Dr IJ Korfage, Department of Public Health, Erasmus MC, PO Box 2040, 3000 CA Rotterdam, the Netherlands. Email firstname.lastname@example.org
Please cite this paper as: Korfage I, van Ballegooijen M, Wauben B, Looman C, Habbema J, Essink-Bot M. Having a Pap smear, quality of life before and after cervical screening: a questionnaire study. BJOG 2012;119:936–944.
Objective To assess the health-related quality of life (HRQoL) impact of cervical cancer screening in women with normal test results.
Design Questionnaire study.
Setting Maastricht, the Netherlands.
Population A cohort of 789 women were followed from screening invitation until after the receipt of screening results. A female age-matched reference group (n = 567) was included.
Methods Questionnaires were sent to the home address of the women before screening, after screening, and again with the screening results.
Main outcome measures Generic HRQoL (SF-12, EQ-5D), generic anxiety (STAI-6), screen-specific anxiety (PCQ), and potential symptoms and feelings related to the smear-taking procedure.
Results A total of 60% of screening participants completed questionnaire 1(n = 924): 803 of these women granted permission to access their files; 789 of these 803 women had normal test results (Pap 1), and were included in the analyses. Generic HRQoL (SF-12, EQ-5D) and anxiety (STAI-6) scores were similar in the study and reference groups. Before screening, after screening, and also after the receipt of test results, screening participants reported less screen-specific anxiety (PCQ, P < 0.001) than the reference group (n = 567), with differences indicating clinical relevance. 19% of screening participants were bothered by feelings of shame, pain, inconvenience, or nervousness during smear taking, and 8 and 5% of women experienced lower abdominal pain, vaginal bleeding, discharge, or urinary problems for 2–3 and 4–7 days, respectively, following the Pap smear.
Conclusion The reduced levels of screen-specific anxiety in screening participants, possibly indicating reassurance, are worthwhile addressing in more depth. We conclude that although considerable numbers of women reported unpleasant effects, there were no adverse HRQoL consequences of cervical screening in women with normal test results.
For a relatively small number of women, cervical cancer screening programmes produce large benificial effects: that is an early diagnosis and treatment of cervical cancer, or its precursor, resulting in a prolonged life. Small adverse effects, for example false-positive test results and unnecessary treatment of inconsequential disease, occur much more frequently.1 As the number of women involved is large, small adverse effects may result in a considerable adverse impact on the health-related quality of life (HRQoL) of the population. Therefore, monitoring the benificial and adverse effects of the screening procedure is important. The impact of an abnormal smear result on women’s HRQoL has been described before, showing increased levels of anxiety or fear in women receiving mildly abnormal smear results.2–4 In a cross-sectional design, Wardle et al.5 found that women with normal results of routine cervical smear tests reported lower anxiety and better general health than women who received abnormal smear results. And one study described the impact of brief training for nurses regarding improvements in the screening procedure on the levels of fear in women participating in the screening.6 However, studies reporting the HRQoL impact of the screening procedure itself and the impact of normal smear results, occurring in over 90% of the cases in most European countries,7 range from scarce to non-existent.
We aimed to fill this gap. We will address the screening procedure as experienced by the large majority of participants: the receipt of an invitation to participate in cervical cancer screening; having the Pap smear test; awaiting the result; and finally receiving a normal test result. We hypothesised that the physical health of women would not be affected by the process of invitation, smear taking and receiving the smear results, that screen-specific anxiety would slightly increase after smear taking and before receiving the smear result, that younger women would report more impact on HRQoL and anxiety scores at invitation, smear taking and on receiving the smear results than older women, and that screening invitees and an age-matched reference group would report similar generic HRQoL and anxiety scores.
Cervical cancer screening in the Netherlands
Since the 1980s, cervical cancer screening with the Pap smear has been offered to the female population of the Netherlands. Since 1996, all women aged 30–60 years receive an invitation to attend screening free of charge once every 5 years. The national uptake rate is 65%,8 and is 71% in the region were this study was conducted. Neither HPV screening nor HPV vaccination had yet been introduced.
We conducted a cohort study in cooperation with the regional screening organisation in Maastricht (the Netherlands) between April and August 2006.
