For Cardiovascular Health, Tackle All the Bad Habits Together

HOUSTON -- All-at-once counseling proved marginally better than the one-by-one approach when helping patients change multiple behavioral risk factors for cardiovascular disease.

HOUSTON, June 11 -- All-at-once counseling proved marginally better than the one-by-one approach when helping patients change multiple behavioral risk factors for cardiovascular disease.

This study found that addressing multiple behaviors one-at-a-time is not superior to, and may be inferior to, a simultaneous approach.

Although the differences were small, said David J. Hyman, M.D., M.P.H., of Baylor College of Medicine here, and colleagues, the results did not support their initial hypothesis that behavioral changes for those given sequential counseling would be cumulative and thus result in greater adoption of multiple risk-reduction behaviors.

The randomized trial in a publicly funded primary-care setting, reported in the June 11 issue of the Archives of Internal Medicine, tested sequential versus simultaneous counseling among 289 African-American patients (67.3% female; ages 45 to 64). All had hypertension, were smokers, and were initially nonadherent to the study's goals.

The study had three behavioral goals: stop smoking, reduce dietary sodium levels to less than 100 mEq/L per day, and increase physical activity by at least 10,000 pedometer steps a week.

Participants were assigned to one of three groups. Those in the first group had an in-clinic counseling session every six months that addressed all three behaviors. They also received seven regularly-spaced telephone counseling sessions over the course of the 18 months.

A second group followed a similar protocol that addressed a new behavior every six months, while a third group (usual care) was given a one-time referral to existing group classes.

At 18 months, of 230 patients (79.6%) who completed the study, less than 10% of the participants in each group met the primary endpoint of changing at least two of the three behaviors. Only 6.5% in the arm that addressed the behaviors simultaneously, 5.2% in the one-at-a-time arm, and 6.5% in the usual-care arm met the primary end point.

However, results for single behavioral goals consistently favored the simultaneous group, particularly in reducing dietary sodium intake, the researchers said.

At six months, 29.6% of those in the simultaneous arm had reached the urine sodium goal, 16.5% in the sequential arm, and 13.4% in the usual-care arm (P=0.01).

At 18 months, smoking abstinence was highest in the simultaneous group and lowest in the usual care group: 20.3% in the simultaneous, 16.9% in the sequential, and 10.1% in the usual care arms were urine cotinine negative, (P=0.08 for trend).

A pedometer-measured increase in baseline level of physical activity followed a similar pattern favoring the simultaneous arm, the researchers said.

However, they added, none of the between-group differences reached the 20% level for which the study was powered.

Although objective measures of participants' behavior rather than self-report were used in this study, there were limitations in these measures, the researchers said.

These include larger variance in pedometer reading than expected, possibly caused by the inaccuracy of standard spring-loaded pedometers. Others have also noted concerns about the reliability of 24-hour urine sodium measurements, but, the researchers noted, the method is still better than self-report.

The smokers eligible for this study may have been a group for whom behavioral change was particularly difficult, they said, acknowledging that it has been noted that almost two-thirds of smokers with hypertension had already quit.

The results of this study suggest that counseling patients to change multiple behaviors is difficult, the researchers said, but that simultaneous counseling is more effective at achieving a change in a least one high-risk behavior compared with addressing the behaviors individually.

Further research is needed to identify the specific target behaviors, demographic groups, and design features (for instance, investigator-assigned versus participant-chosen order of interventions) in which simultaneous interventions are more effective than sequential ones, the researchers concluded.