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 compared different delivery options of a programme addressed at young couples which was aimed at the preservation or adoption of positive behaviours regarding nutrition and physical activity. The options compared were programme delivery using primarily mail-outs (low level) and programme delivery using a combination of mail-outs and interactive group sessions (high level). The intervention was compared with no intervention (control group).

Primary prevention.

Cost-effectiveness analysis.

The study population comprised couples living together for the first time and for less than 2 years in the area of Perth. Only couples who planned to stay in Perth throughout the study period and did not plan a pregnancy during the implementation of the intervention were included in the study. Individuals who suffered from heart disease, diabetes or severe asthma were excluded from the study.

The setting appears to have been the community. The economic study was carried out in Perth, Western Australia.

The dates to which the effectiveness and cost data related were not reported. The price year was also not reported.

The effectiveness data were derived from a single study.

It seems that the costing has been carried out prospectively on the same sample of patients as that used in the effectiveness study.

Power calculations demonstrated that a sample of 16 couples per group was required to detect a difference from controls of 2% in total fat ingestion and energy expenditure equal to quick walking for 45 minutes/week. The calculations assumed an alpha level of 0.05 and a minimum statistical power of 80%. Couples informed through press, radio and television advertisements, and who fulfilled the inclusion and exclusion criteria, were selected to participate in the study. No couples were reported to have refused to participate in the study. No couples were excluded from the initial sample.

Overall, 137 couples entered the study. Forty-three couples comprised the control group, 47 the low-level group and 47 the high-level groups. The mean age was 31.5 (standard deviation, SD=9.6) for men and 29.2 (SD=8.7) for women in the control group, 29.8 (SD=8.7) for men and 27.7 (SD=6.7) for women in the low-level group, and 30.9 (SD=8.3) for men and 27.4 (SD=7.1) for women in the high-level group.

The analysis was based on a single-centre randomised controlled trial. The method of randomization was not reported. The data were collected at baseline and at 16 weeks and 12 months from the start of the programme through self-administered questionnaires. From the initial sample size, 111 (81%) successfully completed the programme at 16 weeks. The reasons for withdrawal were reported. Only 81 (59%) couples from the initial sample were followed up at one year. The reasons for these withdrawals were also reported.

The analysis was conducted on an intention to treat basis. It was reported that there were no statistical differences in demographic and baseline characteristics between the patient groups, or between those patients who completed the study and those who dropped out. The primary outcomes used in the analysis were:

changes in consumption of high-fat energy foods,

consumption of fat,

fibre intake,

fruit and vegetable consumption,

body mass index,

exercise days, and

fitness adjusted for body weight.

The health outcomes included cardiovascular risk factors such as total blood cholesterol, low-density lipoprotein (LDL) cholesterol, and systolic and diastolic blood pressure.

The effectiveness results were reported as mean differences at post-intervention (16 weeks) and 1-year follow-up from baseline.

At 16 weeks, the total blood cholesterol was decreased by 0.09 mmol/L (standard error, SE=0.05) (2%) in the high-level group and by 0.04 mmol/L (SE=0.06) (1%) in the low-level group, while it was increased by 4% in the control group. The difference between the high- and low-level groups was not statistically significant, (p=0.569).

LDL cholesterol was decreased by 0.12 (SE=0.05) (4%) in the high-level group and by 0.05 (2%) in the low-level group. The difference between the two groups was not statistically significant, (p=0.187).

Differences between the two groups were not statistically significant for decreases in systolic and diastolic blood pressure, (p=0.176 and p=0.753, respectively).

At the 1-year follow-up, differences between the two groups were statistically significant for an increase in fitness adjusted for body weight (W/kg), (p=0.021), decrease in total fat (% energy), (p=0.013), and decrease in saturated fat, (p=0.045).

Both intervention groups proved to be more effective than the control group at the end of the 16-week programme, particularly in terms of increasing exercise days (days/week) and decreasing total blood cholesterol and LDL cholesterol. However, only the high-level programme was more effective in decreasing systolic and diastolic blood pressure in comparison with the control group.

At the 1-year follow-up, both intervention groups were more effective in decreasing energy, saturated fat, sodium intake and high-fat food (serves/day) and increasing fruit and vegetable intake (serves/day). Only the high-level programme remained more effective in decreasing total fat intake total blood cholesterol and systolic and diastolic blood pressure, and increasing fitness adjusted for body weight, in comparison with the control group. However, the low-level programme was more effective in increasing exercise days. The statistical significance of differences in outcomes between the intervention and control groups was not reported.

