Physician-delivered nutrition intervention: Recent efficacy studies

Several recent studies have been designed specifically to evaluate differing nonacademic practice models for the detection, evaluation, and treatment of elevated blood cholesterol levels. (30) To date, these have reported only modest success. Caggiula and colleagues evaluated office-assisted (education of physicians about the NCEP guidelines and the provision of materials) and nutrition referral center interventions in a selected group of interested physician practices.(31) As compared to a historical control group, at a short follow-up time of 60 days the office-assisted model had a mean serum cholesterol reduction of 6.7 mg/dl in non-medication users. The nutrition referral center model in which patients were referred for individual and group counseling had a greater change ( 15.6 mg/dl) but 42% of pts in this condition did not accept referral and were not included in the analysis. No dietary change data were reported. Beresford and colleagues evaluated dietary change in 28 primary care physician practices utilizing physician delivery of a 3-minute “motivational message” and a self-help booklet. At 12 months of follow-up there was a significant 1.2% decrease in reported intake of fat (% of energy). Saturated fat intake was not reported. There were no significant changes in either BMI or plasma cholesterol levels. (32) Our own WATCH study of physician-delivered pt-centered counseling plus an office support system is the only reported large, primary care-based study which thus far has demonstrated significant dietary, LDL-C and weight changes (see Section C.1.). (33)

Adjuncts to physician-delivered nutrition intervention

A team approach to lipid management and dietary intervention can make efficient use of physician time, and can enhance the physician’s effect on behavioral change. Such enhancement of effects has been demonstrated in smoking (34) and in diabetes control. (35) Both nutritionist-facilitated group dietary interventions and telephone counseling have demonstrated promise. In each, there is a need for close collaboration between physician and dietitian, and for a structured program that provides direction and allows evaluation of the results. (36)

Group intervention. Because they are less expensive to deliver than one-on-one interventions, group interventions have the potential to be very cost effective. Intervention groups are characterized by specific properties which facilitate change, including didactic instruction, instillation of hope, universality, altruism, interpersonal learning, group cohesiveness, and catharsis.(37) There is therefore a theoretical basis for believing that group interventions may be more effective than one-on-one counseling, especially when dealing with eating and food preparation behaviors, which are social activities. We are aware of no study in which such a comparison was investigated. There are no directly applicable studies available on the efficacy of group interventions to improve diet or reduce serum cholesterol levels in a primary care setting.

However, there are several studies, which, while they either were not conducted in a primary care setting (38-42) or were not primary prevention, (16) do provide support for the efficacy of group nutrition intervention. A large randomized clinical trial, the Multiple Risk Factor Intervention Trial (MRFIT), employed both group (10 sessions) and individual counseling sessions in the special intervention (SI) condition as the format for teaching subjects (all males) and their spouses a low-fat, low-cholesterol eating pattern. At the end of the first year of intervention, about one-half of the study subjects were following the eating patterns at a good or excellent level (trained nutritionist assessment). (38) The MRFIT SI condition experienced an average cholesterol reduction of 7.5%, maintained for over 6 years. (39, 40) The MRFIT group intervention was complex, since it also dealt with smoking and hypertension. A worksite study also demonstrated a significant reduction of cholesterol in the Intervention condition provided an 8-week group intervention program compared to the Control condition. (41)

The Women’s Health Trial (the feasibility study for the Women’s Health Initiative (WHI)) - demonstrated that disease-free women attended group meetings and adhered to a low-fat diet for as long as two years. (42) In our own WHI program we have experienced excellent attendance at the groups (see C.2.5.). The landmark secondary intervention study of Ornish and colleagues demonstrated striking changes in dietary fat and LDL-C levels, but was a very intensive intervention and had a small sample size. (16)

Based on the available data and findings from preliminary studies, we conclude that the challenges of group interventions to effect reductions in serum cholesterol center around: 1. increasing the rate of referral to groups; 2. increasing the acceptance and attendance rates among those referred; and 3. improving the efficacy of the intervention with respect to lowering dietary SFA and serum LDL-C levels. Methods for achieving these aims are described in section D.

Telephone counseling (TC).

TC has been used for recruitment, delivery of interventions, and enhancement of adherence to diagnostic tests and treatment interventions. TC can be an effective mechanism for simultaneously addressing the interplay of educational, psychosocial and practical barriers to adherence (43).

Little information is available on the use of TC as part of nutrition interventions. In a secondary prevention study carried out by DeBusk and colleagues, TC by nurses was effective in reducing SFA intake and serum LDL-C at 1-year follow-up, but the diet change was no greater than in the control condition, and the LDL reduction was primarily attributable to the far greater use of lipid-lowering medication in the intervention condition. (44).

In a recent study seeking to improve the dietary self-care of diabetics, (35) pts received TC calls 1 and 3 weeks following their physician visit in combination with immediate computer-generated feedback, a 20-minute meeting with an intervention staff member, a copy of a mutually developed goal-setting/strategy worksheet, a self-help pamphlet with sections relevant to their goal highlighted, and videos aimed at enhancing self-efficacy. This brief intervention resulted in significant differences in cholesterol levels at 3 mo. follow-up.

There is substantial information on TC for smoking intervention. We demonstrated the efficacy of a TC protocol when used to counsel smokers with CHD in a randomized clinical trial. (45, 46) In a study of a physician-delivered smoking cessation intervention in a primary care setting, there was a trend towards a significant effect for TC. (47) Orleans and colleagues (48) found that TC intervention increased the use of self-help materials, and yielded significantly higher short- and long-term cessation rates. Likewise, in a study by Lando and colleagues (49) the use of 2 TC calls (averaging <15 min. each) led to significant differences in validated six-month cessation rates.

A recent meta-analysis of TC for smoking cessation also supports the effectiveness of counseling calls both at short- and long-term follow-up. (50) In one cited study there was a dose-response effect, with six calls significantly more effective than one call, which in turn was more effective than written materials. TC calls have been well-received in a number of studies (51, 52) including our own, where only 10% of cardiac patients who smoked refused TC. (45, 47).

Since TC calls have demonstrated efficacy in smoking with some support for their use in diet, we have elected to use TC as a major adjunct for physician-delivered nutrition intervention. TC will incorporate our pt-centered counseling model which enhances pts’ adherence and motivation to change and the likelihood that the pt’s goals are realistic (see Section 8.4.1.). (53) The primary objective of TC will be to enhance motivation, help the pt develop and adhere to realistic goals and a plan for change, (54) and facilitate the mobilization of a coping response when needed. Relapse prevention steps will be negotiated and practiced during the calls. Such relapse prevention strategies have shown effectiveness. (55)