Article Text

Download PDFPDF

Psychological effects of diagnosis and treatment of cervical intraepithelial neoplasia: a systematic review
  1. Maria Eiholm Frederiksen,
  2. Sisse Njor,
  3. Elsebeth Lynge,
  4. Matejka Rebolj
  1. Department of Public Health, University of Copenhagen, Copenhagen K, Denmark
  1. Correspondence to Maria Frederiksen, Department of Public Health, University of Copenhagen, Øster Farimagsgade 5, Copenhagen K, DK 1014, Denmark; marfr{at}sund.ku.dk

Abstract

Background Treatment of cervical intraepithelial neoplasia (CIN) is a common minor surgical procedure to prevent uterine cervical cancer. However, news of an abnormality detected at screening for cancer might cause the woman to worry.

Objectives To investigate the psychological consequences of CIN diagnosis and treatment in a systematic review.

Data sources We searched PubMed using Medical Subject Headings (MeSH) terms for articles published from January 1990 to February 2013. We also examined the reference lists of retrieved articles.

Selection criteria Quantitative studies measuring psychological outcomes in women with a histological diagnosis or treatment of CIN, and in women having an outcome other than CIN at cervical screening.

Data collection and analysis We abstracted the data using a pre-specified list of study characteristics and measured outcomes. For studies not reporting statistical testing, we estimated the statistical significance of the differences between the compared groups using unpaired t tests.

Main results From 5099 retrieved abstracts, 16 studies were included. Diagnosis and treatment of CIN were associated with worse psychological outcomes than normal cytology test results, but the impact decreased over time. In several but not all studies, CIN appeared to have similar psychological consequences to abnormal smears. No study showed a difference in psychological outcomes between CIN and cervical cancer diagnosis when these were measured some years after diagnosis.

Conclusions The studies suggested that CIN diagnosis and treatment have a negative psychological impact. However, this conclusion should be viewed in the context of a paucity of rigorously designed studies.

Keywords
  • cervical intraepithelial neoplasia
  • mass screening
  • diagnosis
  • conization
  • psychology
View Full Text

Statistics from Altmetric.com

Keywords

Introduction

Uterine cervical neoplasms are caused by oncogenic types of human papillomavirus (HPV).1 HPV is sexually transmittable and very common. A diagnosis of cervical intraepithelial neoplasia (CIN) can be made in approximately 20% of infected women.2 On average, it takes over 10 years for CIN to progress to cervical cancer.3 In order to prevent this, several countries have implemented mass screening and have achieved varying coverage rates (eg, approximately 75% in Denmark).4 Women with a screen-detected abnormality are monitored or referred for colposcopy for the purpose of diagnosing CIN. Treatment of screen-detected CIN has been highly effective in halting progression to cancer.5 However, as HPV infections and CIN can regress spontaneously, only a minority of CIN lesions are harmful. For example, the most severe grade of CIN, CIN3, progresses to cervical cancer in about 30% of cases in 30 years.6 Progression biomarkers have not yet been identified, so all high-grade CIN (CIN2/3) tend to be treated. CIN treatment is therefore quite common. In Denmark, where the lifetime background risk of cervical cancer is estimated at just above 3%, the lifetime risk of CIN treatment is 14–16%.7

Historically, it was not uncommon to treat CIN with hysterectomy and other techniques that required inpatient care. Nowadays, CIN treatment is a minor surgical procedure undertaken primarily as conisation in the form of large loop excision of the transformation zone (LLETZ) in outpatient settings.7 The side effects of LLETZ are relatively minor.8 Whether modern CIN treatment modalities can potentially lead to adverse obstetric outcomes has been evaluated in a meta-analysis and intensively debated.9 ,10

As explained above, cervical screening can protect some women from developing cervical cancer, but many more will be affected by its side effects. For example, six to eight women undergo CIN treatment per prevented cervical cancer case,7 and 2–10% of screened women have positive screening test results without CIN (false-positive tests) at each screening round.11 Clinically inconsequential screen-detected abnormalities, either false-positive tests or overdiagnosed CIN, are inevitable given that screening tests cannot be 100% accurate, and because CIN often regresses. However, these findings may induce distress in women. In their influential review from 1992, Wardle and Pope called screening-induced psychological side effects the ‘psychological costs of screening’.12 Although they are often seen as having secondary importance compared to the clinical benefits of screening, these effects need to be taken into account in the fine-tuning of screening programmes.