We randomly selected a sample from all women aged 30–60 years living in the Maastricht area, and due to be invited for screening. These women were sent an initial questionnaire, described in more detail below, that was attached to the invitation for screening. A leaflet about the screening programme and the screening test that is routinely sent to all women invited for cervical screening was also attached. In an accompanying letter the women were asked to complete the enclosed questionnaire, a second questionnaire after the smear-taking procedure, and a third one after the receipt of the test result (see Figure 1). Women were assured that non-participation did not have any consequences for their medical care. If the initial questionnaire was not returned within a month, a reminder letter was sent. For logistic reasons women were not sent second or third questionnaires if the time lag between invitation and smear taking was longer than 4 months. Inclusion criteria for the current paper were: (1) completion of the initial questionnaire; (2) participation in cervical cancer screening after the invitation; (3) permission granted to obtain screen results from the screening organization; and (4) normal Pap smear test results (Pap 1). Hence, this paper focuses on women who had a Pap smear taken (‘participants’) and had a normal test result.
To assess age-related non-response bias, the average age of respondents was compared with that of women who were sent a questionnaire, had a Pap smear taken, but did not complete the questionnaire.
This study was part of a comprehensive evaluation of the Dutch cervical cancer-screening programme. The ethics review committee of the Erasmus MC, University Medical Center Rotterdam, approved the research protocol. Participating women gave written informed consent.
Characteristics of respondents
Information on education, employment status, marital status, and having children or not was obtained through the initial questionnaire. Education level was classified as low (primary school or lower technical education), intermediate or high (college/university degree).
Only upon receiving permission from the women involved were individual test results obtained from the screening organisation.
To facilitate the interpretation of findings from our comprehensive evaluation of the Dutch cervical cancer-screening programme, a female age-matched reference group was included. Randomly selected women (aged 30–70 years, stratified in 10-year age groups) who were not due for cervical cancer screening within the following 2 years were sent a questionnaire through the regional screening organization in Maastricht. The content of the questionnaire was similar to the initial questionnaire for screen participants (see below). The overall response rate was 46%. In the analyses for the present study only data from women eligible for cervical cancer screening were included, that is women aged 30–60 years (n = 563).
Younger versus older women
We compared HRQoL and anxiety scores between women aged 30–35 years and women aged 36–60 years.
Content of the questionnaires
Questionnaires 1, 2, and 3 included validated measures on generic HRQoL, generic anxiety, and screen-specific anxiety, which we have also used in earlier studies.4 Generic HRQoL was assessed through the EuroQol classification (EQ-5D) and the 12-item Short-Form Health Survey (SF-12). The EQ-5D classification consists of five items (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression). Classification scores can be linked to a utility score, with zero indicating ‘death’ and one indicating ‘full health’.9 The EQ-5D is complemented by a visual analogue scale on current health, the Valuation of Own Health, which is anchored at the lower end (zero) by ‘worst imaginable health state’ and at the upper end (100) by ‘best imaginable health state’. The SF-12 consists of 12 items in the physical and mental domain. Based on these item scores, summary measures for the physical and mental component (PCS-12 and MCS-12) are constructed,10 using norm-based methods, with a mean of 50 and a standard deviation (SD) of 10 in the general US population. A one-point difference can be interpreted as one-tenth of an SD.11
Generic anxiety was assessed through the STAI-6, a validated short version of the State Trait Anxiety Inventory, containing six items on, for example, feeling at ease or upset. Higher scores (20–80) indicate higher levels of generic anxiety.12,13 A STAI-6 score of over 44 defines an individual as highly anxious.14
To measure the psychological impact of the screening procedure we included the Psychological Consequences Questionnaire (PCQ) in all three questionnaires. The PCQ was originally developed to assess the consequences of breast screening on the dimensions of emotional, physical, and social functioning. The corresponding subscales contain five, four, and three items, respectively.15 Ratings for symptoms within each dimension vary from 0 (not at all) to 3 (quite a lot of the time). The added ratings indicate the level of dysfunction, with higher scores indicating more dysfunction. As the three subscales are highly correlated and measure the same concept of adverse psychological consequences,16 we also report an overall PCQ score (with a score range of 0–36). We used the Dutch version as adapted by Rijnsburger et al.17 These questions were also meaningful to the reference group, who, although they were not invited for screening in the study period, belonged to the targeted age range of the screening programme.
