Effectiveness of Computer-Aided Removable Partial Denture Design

Effectiveness of Computer-Aided Removable Partial Denture Design T e T J Lindquist, DDS, MS,'James M. S. Clany, DDS, MS,2 Lynn A.Johnson, PhD,,.' and Frank J. Wiebelt, DDS, MS4 Purpose: Computer programs have been developed for removable partial denture (RPD) design, but their educational impact has never been investigated. This study measured the effectiveness of computer-based RPD design simulations in a traditional RPD design course for second-year dental students. Materials and Methods: Thirty-six students were randomly assigned to four groups, t w o simulation and two control groups, at the beginning of the RPD design module. A cross-over design compared the t w o strategies during laboratory exercises. First, the simulation group worked with the simulation, while the control group met in small-group seminars with faculty. Post-test 1was given, and then students switched teaching methods and post-test 2 was given. Each post-test had t w o parts. Part 1 involved designing an RPD based on specific patient criteria. Part 2 involved choosing the more correct of the t w o designs. Results: Results showed a significant difference ( p = .0072) in t w o of four test groups. Although it was an initial evaluation, the computer simulation was shown to be equivalent t o faculty-led small-group seminars. Students also responded positively t o a questionnaire on their perceived effectiveness of the simulations. Conclusions: The results suggest that RPD design can be taught as effectively with computerbased simulations as with faculty-led seminars. J Prosthod 1997;6:122-127. Copyright 0 1997 by The American College of Prosthodontists. INDEX WORDS: computer simulation, dental education

R

EMOVABLE PARTIAL DENTURE (RPD) de-

sign is an area of prosthodontic education that has been traditionally difficult for students to learn quickly and effectively. It usually requires a significant amount of one-on-one interaction between instructors and students before an acceptable level of competence is achieved. Kecently, several computer software programs have been developed for various aspects of RPD design. However, none have been formally tested as a potential teaching tool in a predoctoral setting. This article will review an experimental software program, and describe how its utilization in the preclinical removable prosthodontic

'-'ksi.stant Pmjssor, Department gPro.cthodontics, UniiwsiQ of Iowa College $Dentistpi, i o w a City, L4. 2Asson'ate Prgessor, Departmpnt 0J'Pro.cthodontics:C h iimig of Iowa College ofDentistly, Ioua Cig,,!A. 3Rrsistant Professor. Department of Oval Patholog). Radiolou, and Medacine, L'nicersity !flown College ofDentistg1, Iowa Cicv,M. 4ilssocialeYrofi.sro~and Chairman. Department ofRemoiable Pmsthodontics, College of'llentlrt,?: Linioersi& ofOklahoma, Okluhoniu CiQ, OK. AcceptedMarch 4, 1997. Carre@&nrx tz- T PJ. ~Lindpisf, DDS, MS, Department !/" Pra.ct/iodoniicc, Uaiorniy .JIowa, Collqge of DentiJtgi, Iuzoa CI&, L4 52240. Cofiright 0 1997 The American Collqe Ospusthodontirts 1059-941'y/97/0602-000~~~.00l0

122

course at the University of Iowa compared with the traditional method of teaching RPI) design.

Background Computer-aided simulations have bccn used in dental education for over a decade. At the University of Iowa, computer-aided patient simulations have been incorporated into the student curriculum in the Department of Oral Diagnosis, Radiology,and Pathology for several years. Summative evaluations of these simulations have proven them to be effective in teaching clinical problem-solving skills in a costeffective manner.' Several software packages have been developed for various aspects of RPU design. A Macintoshbased interactive program was devcloped by Beaumont in 1989 that facilitated design and writing work a~thorizations.2;~ It utilized a database of 160 W D drawings that could be modified to reflect different clinical situations, and provide a more accurate, standardized authorization form than one drawn by hand. In 1990, Lefebvre presented an interactive RPD design program developed for the Apple IIGS personal computer? This program consisted of two modules: a faculty module used to create design

