|
Human papillomavirus testing for triage of women referred because of abnormal smears: a decision analysis considering outcomes and costs |
Meerding W J, van Ballegooijen M, Burger M P, Marle M E, Quint W G, Habbema J D |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Three strategies for the management of women with dyskaryosis and referred for colposcopy were considered in the study.
Strategy 1: conventional management, where treatment was based on colposcopic and histological assessment.
Strategy 2: women were triaged by high-risk human papillomavirus (HR-HPV) testing. HR-HPV-positive women were treated directly with loop excision of the transformation zone (LETZ) without prior histological assessment, while HR-HPV-negative women followed the conventional management policy.
Strategy 3: direct LETZ treatment without prior virologic or histologic assessment. The HPV test was detected using the GP5/6 general primer-mediated polymerase chain reaction (PCR).
Economic study type Cost-effectiveness analysis.
Study population The study population comprised women aged 30 to 60 years, with either persistent (at least two smears) mild or moderate dyskaryosis or a single smear reported as severe dyskaryosis. The two groups of women were analysed separately.
Setting The setting was a hospital. The economic study was carried out at the university hospital in Groningen, The Netherlands.
Dates to which data relate No dates relating to the effectiveness and resource use data were reported. The price year was 1999.
Source of effectiveness data The effectiveness evidence came from a single study and from the authors' assumptions.
Link between effectiveness and cost data The costing was performed prospectively, mainly on the same sample of patients as that used in the effectiveness study.
Study sample Power calculations to determine the sample size were not reported. A sample of 221 consecutive women was enrolled in the study. Of these, 106 women had persistent mild or moderate dyskaryosis and 115 women presented severe dyskaryosis. No control group was considered in the analysis.
Study design The objective of the study was to compare the three management strategies specified. However, the data for this analysis came from a prospective observational study in which all of the patients received both HPV testing and histologic assessment by colposcopically-directed biopsy. Positive samples were analysed using type-specific primers for low-risk HPV types 6 and 11 and HR-HPV types 16, 18, 31 and 33, separately. How the available data from the observational study were used to compare the different strategies was unclear (see 'Modelling' section). In addition, neither the length of follow-up nor loss to follow-up were reported.
Analysis of effectiveness All of the patients in the initial study sample appear to have been included in the effectiveness analysis. The HPV prevalence, specificity, and positive predictive value (PPV) of PCR in the two samples, and the proportion of women who tested positive for HR-HPV 16, 18, 31 or 33, were directly calculated from the observational study. The number of medical procedures was then modelled per strategy in terms of HPV testing, outpatient visits, colposcopy, biopsy, LETZ, all treatment and years in follow-up.
Effectiveness results Among women with persistent smears of mild to moderate dyskaryosis:
for HPV 16, 31 or 33, HPV prevalence was 59% (range: 46.7 - 71.4), the specificity was 77.8% (range: 65.6 - 89.9) and the PPV was 78.3% (range: 66.3 - 90.2); and
for HPV 18, HPV prevalence was 1.6% (range: 0 - 4.8), the specificity was 93.3% (range: 86 - 100) and the PPV was 25% (range: 0 - 67.4).
For all smears, the PPV was 57.5% (range: 48.1 - 67).
Among women with one smear of severe dyskaryosis:
for HPV 16, 31, or 33, HPV prevalence was 66.7% (range: 57.6 - 75.7), the specificity was 100% and the PPV was 100%; and
for HPV 18, HPV prevalence was 6.7% (range: 1.9 - 11.4), the specificity was 80% (range: 55.2 - 100) and the PPV was 77.8% (range: 50.6 - 100).
For all smears, the PPV was 91.3% (range: 86.2 - 96.5).
Overall, the proportion of women who tested positive for HR-HPV 16, 18, 31 or 33 was 47% for women with persistent smears of mild to moderate dyskaryosis, and 69% for women with severe dyskaryosis.
The predicted number of all treatments (LETZ, conization or hysterectomy) per woman was 0.77 with the conventional strategy, 0.82 with triage and 1.02 with treatment 1.02.
The predicted years in follow-up were 4.32 with the conventional strategy, 4.31 with triage and 4.51 with treatment.
Clinical conclusions Compared with the conventional strategy, the predicted outpatient visits were significantly reduced with direct treatment, but were barely different with HPV triage. The predicted number of colposcopies was reduced with HPV triage and was absent (i.e. not required) with direct treatment. Similarly, the predicted number of biopsies was reduced with HPV triage and was absent (i.e. not required) with direct treatment. Compared with the conventional strategy, the predicted LETZ, all treatment procedures and years in follow-up were barely different for HPV triage, but were slightly higher for direct treatment.
Modelling A modelling process was used to calculate the number of procedures (e.g. biopsy, LETZ, conization and hysterectomy) required with each approach under study. This represented the benefit measure used in the economic analysis. The authors extrapolated the data from the effectiveness study and combined them with some assumptions. Details of the modelling procedure were not reported.
Methods used to derive estimates of effectiveness The authors made some assumptions in order to extrapolate data from the single study to the three management policies considered in the analysis.
Estimates of effectiveness and key assumptions The authors made several assumptions. First, women with histologic low-grade squamous intraepithelial lesions (SIL) were, in principle, not treated. Thus, the number of treatments in these women would be similar to those observed in women with normal histology. Second, women not treated in the conventional and HPV triage policy were treated with LETZ in the direct treatment policy. Third, women with low- or high-grade SIL after treatment were followed with yearly Pap smears for over 5 years. Finally, women with normal histology or with low-grade SIL who were not treated were followed with two (6-monthly) smears and colposcopies over one year.
