- Click here for explanation of classification scheme.

The primary objective of this study was to compare the cost-effectiveness of a cholesterol lowering protocol implemented by RDs with that of cholesterol lowering advice given by physicians in a clinical ambulatory setting. 

1. 21-65 years of age

2. Total serum cholesterol: >5.2 mmol/L <8.84 mmol/L

1. 2° medical conditions that would influence lipid levels: (pregnancy, perimenopausal condition, diabetes, thyroid disease or renal failure).

2. Use of medications that would alter lipids: hypolipidemic meds, thiazide diuretics, beta-blockers or estrogen replacement therapy.

3. Hypertriglyceridemia: >4.52 mmol/L

4. MNT for hypercholesterolemia in the past year.

Recruitment:  Potential study candidates included 1,493 adults identified through the Massachusetts General Hospital (MGH) Clinical Laboratory Information Center.  13 primary care physicians who agreed to participate in this study enrolled 90 of their patients as study volunteers between June 1996 and April 1997.

Design

• Study volunteers were randomly assigned to receive either MNT or usual care (UC) from physicians
• MNT: cholesterol-lowering nutrition counseling using NCEP guidelines for a minimum of 2-3 visits over 2-3 months
• MNT: If serum lipids were not in target range after 2-3 months, 2-3 additional RD visits were provided over the next 3 months.
• UC: advice from physician but not RD
• After 6 months lipid lowering medications and additional MNT could be added at the discretion of the physician for the UC study group.

Blinding Used (if applicable):

Lab tests.  Nutrition interviewers were masked to treatment assignment.

Intervention (if applicable):

Randomly assigned to receive MNT from RDs or usual care from physicians using permuted block randomization.

Statistical Analysis

• Analysis of covariance controlling for baseline value of the variable and gender was used to asssess group differences in the change in dietary intake and lipid levels at 3 months, 6 months, and 1 year.
• baseline comparisons made using t tests for continuous variables and fisher's exact test for categorical variables
• within group changes were calculated using paired t tests
• Pearson product moment correlations were coputed to correlate dietary intake and clinical outcomes
• Wilcoxon test used to compare differences in patient satisfaction between groups at baseline and 6 months

 

 

Timing of Measurements:  Baseline, 6 months, 1 year, and 6 months after the end of the study intervention. 

• Body weight and height, fasting plasma lipids: 0, 3, 6 months and 1 year;
• Diet analysis: 0, 3, 6, months and 1 year using 24-hour recall
• Patient Satisfaction Survey: 0, and 6 months; subjects reported physical activity.

Dependent Variables

• Weight
• Fasting plasma lipid profiles
• Dietary intake of fat and cholesterol, based on random 24-hour recalls
• Patient Satisfaction, survey adapted from the Diabetes Quality of Life measure for the DCCT 

 Independent Variables

• Total RD time with patients
• Cost Outcomes, according to the standard charges at the Massachussetts General Hospital using Transition Systems Inc data from 1995 and changes in the consumer price index for medical care services for urban consumers betwee 1995 and 1997

Control Variables

• baseline values
• gender

 

Initial N:  90 (60 men and 30 women)

Attrition (Final N): 88.  One man from each study group dropped out of the study. There was no difference in demographic characteristics of subjects by study group at baseline.

Age, Ethnicity, Other Relevant Demographics, and Anthropometrics

	MNT Group	UC group
Age, y	49±10	49±11
BMI	27±4	28±4
Male sex, %	67	67
Caucasian, %	93	96
Income  > $40,000, %	84	80
College Graduate, %	82	67
Smoker, %	11	4

 Location:  Boston MA

Outcomes at 6 months

	MNT	Usual Care
Kcals	1679± 796	1850± 675
% fat	25± 10**	29 ±10
% sat fat	7 ±4***	10 ±4
% MUFA	9± 4**	11± 4
%PUFA	7 ±4	5 ±2
Cholesterol,  mg	165 ±132*	239 ±199
Dietary fiber, g	18 ±8	16 ±6
Serum cholesterol, mmol/l	5.77± 0.60*	6.03 ±.065
LDL, mmol/l	3.98± 0.55	4.13 ±0.60
HDL, mmol/l	1.14 ±0.36	1.09± 0.31
Trig, mmol/l	1.47 ±0.73	1.85 ±1.27
Weight, kg	77.7 ±15.4***	83.2 ±15.0
Activity, minutes per week	144 ±130	108 ±109

*P<0.05    **P<0.001     ***P<0.001

MNT counseling time:

RDs spent an average of 90 minutes (60-140 minutes)/subject or an average of 2.5+0.5 visits in the first 3 months and 30 minutes /subject (1.5+0.8 visits) in the second 3 months.

The time spent with the RD by month 3 was correlated with decrease in serum total cholesterol (r= -0.47, P=0.001) and LDL-cholesterol (r= -0.39, P=0.008) and with weight reduction (r= -0.34, P=0.02).

Cost Outcomes:

The additional costs of MNT were $217/patient to achieve a 6% decrease in cholesterol and an additional $98/patient to sustain the decrease.

The cost effectiveness of MNT was $36 per 1% decrease.

Patient Satisfaction:

At baseline there were no differences between study groups. At 6 months, the MNT group reported significantly increased satisfaction levels with clinic visits, dietitians’ vs. physicians' understanding of lifestyle and eating habits, their knowledge about cholesterol, ability to manage cholesterol levels, and eating habits than the UC group. Satisfaction scores were significantly different (P<0.001).

Dietitians can use these research results with physicians, third-party payers, and decision makers in health systems to substantiate the effectiveness of MNT in decreasing fat intake and decreasing serum cholesterol. The greater patient satisfaction that occurred with MNT provides evidence that the quality counseling process is also important.

The significant improvement in all of these outcomes provides evidence that MNT is a reasonable investment of healthcare resources and supports our recommendation that NCEP dietary guidelines should be implemented by RD as much as possible and in an ongoing manner to achieve and sustain maximum health benefits.

MNT by an RD is a reasonable investment of resources because it results in significantly better lipid, diet, activity, weight, and patient satisfaction outcomes than UC.

University/Hospital:

Massachusetts General Hospital

Very nice study with good design. The length of this study was good (1 year), however the protocol for the UC group changed for the second 6 months so the differences in LDL cholesterol were no longer significant at 1 year. Even though the differences between groups were statistically significant for some parameters (weight, LDL cholesterol, % SF in diet), the differences were quite small.