In addition to the aforementioned measures, questionnaires 2 and 3 included items on the burden of the screening procedure. In questionnaire 2, how satisfied women were with aspects of the procedure was assessed through five items on, for example, the way they were received by the general practitioner (GP), the explanation of the screening procedure, and the way privacy was dealt with. Response options were ‘satisfied’, ‘neither satisfied nor dissatisfied’ and ‘dissatisfied’. Four items addressed feelings of shame, pain, inconvenience, or nervousness that may have been experienced during the smear-taking procedure. The four response items ranged from ‘not at all’ to ‘very much so’. The incidence of symptoms that may occur after a smear test were assessed through seven items on lower abdominal pain, vaginal bleeding, discharge, urinary problems, feeling dizzy or sick, and pain during sexual activity. The four response options were ‘no’, ‘yes, 1 day’, ‘yes, 2–3 days’, and ‘yes, 4–7 days’. Whether these symptoms were considered painful or annoying was assessed in two additional items. One item, included in questionnaire 3, assessed how stressful the entire screening procedure had been to the participants. Additionally, questionnaire 3 contained three items on women’s satisfaction with the way the Pap smear result was communicated to them, and the response options were ‘satisfied’, ‘neither satisfied nor dissatisfied’, and ‘dissatisfied’.
In accordance with guidelines,18 missing items in the STAI-6 and the PCQ were imputed by the respondents’ own average score if they had completed at least 50% of the items.
Differences between the respondent groups in demographic variables were assessed using Student’s t-tests for continuous variables and chi-square tests for categorical variables. HRQoL scores tend not to have a normal distribution, with ceiling effects. Therefore we used a non-parametric test, that is the Mann–Whitney U-test, to assess differences in generic HRQoL, generic anxiety, and screen-specific anxiety, both between screening and reference groups, and between younger and older screening participants. For the same reason we used a non-parametric test to assess changes in HRQoL scores across multiple measurements, that is the Friedman test, which is only based on data from the respondents who completed all assessments. For each measurement scale we report how many women completed it at all three time points, and we report on the quality of life and anxiety scores of just those women (Table 3). Statistical analyses were performed using spss for windows, versions 15 and 17. A P value of <0.05 (referring to two-sided statistical tests) was considered to be significant. The minimal important difference (MID), indicating clinical relevance, and defined as the smallest change in a patient-reported outcome that is perceived by patients as beneficial, or that would result in a change in treatment, was operationalised as a difference of at least half an SD.19
Characteristics of respondents
Of all the women addressed, 1551 had a Pap smear taken (67% participation). Of this group, 60% (924/1551) had completed the initial questionnaire that was attached to the screening invitation (Figure 1). And 803 of these 924 women allowed us to access their smear results. Nine out of the 803 women were found to have borderline or mild dyskaryotic (BMD) smear results, three women were referred to a gynaecologist because of more severe abnormalities, and two women were advised to have a repeat test because of unclear results. Of all women who participated in the screening, completed the initial questionnaire, and allowed us access to their smear results, 789 of 803 women (98%) had a normal test result (Pap 1). These women were included in the current analyses. The second questionnaires were completed by 89% (706/789) of those women. The average time between having the screen test and the test being available to the screening organisation was 25 days (SD 6.5 days, range 13–55 days). In almost all cases the test results, to which questionnaire 3 was attached, were forwarded 1 day later by the screening organisation to the screen participants, directly by mail to their home address. These questionnaires were completed by 731 of 789 women (93%). Overall, 683 women completed all three questionnaires.
The average age of the respondents was 45.3 years, compared with an average age of 44.9 years in screen participants who did not complete the questionnaire (P = 0.41).
In respondents as well as in the reference group, the majority had a paid job, were married or cohabiting, and had children (Table 1).
Table 1. Demographic variables of screen participants with normal Pap smear results and the reference group, with scores given as numbers and percentages, unless otherwise indicated
Screen participants n = 789
Reference group n = 563
Job status (%)
Marital status (%)
Living without partner
Average no. (SD)
Generic HRQoL and anxiety compared with the reference group
The screen participants and the reference group reported similar generic HRQoL and generic anxiety scores. Screen-specific anxiety scores were statistically (P < 0.001) and clinically significantly better in screen participants than in the reference group (Table 2).
