|
Lifestyle intervention by group care prevents deterioration of Type II diabetes: a 4-year randomized controlled clinical trial |
Trento M, Passera P, Bajardi M, Tomalino M, Grassi G, Borgo E, Donnola C, Cavallo F, Bondonio P, Porta M |
|
|
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 use of group sessions built up on a systemic education programme for the management of non-insulin-treated Type II diabetic patients. The educational sessions were given by one or 2 physicians and one educationalist. The topics included were the burden of overweight, choosing food, meal planning, physical exercise, checking and improving metabolic control, smoking cessation, assuming medication and preventing complications. These topics were divided into 4 sessions, one every 3 months, during the first 2 years. During years 3 and 4 a total of 7 sessions were held, with deeper discussion and learning. Individual consultations were reserved for emerging medical problems, yearly checks for complications, or the patients' explicit requests.
Type of intervention Treatment and secondary prevention.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised non-insulin-dependent Type II diabetic patients. To be included in the effectiveness analysis, the patients had to be Type II diabetics treated by diet or with oral agents. They also had to be younger than 80 years and have attended the clinic for at least one year.
Setting The setting was a hospital. The economic study was performed at the University of Turin, Italy.
Dates to which data relate The dates to which the effectiveness data related were not reported, but may have been the same as for the cost data. The costing appears to have been from between 1996 and 2000. The price year was not given.
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 population as that used in the effectiveness analysis. Some of the cost data may have been collected prospectively, for example timing of the clinic procedures. Other cost data were collected retrospectively, using a questionnaire for the transportation costs and opportunity-costs of the patients' time spent on treatment.
Study sample There was no report of power calculations being performed in the planning phase of the study to assure a certain power. In total, 120 patients met the inclusion criteria and were randomly selected from the database of the centre where the study was performed. Eight of these were not finally considered either because they refused to participate or because they had to start insulin. The final sample comprised 112 patients. There were 56 in the group session therapy (GST) and 56 in the individual session therapy (IST).
Study design The study was a randomised controlled trial, which was carried out in a single centre. The patients were allocated randomly to either the GST or IST groups using random table numbers. Although the authors reported that the physicians were blinded to the IST patient status to avoid performance bias, they also stated that these physicians actually took charge of the group sessions and, therefore, were not actually blinded, at least to the GST patient status. The period of follow-up was 4 years, 51.1 (+/- 2.1) months in the GST group and 51.2 (+/- 1.8) months in the IST group. In the GST group 3 patients died and 8 moved to other clinics. In the IST group 2 patients died, 17 moved to other clinics and 2 were lost to follow-up.
Analysis of effectiveness The authors reported that the basis for the effectiveness analysis was intention to treat. The health outcomes assessed in the effectiveness analysis were:
the average number of sessions received by GST and IST patients;
measurements of body weight, fasting blood glucose, haemoglobin (Hb)A1c and blood lipids;
knowledge of diabetes, health behaviour and quality of life (QoL);
assessment of diabetic retinopathy, hypoglycaemic medication, microalbuminuria, systolic and diastolic blood pressure, Framingham score for cardiovascular risk, and antihypertensive and lipid-lowering medication.
Knowledge of diabetes was assessed by a questionnaire developed by the Education Study Group of the Italian Society for Diabetes. Health behaviour was measured using a questionnaire that was built specifically for the study. The DQOL questionnaire was used to measure QoL. Diabetic retinopathy was assessed according to EURODIAB and European screening Guidelines procedures. Hypoglycaemic treatment was assessed both as the class of medication and overall dosage modifications (algebraic sum of interventions in which medication was increased minus the sum of interventions in which medication was reduced). Antihypertensive medication was assessed as the number of different classes of drugs administered. Lipid-active medication was assessed as the prescription of fibrates or statins.
The groups were shown to be comparable at analysis in terms of their gender, age, occupation, known duration of diabetes, time attending the clinic before study, family history of diabetes mellitus, self-monitoring blood glucose, smoking and hypoglycaemic treatment. Patients in the IST group, however, had higher education and better knowledge of diabetes when compared with those in the GST group.