Journal ofProsthodontics, Val 6,No 2 (June), 1997:pp 122-127

123

June 1,997, Volume 6iVurnber 2

exercises, and a studcnt module that included a tutorial and multiple design exercises. However, implementation of the program into the student curriculum was not fully realized becausc of the relocation of key personnel in\~olvedwith the project. An IBhWindows-based interactive RPD design program developed by Davenport et al“ has been used in several British dental schools, and for continuing education for gencral practitioners. It is a tutorial based on key chapters from a textbook on RPD design and focuses on task-based learning. Thc software is available commercially for a variety of applica tions. Another IBM-based design program has been developed by Wicks and Penncl16with support from TEREC (a technical research organization of large commercial laboratories in the United States, Canada, and Great Britain). The program requires the user to enter missing teeth and then asks the user questions about the health of potential abutments. Retentive assemblies, indirect retainers, and major connectors arc suggested by the program. There is also a “users choice” leature that prompts the user to enter his or her design philosophy to customize the program’s suggested parameters. The softwarc for this program is commercially available through a private company. An experimental soltware package for facilitating and teaching FU’D design has been developed by Weibelt of the University of Oklahoma, College of Dcntistry. Thc authors of this article obtained the prototype software under special permission [or research. after review and familiarization of the pro-

Figure

1. ‘The program matches thr iooth configuration with its database and provides d suggestrd design.

gram, a study was designed to investigate the effectiveness of using the prograni for teaching the section on RPD design to second-)ear dental students in thc preclinical removable prosthodontic course.

Software Description The software evaluated is a Macintosh-based program written with “Hypercard.” There are actually two different programs: the Atlas and the Clasp Selection hfodulcs. The Atlas is a database of many RF’D dcsigns that are also listed in an KPD text.’ Thc user enters the missing teeth using a conventional point and click interface, and the program matches the tooth configuration with its database and provides a suggested design (Fig 1). The program also lists references in the RF’D text. The Clasp Selection program is divided into molar, premolar, and canine modules. Each utilizes an interface with multiple parameters that the user may modify to accurately describe the characteristics of the intended abutment tooth (Fig 2). Once all the information has been entered, the program will suggest neccssary tooth modifications and an appropriate direct retainer (Fig 2). The user may also customize the criteria and priorities for clasp selcction to fit the philosophy of the user. All graphics &splayed on the screen may be printed if desired. The authors of this article invcstigated the potential of incorporating these programs into a preclinical RE’D course. The goal of the studywas to evaluate the

124

Computrr-Aided Remooable Dentures

Mesial B/L T i l t

M/DTilt

Opposing Occlusion

C/R Ratio

Survey Lines

Lzndquzst el a l

Distal

Mobility

Crown Length

5HOk-r

Kennedy Classification

u

C u r rent Required Restorations Restorations

!,.a + ,. . . ..

DULYIIDH

8

:....

.

NOHt

ii

+, MDHC

Preferred Clasp Type

'i

Figure 2. (Top) Parameters that the use1 may modify to describe the chaiacteristics of thc abutinciit tooth. (Bottom) Kccommendcd tooth modifications and direct retaincr based on the set

advantages of students practicing KPD design exercises without supervision. In addition, the program could provide a very structured approach to teaching all the relevant factors in clasp selection. It also had the potential to be used as a powerful tool for student evaluation.

Methods Second-year dental students enrolled in the preclini-

cd removable prosthodontic course were given a demonstration ofthe W D design program's capabilities. Members of the class wcrc givcn thc choice to participate in the study if they wished. There was no penalty for nonparticipa.tion, and test scores were not

applied to their final grade. Most of the 72 students in the class chose to be considered for participation in the study. The class ranking of the students that chose to participate was used by the college administration to make group assignments. A group of 36 that attempted to reflect a homogeneous sample were assigned to the study. These students were then randomlv assigned to one of four ,groups .~. . utilizing a stratified random assiLpment-two simulation groups that used the computer software, and two control groups that met in small groups with faculty. A cross-ovcr design compared the two strategies during the laboratory exercises (Table 1). The remainder of the class was assigned the same cxcrcises, but only

Table 1. Cross-Over Design Used to Compare the Two Strategies During the Laboratory hcrcises Session 1 Case 1 Case 2 Session I1 Case 3 Case 4 Post-test 1

Conventional strategy All faculty All faculty Convcntioiial strategy All faculty All faculty All Students takc Post-test 1

Computer strategy Faculty A Faculty R Computer stratqgy Faculty A Faculty B

Seminar strategy Faculty B Faculty A Seminar strategy Faculty B Faculty A

Session III Case 5 Case 6 Session IV Case 7 Case 8

Coiivcntional strategy All faculty All faculty Conventional strategy All faculty AIIrac:uity All Students take Post-test 2

Seminar strategy Faculty A Faculty B Seminar strategy Faculty A Faculty B

Computer strategy Faculty B Faculty A Computcr strategy Faculty B Faculty A

Post-test 2 Note. Nonparticipants:

11=

36; study participants: n

= 36; group I: n =

18; group 11: n = 18.

