Abstract
Background: The provision of reinforcements or boosters to interventions is seen as a logical approach to enhancing or maintaining desired behavior. Empirical studies, however, have not confirmed the effectiveness of boosters nor assessed the optimum number of boosters or the timing for their delivery.
Objectives: This randomized controlled trial contrasted the effect of four booster conditions (a) 30 days; (b) 90 days; (c) 30 and at 90 days; and (d) no boosters of the intervention to increase the use of hearing protection devices (HPDs).
Methods: A total of 1,325 factory workers completed a computerized questionnaire and were randomly assigned to one of three computer-based (tailored, nontailored predictor-based, or control) multimedia interventions designed to increase the use of hearing protection devices. After the intervention, colorful boosters specific to the type of training received were mailed to workers' homes. Posttest measures of use were administered at the time of their next annual audiogram 6 to 18 months after the intervention.
Results: Repeated measures of analysis of variance (ANOVA) showed a significant main effect for the booster (after 30 days) in the group that received tailored training (F [3,442] = 2.722; p = .04). However, in the assessment of the interaction between time (pretest and posttest) and boosters (four groups), the ANOVA did not find significant differences in hearing protection device use for any of the training groups. To assess for significant differences between groups, post hoc comparisons were conducted at the pretest and posttest for the total sample and for the subsample of workers who reported using hearing protection devices less than 100% of the time needed. Sheffe contrasts by intervention group, gender, ethnicity, and hearing ability found no significant changes in the mean use of hearing protection devices for the booster groups.
Conclusions: Although the provision of boosters represented a considerable commitment of resources, their use was not effective in this study. However, it would be premature to eliminate boosters of interventions. Further study is needed to explore the effects of different booster types for increasing the use of hearing protection devices, and to assess carefully the effects of boosters on other health behaviors in studies with controlled designs.
Achieving positive health behaviors can reduce illness and premature death. Although nursing interventions may be effective in influencing behavior change in the short term, maintenance of behavior change is more challenging. Intuitively, the provision of reinforcements or boosters to interventions is seen as a logical approach for enhancing or maintaining the desired behavior.
Since the late 1960s, boosters have been studied as a method of more effectively influencing health behavior (Stuart, 1967). Boosters have been studied in the context of widely ranging health behaviors including weight loss (Craighead, Stunkard, & O'Brien, 1981;Hall, Hall, Borden, & Hanson,1975;Perri, Shapiro, Ludwig, Twentyman, & McAdoo, 1984;Richmond, 1996), smoking cessation (Jason, Salina, McMahon, Hedeker, & Stockton, 1997;Minneker-Hugel, Unland, & Buchkremer, 1992;Pokorski, Chen, Pigg, & Dorman, 1995;Prochaska, DiClemente, Velicer, & Rossi, 1993), adult assertiveness training (Baggs & Spence, 1990), breast self-examination (Strickland et al., 1997), and arthritis self-management (Lorig & Holman, 1989). The nature of the initial treatment in these studies varied, and each compared the results of a booster treatment with a nonbooster condition.
Although numerous studies have been conducted to determine the usefulness of boosters, their results have not been consistent. Stuart's (1967) initial results encouraged the use of boosters to maintain weight loss, but subsequent studies using boosters have been discouraging. In comparing the frequency and content of boosters, multiple studies found no significant difference in effect between groups that received boosters for the maintenance of weight loss and those that did not (Ashby & Wilson, 1977;Craighead et al., 1981;Hall et al., 1975). However, in a study comparing the effects of relapse prevention training and boosters, Perri et al. (1984) found that the effects of boosters were mixed. The group receiving behavioral treatment and postrelapse prevention training maintained weight loss better than comparison groups, whereas other booster groups showed no difference. In a study comparing the results of group-based booster sessions, Kingsley and Wilson (1977) found that although the results of booster use were encouraging at 3 and 6 months, these effects were not retained at 12 months.
The results also have been mixed for studies that used boosters as an approach to smoking cessation. Two previous studies (Minneker-Hugal et al., 1992;Pokorski et al., 1995) found that boosters included in smoking cessation programs showed less favorable results than those of alternate intervention or control groups. More encouraging effects were found by Prochaska et al. (1993), whose findings showed that smoking cessation programs containing boosters (mailed progress reports or personal telephone calls) were more effective than programs without boosters. In a study of worksite-based smoking cession programs, Jason et al. (1997) found that participants active in 14 support group meetings over 6 months were more successful than participants without such follow-up support.
The effects of boosters on other target behaviors have been more encouraging. A program to improve assertiveness among adults conducted by Baggs and Spence (1990) showed enhanced effects of therapy among members of a group that received 6 months of booster sessions. Similar favorable results were found in a study by Strickland et al. (1997) that compared the effects of a follow-up telephone message and post card on rates of breast self-examination. Likewise, Lorig and Holman (1989) found that participants of an arthritis self-management class that received four biweekly editions of a newsletter had lower depression scale scores than participants who had not received such a newsletter.
In the studies reviewed, content, mode of delivery, and frequency of booster sessions varied considerably. However, none of these previous studies provided a theoretical basis or other rationale for the selection of booster sessions. Overall, these reports show that the effects of boosters, when used as part of behavioral intervention studies, were inconsistent. According to the current literature available, it is unknown which booster intervention components (frequency, content, duration, or medium) and for which behaviors booster interventions may be most effective.
Worksite intervention programs to promote health and reduce illness and injury costs have shown positive effects (Lusk, 1999;Pelletier, 1993;Pelletier, 1996). Most of these studies, however, measured the effects of a one-time intervention after a relatively short time and did not test the effects of boosters. This article describes a project designed to test the effect of the number (one or two) and timing (30 and/or 90 days after the intervention) of boosters on factory workers' use of hearing protection devices (HPDs). Although no other studies in the literature examined the effect of boosters with regard to the use of HPDs, the aforementioned studies were used in decisions regarding the design of this project.