Treatment of High-Risk Patients with Diabetes: Motivation and Teaching Intervention: A Randomized, Prospective 8-Year Follow-Up Study

Materials and Methods

A total of 167 patients who had type 2 diabetes, hypertension, and dyslipidemia and referred for consultation to the diabetes clinic of Meir Hospital during the years 1995 to 1996 were randomized to a standard consultation (SC) or to a patient-participation program (PP). The inclusion criteria were age 40 to 70 yr; type 2 diabetes of <10 yr duration; BMI ≤35 kg/m²; BP values ≥140/90 mmHg; LDL ≥120 mg/dl; serum creatinine ≤2 mg/dl (176 μmol/L); albumin/creatinine ratio <200 mg/g; and no history of myocardial infarction, angina pectoris, vascular surgery, stroke, or any systemic or malignant disease.

A total of 154 patients signed informed consent forms, 13 patients were excluded, and 141 were randomized by the help of computer-generated random numbers to either SC (n = 70) or PP (n = 71) programs. The flow of the patients throughout the study is outlined in Figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Flow of participants through the trial.

Patients of both groups were initially interviewed and examined twice over 2 wk and annually thereafter. Consultation letters were written to the family physicians. No prescriptions were ever issued by members of the consultation team. The patients of the SC group attended standard consultation visits, whereas the patients of the PP program were given two 2-h teaching sessions about ways to achieve tight control of the modifiable risk factors including also an individualized plan of lifestyle modification and a fitness program, instructed to measure BP weekly, keep records of the results of the laboratory investigations, and urge their physicians to change or intensify treatment if the target values of BP (130/80 mmHg), LDL cholesterol (100 mg/dl), and hemoglobin A_1c (HbA_1c) (7%) were not reached. These patients were encouraged to call the consultants when they believed that they needed advice.

Sixty-five patients in the SC group and 64 in the PP group completed the first 4-yr phase. The follow-up was continued for an additional 4 yr, bringing the mean follow-up period to 7.7 ± 0.2 yr. Data on clinical outcome events were obtained from the letters of the primary care physicians, hospital discharge summaries, or other consultants’ reports. The protocol was approved by the institutional ethical committee. The hospital laboratory serves the whole area; all laboratory examinations therefore were done centrally. The assays were unchanged during the study period. The analytical methods were previously described (10). Renal outcome was assessed by estimated GFR (e-GFR) (11).

Results

The baseline characteristics of the patients of the PP and SC groups were previously described (10) and are shown in the first column of Table 1. There was no difference in any of the baseline characteristics between the two groups. Over 8 yr, the patients of the PP group initiated 826 additional visits to the consultation clinic, an average of 1.2 ± 0.8 visits/patient per yr. The main reason given by the patients was a failure to meet one or more of the target values of the risk factors.

The patients in the PP group were more likely to receive angiotensin-converting enzyme inhibitors or angiotensin II antagonists than those of the SC group. Also, all PP patients were prescribed lipid-lowering medication versus one half of the SC patients. Glucose-lowering regimens were not different in the two groups. The prescribed daily doses of most of antihypertensive agents and those of statins were higher in the PP than in the SC patients. The data on drug treatment are summarized in Table 2.

Nephropathy and Retinopathy

The average annual decline in estimated e-GFR in the SC and the PP groups was 4.6 ± 2.1 and 3.0 ± 1.8 ml/min per 1.73 m², respectively (P = 0.01). Overt nephropathy (albumin/creatinine ratio >300 mg/g) developed in 14 (22%) patients in the SC group versus 7 (12.5%) in the PP group (P = 0.01). The numbers of patients who developed ESRD were four and one in the SC and PP groups, respectively. Retinopathy was found in 35 versus 21 patients in the SC and PP groups, respectively (P = 0.03). The relative risk reduction of microvascular complications in the intervention (PP) versus the control (SC) group was 0.57 (95% confidence interval, 0.91 to 0.28). These data are listed in Table 1 and shown in Figure 2. No standard method to assess neuropathy was used.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Mean values of estimated GFR (e-GFR; see Results) and albumin/creatinine ratio (ACR). Standard care (SC) and patient participation (PP) groups during 8 yr.