125

June 1997, Volume 6, Number 2

met with faculty as needed to receive feedback on their designs. Four lab periods were allotted for the exercises. During this timc, all students were given eight partially edentulous casts and were required to survey them, design a RPD, draw the design on the cast, and write an appropriate work authorization for each. Initially, the simulation groups worked with the computer simulations on specific cases with a faculty monitor present to assist with any questions regarding the software. The control groups met for an identical period of time and worked one-on-one with a faculty member on the same cases. The faculty switched strategies after each session. Halfway through the module, a post-test was given to all students. Next, all thc students in the study switched teaching methods for the second half of the exercises. Asecond post-test was given to all students at the end ofthe exercises. Each post-test had two parts. Part 1 required designing an appropriate RPD for a patient with specific criteria. Part 2 involved choosing the more correct of two RPD designs (Fig 3). Results of the post-tests were recorded and subjected to statistical analysis. A reliability measure (KR-21 = .43) was determined by giving the post-test to third-year dcntal students (N = 16) 2 months before the start of thc study. The SAS statistical program on the mainframe computer at the University of Iowa was utilized for the analysis of the study results.

Results A total of 15 points were possible on Part 1 of each test. A total of 10 points were possible on Part 2 of each test. The mean and standard deviation of each group were calculated for each part. Tests 1 and 2 were evaluated separately. Parts 1 and 2 of each test were also examined separatcly for each group. Student’s t tests were used for data comparison. Two of

W

Figure 3. Part 2 of thc exam: choosing thc more correct of thc two RPD dcsigns.

the four groups (one small group and one computer group) showed a significant difference in part 2 of Test 1 (p = .0059) (Table 2). There was no significant difference for any of the groups at Tcst 2. These results suggest that the computer simulation, as a teaching aid, was equivalent to small group discussions in the RPD learning process. Additional t tests were completed comparing the scores of the participating students and nonparticipating students in the class. The comprehensive scores for each tcst were evaluated. Test 1 showed a significant difference between the nonparticipating students and study participants (p = .0276). Test 2 showed a difference betweent the nonparticipating students and study participants, but the difference was not significant (p = .2267) (Table 3). The number of subjects who took each dropped from Test 1 to Test 2. This was due to absenteeism and a breakdown in the group-identification coding process in Test 2. A participant questionnaire was given to the students after Test 2. In general, the students responded positively to the questionnaire on their perceived effectiveness of the simulations. The written comments implied a wide range of feeling for the simulation program. The biggest drawback with the simulations was felt to be a lack of feedback, with comments such as: “the computer enhances skills, doesn’t teach them” and “I didn’t learn concepts from simulations.” However, other students noticed assets such as “organized thought processes involved,” “good reinforcement for doing the exercises yourself,” and “could play with variables.”

Discussion During the course of this investigation, several unanticipated observations were made by the students and/or the faculty. First, once the students had become competent with the basics of RPD design, they found the computer program somewhat cumbersome. They felt that setting the parameters for each individual abutment that was to receive a clasp was very time-consuming. It became evident that, during the design process, the more experienced students integrated many factors simultaneously. However, the same students felt that, initially, setting all the parameters was a good learning process. Another observation was that the students seem to enjoy learning RPD design the most in a struc-

126

Comfluter-Aided Removable Dentures

Lindquist et a1

Table 2. Results From Tests 1 and 2 Student Group

Method

Testl Questzons

Sesszons

Mean

T

SD

IX

€‘rob

> IT

1

~

1 2 1 2 1

2 1 2 3 4 3 4 3 4 3 4

Computer Small group Computer Small group Small group Computcr Small group Computer Small group Computer Small group Computcr Computer Small group Computer Small group

