0.01) and Fanos and Johnson (1995b) reported that during interviews, sibling carriers retrospectively redefined health problems as related to cystic fibrosis, although the authors do not report report how many.

Authors of four studies found that perception of health did not alter after learning one’s carrier status, using measures such as the Tennessee Self Concept Scale (McConkie-Rosell et al. 2000), a multiple choice questionnaire (Bekker et al. 1994), the HOS (Gordon et al. 2003) and a five point Likert-scale (Lakeman et al. 2008). However, both McConkie-Rosell et al. and Bekker et al. do provide some anecdotal evidence to suggest that carriers might attribute previous health problems to their carrier status. McConkie-Rosell et al. found that 12% of participants at Time 1 and 20% of carriers at Time 2 reported feeling they had mild clinical features of fragile X. They felt that perhaps if they were carriers it would explain why they had to “study hard in school” (p. 340). A participant in the Bekker et al. cohort wondered whether her allergies and chest colds were in some way linked to her carrier status. Therefore, even though perception of health did not alter when measured quantitatively, during qualitative interviews there were some indications that it did in fact occur in a small number of cases. In fact, in the case of fragile X, it is possible that carriers did experience a mild manifestation of the disease due to skewed X-inactivation (Skirton et al. 2005). Furthermore, this finding may also be attributable to the repeat length itself which appears to be associated with toxicity due to elevated mRNA levels (Koldewyn et al. 2008).

Self-stigmatization

There is evidence from four studies to indicate that self-stigmatization occurred in carriers to some extent (Gordon et al. 2003; James et al. 2006; McConkie-Rosell et al. 1997, 2000). Gordon et al. identified that carriers experienced less positive feelings; more afraid, worse, weaker, less relieved, less happy, more marked (although the authors do not explain what is meant by this) and angrier, compared to those who tested negative, on the HOS. Similarly, James et al. found that carrier status is associated with stigma and is significantly associated with mode of inheritance using the same scale. The only other study (Pastore et al. 2008) specifically looking at stigma using the HOS consisted of just one carrier, and therefore findings were not significant. Stigma was also evident in two of the qualitative studies. Just under half (9/19) of the fragile X carriers in one study (McConkie-Rosell et al. 1997) indicated that there had been a negative change in the way that they viewed themselves. One reason cited for this change was a “feeling of being abnormal or inferior” (p. 64), a statement indicative of self-stigmatization.

Social Stigmatization

Evidence of social stigmatization was evident in two studies, one quantitative (Gordon et al. 2003) and one mixed methods (McConkie-Rosell et al. 2001). Gordon et al. found that carriers and non-carriers attributed significantly more negative feelings to cystic fibrosis carrier status than non-carrier status. This finding was significant for all emotional scales on the HOS (p < 0.001).

Active Coping Mechanisms

Use of active coping mechanisms was identified in five studies, out of a possible seven studies in which qualitative research techniques were employed. These studies included participants from the general population without affected children (Anido et al. 2007) and participants with a family history (Anido et al. 2005; McConkie-Rosell et al. 1997, 2000, 2001). McConkie-Rosell et al. (2001) found no change in the level of distress or perceived seriousness of fragile X when women were “at risk” of being a carrier as when they were found to be carriers. The increase in perception of seriousness only occurred in the non-carriers when the threat was no longer present. This possibly indicates that “threat minimization” was used by the participants as an active coping mechanism in both situations. McConkie-Rosell et al. (2001) also found, during in-depth interviews, that 11 out of 20 (55%) carrier women used spontaneous coping statements such as “life goes on” (p.41) and “If I am, I am. I’ll deal with it” (p.41). Coping behavior statements were also evident during interviews in the study conducted by Anido et al. (2007).

For carriers identified in the study by Anido et al. (2007), most appeared to be considering their carrier status over the course of the interview, having not given the subject much thought previously. The authors postulated that this attitude is consistent with the coping mechanism known as “just-in-time” learning, as described in Adult Learning Theory (Wlodkowski 1999), wherein adult learners process information which is relevant and applicable to them at the time they need it.

