Improving women's experience during speculum examinations at routine gynaecological visits: randomised clinical trial

Introduction

Compliance rates with cervical screening vary, and embarrassment and fear of pain during examination have been reported as potential barriers to screening.^1–3 Reducing these barriers may improve screening rates.3 Whereas in countries such as the United Kingdom, Australia, and New Zealand speculum examinations are done without stirrups with the woman in the dorsal or lateral positions, in the United States medical providers are trained to do gynaecological examinations with women in the dorsal lithotomy position in stirrups. We examined the use of an alternative positioning to improve women's experience with speculum examinations at routine gynaecological visits.

Methods

Assignment, masking, and flow—Women aged 3 18 years presenting for a routine gynaecological examination in a family medicine clinic were eligible for inclusion. We excluded women who were pregnant, had vaginal complaints, or were unable to consent for themselves. Participants were enrolled from January to August 2005. Women were randomised into “stirrup” and “no stirrup” groups by assigning study numbers to one of the two groups using an online random number generator and enclosing the assignments in envelopes. The envelopes were opened, revealing the method to be used, after the woman agreed to participate (fig 1). This study was powered to have an 80% chance of detecting a 5 mm difference in sense of vulnerability or discomfort, with an SD of 22 mm (determined by pilot study), on a 100 mm visual analogue scale. We originally intended to include 300 women but we stopped after a planned interim analysis of 200 patients.

Fig 1

Patient flow through the study

• Download figure
• Open in new tab
• Download powerpoint

Protocol—Patients in the control group were examined with their heels placed in uncovered metal stirrups at 30-45° angles to the examination table. The intervention group were examined with their feet placed on the corners of a fully deployed table extension. All women were draped with a full sized sheet in a standardised manner that maximised coverage of the woman and allowed visualisation of the perineum (fig 2). Medium Graves' plastic speculums and an external light source were used for all examinations.

Fig 2

Positioning of women with draping for examination without stirrups (drawn by Jordan Mastrodonato)

• Download figure
• Open in new tab
• Download powerpoint

Provider training—We recruited eight providers from within the clinic: five staff physicians, two resident physicians, and one family nurse practitioner (five men and three women). All providers were taught how to position and cover the women with a combination of individual teaching, still photographs, and a short video.

Outcomes measured—We collected demographic data for each participant. Race, as identified by the women from a list provided, was recorded because it has often been looked at in relation to satisfaction or compliance with cervical smears. Afterwards, women completed a questionnaire regarding the examination. We used 100 mm visual analogue scales to measure physical discomfort, sense of vulnerability, and sense of loss of control. Study identification numbers were used to preserve anonymity of the respondents. We had previously piloted the questionnaire on 25 women. We assessed the quality of smears according to adequacy and the presence or absence of endocervical cells reported in the pathology reports.

Data handling—Two researchers (DAS and DRJ) entered the demographic data and results of the smear tests on a spreadsheet. An independent observer measured and entered the visual analogue scores on a separate spreadsheet in a blinded manner.

Analysis—We used SPSS version 11.0 for analysis. In the preliminary analyses we used χ2 tests for categorical data and t tests for independent samples for continuous data. We tested the hypothesis with the t test for independent samples to determine differences in outcome variables.

Results

We recruited and randomised 204 women. After randomisation, seven women were removed from the study: five did not meet enrolment criteria (three were pregnant and two were adolescents), one woman had labial adhesions that prevented insertion of the speculum, and there was one equipment failure caused by a faulty stirrup. Table 1 shows the baseline characteristics of the remaining women (table 1).

Physical discomfort and sense of vulnerability were both significantly lower in the women examined without stirrups (table 2). Mean sense of vulnerability was 44% lower and mean physical discomfort was 43% lower. Sense of loss of control was not significantly different.

There were no significant differences in the quality of smears between the two groups. Only two smears were inadequate for evaluation (both from the no stirrups group, P = 0.16). Sixteen of the smears from the women examined without stirrups and 15 from the women examined with stirrups did not contain endocervical cells (P = 0.84).

Discussion

Women undergoing gynaecological examination without stirrups report less physical discomfort and a reduced sense of vulnerability than women in stirrups.