Effectiveness results The average number of sessions received by patients was 15.8 (range: 13 - 17) in the GST group and 12.5 (range: 6 - 17) in the IST group.
For the GST group, the difference between the body weight at the end of the study (75.2 +/- 13.0 kg) and the baseline value (77.8 +/- 13.6 kg) was -2.6 kg, (p<0.001). This difference was not statistically significant for the IST group.
The body mass index was also significantly lower (-1.0 kg/m2; p<0.001) after 4 years of follow-up for patients in the GST group (28.7 +/- 4.0), compared with the baseline value (29.8 +/- 4.5). This difference was not significantly different for IST patients.
The HbA1c level remained stable between baseline and year 4 among the GST patients. However, it worsened among the IST patients, from 7.4% (+/- 1.4) to 8.6% (+/- 2.1) of the total Hb.
Blood pressure decreased between baseline and year 4 in both the GST and IST patients, although the difference was only significant for diastolic blood pressure, (p<0.001). This decreased from 95 (+/- 11) mmHg to 88 (+/- 7) mmHg for GST patients, and from 92 (+/- 10) mmHg to 86 (+/- 9) mmHg for IST patients.
Blood urea nitrogen increased significantly among the IST patients, from 13.56 (+/- 3.57) mmol/L at baseline to 15.74 (+/- 5.78) mmol/L at year 4, (p<0.05).
HDL cholesterol increased significantly among the GST patients, from 1.27 (+/- 0.31) mmol/L at baseline to 1.42 (+/- 0.31) mmol/L at year 4, (p<0.001).
The scores obtained for the knowledge of diabetes increased for the GST patients, from 14.9 (+/- 7.9) at baseline to 27.1 (+/- 6.6) at year 4, (p<0.001). However, these scores worsened for the IST patients, from 20.4 (+/-7.8) to 17.2 (+/-8.7), (p<0.05).
The scores related to health behaviour improved for GST patients, from 11.0 (+/- 2.7) at baseline to 16.5 (+/- 2.9) at year 4, (p<0.001). However, they worsened among the IST patients, from 12.3 (+/- 4.2) at baseline to 10.2 (+/- 3.9) at year 4, (p<0.001).
The scores for QoL improved for GST patients, from 67.6 (+/- 19.0) at baseline to 44.0 (+/- 7.5) at year 4, (p<0.001). However, they worsened among IST patients, from 70.5 (+/- 21.7) at baseline to 89.8 (+/- 28.1) at year 4, (p<0.001). (The reader should note that higher scores for the DQOL questionnaire imply a worse QoL).
The authors reported that, after adjusting by age, duration of diabetes and education, the only differences between baseline values and values at year 4 that remained statistically significant across the groups were those related to HbA1c levels, urea nitrogen levels, knowledge of diabetes, health behaviour and QoL.
When the increases and decreases of hypoglycaemic medication were compared between GST and IST patients, significant differences were found that favoured the GST group, (p<0.001).
The absolute cardiovascular risk did not change over 4 years in either group (24.6 +/- 8.9 among GST patients and 25.8 +/- 12.0 among IST patients). The relative cardiovascular risk was reduced in both the GST and IST groups by 0.2, (p<0.05).
In terms of retinopathy, there was a significant difference in the number of patients who improved versus those who worsened, (p=0.009), favouring the GST patients. In the GST group, 6 of the 12 patients improved from mild retinopathy to no retinopathy, while 4 of the 33 patients worsened from no retinopathy to mild retinopathy. In the IST group, only one of the 14 patients improved from mild to no retinopathy, while 3 out of 14 patients worsened from mild to severe retinopathy, and 10 out of 18 worsened from no retinopathy to mild retinopathy.
Clinical conclusions GST was shown to be more effective than IST in terms of promoting better health behaviours, better knowledge of diabetes, improvements in metabolic control and QoL. It also reduced progression to retinopathy and the use of hypoglycaemic agents.
