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.

Nutritional counselling in general practice for overweight individuals and patients with hypertension and type II diabetes.

Treatment.

Cost-effectiveness analysis.

Male and female consecutive patients aged 25-65 years, screened opportunistically on arrival at general practice, for a consultation about one or more of the following, obesity, hypertension and type II diabetes.

Primary care. The economic study was carried out in an outer suburb of Perth, Australia.

Effectiveness data relate to the period November 1992 to May 1994. Cost data relate to 1993/94 values.

Effectiveness data were derived from a single study.

Costing was undertaken prospectively on the same patient sample as that used in the effectiveness analysis.

The initial study sample comprised 296 consecutive patients, of whom 23 refused screening and 21 declined to participate in the study after screening. 273 patients took part in the study (75 males, 198 females). Each patient was allocated to either a dietician only or doctor/dietician for 6 nutritional counselling sessions (nutrition and exercise advice) over a 12-month period, or no counselling in the control group. In the doctor/dietician group it was the doctor who invited the patients to join the study and reviewed their progress at two of the 6 sessions. The dietician invited the patients in the dietician-only group. 78% of patients were in the lower two socio-economic quartiles, with 56% describing their occupation as home duties, and 78% had a partner. Power calculations were performed based on an expected 5% weight reduction in the dietician group and 10% in the doctor/dietician group. A minimum of 35 overweight patients per group was required to achieve a power of 0.9, that the null hypothesis would be rejected at the 0.5 level. Exclusion criteria included patients who were mentally ill, intellectually handicapped, terminally or acutely ill or participating in other education programmes. Following screening, a diagnosis of overweight alone accounted for 62% of patients. A further 31% of patients were overweight and hypertensive, 2% were overweight and diabetic and 4% had all three conditions. The remaining 1% had either hypertension or diabetes.

This was a prospective randomised controlled trial, based in a single centre. Patients were randomised using random number tables following screening. At the end of the 12-month period, of the initial 273 patients recruited, 177 (65%) completed all 6 sessions, with the drop out rate for the overweight in the dietician group (45%) being significantly greater than the 29% for both the doctor/dietician group (p=0.022) and the control group (0.021).

The analysis of effectiveness was based on intention to treat. Groups were comparable with respect to age, sex, socio-economic status and occupation. Main health outcomes were reductions in weight and blood pressure, glycated haemoglobin levels and medication use. No side effects were examined. There was no difference between the control and doctor/dietician groups with respect to drop out rate in the hypertension group, but significantly fewer patients attended all sessions compared with the control group, (p=0.035).

With regard to weight reduction, patients in the doctor/dietician group on average lost an extra 6.7 kilograms over the 12-month period or 7.3% of screening weight, compared to the control group (95% CI: 6.5% - 8.3%) and the dietician only group lost 5-6 kilograms or 6.6% of screening weight (95% CI: 5.8% - 7.6%). The doctor/dietician group lost 1.1 kilograms more than the dietician only group, but this result was not statistically significant. Analysis only of patients who attended all sessions showed doctor/dietician patients were 8.1 kilograms or 8.8% lighter than their screening weight (95% CI: 8.0% - 9.6%) and dietician patients lost 7.7 kilograms or 9.1% (95% CI: 8.0% - 10.2%)

With respect to blood pressure, the doctor/dietician and dietician groups both had significant changes in final mean blood pressure compared to the control group with falls of 12mmHg or 12% (95% CI: 9% - 15%), and 7mmHg or 7% (95% CI: 4% - 10%) respectively. The doctor/dietician group lowered blood pressure by 5mmHg or 5% (95% CI: 2% - 8%) more than the dietician group. Analysis only of patients who attended all sessions showed doctor/dietician patients were 14mmHg or 12% less than their screening mean blood pressure (95% CI: 8% - 16%) and dietician patients were 7mmHg or 7% lower (95% CI: 1 - 13).

No significant difference was found between the control and intervention groups for glycated haemoglobin for patients with diabetes.

No significant difference was found in the average defined daily dose of cardiovascular medication use by patients in all three groups on recruitment or at the final counselling session. The intervention of nutritional counselling in general practice can create weight loss and a reduction in blood pressure for patients who are overweight and/or hypertensive.

No modelling was carried out.

The measures of benefit used in the economic analysis were kilograms lost and blood pressure reduction.

Costs and resources were not reported separately. Direct costs related to activities associated with the provision of the nutritional counselling in the general practice including, initial screening, organisation of patient appointments and patient files, data entry and filing. Associated resource use related to resources required for these were materials used by the dietician, room rental, practice overheads and dietician and doctor time. Costs were not discounted, as the study was for a 1 year period only. No specific information was provided on the level of resources used and their particular costs. The quantity/costs boundary estimation is based on actual data. The price year was 1993/94.

No statistical analysis of costs was reported.

No indirect costs were reported.

Australian dollars (Aus$). No currency conversion was reported.

No sensitivity analysis was reported.

The additional weight change per patient was -6.71 kg in the doctor/dietician group and -5.63 kg in the dietician group.

The total cost of the intervention was Aus$8,240.30 for the doctor/dietician group and Aus$5,715.06 for the dietician group, compared to the control group at Aus$2,103.53. The cost per patient was reported to be Aus$23.12 for the control group, Aus$88.61 for the doctor/dietician group and Aus$64.21 for the dietician group. This represented an additional cost per patient of Aus$65.49 for the doctor/dietician group and Aus$41.09 for the dietician group.

Compared with the control group, the cost for an extra kilogram of weight loss was Aus$9.76 for the doctor/dietician group and Aus$7.30 for the dietician group.

General practice can provide health promotion to reduce weight and hypertension in patients, at a reasonable cost over a 12-month period, with a combination of a doctor and a dietician. Although the GP contribution increases the cost of health promotion, it has fewer patients dropping out and better outcomes than those achieved by the dietician alone.

The reason for the selection of comparators is clear. However, there is the potential for selection bias in that consecutive patients were recruited to the study on presentation at the general practice reception, and, where two individuals arrived at the same time, only one was selected. You, as a user of this database should decide whether these health technologies are relevant to your setting.

The measure of benefit was appropriate for the type of intervention. However, there are additional benefits that have not been included in the study, such as information on whether there were any reduced episodes of illness due to weight reduction, as compared with the control group.

Information on the quantities of resource use and specific costs was not reported. It is not possible to examine whether direct costs were calculated appropriately, as there is insufficient information on resource use. No information was provided on whether there are reduced costs as a result of fewer episodes of illness due to weight reduction, compared with the control group. Costs are likely to be difficult to generalise due to local values being applied. Further, it could have been useful to examine the costs incurred by patients due to nutritional and exercise changes, but no indirect costs were reported. No sensitivity analysis was reported making it difficult to examine any assumptions made about resource use.

The authors' conclusions appear fairly justified given the results presented, and appropriate comparisons were made with other studies in this area of research. However, while the nutritional counselling was shown to be cost-effective compared to a control, a comparison with hospital based nutritional counselling would have to be made in order to examine whether general practice is preferable to hospital care, where nutritional counselling usually takes place. Further, follow-up of the patients after the 12-month study period is required to examine whether the nutritional counselling had influence after the intervention period. Finally, it is possible that the study may only be applicable to patients from socio-economically disadvantaged areas.

Further research is required to examine the nutritional intervention in more than one general practice. Further, it would be useful to examine whether there were any changes in hospitalisation rates, due to the weight and blood pressure reductions.

Research funded by a grant from the Western Australian Health Promotion Foundation.