Impact on Reproductive Issues

The impact of carrier status on participants’ views on reproductive issues varied depending on their life stage, their views on prenatal testing and abortion, whether their partners were also carriers, and whether they were carriers of an X-linked or recessive condition. Authors of four studies (Callanan et al. 1999; Henneman et al. 2002; Lakeman et al. 2008; Watson et al. 1992) of cystic fibrosis carriers identified from both high risk groups and the general population who did not have affected children, all reported that the majority of carriers showed no change in reproductive plans after testing, as measured on questionnaires which included multiple-choice options (Callanan et al. 1999; Watson et al. 1992) or a five point Likert-scale (Henneman et al. 2002; Lakeman et al. 2008). Reasons given included the availability of prenatal diagnosis (Henneman et al. 2002; Lakeman et al. 2008; Watson et al. 1992) and having completed their families (Watson et al. 1992). Furthermore, in two of the studies (Cheuvront et al. 1998; Henneman et al. 2002), only carrier by non-carrier couples were included. If one partner tests positive and the other negative, the risk of having a child with CF is about 1 in 640 (Watson et al. 1992).

However, in two interview-based studies, females carrying X-linked mutations, many of whom were mothers of affected children, were more likely to indicate their carrier status had caused a change to their reproductive plans (Anido et al. 2005; McConkie-Rosell et al. 1997). In the study conducted by McConkie-Rosell et al. (1997) 19 out of 28 (67%) fragile X carriers stated that they would not have any more children because of their carrier status, and 25 out of 28 (89%) would have either reduced the size of their families or not had any biological children, if they had known earlier. Anido et al. (2005) also found through in-depth interviews that many women with fragile X children stopped planning to have more children after receiving their test results. Furthermore, those without affected children expressed a strong desire “to figure out a way to end it with me” (p. 301). Dunn et al. (2008) also reported findings from open-ended questions that revealed some respondents felt they might not have had as many children if they had known their carrier status earlier.

Findings differed however, in the study conducted on fragile X carriers identified from the general population (Anido et al. 2007). Many carriers expressed that although the information could be relevant in the future, it was not relevant at this stage of their lives in terms of family planning. Some had not really considered the implications for family planning and their thoughts about prenatal testing, but for those that had, carrier status did not have an apparent effect on their attitudes about termination. The issue of premature ovarian failure appeared to be more prominent than the risk of having children affected with fragile X.

Disclosure of Test Results and Family Relationships

In six studies in which disclosure of test results was assessed, the researchers found that participants did share their test results with others, although this disclosure was selective (Anido et al. 2007; Dunn et al. 2008; Henneman et al. 2002; McConkie-Rosell et al. 1997; Watson et al. 1992; Williams and Schutte 1997). Anido et al. (2007) found that providing information to partners primarily depended on the seriousness of the relationship. Watson et al. (1992) found that 89% (47/53) of CF carriers informed their partners of their test results, 83% told their parents, 82% their siblings and 48% told other relatives. Henneman et al. (2002) reported that most CF carriers shared the information with parents and siblings. All but one of the carriers whose parents were still alive had told them about their test results. Ten carriers had shared the information with their brothers and sisters, but two had not. With respect to participants who did not disclose carrier information to other family members, their reasons included not wanting to disclose results to relatives who had affected siblings, and not wanting to cause feelings of guilt (Williams and Schutte 1997).