Apart from the rare findings of cervical cancer, histologically confirmed CIN is the worst screening diagnosis. The psychological consequences of CIN diagnosis and treatment have been studied for decades. In 1982, Kilkku et al13 analysed several aspects of sexual functioning after CIN treatment. One year after treatment, libido, experience of orgasm, coital frequency and overall satisfaction with sex remained similar as in the period around treatment. In 1988, Posner and Vessey reported that CIN treatment caused considerable distress for almost half of the women treated.14 Unfortunately, neither of these two important early studies used control groups, so the effects of CIN could not be distinguished from competing sources of emotional disturbance. Other studies have since been published on psychological outcomes in women with a CIN diagnosis treated with modern modalities. These also included various types of control groups (eg, women with normal smears, abnormal smears or cervical cancer), so that the effects of CIN could be better understood than in the early studies.

The aim of this systematic review was to investigate the psychological consequences of being diagnosed and/or treated for CIN, by comparing these women to others with different cervical screening outcomes.

Methods

Literature search

In collaboration with an expert librarian, we searched PubMed using Medical Subject Headings (MeSH) terms for studies published between January 1990 and February 2013 (see online supplementary appendix S1). Two researchers (MEF, MR) independently screened the retrieved abstracts, and examined the reference lists of all included manuscripts to identify any additional studies. Both researchers independently retrieved the data from the identified studies and evaluated their quality. Consensus was sought in case of disagreement.

Inclusion criteria

Included manuscripts described original quantitative studies of women who had been informed of their histologically confirmed CIN diagnosis excluding any other cervical diagnoses (eg, condyloma, cervical cancer) or had received CIN treatment, and did not include women with significant comorbidities, for example, HIV infection. These women were defined as our study group. We included longitudinal studies and cross-sectional studies. For longitudinal studies, data had to be reported for both before and after CIN diagnosis for the same woman so that each woman served as her own control. For cross-sectional studies, study groups were compared to control groups, defined as women with a cervical screening diagnosis other than CIN (normal or abnormal cytology, cervical cancer; a diagnosis of external genital warts (EGW), although not strictly a screen-detected diagnosis, was also accepted), or a group of women undergoing non-therapeutic procedures typical of cervical screening (screening itself, colposcopy). Psychological outcomes (including, for example, anxiety, depression, anger, distress, well-being) had to be measured in both study and control situations (longitudinal studies), or in both study and control groups (cross-sectional studies). Although psychological and physiological aspects are sometimes difficult to distinguish from each other,15 sexual outcomes were considered to be an indicator of psychological outcomes. Other outcomes, such as preferences for, or satisfaction with, procedures and somatic consequences (pain, bleeding, discomfort, disability) were not considered here.

Manuscripts reporting only the significance of the differences between the situations/groups but not disclosing the absolute outcomes were excluded. In several identified studies, a comparison of psychological outcomes between the situations/groups was not the primary objective. If, nevertheless, psychological outcomes were measured and reported, those studies were included in our review. No other criteria were imposed.

Statistical analysis

We abstracted the data using a pre-specified list of study characteristics (tables 14; results on psychological outcomes are further detailed by study design, type of outcome, study and control groups in online supplementary tables S1 and S2). For studies that did not report statistical testing for differences in psychological outcomes between the study and control groups, we assumed that the data were normally distributed. Using the reported information on the mean and the SDs, we calculated p values using unpaired t tests (results reported in parentheses in tables 3 and 4, or in square brackets in appendix file).

Table 1

Study designs and participants

Table 2

Socio-demographic characteristics of the participants

Table 3

Narrative summary of psychological outcomes after CIN in longitudinal studies

Table 4

Narrative summary of psychological outcomes after CIN in cross-sectional studies

We assessed the methodological quality of the studies following the criteria proposed by Downs and Black (see online supplementary appendix and table S1).16 Some of their criteria, however, had to be ignored, particularly because they were only appropriate for (randomised) intervention studies. Nevertheless, all categories on the checklist (reporting, external validity, internal validity—bias, internal validity—confounding (selection bias), power) could be assessed.

