Effect of Tool Design on Hand Pain in Dental Practitioners

This is a randomized controlled trial to evaluate the effects of a new dental tool handle design on hand and arm symptoms among dentists and dental hygienists who perform dental scaling. This will be a 4 month intervention in the practitioner's workplace. The study may identify a tool handle design that reduces the risk for developing upper extremity occupational musculoskeletal disorders (e.g., carpal tunnel syndrome) among dentists and dental hygienists.

The specific aims of the project are to determine whether hand/wrist, forearm/elbow and shoulder symptoms (e.g., pain, fatigue, paresthesias) and hand function are less severe among dentists and dental hygienists who use a new dental tool in comparison to those who use a more conventional tool for performing scaling. One hundred twenty (120) subjects will be randomized to use one of 2 handle designs for a 4 month period and during this time will perform their usual dental scaling work with the new tools. The effects on outcome measures will be adjusted for potential covariates (e.g., age, gender, hours scaling per week, etc.). The primary null hypothesis is: there is no difference in change in pain severity between subjects who use a new tool in comparison to those who use a conventional tool.

RESEARCH DESIGN AND METHODS

In this study, 120 participating dentists and dental hygienists will be randomized to receive one of 2 scaling tools and will use the tools for 3 months while performing their usual dental scaling work. Randomization will be at the level of the clinic to avoid participant contamination. Participants will complete a weekly questionnaire to assess symptom severity. The data analysis will use an intention-to-treat approach and will control for important covariates.

Interventions Based on the pilot data, the investigators have selected 2 tool handle designs that are similar to conventional tools but have features that exploit the differences identified in our preliminary data. The dependent measures of pinch force and muscle activity were most sensitive to diameter, followed by weight, followed by shape. The tool characteristics selected for this proposal are based on the data from the laboratory studies presented in the Preliminary Studies section and the Appendix (Dong et al., JADA, 2006; Dong et al., Appl Ergo, 2007). The tool parameters were selected to provide the greatest difference in peak pinch force, yet be as close as possible to the look and feel of a conventional instrument. Half of the participants will be provided with Tool 1 and half with Tool 2. Tool 1 is the closest to a conventional tool in diameter and weight while Tool 2 has a larger diameter and is lighter. The differences and similarities between Tools 1 and 2 are listed below.

Tool 1 Tool 2 Color black black Material steel Dacron Surface texture medium diamond medium diamond Diameter 7.5 mm 10.8 mm Weight 20 g 12 g Shape round round

The investigators expect participants to apply less force with Tool 2 than Tool 1 and, therefore, decrease their risk for hand pain and disorders.

Subjects

Dentists and dental hygienists (N=120) will be recruited from community and private clinics in the Bay Area.

Recruitment

Dentists will be recruited at county dental society meetings and through mailings to their clinic (mailing list provided by the CDA and The Zenith). Dental hygienists will be recruited through mailings to their homes or places of work (mailing list provided by regional dental hygiene association) or at informational meetings at their regional dental hygiene meetings or at dental clinics.

At recruitment meetings, potential participants will be handed the consent form and the study. The inclusion criteria will be explained by one of the co-investigators and questions will be answered. Potential participants will complete a brief, anonymous survey to indicate whether they are eligible and whether or not they are interested in participating. If they are interested and eligible in participating, they will sign the consent form and be enrolled in the study.

Mailings will include a description of the study, the consent form and a participation postcard. The postcard will have 3 check boxes: (1) yes I am interested in participating, please call me, (2) I have more questions about the study, please call me, or (3) I am not interested in participating, please do not contact me again. Those who agree to participate will be instructed to sign the consent form and return it.

Baseline Questionnaire

Participants will be mailed a 30 minute baseline questionnaire to collect the following demographic information and covariates: recent or current symptoms of an MSD; occupational history; current job information (e.g., hours per clinic per week, hours scaling per week, glove type used); medical history including previous musculoskeletal injury or illness, pain medication usage, systemic diseases related to MSDs; tobacco use; personal demographics; and psychosocial factors. The investigators plan to use the standardized Job Content Questionnaire to measure job control, psychological demands, job strain, social support, and job satisfaction (Karasek et al., 1998; Rempel et al. 2006).