The effects of sharing information about one’s carrier status with a partner and/or family members varied across the studies. Positive experiences related to disclosure of test results were documented by Dunn et al. (2008) and McConkie-Rosell et al. (1997). Of the 18 carriers who indicated in a change in their relationship with their husband in the McConkie-Rosell et al. (1997) study, 13 carriers (72%) indicated this change had been positive. Seventeen (61%) felt that there had been an improvement in their relationship with their siblings. Difficult or distressing experiences were highlighted in three studies (Dunn et al. 2008; McConkie-Rosell et al. 1997; Williams and Schutte 1997). Dunn et al. and McConkie-Rosell et al. identified a negative effect on the relationship with the partner in 13/31 (42%) and 5/18 (27%) of cases, respectively. Reasons cited included anxiety and anger from the male partner (Dunn et al.) and feeling blamed by their spouse (McConkie-Rosell et al.). In cases where the experience had a positive effect on the relationship (in 4/31 and 13/18 of cases respectively), the carrier felt completely accepted by her partner (Dunn et al. 2008) and there was an increase in understanding and communication. Henneman et al. (2002) found the majority of participants (98%) perceived no impact of carrier testing results on the relationship with their partner. For the majority of participants in the Anido et al. (2007) study, providing information about fragile X carrier status to family members was not problematic. However, providing the information to partners depended on the seriousness of the current relationship.

Discussion and Conclusion

Discussion

This review is useful in that it identifies a number of factors that seem to influence the emotional consequences of carrier testing. These include population group, whether the carrier has an affected child, stage of life, psychological coping mechanisms, and mode of inheritance. In this respect the results of this systematic review provide some interesting insights into how genetic testing for different conditions may have a varying psychological impact that is dependent on the context in which testing occurs.

Anxiety, an emotion frequently measured in studies investigating the impact of carrier testing on individuals, dissipated in the long term for the majority of participants in all studies. In addition, the reasons suggested by authors, another reason may have been because none of the participants were pregnant at the time of receiving their carrier test results and were therefore not anxious about the possibility that the fetus was affected. For carriers, knowledge that reproductive options were available to them if there was a risk of having an affected child may also have overridden any initial anxiety. Furthermore, good quality genetic counseling services may have lessened the impact of the test results.

Variables including mode of inheritance, gender and whether the carrier already had a child affected by the condition appear to be strongly linked to the issue of guilt. The finding that guilt was more dominant in women than men, indicates that it may be strongly connected with what Peters and Jackson (2009) describe as a unique emotion concerning a mother’s relationship with her affected child. Guilt also appeared to be more commonly reported by mothers of children with X-linked conditions. One possible explanation lies in the close association of guilt and blame. In the case of X-linked conditions, it only takes a carrier mother to pass along an X-linked condition rather than having both parents contribute the “faulty” gene. Therefore the burden of having passed on a faulty gene cannot be shared with a partner. In these cases men may “externalize their emotional response to devastating news and blame, while women are likely to internalize their responses and to accept this blame” (James et al. 2006). Mothers are also more likely to self blame (Peters and Jackson 2009).

Guilt may also be an emotion linked to family history. Anido et al. (2005) found that women who did not have affected children but had the condition in the family, expressed feelings of guilt, which may indicate a form of “survivor guilt.” Survivor guilt has also been identified in CF families. In a study (which was excluded from this review as it contained women who were pregnant at the time of testing) in which barriers to carrier testing for adult cystic fibrosis siblings were identified, carrier status served an important function in binding guilt, with 15% of siblings either hoping they were carriers or feeling guilty they were not (Fanos and Johnson 1995a).

There are conflicting results in the literature regarding the issue of perceptions of health. Yet even those studies in which carriers did not indicate feeling less healthy on surveys or questionnaires, during in-depth interviews, some participants reported clinical features of the disorder for which they were being tested or for which they were found to be a carrier. In the case of fragile X, it is possible that carriers did experience a mild manifestation of the disease due to skewed X-inactivation (Skirton et al. 2005). Furthermore, this finding may also be attributable to the repeat length itself which appears to be associated with toxicity due to elevated mRNA levels (Koldewyn et al. 2008). However this finding also suggests that participants may have been seeking support for beliefs they held about themselves.