Results

Search results

The search identified 5099 abstracts. Based on titles and abstracts, we retrieved 44 full-text manuscripts. Of these, 15 were consistent with the inclusion criteria; two were from the same study.15 ,17–27 ,30–32 Examination of reference lists identified two additional manuscripts published since 1990.28 ,29 In total, therefore, 17 English-language articles describing 16 studies and published between 1993 and 2011, satisfied our inclusion criteria (see online supplementary figure S1). We were not able to identify any other systematic reviews on the same topic.

Four studies measured psychological outcomes longitudinally, while 15 studies were cross-sectional (table 1). The number of participants varied from 30 to 2605. Women's characteristics (age, education, marital status) are summarised in table 2. In the studies with reported data, most women were married, while educational level varied substantially by study. Both disease-specific and general measures were used for assessment of outcomes, leading to a wide variety of evaluated psychological indicators (anxiety, distress, ‘psychosocial impact’, etc). Studies generally used well-validated questionnaires, including the Spielberger State Anxiety Inventory, State-Trait Anger Scale, and Hospital Anxiety and Depression Scale, although a minority of studies surveyed the women using questionnaires developed by the researchers themselves (see online supplementary tables S2 and S3). Eleven studies were undertaken in predominantly English-speaking countries (USA, Canada, UK, Ireland, Australia), while two were undertaken in Thailand, and one each in Sweden, Taiwan and China.

Results of longitudinal studies

All longitudinal studies compared psychological outcomes before and after CIN treatment (table 3). Palmer et al17 found that distress and state anger did not differ between about a week after treatment and about a week after the women had been informed that they had CIN. Walsh et al19 detected more anxiety before colposcopy compared to after CIN treatment, while anxiety further decreased in the months following treatment. In the study by Hellsten et al,21 frequency of sexual intercourse, spontaneous interest in sex and sexual arousal appeared less frequent 4 months after CIN treatment than in the 3 months before the initial colposcopy. Interestingly, negative feelings towards sex decreased after treatment. By 2 years after treatment, only spontaneous interest in sex remained at a lower level than before the initial colposcopy. Inna et al15 reported that 3 months or more after CIN treatment, women on average had less overall sexual and orgasmic satisfaction and reported less vaginal elasticity than before treatment. Frequency of sexual intercourse, dyspareunia, sexual desire, vaginal lubrication, sex-related anxiety, and perceived partner satisfaction did not change.

Results of cross-sectional studies

CIN diagnosis

As explained above, most identified studies were cross-sectional, establishing psychological outcomes at a particular time point (eg, x days/weeks after the diagnosis). These studies surveyed various control groups, ranging from women with a normal smear (the least burdensome and most frequent outcome in cervical screening) to women with cervical cancer (the most severe and least common outcome; table 4). The differences were, as expected, most pronounced when women with a CIN diagnosis were compared to women with normal cytology. This was the case for distress and state anger in a study by Palmer et al,17 anxiety about cervical cancer in a study by Salz et al,30 and ‘psychosocial impact’ (combining, for example, worry about future health and fertility, depression, anxiety, sexual and partnership problems, deterioration of self-image, effect on sleep) in studies by Mast et al28 in the USA, Pirotta et al27 in Australia, Wang et al31 in China and Wang et al29 in Taiwan. Measured with the latter instrument, the psychosocial impact of a CIN diagnosis was less negative than that of EGW in the two studies undertaken in Chinese-speaking populations and in the US study.28 ,29 ,31 In the Australian study, no differences were found for psychosocial impact and quality of life between women with CIN2/3 and women with EGW.27

The psychological effect of a histologically confirmed CIN diagnosis was compared to that of abnormal cytology in several studies. In the study by Pirotta et al,27 the psychosocial impact in women with histologically confirmed CIN2/3 was not worse than in women with low-grade squamous intraepithelial lesions (LSIL) or worse on cytology. The psychosocial impact was worse for women with histologically confirmed CIN1 than for women with atypical squamous cells of undetermined significance (ASCUS) on cytology in the study by Mast et al.28 In the same study, the psychosocial impact of histologically confirmed CIN did not differ from that of LSIL and high-grade squamous intraepithelial lesions (HSIL) on cytology. After abnormal cytology, however, women appeared to be reassured by a negative HPV test result, and reported a much lower psychosocial impact than other women with screen-detected abnormalities. The opposite was seen in women whose abnormal cytology samples were followed with an HPV-positive test. Also in the Taiwanese study by Wang et al,29 receipt of a positive HPV test after abnormal cytology had a similar psychological impact as a histologically confirmed CIN diagnosis. In contrast, the large Chinese study by Wang et al31 reported that a histological CIN diagnosis had a significantly worse psychosocial impact compared to a diagnosis of an abnormal smear.