Measure of benefits used in the economic analysis No summary measure of benefit was used in the economic analysis. Hence, a cost-consequences analysis was conducted.
Direct costs Discounting was not performed since the costs were incurred in less than two years. The unit costs were reported separately from the quantities of resources used. The health services included in the economic analysis were the Pap smear, HPV testing, (primary and secondary) colposcopy, biopsy, LETZ, conization, hysterectomy, outpatient visit, hospital stay, and time spent for outpatient and general practitioner visits. The cost/resource boundary adopted in the study appears to have been that of society. Resource use was mainly estimated using data derived from the records of patients involved in the effectiveness study, although the authors' assumptions were also used. The unit costs were estimated from published studies and local sources. The price year was 1999.
Statistical analysis of costs The costs appear to have been treated deterministically.
Indirect Costs The time costs for women, in terms of outpatient and general practitioner visits, were included. These were derived using the average hourly labour wage for women. The price year was 1999.
Currency US dollars ($). The conversion rate from Dutch guilders (Dfl) to $ was $1 = Dfl 2.07.
Sensitivity analysis Sensitivity analyses were not conducted.
Estimated benefits used in the economic analysis See the 'Effectiveness Results' section.
Cost results In the sample of women with persistent smears of mild to moderate dyskaryosis, the estimated extra costs per women with HPV triage in comparison with the conventional policy were $134. Direct treatment led to cost-savings of $114 in comparison with the conventional policy.
In women referred with one smear of severe dyskaryosis, HPV triage led to additional costs of $133 per women in comparison with the conventional policy. Direct treatment saved $103 per woman compared with the conventional policy.
Synthesis of costs and benefits Not relevant as a cost-consequences analysis was conducted.
Authors' conclusions The choice of the most appropriate approach for the management of women with persistent smears of mild to moderate dyskaryosis or severe dyskaryosis depended on the relative burden attributed to invasive procedures (such as colposcopies and biopsies) in comparison with direct treatment (loop excision of the transformation zone, LETZ). Since pain and discomfort may be strongly associated with all of these procedures, direct treatment could represent the strategy of choice in severe dyskaryosis, as this strategy reduced the number of diagnostic procedures. A literature review to identify relevant studies evaluating the approaches examined in the present article found similar results when the populations and characteristics of polymerase chain reaction (PCR) testing were comparable.
CRD COMMENTARY - Selection of comparators The three approaches under evaluation were selected to represent feasible strategies for the management of women referred because of abnormal smears. In particular, strategy 1 represented the current management policy in The Netherlands, strategy 2 was the strategy under examination, and strategy 3 was included to take into account a more aggressive approach. You should decide whether they represent widely used approaches in your own setting.
Validity of estimate of measure of effectiveness The analysis of effectiveness largely used data derived from prospective observational studies at the authors' institution. Few details on the studies were reported. The patient demographics and clinical characteristics were not mentioned. The length of and loss to follow-up were not reported. There may well have been bias in allocating the treatment since the study management strategies were different to that in the observational study. It was unclear how the predicted procedures were derived. In addition, how the predicted outcomes were derived from the observational study was not well explained.
Validity of estimate of measure of benefit No summary benefit measure was used in the economic analysis. The analysis was therefore categorised as a cost-consequences study.
Validity of estimate of costs The authors stated that a societal perspective was adopted in the study and appropriate costs seem to have been included. The unit costs were appropriately reported separately from the quantities of resources used and the price year was mentioned. This enhances the reproducibility of the economic analysis in other settings. Statistical analyses of the costs were not reported and no sensitivity analyses were conducted. The cost estimates were obtained using data derived from national sources, and appeared to be specific to the study setting. Some assumptions were required to estimate resource use. The authors noted that the unit cost of HPV testing represented a critical variable in the economic analysis.
Other issues The authors carried out a literature review to identify published studies and compared them with their own results. The details of the literature review and the study samples involved in each study were reported. However, the authors did not address the issue of the generalisability of the study results to other settings and did not perform sensitivity analyses. Thus, the external validity of the analysis was low. The study referred to women with either persistent smears of mild to moderate dyskaryosis or severe dyskaryosis and this was reflected in the conclusions of the analysis. The authors appear to have presented their results selectively.
Implications of the study The clinical outcomes of the study suggest that quality of life issues should form the basis for making decisions about the best approach for the management of women with either persistent smears of mild to moderate dyskaryosis or severe dyskaryosis. The authors stressed that future studies should be aimed at the identification of those HR-HPV types that are predictive for high-grade SIL. Further, the role of age as a possible additional triage criterion should be explored in populations with mild cytology.
Source of funding Funded by grant OG91-053 from the Fund for Investigative Medicine of the National Health Insurance Council.
Bibliographic details Meerding W J, van Ballegooijen M, Burger M P, Marle M E, Quint W G, Habbema J D. Human papillomavirus testing for triage of women referred because of abnormal smears: a decision analysis considering outcomes and costs. Journal of Clinical Epidemiology 2002; 55(10): 1025-1032 Indexing Status Subject indexing assigned by NLM MeSH Adult; Cervical Intraepithelial Neoplasia /prevention & Decision Support Techniques; Female; Humans; Male; Middle Aged; Papillomaviridae /isolation & Papillomavirus Infections /diagnosis; Prospective Studies; Triage; Tumor Virus Infections /diagnosis; Uterine Cervical Dysplasia /pathology; Uterine Neoplasms /prevention & Vaginal Smears; control; control; purification AccessionNumber 22002001983 Date bibliographic record published 30/11/2003 Date abstract record published 30/11/2003 |
|
|
|