In interpreting this finding, McConkie-Rosell et al. (2000) refer to the theory of self-concept as described by Shavelson et al. (1976). Shavelson et al. hypothesize that self-concept is hierarchical, with perception of personal behavior in specific situations at the base of the hierarchy, inferences about the self in broader domains (e.g., social, physical) at the middle, and a global, general self-concept at the apex. Global self-concept is stable, but as one descends the hierarchy self-concept becomes increasingly situation specific and less stable. Seeking clinical features related to actual or possible carrier status might be indicative of situation-specific changes in feelings about self. Additionally, it may be the case that scales such as the HOS (used by Gordon et al. 2003) and TSCS (used by McConkie-Rosell et al. 2000) are not sensitive enough to detect the subtleties concerning how carriers perceive their own health, which are more likely to be expressed during in-depth interviews.

Reproductive intent also appeared to be closely linked to mode of inheritance, stage of life and whether the participant already had an affected child, with the greatest impact being identified for carriers of X-linked conditions with affected children. This group was most likely to refrain from having more children. One possible reason involves the documented psychological difficulties of raising a child with fragile X (Abbeduto et al. 2004; Lewis et al. 2006). When Anido et al. (2007) interviewed fragile X carriers who did not have affected children and were from the general population, the information did not appear to have an impact on family planning with many not having considered the issue. This is likely to be because they did not have any experience, either themselves or through other family members, of raising a child with the condition. It may be that these carriers would experience increased distress as they consider reproduction more seriously. Similarly, carriers of cystic fibrosis in the general population did not change their reproductive plans as a result of their carrier status. Participants in these studies did not have affected children, and even as a carrier, there would only be a risk to future children if the partner was also a carrier.

Active coping mechanisms, such as “threat minimization,” significant changes to reproductive intentions and the use of active coping statements, were identified in those participants at an increased risk of carrying the fragile X gene. Lazarus and Folkman (1984) describe coping as consisting of two different strategies, problem-focused coping and emotional-focused coping. The findings from this systematic review suggest that women at high risk of being a carrier of fragile X engaged in problem-focused coping by managing their health threat through genetic testing, and if found to be carriers, by changing their reproductive intentions. They engaged in emotional-focused coping through threat minimization and active coping statements.

In addition to these coping strategies aimed at lessening distress, Lazarus and Folkman describe a smaller group of cognitive strategies directed at increasing distress. For some individuals, there is a need to feel worse before they can feel better. Self-blame, a coping mechanism found to be used by carriers of X-linked conditions, is one such form of self-punishment individuals may use. This deliberate emotional distress may mobilize individuals into action. Evidence that women use self-blame as a coping strategy has been identified in other studies; for example, self-blame was found to be significantly correlated with both problem-focused and emotional-focused coping strategies in a study of patients with diabetes (Tuncay et al. 2008). Self-blame was also used as a strategy to cope with depression in a study of how primary care patients manage their illness (Brown et al. 2007).

Other studies, in which participants became aware of their carrier status through family history or newborn screening, have identified similar psychosocial issues to those in this review. Fanos and Mackintosh (1999) recognized a number of coping mechanisms used by parents of children with ataxia-telangiectasia, including rationalizing their child’s condition as a “statistical quirk” (p.417), and imbuing the occurrence with meaning and significance through connecting it with the wider sphere of human suffering or to the spiritual world. Guilt was not however a common finding in their study, and surprisingly when it was mentioned, it was in reference to fathers. Undue concerns about the health of carriers was also identified in a minority of parents in a study assessing the impact of carrier status information following newborn screening (Kai et al. 2009), as was a sense of responsibility to share carrier status information with extended families. Stigmatization was also evident in a study which included participants from high risk CF families who did not want to learn their carrier status (Fanos and Johnson 1995a). For example, one untested woman was worried that she would be “less desirable” (p. 88) to men if they knew she was a carrier.