The analysis demonstrated that both interventions are more effective than no intervention (control group). However, only a few outcome variables sustained increased effectiveness after 1 year.

The authors did not use a summary measure of benefit. In effect, a cost-consequences analysis was performed (see 'Effectiveness Results' section).

The health service costs included in the analysis were personnel costs and other costs. Personnel costs comprised staff salaries plus "on-costs" (3-month programme development, 1-month programme review, 1-month programme improvement training) and the costs of a dietician (workshops, data collection and administration), exercise physiologist (workshops, fitness test, data collection and administration), nurse (data collection) and data analyst (data processing and analysis). Other costs included the costs of printing, postage, refreshments administered at workshops, workshop venue rental and capital costs of 2 sphygmomanometers, 2 exercise bicycles and a computer. It was reported that overhead costs were included in the analysis.

The costs and the quantities were not analysed separately, and the unit costs were not reported. The quantities of resources used were derived from the single study, while the sources of the cost data were not explicitly reported. Discounting was not relevant since the costs were incurred during less than 2 years. However, equipment costs were treated as one-time capital expenditure and the value was annualised.

It appears that the costs have been treated deterministically.

The indirect costs were not included in the analysis.

Australian dollars (AUD).

The authors conducted a one-way sensitivity analysis to test the robustness of the results to variability in the data. The impact of variability in the costs on the results was investigated in a sensitivity analysis by varying the completion rates of participants, based on the rationale that average costs depend on the number of participants and the interventions become more costly with falling completion rates. A 90%, 75% and 50% participation rate at post-intervention and a 60%, 50% and 40% participation rate at follow-up were tested in sensitivity analyses. However, the method used to select the ranges was not reported.

See the 'Effectiveness Results' section.

The total cost of the high-level intervention was AUD 41,854.34 (AUD 445.30 per participant and AUD 111.33 per participant per month).

The total cost in the low-level group was AUD 41,847.26 (AUD 445.18 per participant and AUD 111.30 per participant per month).

The total cost in the control group was AUD 38,372.55 (AUD 446.19 per participant and AUD 111.55 per participant per month).

The authors produced average cost-effectiveness ratios for 14 clinical outcomes. They also produced incremental cost-effectiveness ratios comparing the two interventions for every clinical outcome.

There were too many ratios to report here.

The sensitivity analysis demonstrated that the results were sensitive to all parameters investigated.

The intervention with support from interactive workshops was more cost-effective over 12 months than a mailed programme or simply changing awareness.

The authors compared two delivery methods of a diet and physical activity programme versus no intervention. No intervention seems to have represented standard practice in the authors' setting. You should decide if the comparators used comprise a valid health technology in your own setting.

The analysis was based on a single-centre randomised controlled trial, which seems to have been appropriate given the study question. The study sample was representative of the study population and the patient groups were shown to be comparable at analysis. The method of randomisation was not reported. Power calculations were undertaken but, although an adequate sample size was used to detect statistically significant differences from the controls, the study was not of sufficiently powered to detect statistically significant between-group (intervention groups) differences.

The authors did not use a summary measure of benefit. In effect, a cost-consequences analysis was performed.

The perspective adopted in the economic analysis was not explicitly reported. However, it appears to have been that of the health care system. The costs and the quantities were not reported separately, which will not enable the analysis to be easily reworked for other settings. The quantities of resources were obtained from the single study, but no statistical analysis on the quantities was performed. The costs were evaluated alongside the clinical trial. The unit costs were treated deterministically, but no sensitivity analysis on the cost estimates was conducted to asses their robustness. In addition, the price year was not reported.

The authors compared their findings with those from other studies and found them generally to be in agreement. The issue of generalisability of the results to other settings was not directly addressed. The authors do not appear to have presented their results selectively. There appears to be little value in the authors' cost-effectiveness analyses since it was a cost-consequences analysis. The study enrolled couples cohabiting for the first time but the authors generalised their conclusions across all individuals.

The authors reported a number of limitations to their study. First, the time horizon of the study (12 months) was too short to allow for improvements in cardiovascular risk factors to be observed. Second, the cost categories and cost data used in the analysis were an overestimation of the actual costs incurred for the implementation of the programme in a community environment. In addition, higher socioeconomic status was over-presented in the study sample and this might have introduced some bias. However, the participants did not follow a healthier lifestyle in comparison with the general Australian population.

The authors did not make explicit recommendations for changes in policy or practice or for further research. However, the discussion highlighted areas where more research-based information could be useful.

None stated.