|
Cost and cost effectiveness of venous and pressure ulcer protocols of care |
Kerstein M D, Gemmen E, van Rijswijk L, Lyder C H, Phillips T, Xakellis G, Golden K, Harrington C |
|
|
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 The health technologies in the study were represented by protocols of wound care for pressure ulcer and venous ulcer. The three care protocols for pressure ulcer were saline moistened gauze, hydrocolloid Comfeel dressing, and hydrocolloid DuoDERM dressing. The three care protocols for venous ulcer were paraffin-, saline-, or zinc oxide-impregnated gauze, hydrocolloid DuoDERM dressing, and human skin construct.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised patients suffering from venous and pressure ulcers. Patients with ulcers secondary to diabetes mellitus were excluded from the study.
Setting The setting was not unclear, but was consistent with that of the community or a hospital. The economic study was carried out in the USA.
Dates to which data relate The effectiveness and resource use data were derived from studies published between 1984 and 1999. The price year was 2000.
Source of effectiveness data The effectiveness evidence was derived from a review of published studies.
Modelling A decision model was set up to assess the costs and the outcomes of the three care protocols for both pressure and venous ulcers. A hypothetical managed-care plan with a population of 100,000 covered lives was considered. The data used to populate the model were derived from a review of the literature and were validated by an expert panel. The time horizon of the model was 12 weeks.
Outcomes assessed in the review The outcomes assessed in the review were the average (weighted) proportions of pressure and venous ulcers completely healed at different timeframes and with the different care protocols. Data on the prevalence of venous and pressure ulcers were also reported.
Study designs and other criteria for inclusion in the review The bulk of the study designs were controlled clinical trials (24 out 26 studies). Several inclusion and exclusion criteria were reported. Studies that did not include information on patient demographics by treatment modality, wound healing assessments, or methods of care (type of dressing used) were excluded. Studies of Stage I pressure ulcers only (intact epidermis) were also excluded. Studies that did not report the proportion of ulcers healed in at least one of a number of specific timeframes (4, 6, 8 and 12 weeks for pressure ulcer, and 6 and 12 weeks for venous ulcer) were omitted from further analysis. Finally, only treatment modalities that reported the above-mentioned criteria on a pooled total of at least 100 ulcers were maintained for further analysis to ensure an adequate power for the study. Details on the design of primary studies providing data on the prevalence of disease were not reported.
Sources searched to identify primary studies MEDLINE and CINAHL were searched for publications in the English language. The search used the subject terms "clinical study" or "randomised controlled clinical study", "leg ulcer" (or "venous ulcer") or "pressure ulcer" (or "decubitus ulcer"). All searches were conducted with and without the subject "dressing". The references in the retrieved publications were also examined, and those that might meet the inclusion criteria were obtained.
Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included The effectiveness evidence relating to wound care modalities were obtained from 26 primary studies. The prevalence data were derived from 7 published studies.
Methods of combining primary studies Data for each ulcer model were first pooled and then weighted averages were obtained. Where data were unavailable for a given timeframe, it were obtained by linear growth interpolation or extrapolation. A meta-analysis was also conducted on the proportion of ulcers healed for each topical-care group to combine the estimations derived from primary studies. However, meta-analytic averages and 95% confidence intervals were not used in the decision model, because almost half of the studies were reported to have failed to meet the test of homogeneity in the later timeframes (8, 10 and 12 weeks). The method used to combine the data on prevalence was not reported.
Investigation of differences between primary studies A homogeneity test was conducted to take into account differences among the primary studies. The homogeneity was not met by almost half of the primary studies included in the meta-analysis.
Results of the review In the case of pressure ulcers, the average proportion of ulcers completely healed with saline gauze was 24% (range: 0 - 75) after 4 weeks, 28% (range: 0 - 83) after 6 weeks, 36% (range: 0 - 94) after 8 weeks, 40% (range: 0 - 94) after 10 weeks, and 51% (range: 0 - 100) after 12 weeks.
The average proportion of ulcers completely healed with hydrocolloid C was 11% (range: 8 - 20) after 4 weeks, 28% (range: 15 - 56) after 6 weeks, 37% (range: 20 - 65) after 8 weeks, 42% (range: 25 - 75) after 10 weeks, and 48% (range: 29 - 80) after 12 weeks.
The average proportion of ulcers completely healed with hydrocolloid D was 24% (range: 11 - 89) after 4 weeks, 32% (range: 17 - 89) after 6 weeks, 41% (range: 22 - 100) after 8 weeks, 51% (range: 28 - 100) after 10 weeks, and 61% (range: 33 - 100) after 12 weeks.