End-of-Week Questionnaire: Pre-intervention

They will also begin completing a weekly, 5 minute, end-of-week questionnaire (see appendix) which assesses work hours for the week, number of hours scaling for the week, medication use for pain, hand function, and acute injury events to the upper extremities (e.g., laceration, bruise) during the week (If an elevated pain score is due to an acute traumatic event, the symptom score for that week will be excluded from the analysis). Three body regions (1) the right hand/wrist, (2) right elbow/forearm and (3) the right shoulder (or left if the subject performs scaling with the left hand), will be assessed for the worst pain during the preceding 7 days using a 0 to 10 point scale (0=no pain; 10=unbearable pain) (Gerr et al. 2002).

Randomization and intervention

The dental office that the participant works in will be randomized to either Tool 1 or Tool 2. A computer generated random number sequence will be used. All participants at the same dental office will use the same tool to avoid contamination.

After completing the end-of-week questionnaire for four weeks the participant will receive the new set of 8 assigned dental tools and will be instructed on their use, tip replacement, tip sharpening and sterilization. If there is more than one participant at an office, the tools for each participant will be separately labeled with color tape to identify the owner.

End-of-Week Questionnaire: Post-intervention

The same end-of-week questionnaire will be administered for the next 3 months. However, the questionnaire will contain an additional question which asks what percent of time they used the new tools when scaling. This question will allow us to determine if the assigned tools are being used.

Data Analysis and Sample Size Calculations

The analysis will follow an intention-to-treat approach. For the analysis of the effect of intervention on pain in the three body regions, the primary outcome measure will be the pain score change (difference between the mean post-intervention pain level, collected during the last 4 weeks of their participation in the study, minus the mean pre-intervention pain level). Because the analysis uses an intention to treat approach, if the subject drops out of the study early, the investigators will use the pain data from the 4 weeks prior to the drop out point to estimate the mean post-intervention pain score. Therefore, the investigators collect weekly symptoms data throughout the intervention period. The subjects studied are in relatively stable occupational settings, so the investigators do not anticipate that many subjects will leave early. Ultimately, if there is a change in pain scores after the intervention, all of the weekly data will be used graphically to explore the time course of change. Missing weekly pain scores will be imputed by the mean of the scores just prior to and following the missing data.

General linear models will be used to calculate the beta-coefficient and 95 percent confidence intervals for the intervention for each of the two body regions. The mean pre-intervention pain score will be forced into all models to reduce regression to the mean. The other possible covariates (e.g., gender, age, occupation, hours per week of scaling, psychosocial factors, etc.) will each be examined, one at a time, in multivariate models which include the pre-intervention pain score. They will be included in the final models only if they are modestly (p<0.2) related to both the dependent variable (e.g., pain score change) and intervention group. The investigators expect effective randomization so it is likely that few potential covariates will be included in the final model. The potential problem with the unequal gender distribution between occupational groups will be addressed by also testing the effect of a new single variable with 4 categories (male dentists, female dentists, male hygienists, female hygienists) on the model. The analysis of the hand function data will be similar to the analysis of pain.

The sample size calculations are based on the estimated mean hand pain score of 2.9 (SD 1.6; scale of 0-10) (based on values collected from subjects in the preliminary studies). For type I error of 5%, the investigators would need a total of 82 subjects (41 each group) to have 80% power to detect a mean pain score change difference of 1.0 (33% difference in pain score change). Covariates in the final model will increase the required sample size. To account for covariates and dropouts the investigators will recruit a total of 120 subjects.

The selection of a difference of 1.0 for pain score change is based on the affects from the expected difference of 27 to 40% in pinch force. Unfortunately, the investigators do not have good dose-response data linking reduction in a risk factor level, such as pinch force, to a reduction in pain levels. If the dose-response relationship is linear, the investigators might expect a similar reduction in pain intensity. In 3 prior RCT intervention studies, although the interventions were different than in this proposal, the investigators observed such effect levels. The studies were among engineers, customer service, garment workers with study durations of 6, 12 and 4 months. The observed effects on upper extremity pain scores were 41% (1.2/2.9), 28% (0.7/2.3), 42% (1.0/2.4) (Tittranonda et al., 1999; Rempel et al., 2006; Rempel et al., 2007).