Table 2. Average generic quality-of-life scale scores (SDs) of screen participants with normal Pap smear results versus a reference group. The p-values indicate the significance level of differences in observed scores
Screen participants n = 789
Reference group n = 563
Meaning of HRQoL scores for screen participants compared with reference group
*EuroQol and SF-12: lower scores indicate worse functioning.
**The size of the difference is too small to be reflected in the number of currently used digits.
***STAI-6 and PCQ: lower scores indicate better functioning.
****Differences exceed the minimal important difference (MID), which indicates clinical significance.
Generic health-related quality of life
Better generic HRQoL
Rating of own health (0–100)
Similar self-rated health
Sum score physical (PCS-12)
Better physical health
Sum score mental (MCS-12)
Worse mental health
STAI-6 (State–Trait Anxiety Inventory, 20–80)***
Similar generic anxiety
PCQ (Psychological Consequences Questionnaire)***
Emotional scale (0–15)
Lower levels of screen-specific anxiety
Physical scale (0–12)
Social scale (0–9)
Total score (0–36)
Generic HRQoL and anxiety before and after screening
Over time, the EQ-5D ‘rating of own health’ increased among screen participants, indicating a small but statistically significant improvement (P = 0.02). Also, the mental health score (MSC-12) increased, indicating better mental health (P < 0.001), and the generic anxiety score (STAI-6) decreased, indicating less anxiety (P < 0.001). In summary, these changes all showed improvements over time. However, these changes were small and not clinically relevant.
The PCQ scores, however, increased initially (indicating a small increase of screen-specific anxiety), followed by a decrease, indicating a decrease of screen-specific anxiety after the receipt of the Pap smear results. Only changes in the emotional subscale were statistically significant (P = 0.03), and none of the changes over time in the PCQ-scale scores were clinically relevant (Table 3).
Table 3. Average HRQoL scores and anxiety scores (SDs) of screen participants with normal Pap smear results. P scores relate to differences between the three assessments
Before Pap smear n = 789
After Pap smear n = 706
After receipt of favourable Pap smear result n = 731
No. of women that completed the scale at all three assessments
Meaning of change over time
*Statistical significance of differences was calculated using Friedman tests, including only respondents that completed all three assessments. HRQoL and anxiety scale scores are reported of those who completed that specific scale at each assessment.
**EuroQol and SF-12: higher scores indicate better functioning.
***STAI-6 and PCQ: higher scores indicate worse functioning.
Generic health related quality of life
Stable generic HRQoL
Rating of own health (0–100)
Better self-rated health
SF-12 (Short Form 12, 0–100)**
Sum score physical (PCS-12)
Stable physical health
Sum score mental (MCS-12)
Better mental health
STAI-6 (State Trait Anxiety Inventory, 20–80)***
Less generic anxiety
PCQ (Psychological Consequences Questionnaire)***
Emotional Scale (0–15)
Overall a slight initial increase of screen specific anxiety, then a decrease to pre-screen levels
Physical Scale (0–12)
Social Scale (0–9)
Total Score (0–36)
Younger versus older women
Comparing HRQoL and anxiety scores between younger (30–35 years) and older (≥36 years) women, we found that most scores at most time points were similar. We did find a number of differences, indicating better physical health in the younger group at all time points (PCS-12, P ≤ 0.005). Also, younger women reported worse mental health at baseline (MCS-12, 50 versus 52, P = 0.012), more anxiety in questionnaire 2, after the Pap smear was taken (STAI, 34 versus 32, P = 0.028), and better generic HRQoL after having received the test result (EQ-5D, 0.93 versus 0.90, P = 0.001). None of the changes were clinically relevant.
Burden of the screening procedure
Before they received their smear results, 80% of women were satisfied with all aspects of the procedure. After the receipt of their smear results a majority (66%) remained satisfied about all aspects of the smear-taking procedure: the way they were received by the general practitioner; the explanation of the screening procedure; the way privacy was dealt with; and the way the Pap smear result was communicated to them. A total of 122 screen participants (19%) reported to have been quite or very much bothered by feelings of shame, pain, inconvenience and/or nervousness during the smear-taking procedure (Table 4). And 177 screen participants (27%) reported experiencing at least one of the following symptoms during at least 1 day after the smear had been taken: lower abdominal pain; vaginal bleeding; discharge; urinary problems; or feeling sick. For another 84 women (13%) these symptoms lasted at least 2–3 days. The symptoms were considered to be (very) painful by 79 women (12%) and (very) annoying by 81 women (12%). After receiving the smear result, 219 participants (32%) reported considering the entire screening procedure to be fairly stressful, and seven women (1%) considered it to be very stressful (Table 4).