In a small study, Pruitt et al24 found no differences in perceived seriousness of and worry about the condition between women with a CIN1 diagnosis and women with a normal cervical biopsy.

After an unspecified period since the diagnosis in the study by Taneepanichskul et al,32 the quality of life as well as the social, functional and emotional well-being of Thai women with a previous CIN diagnosis was similar to that of women with a previous diagnosis of cervical cancer.

CIN treatment

Studies on women with CIN treatment also included several different control groups, as did the studies on women with a CIN diagnosis. In the studies by Freeman-Wang et al18 and Le et al,20 women appeared to be as anxious and distressed before undergoing CIN treatment as when about to undergo the initial colposcopy; in the study by Le et al,20 women received additional counselling immediately after colposcopy. Women surveyed by Kola and Walsh25 reported feeling more unpleasantness while undergoing CIN treatment in a see-and-treat setting compared to women who underwent colposcopy and no treatment. However, anxiety in a study by Walsh et al19 and embarrassment in the study by Kola and Walsh25 did not differ between the treatment and no-treatment groups referred for colposcopy because of abnormal cytology.

During the 2 years after CIN treatment, state anxiety in the study by Walsh et al19 and sexual functioning in the Swedish study by Hellsten et al21 were similar to findings after colposcopy that did not lead to treatment. Nevertheless, in the same study treated women appeared calmer than untreated women.22 Finally, several years after CIN treatment, psychological outcomes such as anxiety, depression and psychological impact (concerns regarding sexual activity, fertility and developing cervical cancer) in a study by Cairns23 (measured over 6 years or less) and quality of life and self-esteem in a study by Bartoces26 (measured over 5–8 years), appeared to be similar to those in women who were diagnosed with cervical cancer.

Discussion

Main findings

Psychological outcomes in women with a CIN diagnosis or after CIN treatment have often been studied, but the results are difficult to interpret because studies used a broad range of measures assessed at different stages of post-screening care. Nevertheless, some consistent patterns emerged. First, a CIN diagnosis is associated with short-term distress, anxiety and anger, and has a negative psychosocial impact which gradually wanes. However, no study compared the long-term impact of CIN with that of normal cytology, so it is not known whether over time the psychological burden reduced to background levels. Second, visible EGW tended to be associated with poorer psychological outcomes than a CIN diagnosis. Interestingly, women considered a CIN diagnosis to be as bad as one of cervical cancer; however, all studies assessed survivors years after diagnosis, after women had adapted to the situation. Third, treatment did not appear to have an additional negative psychological impact compared to a CIN diagnosis. This impact affected specific aspects of sexual functioning for at least 2 years, but was only evaluated in one study.

Strengths and limitations

A limitation of our review was that the search was undertaken in only one database. To (partially) compensate for this limitation, we examined the reference lists of retrieved studies, which helped us to identify an additional two relevant studies.

The study quality checklist16 revealed several methodological issues regarding the studies. First, external validity was often impossible to determine because selection processes were not reported in sufficient detail. Second, several studies used relatively small samples, and only two supported their study size by reporting a power calculation. Third, some studies did not compare the study groups on socio-demographic characteristics, while only a small number of studies adjusted for such characteristics in their statistical analyses.

Other limitations should also be considered. In several studies, statistical testing was not always undertaken to compare the groups that we were interested in. We performed statistical testing assuming normally distributed data. In some studies, women with CIN differed from women in control groups regarding socio-demographic characteristics, but we were not able to adjust for this. Hence, our p values may vary from those that would have been obtained on the actual data, and all statistical testing denoted by parentheses in tables 3 and 4 should be interpreted with caution.