Modelling A prediction model assessed the cardiovascular risk, using a Framingham population sample matched by age and sex (see Other Publications of Related Interest). The risk factors considered were age, gender, systolic and diastolic blood pressure, total and high-density lipoprotein (HDL) cholesterol, smoking and diabetes.
Measure of benefits used in the economic analysis The measure of health benefit used in the economic analysis was the DQOL score obtained in the effectiveness analysis (see the 'Effectiveness Results' section).
Direct costs Only some resource quantities were reported separately from the costs. Additionally, some of the resource quantities could be inferred from the description of the therapies under study. The direct costs included in the study were those of the health service and the patients. The health service costs were for staff, educational material, clinical procedures and pharmacological treatment. The patients' costs were for transportation of the patients and their caregivers to attend the clinic. The transportation time incurred by the patients was calculated from a questionnaire completed by the patients. Timing of the clinic was measured during 12 group sessions. The sources of the costs imputed to resource utilisation were not reported. It was not reported whether discounting was performed, although it would have been necessary given that the follow-up was longer than 2 years. The only date reported was that related to the staff costs (1999). The price year was not given.
Statistical analysis of costs No statistical analysis of the costs was reported.
Indirect Costs The indirect costs included in the analysis were the opportunity-cost value of time of patients in either the GST or the IST groups. The resource quantities and the costs were reported separately. The authors reported that a questionnaire was used to obtain a subjective evaluation of these opportunity costs. It was not reported whether discounting was performed, although it was relevant because the time horizon for the study was longer than 2 years. No price year was reported for the indirect costs.
Currency US dollars ($). Costs converted from Italian lira (L) at a rate of L 1,000 = $0.46985 (a weighted average of the annual official rates from 1996 to 2000)
Sensitivity analysis No sensitivity analysis was reported.
Estimated benefits used in the economic analysis There was a reduction of 23.60 points in the DQOL score between baseline and year 4 of among the GST patients. For IST patients the DQOL increased by 19.20 points. The difference in the DQOL scores between the GST and the IST patients was -42.80 points.
Cost results The total costs reported by the authors (including both direct and indirect costs) were $756.54 per GST patient and $665.77 per IST patient. Therefore, the cost per patient was $90.77 higher in the GST group than in the IST group.
Synthesis of costs and benefits The costs and benefits were combined using an incremental cost-effectiveness ratio (ICER). This was calculated as the ratio between the differential total costs and the differential DQOL scores. This ICER was $2.12 per incremental improvement (i.e. a reduction of 1 point) in the score of the DQOL questionnaire with the GST strategy when compared with the IST strategy.
Authors' conclusions Group session therapy (GST) can be a good way to implement lifestyle intervention in Type II diabetic patients because of the improvements obtained in quality of life (QoL), health behaviours, knowledge of the illness, and metabolic control. No conclusions about the costs and the incremental cost-effectiveness ratio (ICER) were drawn.
CRD COMMENTARY - Selection of comparators The comparator was justified on the grounds that traditional individual diabetes care was the current practice in the authors' setting. You should consider if this is a widely used health technology in your own setting. Also, whether there are alternative strategies to care for non-insulin-dependent Type II diabetic patients.
Validity of estimate of measure of effectiveness The analysis used a randomised controlled trial in which the patients were allocated to groups using random number tables. This design was appropriate for the study question. The patient groups were shown to be comparable at analysis in terms of some of the relevant risk factors associated with the disease, although they were different for education, which might be related to the presence of selection bias. The authors reported that the physicians were blinded, but this was only the case for IST patients, not GST patients. Multivariate analysis was used to take account of confounding factors. Although the authors reported statistical analyses comparing values at baseline and at year 4 for both GST and IST groups, they did not report statistical analyses comparing the results for the GST and the IST groups. Some of the effectiveness outcomes were based on Italian questionnaires, which may be appropriate for use in Italy but may not be generalisable to other settings. It is not clear whether the assessment of some outcomes, such as the hypoglycaemic treatment or the health behaviour, was appropriate, because they were assessed following specific methods developed for this study.
