|Cost-effectiveness of strategies to enhance mammography use
|Fishman P, Taplin S, Meyer D, Barlow W
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.
The use of alternative reminder strategies to encourage women, who failed in scheduling a mammography 2 months after an invitation sent by mail, to participate in mammographic screening. Three strategies of reminding the women were compared:
a postcard (strategy 1);
a telephone call from a trained caller (strategy 2); and
a motivational telephone call from a trained caller (strategy 3).
Strategy 1 was compared with no reminder (current practice), strategy 2 was compared with strategy 1, and strategy 3 was compared with strategy 2.
Economic study type
The study population comprised women aged 50 to 79 years who were due to receive mammography through the GHC's breast cancer screening programme.
The setting was community care. The economic study was carried out in western Washington State, USA.
Dates to which data relate
The effectiveness and resource use data were derived from a randomised trial published in 2000. The year when the study was performed was not specified. The unit costs were based on local markets and 1998 prices were used.
Source of effectiveness data
The effectiveness data were derived from a single study.
Link between effectiveness and cost data
The costing was undertaken on the same patient sample as that used in the effectiveness study. It was not stated whether the costing was carried out prospectively or retrospectively.
A total of 11,570 women aged 50 to 79 years were due for a mammogram, of which 6,147 were randomly recruited for the trial. Of this group, 703 women had already scheduled a mammography and 382 were not eligible for other reasons (no further information provided). All of the women who completed a survey (n=3,743) received an invitation by mail to schedule a mammography within a 2-month period. Only 1,965 of the 3,743 women sent an invitation scheduled a mammography within 2 months. Women who failed to respond within 2 months were selected for the trial (n=1,778). Thirteen women (0,7%) withdrew before they were randomly assigned to the alternative intervention groups in the trial. Finally, 1,765 women were included in the trial. There were 590 (33.4%) in the postcard reminder group, 585 (33.2%) in the (telephone) reminder call group and 590 (33.4%) in the motivational call group. It was not stated whether power calculations were used to determine an appropriate sample size.
The study was a randomised controlled trial that was conducted alongside an ongoing breast cancer-screening programme. Information on the setting in which the trial was carried out was not explicitly given. In addition, the method of randomisation was not reported. The full details of the study were published elsewhere (see Other Publications of Related Interest). The effectiveness of the alternative promotional strategies was determined at 12 months. The telephone callers were unable to reach 18% of the women assigned to the reminder call group and 17% of the women assigned to the motivational call group.
Analysis of effectiveness
An intention to treat analysis was carried out. The primary outcome was the proportion of women who scheduled a mammogram within 12 months of receiving a reminder. The alternative intervention groups were shown to be comparable in terms of their age, ethnicity and history (prior mammograms). All of the baseline characteristics were provided.
After the initial recommendation letter, 18% of the women had scheduled a mammography within 2 months.
Of the women who received a reminder by postcard, 36.1% had successfully scheduled a mammography at 12 months (upper bound 95% confidence interval, CI: about 40%). This compared with 52.6% of those receiving a reminder call (upper bound 95% CI: about 58%) and 50.3% of those receiving a motivational call (upper bound 95% CI: about 58%).
The difference between the two telephone groups was not statistically significant. However, the results in the telephone groups were significantly higher (no figures were reported) than the results in the postcard groups.
In the group of women who had had prior mammograms, 22.75% scheduled a mammography after receiving the initial letter. Then, among those women randomised to the interventions, 45.5% scheduled by 12 months in the postcard group versus 62.5% in the reminder call group and 60.4% in the motivational call group.
The results for the group of women with no prior mammograms were 5.65% for the initial letter, then 11.3% for the postcard group, 26.5% for the reminder call group and 24% for the motivational call group.
The effectiveness study showed that telephone reminding was a more effective means of ensuring that women scheduled a mammogram within the 12-month timeframe.
Measure of benefits used in the economic analysis
The outcome measure used was the proportion of women who scheduled a mammography within 12 months following the receipt of a reminder.
The resource quantities and the costs were not reported separately. The data were derived from information from the project team's report of resources used to deliver the intervention. The direct costs were for office space and equipment from the GHC's screening programme that was used for the trial, personnel, hardware and materials necessary to implement and conduct the promotional strategies. The costs of building and maintaining the screening programme information system were not included. The costs for the postcard group included the development of the text on the postcard, and its design, production and mailing. The costs for the telephone call groups included personnel costs of time spent on the telephone and the training of personnel conducting motivational calls.
The costs were divided into the fixed and variable costs used in each strategy allocated on a per-participant basis. The costs for personnel, office space, furnishing and equipment were estimated on the basis of local market costs. Production and mailing costs were based on bulk mailing charges and telephone costs were based on average cost per minute of contact time. Adjustments for the protocol-driven costs were not reported. The total costs for each strategy per participant, the marginal costs per participant per strategy, and the marginal cost-effectiveness estimates were presented. Discounting was not performed, which was appropriate since the follow-up was for 12 months only. The prices used were 1998 US dollars.
Statistical analysis of costs
The costs were treated as point estimates. Statistical analyses of the costs were not performed.
The indirect costs were not included.
In the base-case analysis, it was assumed that 50% of the women who scheduled mammography after receiving the reminder postcard would have done so within 10 months even without this reminder. One-way sensitivity analyses, in which this percentage was varied from 0 to 80%, were performed. The authors also looked at whether the cost-effectiveness results were sensitive to mammography history.