Cardiovascular Disease

In the SC group 12 patients died, eight of cardiovascular causes (five MI, three strokes). Among the PP patients, there were nine deaths, five of which were cardiovascular (three MI, two strokes).

The number of nonfatal cardiovascular events was 72 in 45 patients of the SC group versus 47 in 31 patients of the PP group. Together with the cardiovascular fatal events, the total number of cardiovascular events was 80 versus 52 (P = 0.001). There were significantly fewer events of stroke in the PP than in the SC group (fatal + nonfatal stroke, 8 in the PP versus 17 in the SC group; P = 0.01). There were also fewer coronary artery events and interventions in the PP than in the SC patients (fatal + nonfatal MI + coronary artery bypass graft + PCI, 56 versus 41; P = 0.046). Also, the participants of the PP program had fewer amputations and bypass surgeries than their SC counterparts (3 versus 7).

A Kaplan-Meier estimation of event-free survival in the SC and PP groups is shown in Figure 3. The relative risk reduction of the main cardiovascular parameters is outlined in Table 3.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Kaplan-Meier estimation of combined cardiovascular event-free survival in SC and PP groups (P = 0.004 by the log-rank test). Cum survival, cumulative survival; GRP, group.

View this table:

• View inline
• View popup

Table 3.

Relative risk reduction (95% CI) of the main cardiovascular parameters in the intervention (PP) versus the control (SC) groups^a

Discussion

The patients with diabetes were provided with in-depth information about their disease, motivated to actively pursue their therapeutic goals, and given tools to monitor their modifiable risk factors. This was achieved via individual teaching sessions and enabling contact between patients and consultants. During 8 yr of follow-up, the cardiovascular morbidity was significantly reduced in the PP versus the SC patients. The progress of microvascular disease expressed by the annual decline in e-GFR, the rise in albuminuria, and the number of patients with retinopathy was significantly retarded already after 4 yr; at 8 yr, this difference grew in favor of the PP group.

The more favorable clinical outcome of the PP patients is borne out by the significantly lower values of the major risk parameters. Thus, the PP patients had lower BMI, lower systolic (7 mm) and diastolic (5 mm) BP, lower LDL (116 versus 128 mg/dl), and lower HbA_1c values (8.3 versus 9.2%). These differences may be partially accounted for by some differences in therapeutic policies. All PP patients received renin-angiotensin system blockers throughout the study period as compared with 54 and 74% of the patients in the SC group during 4 and 8 yr, respectively. All PP patients were on lipid-lowering medications as compared with only 34 to 59% of the SC patients at 4 and 8 yr. Furthermore, the average daily dose of angiotensin-converting enzyme inhibitors and statins was higher in the PP than in the SC patients. Also, more PP patients were on three antihypertensive drugs than their SC counterparts. The differences in glucose-lowering therapy between the groups were less conspicuous. The better compliance, perseverance, and strictness in recording personal data of the PP patients are, most probably, the main reasons for the lower values of the main risk factors in this group. The results of this study compare favorably with early as well as recently published intensive care interventions in high-risk patients with diabetes (4–9).

Most randomized, controlled studies that have compared intensive and less intensive regimens (4,7,12–15) concentrated on a single intervention, lowering BP, lipids, or glucose. The favorable outcome of the intensive versus less intensive patient groups in such studies compares well with the results of similar studies that used multiple interventions (6,8–10). This similarity is probably due to better compliance of patients who participate in clinical trials also with other prescribed medications on top of the trial drug. Every intensive intervention has a dual effect: A direct effect of the trial intervention itself and an indirect, perhaps not less powerful, effect of improved general motivation and compliance. The present study isolates and highlights the indirect effect because there was no direct pharmaceutical intervention in either of the groups.

The power of the patients to attenuate the clinical course of their disease stands out in view of failures of educational programs previously published (1,16–18). The difference lies perhaps in the availability of the contact with the consultants throughout the study period and the reinforcement of education and motivation of the patients during the repeated annual consultation sessions.