Impact of alternative management policies on health-related quality of life in women with low-grade abnormal cervical cytology referred for colposcopy

https://doi.org/10.1016/j.ejogrb.2013.02.006Get rights and content

Abstract

Objective

To assess the health-related quality of life (HRQoL) over 30 months among women who received colposcopy following low-grade abnormal cervical cytology.

Design

751 women with low-grade abnormal cytology were seen for colposcopy. Of these, 350 additionally underwent either immediate treatment by large loop excision of the transformation zone (LLETZ) or investigation by punch biopsy followed by treatment of high-grade cervical intraepithelial neoplasia if present. The HRQoL of the women was assessed on seven separate occasions over 30 months by means of the EQ-5D instrument. Outcomes for women receiving colposcopy only, colposcopy with biopsy and colposcopy with LLETZ were compared.

Results

Women experienced modest short-term increases in HRQoL, statistically significant for the colposcopy only and biopsy groups but not for the LLETZ group. HRQoL in all three groups thereafter fell until 12 months following recruitment, significantly so for the LLETZ group. Changes in EQ-5D index score arose primarily from changes in severities in the “pain and discomfort” and “anxiety and depression” domains. Changes in the visual analogue scale (VAS) representations of quality of life corresponded closely to those of the index scores and most of the VAS scores themselves did not differ from VAS population norms. All inter-assessment changes in index scores were smaller than the minimum important difference for the instrument. Beyond 18 months from recruitment, HRQoL stabilised in all three groups.

Conclusion

Women referred to colposcopy following a low-grade abnormal smear test result experienced a short-term improvement in their health-related quality of life, but the long-term effect was insubstantial. HRQoL over the post-recruitment follow-up period did not vary by intervention.

Introduction

The optimal management for women with low-grade abnormal cervical cytology remains contentious [1]. The UK's Trial of Management of Borderline and Other Low-grade Abnormal smears (TOMBOLA) was the first randomised controlled trial to evaluate two of the alternatives, namely, colposcopic investigation followed, if required, by (i) immediate treatment by large loop excision of the transformation zone (LLETZ), or (ii) targeted punch biopsies with selected recall for treatment of high-grade cervical intraepithelial neoplasia (CIN) only. Women recruited into TOMBOLA had recently recorded either borderline nuclear changes or mild dyskaryosis at screening. Full details of the trial, including the management protocol, subject characteristics and main findings, have been presented previously [2], [3]. TOMBOLA found that the rates of detected CIN2 or more severe disease over three years did not differ significantly between the two alternatives, although the rates of over-treatment and adverse side-effects were higher in the LLETZ arm [4]. When analysed in terms of intention to treat, no significant differences in the prevalence of anxiety and depression between the two trial arms were detected [5].

Although TOMBOLA measured psychosocial outcomes using specific instruments, the impact of colposcopy and interventions on health-related quality of life (HRQoL) was assessed using a generic instrument. Generic measurement of change is important for several reasons [6], [7], [8]. First, a specific measure typically focuses narrowly on changes in a single health characteristic and therefore risks missing concomitant changes in others. By contrast, generic measures are inclusive of characteristics and a holistic perception of HRQoL corresponds more closely to that of the patient than does a disaggregated, characteristic-by-characteristic description. Second, measuring outcomes generically facilitates the comparison of effectiveness between interventions of different types, whose outcomes would remain incomparable if assessed only specifically. Third, generic instruments entailing valuations over states of health can measure the utilities associated with the interventions, and utilities can be employed in cost effectiveness calculations to inform resource allocation decisions. Finally, specific instruments may, in effect, be over-sensitive. Considered in isolation, a large change in a single health characteristic might seem a matter of importance, but this importance will recede if a generic measure reveals the characteristic to be an insubstantial component of HRQoL overall. Subjects might consider a change in outcome too trivial to merit a change in valuation.

In this paper, we report the impact of colposcopy and subsequent management (biopsy or LLETZ) on health-related quality of life for a sample of women in TOMBOLA. Impact was assessed using the EQ-5D generic instrument. The EQ-5D is one of the most widely used methods for measuring health state utilities [9], and the UK's National Institute for Health and Clinical Excellence has adopted it as the “preferred measure of HRQoL in adults” [10, p. 38]. We calculated and compared average utility scores over time according to the intervention which women actually received.

Section snippets

Methods

The sample for this study comprised 751 women with borderline or mild screening results who had been randomised to colposcopy accompanied by either immediate LLETZ or 2–4 targeted punch biopsies with recall for LLETZ if these biopsies indicated CIN2/3. Irrespective of their randomisation, many women were found on colposcopy to have normal transformation zones. They were therefore discharged without further treatment at that time, and were followed up by annual cytology tests in primary care.

Results

The socio-demographic characteristics of the sample are presented in Table 1. As would be expected, those receiving a LLETZ were more likely to have recorded mild dyskaryosis rather than borderline nuclear changes on their smear test at trial entry. Women in the LLETZ group were significantly younger on average than those in either the colposcopy only (analysis of variance, p < 0.01) or the biopsy (p = 0.03) groups. The mean ages of the latter two groups did not differ significantly (p = 0.10). The

Discussion

Observed changes in the generic measure were consistent with results from specific measurement. Subjects’ distress was measured by the Hospital Anxiety and Depression Scale [19] at the same seven assessment points [5]. The distress profile mirrored that of HRQoL, in that the lowest prevalences of anxiety and depression were recorded at assessment A3. This particular analysis, however, failed to identify a significant difference in the time paths between the LLETZ and biopsy subjects. This was

References (26)

  • J. Dowie

    Decision validity should determine whether a generic or condition-specific HRQoL measure is used in health care decisions

    Health Economics

    (2002)
  • J. Brazier

    Valuing health states for use in cost-effectiveness analysis

    Pharmacoeconomics

    (2008)
  • National Institute for Health and Clinical Excellence

    Guide to the methods of technology appraisal

    (2008)
  • 1

    On behalf of the TOMBOLA Group.

    View full text