Estimated benefits used in the economic analysis
Marginal effectiveness figures were presented for each strategy compared to the alternative. The marginal effectiveness was 0.1805 for strategy 1 compared to current practice, 0.165 for strategy 2 compared to strategy 1, and -0.023 for strategy 3 compared to strategy 2. The corresponding figures for the group of women with prior mammography were 0.2275, 0.17 and -0.021, respectively. For the group of women without prior mammography, these figures were 0.0565, 0.152 and -0.025, respectively.
For each alternative strategy, the total costs per participant (including cost of the initial recommendation letter) were presented. The initial recommendation letter cost $2.07, strategy 1 (reminder postcard) cost $6.02, strategy 2 (reminder call) cost $21.22 and strategy 3 (motivational call) cost $25.99
The marginal costs per woman were also presented. The marginal cost was $3.95 for strategy 1 compared to no reminder, $15.20 for strategy 2 compared to strategy 1, and $4.77 for strategy 3 compared to strategy 2. The same figures were reported for the group of women with and without prior mammography.
Synthesis of costs and benefits
Incremental analyses were performed for each strategy.
The marginal cost-effectiveness for all women was $21.88 for the postcard reminder and $92.12 for the reminder call. The motivational call was dominated (more expensive and less effective than the reminder call).
For the group of women with prior mammography, the marginal cost-effectiveness was $17.36 for strategy 1 and $89.53 for strategy 2. The motivational call was dominated.
For the group of women without prior mammography, the marginal cost-effectiveness was $69.91 for strategy 1 and $99.80 for strategy 2. The motivational call was dominated.
Varying the percentage of women who were assumed to have scheduled a mammography without any prompting affected the marginal cost-effectiveness of strategy 1. If 68% or more women scheduled a mammography after the initial invitation, strategy 2 became more cost-effective than strategy 1.
The reminder postcard and the reminder telephone call are both cost-effective strategies to encourage those women who do not respond to an initial mailed recommendation to participate in breast cancer screening. The reminder postcard was more cost-effective than the reminder telephone call, as the costs were lower and the relative effectiveness was higher. However, the results for the reminder postcard were less favourable for the group of women with no history of mammography.
CRD COMMENTARY - Selection of comparators
The comparator used to examine the cost-effectiveness of the alternative promotional activities was implicitly justified, as it seems that a mailed invitation to participate in a screening programme reflects current practice in breast cancer screening. However, you should decide if this is valid in your own setting.
Validity of estimate of measure of effectiveness
The analysis used a randomised trial, which was appropriate for the study question. The method used to recruit women for the trial was not described. Women who completed a survey were included and it is likely that the population may have been biased as a result of this selection (selection bias). To fully understand the trial, the reader is referred to the original trial publication. The patient groups were shown to be comparable at analysis and the outcomes were analysed on an intention to treat basis. It is likely that the study sample is representative of the population within the Health Maintenance Organisation where the trial was conducted. In the base-case analysis, it was assumed that 50% of the women who received a postcard reminder would have responded to the initial recommendation within 10 months even without this reminder. The authors did not explicitly state that this assumption was also made for the two telephone groups. The figure was subjected to a sensitivity analysis to assess whether it was robust.
Validity of estimate of measure of benefit
The estimation of benefits was obtained directly from the effectiveness analysis. This choice was (implicitly) justified.
Validity of estimate of costs
All the categories of cost relevant to the perspective adopted were included in the analysis. The costs and the quantities were not reported separately, thus hindering the reproducibility of the results obtained. Resource use was taken from the trial. However, the unit costs were taken from the authors' setting using local market costs in an attempt to make the results more generalisable to other Health Maintenance Organisations. No statistical analysis of the quantities or costs was performed. The price year was reported, which will aid future reflation exercises. Discounting was, appropriately, not performed. The authors also acknowledged that the results were limited to the health plan that conducted the trial.
The authors did not compare their findings with those from other studies. Some sensitivity analyses were conducted. Nevertheless, the authors acknowledged that the results may not be generalisable to other settings and to other populations since the study was performed in one single health plan and may have enrolled a specific population. The authors did not present their results selectively and their conclusions are within the scope of the analysis.
Implications of the study
All the alternative strategies increased mammography use at different average and marginal costs. Choices about how to promote mammography ultimately depend on the willingness to invest in promotional activities of the health plan. Future research should evaluate the cost and effectiveness of promotional strategies over time as women gain more experience with screening mammography.
Source of funding
Supported by a grant from the National Cancer Institute.
Fishman P, Taplin S, Meyer D, Barlow W. Cost-effectiveness of strategies to enhance mammography use. Effective Clinical Practice 2000; 3(5): 213-220
Other publications of related interest
Taplin SH, Barlow WE, Ludman E, et al. Testing reminder and motivational telephone calls to increase screening mammography: a randomized study. Journal of the National Cancer Institute 2000;92:233-42.
Ludman EJ, Curry SJ, Meyer D, Taplin S. Implementation of outreach telephone counselling to promote mammography participation. Health Education and Behavior 1999;26:689-702.
Comment: Effective Clinical Practice 2000;3:250-5.
Subject indexing assigned by NLM
Aged; Breast Neoplasms /radiography; Cost-Benefit Analysis; Female; Health Maintenance Organizations; Health Promotion /organization & Health Services Research; Humans; Mammography /economics /utilization; Middle Aged; Organizational Objectives; Patient Acceptance of Health Care; Reminder Systems /economics; Washington; administration
Date bibliographic record published
Date abstract record published