Another problem was related to study designs. Cross-sectional studies compared groups of women who might have differed on factors other than the screening outcome. Ideally, selection bias should have been controlled for by also studying both groups before CIN was diagnosed. Furthermore, many psychometric measures are prone to recall bias. It is also unclear whether the psychological reactions measured by psychometric instruments actually caused life changes, for example, whether problems in sexual functioning led to dissolution of partnerships, or whether anxiety was treated with psychotherapeutic drugs. Methodological improvements accounting for these problems should be implemented in future research.

Interpretation

The most interesting finding of this review was that the psychological outcomes of women with CIN were often similar to those of women with abnormal cytology, that is, screen-detected abnormalities not necessarily needing treatment. In the studies using the HPV Impact Profile (HIP) questionnaire,27–29 ,31 women with CIN were compared to women with normal and with abnormal cytology. Results showed that the psychosocial impact scores tended to be somewhat lower for women with abnormal cytology than for women with CIN. However, the scores of these two groups were more similar to each other than the scores for women with abnormal and women with normal cytology. This was the case even in the large Chinese study where the psychosocial impact of CIN was significantly worse than that of abnormal cytology.31

A plausible explanation for this similarity is that women with abnormal cytology were anxious primarily because of uncertainty about their true disease status. In the study by Hellsten et al,22 6 months after learning that they did not need treatment, 42% of women still reported that they did not feel calm. Recently, Korfage et al33 reported that women with abnormal cytology and normal histology follow-up reported similar health-related quality of life and anxiety as those who received treatment for high-grade CIN. Hence, Hellsten et al21 ,22 and Salz et al30 concluded that lesion severity did not determine psychological impact but, rather, that the detection of any abnormality at screening seemed to be a stressful event. These findings suggest that efforts to avoid false-positive screening tests would likely reduce screening-induced anxiety.

In women with positive screening results, anxiety levels were much higher than in the general population, and were often comparable to those after a traumatic event.17 ,18 ,34–36 It has been proposed that women's worries should be reduced by offering information, education and/or counselling. These strategies improve women's knowledge of the abnormalities, but they have limited value in decreasing women's anxiety, particularly when women avoid seeking information (‘blunters’) or are very anxious.34 ,35 ,37–39 Measures to reduce anxiety specifically in women with CIN, for example, while waiting for treatment, have been studied only rarely. It has been argued that a see-and-treat approach can reduce anxiety.36 It should, however, be noted that see-and-treat can lead to over-treatment. During CIN treatment itself, Cruickshank et al40 found that putting the woman in charge of administering analgesic gas was helpful, particularly in clinically anxious women. Clearly, more research is needed to identify anxiety-reducing interventions for women with CIN.

Conclusion

The reviewed studies suggested that diagnosis and treatment of CIN have a negative psychological impact. The impact appeared to be similar to that of other cervical abnormalities found at screening. Nevertheless, these conclusions should be viewed in the context of a paucity of rigorously designed and reported studies.

Key messages

  • Findings of the reviewed articles indicated that cervical intraepithelial neoplasia diagnosis and treatment appeared to have a negative psychological impact on women.

  • This impact appeared to be similar to that of other positive cervical screening results, but its strength decreased over time.

  • However, improved study methodology in the future could enable a more precise evaluation of the nature of the psychological impact.

Acknowledgments

The authors would like to thank Lene Borrits, MD (Danish Royal Library) for help with the PubMed search.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

  • Abstract in Danish

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Handling editor Jackie A Cassell

  • Contributors MEF and MR: designed the study, analysed the data, interpreted the results, and wrote the manuscript. SN: contributed to the analyses and interpretation of the data, and participated in writing the final manuscript. EL: participated in the interpretation of the data, and in writing the final manuscript. All authors made the decision to submit. All authors had full access to all of the data in the study.

  • Funding MEF, SN and MR were financed by the Danish Strategic Research Council. The researchers worked independently of the funders. Grant number: 10-092793.

  • Competing interests MEF and SN declare no potential conflicts of interest. MR and her employer received honoraria for lectures from Qiagen. EL served as an unpaid advisor to Hologic/Gen-Probe and Norchip. MR and EL work on a study which includes collaboration with four manufacturers of HPV assays: Qiagen, Roche, Hologic/Gen-Probe and Genomica. None of the authors received salary or bonuses from any of the companies.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement As this is a systematic review of peer-reviewed studies, all data are publicly available.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.