In the case of venous ulcers, the average proportion of ulcers completely healed with impregnated gauze was 11% (range: 6 - 20) after 4 weeks, 19% (range: 9 - 30) after 6 weeks, 27% (range: 12 - 40) after 8 weeks, 32% (range: 15 - 50) after 10 weeks, and 39% (range: 18 - 60) after 12 weeks.
The average proportion of ulcers completely healed with hydrocolloid D was 17% (range: 4 - 51) after 4 weeks, 27% (range: 5 - 76) after 6 weeks, 36% (range: 7 - 100) after 8 weeks, 44% (range: 9 - 100) after 10 weeks, and 51% (range: 11 - 100) after 12 weeks.
The average proportion of ulcers completely healed with human skin construct was 12% after 4 weeks, 22% after 6 weeks, 35% after 8 weeks, 39% after 10 weeks, and 45% after 12 weeks.
The prevalence rate was 3.5% (range: 3.5 - 29) for pressure ulcers and 1% (range: 0.2 - 1.3) for venous ulcers.
Measure of benefits used in the economic analysis The benefit measure used in the economic analysis was the number of patients healed or not healed in an hypothetical managed-care plan with 100,0000 covered lives after 12-week treatments. The benefit measure was derived from the effectiveness analysis by multiplying the proportion healed by the prevalence.
Direct costs Discounting was irrelevant because the costs were incurred over a time period of 12 weeks. The resource quantities and the unit costs were only reported separately for antibiotic use. The physician and nursing times were given. The cost/resource boundary adopted was not reported. The analysis of the costs included dressing materials, nursing, and physician debridement. The costs were estimated from actual data. In particular, the dressing costs were obtained from the 2000 Drug Topics Red Book, the physician costs (including debridement costs) were derived from the 2000 Medicare Physician Fee schedule, and the nursing costs were obtained from the March 1996 National Sample Survey of Registered Nurses. The costs relating to hospitalisation or nursing home stay were not included in the analysis because ulcers were seldom a primary cause of admission. The costs relating to supportive care and vascular assessment were also excluded from the study. The resources were estimated mainly from the opinions of experts, which were used to confirm the outcomes values obtained from the literature.
A group of four experts was asked to complete a questionnaire and to comment upon the results of a preliminary review of the literature, in order to provide some information relating to the resources used and the treatment protocols. The four experts comprised a dermatologist, a vascular surgeon, a family practitioner, and a registered nurse with a doctorate in geriatrics. Some of the estimates derived from the experts' panel were reported: the frequency of the physician visits was once every 4 weeks for pressure ulcers and once a week for venous ulcers; the nursing time to change dressing and assess wounds was 15 minutes for pressure ulcers and 30 minutes for venous ulcers; approximately 50% of the ulcers required, on average, 12 weeks for healing (length of care). The experts also confirmed that pressure ulcer care takes place most often in inpatient care settings, whereas venous ulcers are generally treated on an outpatient basis. A further assumption was that, regardless of the type of dressing, all patients with pressure ulcers required supportive care (pressure relief, nutritional support, and incontinence management) and venous ulcer patients required vascular assessment.
All costs were inflated to the year 2000 using the Medical Component of the Consumer Price Index.
Statistical analysis of costs No statistical analysis was reported, the costs being treated deterministically.
Indirect Costs The indirect costs were not included.
Sensitivity analysis No sensitivity analysis was carried out.
Estimated benefits used in the economic analysis After 12 weeks, in the case of pressure ulcers, 511 patients were healed with saline gauze, 358 with hydrocolloid C, and 696 with hydrocolloid D.
In the case of venous ulcers, 200 patients were healed with impregnated gauze, 241 with hydrocolloid D, and 232 with human skin construct.
Cost results In the pressure ulcer group, the average costs for 12 weeks were:
for saline gauze, $92.43 for dressing materials, $996.05 for nursing, and $338.87 for physician debridement;
for hydrocolloid C, $270.05 for dressing materials, $170.37 for nursing, and $338.87 for physician debridement; and
for hydrocolloid D, $260.06 for dressing materials, $151.63 for nursing, and $338.87 for physician debridement.
The total costs for 12 weeks amounted to $3,868,664 for saline gauze, $2,113,305 for hydrocolloid C, and $1,944,646 for hydrocolloid D.
In the venous ulcer group, the average costs for 12 weeks were:
for impregnated gauze, $11.97 for dressing materials, $558.73 for nursing, and $710.34 for physician debridement;
for hydrocolloid D, $223.23 for dressing materials, $227.37 for nursing, and $710.34 for physician debridement; and
for human skin construct, $6,130.02 for dressing materials, $138.23 for nursing, and $710.34 for physician debridement.