Validity of estimate of measure of benefit The estimation of the benefits was proxied directly by the estimate of effectiveness related with QoL (which was the DQOL score). This choice of estimate, however, was not justified. Additionally, although the DQOL has been tested and validated with Type II diabetic patients, it was designed specifically for Type I diabetics with intensive insulin treatment. Thus, it may have deficiencies in assessing the QoL for non-insulin-dependent diabetic patients. In addition, it is unclear whether the use of the DQOL scores is an appropriate way to calculate the ICER.
Validity of estimate of costs All the categories of costs relevant to the perspective adopted seem to have been included in the analysis. Some relevant costs were omitted from the analysis. For example, the costs of blood and urine tests were not included because they were similar for both GST and IST patients. The costs of mortgage rates for premises and equipment were not included either, but in this case the authors justified their exclusion because they were negligible. These omissions appear unlikely to affect the authors' conclusions. Some, but not all, of the resource quantities were reported separately from the costs or could be inferred from the description of the health technologies. No statistical analyses of the quantities or costs were reported, which introduces uncertainty into the reliability of the conclusions. Moreover, the dates to which the costs related and the price year were not given, thus hindering reflation exercises to other settings.
Other issues The authors made appropriate comparisons of their results with those from other studies. They showed that patients receiving the GST therapy obtained better results in terms of metabolic control, compared with the results obtained by other studies of different interventions for Type II diabetic patients. The authors pointed out that patients included in this study were moderately overweight and the results obtained here may not be the same for obese people. In this respect, the issue of generalisability of the results to other settings was addressed. The study enrolled non-insulin-dependent Type II diabetic patients and this was reflected in the authors' conclusions.
Implications of the study The GST strategy may encourage patient involvement more intensively than the IST strategy, obtaining better health outcomes, while the increase in costs does not appear to be high. The study shows a general improvement in the health of GST patients over the study period considered, but not for IST patients. The authors highlight the fact that health education should not be confined to providing information on disease and treatment options, because most notions are either not retained or are easily forgotten.
Source of funding Supported by a grant from Turin University.
Bibliographic details Trento M, Passera P, Bajardi M, Tomalino M, Grassi G, Borgo E, Donnola C, Cavallo F, Bondonio P, Porta M. Lifestyle intervention by group care prevents deterioration of Type II diabetes: a 4-year randomized controlled clinical trial. Diabetologia 2002; 45(9): 1231-1239 Other publications of related interest Anderson RM, Funnell MM, Butler PM, Arnolds MS, Fitzgerald JT, Feste CC. Patient empowerment: results of a randomised controlled trial. Diabetes Care 1995;18:943-9.
Gaede P, Vedel P, Parving H-H, Pedersen O. Intensified multifactorial intervention in patients with type 2 diabetes mellitus and microalbuminuria: the Steno type 2 randomised study. Lancet 1999;353:617-22.
Mazzuca SA, Moorman NH, Wheeler ML, et al. The diabetes education study: a controlled trial of the effects of diabetes patient education. Diabetes Care 1986;9:1-10.
Norris SL, Engelgau MM, Venkat Narayan KM. Effectiveness of self-management training in type 2 diabetes. A systematic review of randomized controlled trials. Diabetes Care 2001;24:561-87.
Wilson PW, D'Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:1837-47.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Aged, 80 and over; Blood Glucose /metabolism; Costs and Cost Analysis; Diabetes Mellitus, Type 2 /economics /physiopathology /rehabilitation; Disease Progression; Educational Status; Female; Humans; Italy; Life Style; Male; Middle Aged; Patient Care Team /organization & Patient Education as Topic /methods; Reference Values; Smoking; Time Factors; administration AccessionNumber 22002001846 Date bibliographic record published 31/08/2003 Date abstract record published 31/08/2003 |
|
|
|