Table 4. Satisfaction with aspects of the screening procedure and symptoms after smear taking among participants with a normal Pap smear result
After Pap smear procedure n = 706
After receipt of the favourable test result n = 731
*Assessed aspects of the screening procedure: way women were received by the general practioner; the explanation of the procedure in general both before and after the Pap smear procedure; the explanation of Pap smear taking itself; and explanation of the meaning of the Pap smear result, in particular, the time needed to have a Pap smear taken; the way privacy was dealt with; and the way the Pap smear result was communicated.
Satisfaction with the screening procedure
Satisfied with all aspects*
Neutral to at least one of the aspects
Dissatisfied with at least one of the aspects
Burden of the smear taking itself
The extent to which women were bothered by feelings of shame, pain, inconvenience, and/or nervousness
A little/not at all
Symptoms during the days after the smear taking
Women who reported lower abdominal pain, vaginal bleeding, discharge, urinary bother, feeling sick, feeling dizzy, and/or painful sexual activity
For at least 1 day
For at least 2–3 days
For 4–7 days
Women who reported no symptoms at all
Painfulness of these symptoms
Very painful/fairly painful
Not painful/no symptoms
Discomfort because of these symptoms
Very annoying/fairly annoying
Not annoying/no symptoms
Burden of the entire procedure
Not at all stressful
Between being invited for cervical cancer screening and receiving the test results, no clinically relevant changes in HRQoL or anxiety scores were reported by our cohort of screen participants with a normal screening result. Compared with a reference group of age-matched women, screen participants reported similar generic HRQoL and anxiety, and, unexpectedly, less screen-related anxiety. Possibly cervical screening offered reassurance to those who participated or participants represent a selection of women who are more relaxed about getting screened and the risk of being diagnosed. As ours is the first study into the HRQoL effects of cervical cancer screening, this finding needs to be explored in further (qualitative) research.
A considerable number of screen participants were bothered by feelings of shame, pain, inconvenience, or nervousness during the screening procedure, or experienced lower abdominal pain or vaginal bleeding in the days following the smear test, for example. The extent to which having a Pap smear test is burdensome could possibly be reduced by paying more attention to the women’s physical position during smear taking,20 and through the provision of information and by adopting respectful and engaged behaviour during gynaecological examinations.21
Of the women who completed the questionnaires and had a Pap smear, 98% had a normal test result (Pap 1). This corresponds with the national rate at the time this study was conducted.8 Lately this percentage has been decreasing, to about 96% (see http://www.rivm.nl/bevolkingsonderzoeknaarkanker/images/2slagfolder_final.pdf). In many other countries the percentage of women with Pap 1 smear results is also below 98%.7 We do not expect this change in rates of normal test results to have consequences for the findings in our study, only on the population of women that they apply to.
Comparing HRQoL and anxiety scores of younger (30–35 years) versus older (≥36 years) women, we found mostly similar scores and some differences. Younger women had better physical health, which seems logical, and some indications were found of an increased impact of the screening procedure on HRQoL. This increased anxiety and decreased mental health may not to be related to age per se, but alternatively may be related to the number of times a woman has participated in the screening programme. We note that none of the differences were clinically relevant.
Although the impact of cervical screening has not been reported upon before, the findings in a review on the psychological impact of mammographic screening, one of the few other well-established screening programmes, were similar to our results. Mammographic screening does not appear to have adverse psychological impact on women who were given a clear result after a mammogram, and are subsequently placed on routine recall.22
New ways to reduce cervical cancer-specific morbidity and mortality are currently being developed and introduced, for example the application of HPV testing within cervical screening programmes and vaccines against human papillomavirus (HPV). In deciding whether or not to implement these new techniques, data are needed about the HRQoL effects of screening, diagnosis and treatment processes related to cervical cancer and its pre-cancerous stages. In previous papers we reported on the long-term HRQoL in cervical cancer survivors,23 and on the impact of a borderline or mild dyskaryotic (BMD) Pap smear.4 Data from these papers and the current one are useful for national and international analyses, which increasingly incorporate not only life years (LYs) gained but also quality-adjusted life years (QALYs) as outcome measures. The results of the current study show that although women with normal test results reported unpleasant effects of smear taking, there were no adverse HRQoL consequences of the screening procedure. Our study also shows that in women who participate in the screening programme, screening experiences will probably not play a big role in women’s decisions about their participation in the subsequent screening round.