*Shows a significant difference @

1 and 2 1 and 2 1 and 2 1 and 2 3 and 4 3 and 4 3 and 4 3 and 4 1 and 2 1 and 2 1 and 2 1 and 2 3 and 4 3 and 4 3 and 4 3 and 4

1Part 1, 1-3 l/Part 1, 1-3 Wart 2, 1-10 1Part 2, 1-10 2Part 1, 1-3 2Part 1, 1-3 2/Part 2, 1-10 2Part 2, 1-10 Wart 1, 1-3 Wart 1, 1-3 l/Part 2, 1-10 l/Part 2, 1-10 2/Part 1, 1-3 2Part 1, 1-3 2Part 2, 1-10 2/Part 2, 1-10

10.22 10.33 4.78 7.67 10.43 8.5 7.14 6.5 9.89 10.25 6.56 5.50 9.14 7.17 7.28 7.67

2.33 3.16 2.33 1.41 3.46 1.87 1.57 1.05 3.37 1.98 2.65 3.07 3.02 1.94 2.06 1.03

-0.0848

16

,9335

-3.1764

16

.0059*

1.23

11.0

.2490

0.8494

11.0

,4138

15.0

,7950

0.7610

15.0

.4584

1.37

11.0

.1973

-0.4095

11.0

,6901

-0.2645

< .05).

tured setting. Both the simulation groups and the control groups were required to focus on specific cases at specific times during the lab periods. The other half of the class was required to finish all the cases by the end of the four sessions, but was not required to meet in small groups and concentrate on a particular case. Feedback from the students was much more positive from the half of the class that was more structured within the study. The students who used the W D software had several suggestions for the improvement of the program. They felt it would be helpful if the Atlas section and thc Clasp Selection modules could be integrated together. They also felt that time would be saved if it were possible to enter some parameters for multiple abutments in the Clasp Selection modules. For example, if the “patient” had all abutments tipped lingually, it would save time if this could be entered once rather than separately for each abutment. The obvious advantages of computer-assisted teaching include the reduction of the number of faculty necessary to teach the course, the flexibility to

allow students to progress at their own pace, and thc potential for totally objective testing and evaluations. The main disadvantage for using the computer for teaching RPD design is the cost of providing enough hardware and support for a full class of students. However, many schools are discussing the idea of requiring incoming students to provide their own laptop computer, which would alleviate much of that cost.

Conclusions Although this study can only be considercd an initial evaluation, the results suggest that computer simulations can be used successfully for teaching W D design. Students felt that the computerized RPD design exercises were good for initial learning, but future programs should have the capability to progressively combine parameters as students learn to integrate the multiple tasks required for RPD design. A side-effect of this study was the discovcry that, by

Table 3. Test 1 and 2 t Test Results Comparing Study Participants and Nonparticipants Test

Group

N

Mean

SD

7‘

DF

Prob > IT1

1

Nonparticipants Study participants Nonparticipants Study participants

21 35 24 26

12.2857 14.4857 12.9166 14.0769

4.0638 3.1564 3.626 3.0714

-2.2644

54.0

.0276*

48.0

,2267

2

*Shows a significant difference 0, < .05).

-

1.2247

JunP 1997. Volume 6; iVumber 2

developing a sn1all-group, structured teaching strategy, instruction could also be improved.

References I . Johnson 1A: Tools lor technology assisted learning: Teaching problem solving skills uith patient simnlations, in Dcntal Irrforrnatics: Integrating Technoloby into the Dental Emironment (ed 1). Ncw York, -W,Springcr-Vrrlag, I992

127

2. Beaumont .4J, Bianco HJ: Microcomputer-aided removable partial denture design.J Prosthet Dent 1989;62:417-421 3 . Beaumont AJ: Microcomputer-aidsd rrtriuvablc partial derrtur-e design: the next evolution.J Prusthet k i t 1989;62:551-556 4.Lefebvre C4, Kicliardsu~iGB. Taylor RI,: Computer-assistcd reniovatile partial denture design.J Dent Educ I990;53:36 5. I Iammond P, Davenport JC, Potts AJ:Knowledge-based design uf Ierriovnble partial dentures using direct manipulation and critiquing. J Oral Rchahil 199:3;20:135-123

6. Ltiic'icks =,,, Pennell hE: A computer assisted design guide for removable parrial drnture frameworks. Dent Prac 199U;ti:51-53