The total costs for 12 weeks amounted to $1,132,334 for impregnated gauze, $953,972 for hydrocolloid D, and $6,834,270 for human skin construct.
Synthesis of costs and benefits The costs and the benefits were combined in a calculation of the cost per healed patient after 12 weeks of treatment. In the case of pressure ulcers, the average cost per patient healed was $2,179 for saline gauze, $1,267 for hydrocolloid C, and $910 for hydrocolloid D. In the case of venous ulcers, the average cost per patient healed was $2,939 for impregnated gauze, $1,873 for hydrocolloid D, and $15,053 for human skin construct.
Authors' conclusions The analysis has confirmed the study hypothesis that a choice made on the basis of the cheapest dressing material does not lead automatically to the selection of the most cost-effective treatment. In the study, "the product with the lowest purchase price (gauze) was not only expensive as a result of increased nursing costs associated with a high frequency of dressing changes, but 49% of pressure ulcers and 60% of venous ulcers managed with this method remained unhealed after 12 weeks of care". Hydrocolloid D was the most effective and cheapest treatment for both venous and pressure ulcer patients.
CRD COMMENTARY - Selection of comparators The reason for the selection of the different venous and pressure ulcer protocols was justified implicitly by the authors. The three selected treatments for each group of pressure and venous ulcers represented the treatment modalities remaining after the studies were selected from the literature. You should consider whether they are widely used technologies in you own setting.
Validity of estimate of measure of effectiveness The effectiveness evidence was derived from a well-conducted systematic review of the literature. The search methods, and the criteria for the inclusion and exclusion of the primary studies, were clearly reported. In addition, many of the primary studies were prospective clinical controlled trials.
The results of the meta-analysis were not used to populate the decision model, but weighted averages of the estimate values obtained from the primary studies were computed. This enhanced the internal validity of the analysis. However, the method used to weight and combine the studies was unclear. Further, due to the lack of some estimations, missing values were obtained through interpolation techniques.
The authors also acknowledged other limitations of the effectiveness analysis. First, it was difficult to compare the primary study populations in terms of the percentage of ulcers healed according to ulcer stage, and ulcer location. Second, the time horizon of the study was short (12 weeks). Finally, adverse events were excluded.
Validity of estimate of measure of benefit The benefit measure (the number of healed and non-healed ulcers) was modelled. Few details of the decision model were provided in the study. Also, the number of healed ulcers would be inadequate for a direct comparison with other technologies. A more generic measure, for example the quality-adjusted life-years, would be required.
Validity of estimate of costs The costs were estimated from actual data, which appears to have been quite specific to the study setting. Costs categories seem to have been appropriate to the setting, although statistical analyses of the costs and the quantities were not reported. The authors acknowledged that some cost items (hospitalisation, nursing home stay, and supportive care and vascular assessment) were omitted. Their impact on the authors' conclusion was unclear. The costs related to adverse events were also excluded, mainly because the length of follow-up reported in the primary studies was too short.
Other issues The authors did not compare their findings with those from other studies. In addition, the issue of the generalisability of the study results to other settings was not addressed. Therefore, the external validity of the study was quite limited. The authors also presented the study results selectively, with much of the costing detail being absent. Finally, the authors noted that a limitation was the lack of available effectiveness data.
Implications of the study The main implication of the study was that the choice of strategy depends on the total costs and effectiveness, and not simply on material costs. In the case of a managed-care plan with 100,000 covered lives, the difference between the least and the most cost-effective treatment was about $5.8 million for venous ulcers and $1.9 million for pressure ulcers. Further research should focus on well-designed studies, using appropriate methods for randomisation and statistical analysis. These conclusions should be viewed in the light of the caveats identified.
Bibliographic details Kerstein M D, Gemmen E, van Rijswijk L, Lyder C H, Phillips T, Xakellis G, Golden K, Harrington C. Cost and cost effectiveness of venous and pressure ulcer protocols of care. Disease Management and Health Outcomes 2001; 9(11): 651-663 Indexing Status Subject indexing assigned by CRD MeSH Adult; Aged; Bandages; Colloids; Cost of Illness; Costs and Cost Analysis; Humans; Managed Care Programs; Pressure Ulcer /prevention & Ulcer /prevention & Varicose Ulcer /prevention & Wound Healing; control /therapy /economics; control /therapy /economics; control /therapy /economics AccessionNumber 22001002196 Date bibliographic record published 30/09/2002 Date abstract record published 30/09/2002 |
|
|
|