Considering the balance between beneficial and adverse effects, there is no need to de-intensify the current screening programme. We want to stress, however, that if changes to the programme are considered (e.g. to replace cytology by HPV testing as the primary test, as has been recommended by the Dutch Health Council),24 a re-assessment of the balance between beneficial and adverse effects will be necessary. The HRQoL data as reported in the current screening programme can serve as benchmark data.
The lower levels of screen-specific anxiety in screen participants, which may indicate a reassurance effect or a selection effect, need to be addressed in more depth.
Our study has its limitations. The response rate of 60% among screen participants was reasonable, but may nonetheless have resulted in a non-response bias. Although screen participants who did not respond to the questionnaire were of similar age as those who did respond, other characteristics may not have been similar. We do not know if non-responding women experienced adverse HRQoL effects.
Our cohort had a mean age of 45 years, most of them were educated, had paid jobs, were married or cohabiting, and had children. This reflects characteristics of the Dutch screening programme, which starts at 30 years of age, has a national uptake of 65%,8 and has lower uptake percentages among the youngest age groups. The average age of screen participants that completed our questionnaire (45.3 years) did not significantly differ from that of screen participants who also received a questionnaire but did not complete it (44.9 years, P = 0.41), indicating that younger screen participants were not under-represented in the questionnaire study. The response rate in the reference group of 46% was low, which we acknowledge is a limitation. Possibly only women who were worried about screening bothered to complete the questionnaire.
The strengths of our study are: the setting in a ‘real life’ well established, routinely accepted screening programme, instead of an experimental setting; the large number of respondents from the target group of the screening programme; the low numbers of dropout; the use of validated measures to assess HRQoL and screen-specific anxiety; and the availability of information about the actual screen uptake and smear test result, which enabled us to select women who participated and had a normal screen result.
We found lower levels of screen-specific anxiety in screen participants, which may indicate a reassurance effect or a selection effect, which needs to be addressed in more depth. Nineteen percent of women were bothered by feelings of shame, pain, inconvenience, and/or nervousness during smear taking, and 13% of women experienced lower abdominal pain, vaginal bleeding, discharge, and/or urinary problems for at least 2 days following the smear. We conclude that although considerable numbers of women reported unpleasant effects of smear taking, there were no adverse HRQoL consequences of the screening procedure.
Disclosure of interests
Until 2008, Dr van Ballegooijen was principal investigator for a project on the cost-effectiveness of human papilloma virus vaccination, financed by GSK (a pharmaceutical company that produces human papilloma virus vaccines against cervical cancer). The other authors have no conflicts of interest to declare.
Contribution to authorship
MvB, DH, and MLE-B conceived the idea for the study; MvB and MLE-B designed the protocol; MLE-B supervised the execution of the study; IK, MvB, and MLE-B designed the questionnaire; IK and HH organized the data collection through questionnaires; IK was responsible for the database design and data entry; BW collected the clinical data; IK, MvB, CL, and MLE-B made the statistical design; IK performed the analyses; IK drafted the report; and all of the collaborators listed above except HH contributed to or edited the article. IK is the guarantor of the study. All authors have full access to all of the data in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis. HH is now retired, and has stopped contributing to this study. We thank him for his valuable input.
Details of ethic approval
The ethics review committee of the Erasmus MC, University Medical Center Rotterdam, approved the research protocol (Medisch Ethische Toetsings Commissie Erasmus MC, 22 June 2004, MEC-2004–099). Participating women gave written informed consent.
We gratefully acknowledge the Dutch Cancer Society for funding our study (EMCR 2003–2775). The authors were completely independent from funders in conducting and writing this study.
We are grateful to the participants in the study. We thank the regional screening organisation in Limburg (BOZ) for their friendly cooperation. We thank H. Huveneers for his valuable input.