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 study examined the critical pathway approach to treating diabetic patients with pedal infections. The critical pathway approach involves a multidisciplinary team taking an aggressive approach to treatment, whilst ensuring similar patients are treated consistently. Within this programme, the critical pathway was initiated in the emergency department using a committee-approved standing physician's orders, and clinical progress records, to ease the transition between departments. No further details of the approach were provided.

Treatment.

Cost-effectiveness analysis.

The study population consisted of patients with diabetic foot infections who were treated in an inpatient setting.

The setting was secondary care. The economic study appears to have been carried out at Roger Williams Medical Centre, the location of which was unspecified. However, it was not explicitly stated that this was the case.

The effectiveness data were collected during several time periods. These were January to June 1993, January to June 1995, and October 1995 to September 1996. The costs were evaluated as charges incurred over the same periods, although the dates during which the costs were collected were not explicitly stated. No price year was reported.

The effectiveness data were derived from a single study.

The costing was carried out retrospectively on the same patient sample as that used in the effectiveness study.

The study sample consisted of all patients who had a primary diagnosis (according to ICD-9 codes) of diabetic foot infection and who were admitted to the authors' institution between January and June 1993, January and June 1995, and October 1995 and September 1996. The authors did not report whether power calculations were carried out to ensure sufficient patients in each sample to rule out the influence of chance. The study sample was appropriate since it included patients admitted for the treatment of pedal infections. Patients with a secondary diagnosis of pedal infection were excluded. In total, there were 60 critical pathway patients, 25 non-pathway patients, and 30 historical controls and non-pathway patients.

The analysis was a prospective and retrospective cohort study, with groups defined by the treatment received. The study was carried out in a single centre. The length of follow-up was not explicitly stated, although the authors defined readmission within 6 months as a primary outcome. Therefore, the reader can infer that the follow-up was at least 6 months. No loss to follow-up was reported. No attempts to blind the analysts or patients were reported either.

The basis of the analysis was intention to treat. The primary health outcomes were defined as amputation level (toe, transmetatarsal, below knee and above knee), and readmission within 6 months for the same problem.

The authors reported information on the percentage of males in each group and the average age of each group, but these statistics were not discussed. Of the historical controls, 60% were male and the average age was 72.6 years (range: 53 - 91). Of the critical pathway patients, 61% were male and the average age was 63.7 years (range: 32 - 93). Of the non-critical pathway patients, 46% were male and the average age was 70.2 years (range: 42 - 95). The average age of critical pathway patients seems to have been substantially below that in the other two groups. This may represent a statistically significant difference, yet the authors did not discuss age or any other variables as potentially confounding factors.

The average length of stay was 14.4 for the historical controls, 4.4 for the critical pathway patients and 8.3 for the non-pathway patients.

The rate of readmission was 20% for the historical controls, 12% for the critical pathway patients and 16% for the non-pathway patients.

The rate of major amputations (above knee and below knee) was 27% for the historical controls, 7% for the critical pathway patients and 20% for the non-pathway patients.

The rate of minor amputations was 30% for the historical controls, 30% for the critical pathway patients and 20% for the non-pathway patients.

The authors reported in the written description of the results that:

the decrease in length of stay for patients in the critical pathway group, compared with the historical controls, was significant, (p<0.05);

the decrease in the proportion of major amputations in 1995 and 1996 (pathway and non-pathway groups), compared with the historical controls, was significant, (p=0.2); and

the decrease in the proportion of major amputations during 1995 and 1996 for patients treated with the pathway model, compared with the non-pathway model, was significant, (p<0.001).

The authors concluded that, over an 18-month period, the inpatient critical pathway decreased length of stay and pedal morbidity (amputation above the ankle) in comparison with patients treated without the pathway.

The authors did not derive a summary measure of benefit. The study was therefore categorised as a cost-consequences analysis.

A perspective for the cost analysis was not reported. The costs were estimated by evaluating the total charges. The costs and the quantities were not reported separately. The authors did not state which cost aspects were included in the study. Therefore, it is not possible to comment on whether all the likely costs were evaluated. The authors did not make inflationary adjustments. Discounting was not reported to have been carried out. However, since the costs seem to have been estimated over the 18 months of the clinical trial, discounting was not necessary. As the costs were based on charges, the total cost was derived from actual data. The reader might infer that data were collected from patient records taken during the clinical trial (January to June 1993, January to June 1995, and October 1995 to September 1996), although this was not explicitly stated.

The authors used an unpaired t-test to compare costs between the pathway and non-pathway patients.

The authors did not report any indirect costs. However, the indirect costs to society may have been relevant in this instance since the length of stay generally exceeded one week, suggesting some implications for economic productivity over this period.

US dollars ($).

No sensitivity analysis was reported.

See the 'Effectiveness Results' section.

The average charge was $16,848 (range: 2,153 - 47,095) in the historical controls, $6,240 (range: 609 - 13,949) for the critical pathway patients and $11,903 (range: 1,856 - 38,381) for the non-pathway patients.

The costs and benefits were not combined.

Over the 18-month trial period, the inpatient critical pathway decreased length of stay, charges and pedal morbidity (defined as amputation level above the knee) in comparison with the non-pathway patients.

The authors compared treatment with and without the critical pathway. It was unclear which alternative represented current practice in the authors' setting, but the choice of the comparator was well justified with reference to targets set by the US Department of Health and Human Services. You should decide if it represents a valid comparator in your own setting.

The analysis used both a prospective and retrospective cohort study. This was appropriate for the stated aim, which was to evaluate the pathway approach to treatment. The study sample was representative of the study population since it comprised diabetic patients treated for pedal infection. There was relatively little comparison of the groups prior to analysis, and no statistical analysis of the differences was reported. Therefore, it is not possible to assess the comparability of the treatment groups. The authors did not discuss potential confounding factors and no sensitivity analysis was conducted. Due to these limitations, the internal validity of the study may be quite low.

The authors did not derive a summary measure of health benefit. The analysis was therefore categorised as a cost-consequences study.

A cost perspective was not explicitly stated. However, the authors assessed charges, suggesting that the perspective may have been that of the third-party payer. The costs were not broken down into their component parts. Therefore, it is not possible to tell whether any potential costs were omitted from the analysis. The costs and the quantities were not reported separately, and no price year was reported. These factors limit the reproducibility of the results to other settings.

The authors appropriately compared their findings with those from other studies. Some differences between the results were highlighted and possible reasons for these differences were discussed. The issue of generalisability to alternative settings was not discussed. The authors did not report their results selectively.

The authors' conclusions accurately represented the scope of the study and the results presented. However, they also claimed that their results provided evidence that the critical pathway group experienced earlier recognition evaluation and clinical intervention. Whilst this claim may be true, the evidence presented did not support it. No limitations of the study were discussed.

The authors suggest that "in order to reach the goal of the US Department of Health and Human Services...diabetic infections must be treated in a much more aggressive and time-efficient way". This comment lends support to wider implementation of the critical pathway treatment. There were no suggestions for further work.

Funded by the Wellcome Trust and the UK Medical Research Council.