While this review provides an overview of the psychosocial experience of living as a carrier, it is important to keep in mind the limitations of making comparisons across different conditions, in particular cystic fibrosis and fragile X (the major conditions included in this review). These two conditions vary greatly in terms of their effects on the affected individual, the implications for the health of the carrier, and risk of the carrier having an affected child. Furthermore, variations in study design, the different population subsets compared, and the obvious complexities of comparing qualitative and quantitative data, mean that the findings should be interpreted with some degree of caution. For example, there were indications from some studies using validated scales of no changes in perception of health. However, when the authors used in-depth interviews, changes in health perception were evident (Bekker et al. 1994; McConkie-Rosell et al. 2000). Some authors used the STAI to measure anxiety, whereas others using qualitative methods relied on participants’ own terminology. Studies using the HOS were much more likely to identify evidence of stigmatization that those that did not use this scale, as this scale specifically measures aspects of self image. Future systematic reviews may therefore benefit from the inclusion of samples involving population groups which are more similar in kind in terms of risk to offspring, severity of the condition or family history. Future research studies may be better summarized if the studies focus on using similar groups of patients and validated tools.

Yet this does not necessarily mean the findings of the present review fail to provide valuable insight into the psychosocial experience of living as a carrier. In particular, the review provides an overview of the commonality of experiences across conditions with different inheritance patterns. Furthermore the overview identifies a number of issues that collectively apply to carriers as a group, because of the familial nature of genetics.

Strengths and Limitations

As stated previously, findings from the review should be considered in light of the difficulties and limitations of combining studies undertaken with different study designs, subsets of the population, measures and outcomes. These factors may have diluted the strength of the comparisons. Furthermore, many of the studies lacked theoretical models or presentation of a conceptual model to help place the variables and their possible interactions in context (Henneman et al. 2002; Pastore et al. 2008; Watson et al. 1992). Such omissions possibly weaken the validity of the results. Nevertheless, in the present systematic thematic analysis, the findings were able to be explained within established theoretical models of coping and self-concept (Lazarus and Folkman 1984; Shavelson et al. 1976).

The systematic review does have notable strengths. Seven databases were used to retrieve studies to maximize the chance of finding all relevant research. In addition, several iterations of the search were conducted using different combinations of keywords, to ensure the search was rigorous. At the present time there does not appear to be another systematic review in the literature that compares the psychosocial experience of carrier testing for autosomal recessive and X linked conditions; thus, this review provides unique and useful information.

Conclusion

The findings from this systematic review provide insight into the variety of psychosocial emotions experienced by individuals undergoing carrier testing and a general overview of the psychosocial impact of living as a carrier. Prominent themes that occur in the literature include anxiety, guilt, relief, effect on self image, active coping mechanisms, impact on reproductive issues and disclosure of test results. Variables that influence the psychosocial effects of carrier testing include whether the carrier has an affected child, mode of inheritance, genetic counseling and life stage. A key finding concerns the different emotions experienced by carriers who already had an affected child compared with carriers who did not. Studies indicated that carriers with affected children were more likely to experience guilt and self-blame. Furthermore, fragile X carriers with affected children were more likely to indicate that carrier status had affected their reproductive plans. In contrast, carriers identified from the general population did not change their reproductive plans as a result of their carrier status. Due to the commonality of experiences identified through this systematic review, it would appear that we can make certain assumptions about the psychosocial impact of living as a carrier. Yet at the same time it is important to bear in mind the limitations of making generalizations across different population groups and condition types.

Practical implications

Genetic counselors and other health professionals offering genetic testing should pay attention to the variety and complexity of psychosocial experiences that may be encountered by individuals undergoing carrier testing. One key finding from this systematic review is that carriers who already have an affected child often react differently when receiving their test results than carriers who do not. For those carriers who already have an affected child, the impact of receiving the test results in these cases may reinforce feelings of guilt, self-blame and maternal blame in the case of X linked conditions. Counselors therefore need to be aware of these issues when testing parents of affected children as these psychological issues may need to be addressed both before and after testing. In addition, counselors should look to address misconceptions related to health and carrier status; some individuals may seek support for beliefs they have about their health by identifying clinical features of the disorder for which they are being tested or are found to be a carrier. Furthermore, while some clients will effectively manage anxiety and their carrier status through threat minimization and other active coping mechanisms, professionals should ensure that those who appear to be managing well do not minimize their threat to the extent that they disengage from protective health actions, particularly when it comes to reproductive issues.