Pilot study – data extraction
form

1Introduction

The studies found in the rapid scoping exercise presented the outcome, or change in treatment plan, in different ways. A common measure is an increase or decrease in length or width of a planned dental implant. Nevertheless, a difficulty arises from the type of implant system used. Some studies measured the change in implant length or width. The difficulty arises because implants change dimensions in steps and this is dependent on the type of implant system used. For example, in one study, Brӓnemark system implants are available in widths of 3.3mm, 3.75mm, 4mm and 5mm. In another, Straumann implants are available in widths of 4.1mm and 4.8mm. Other studies record change in treatment decisions about bone grafting or sinus augmentation surgery. It was, therefore, necessary to design a data extraction form that captured all variations and commonality of outcome between the studies so that analysis could be planned. In addition, other characteristics of the studies had to be recorded to enable synthesis. The intention was that further referral to original publications would be kept to a minimum once data extraction had been carried out.

2Materials and methods

The starting point was the data extraction form used by Albon et al in their similar neuroimaging review. This is reproduced in Appendix A. Preliminarywording changes were made appropriate to the current review. The form was labelled “Data Extraction Form v1” and was piloted on the five rapid scope studies as follows:

Diniz 2008 4, Frei 2004 5, Reddy 1994 6, Schropp 2001 7and Schropp 2011 8

The separate section on subgroup analyses was not thought to be necessary since it largely reproduced data which was captured in other parts of the form.

Data Extraction Form v1 is reproduced in Appendix B

3Results

Several problems were identified with Data Extraction Form v1:

Only one of the studies in the rapid scoping exercise used true three dimensional imaging. In this case, medical computed tomography. Other studies used conventional spiral tomography. Whilst this produces a cross sectional image, it may be considered not to be true three dimensional imaging. Therefore the wording was changed from “3D imaging” to “X sectional or 3D imaging”

The question on age was very specific, asking for the mean, range and standard deviation. Not all of this information was available in all the studies. The question was therefore amended to request whatever information was provided by authors.

The inclusion and exclusion criteria reported in the studies were often extensive, requiring more space than available on the form. This answer box was therefore expanded.

The number of patients in the study was asked twice. One of these questions was removed to prevent duplication.

The lengths and widths of implants available for selection was important for the reasons discussed in the introduction. This was not recorded in the form and so was added.

The question on bone augmentation could cover both the addition of bone and, for example, sinus augmentation surgery. One of the studies made a distinction between bone grafting and other surgical procedures. This was, therefore, amended to ask two separate questions. The first concerned bone grafting and the second “other surgical procedures” which would include sinus augmentation.

Where a question was not addressed in a study it was convenient to write “NS” rather than “Not stated”. In the revised version, therefore, the reviewer was prompted to do this.

In the rapid scoping studies, the reason given for undertaking imaging was always to assess available bone prior to implant placement in the partly or fully edentulous patient. Therefore, it was expedient to answer the question about the reason for scanning only if this was different. The revised form advises accordingly. The question about the “Presenting problem” was also removed as it was a duplication of the same information.

The question about clinical information was changed so that it was only answered if no clinical examination had been carried out.

It was also considered that the section on outcomes should be expanded to capture all variations of outcome measure. It was also amended to record more global, dichotomous outcome measures such as whether implant size changed or not. This was an attempt to capture data that may be common to the largest number of studies.

Finally, a version of the form was produced in Microsoft Excel so that data could be more expediently copied, manipulated and analysed. The printed version of the form is presented in Appendix C.

4Conclusion and discussion

The data extraction form does now seem to capture the required data. This should now be trialled with a small number of co-reviewers to assess its clarity, ease of use and effectiveness.

5Post script

Following discussion, a further refinement was made to the data extraction form to include a question about wash out periods. In other words, the time between the “Before” and “After “ parts of the study. The full form in its Microsoft Excel form is shown in Appendix D.

Appendix A – Data extraction form of Albon et al.

Appendix B Data Extraction Formv1

(Adapted from Albon et al.)

Study details

Author; year, trial name
Country(ies) and years of recruitment
Study design
Area of the mouth studied
Reason for scanning given
Conventional imaging technique used
3D imaging technique used
Setting (practice, hospital etc)
Comments:

Patient characteristics

Population
Number of patients
Age (years) Mean (SD) [range]
Sex Proportion male (%)
Presenting problem
Inclusion/exclusion criteria
Comments:

Evaluator characteristics

Number of evaluators
Types of evaluator (surgeon, radiologist etc)
Comments:

Interventions

Number of patients
Number of exclusions
Implant manufacturer(s)
Number of implants placed
How many anterior mandible?
How many posterior mandible?
How many anterior maxilla?
How many posterior maxilla?
Did evaluators carry out a clinical examination?
Was other clinical information available to evaluators (study casts, clinical findings etc)
Comments:

Outcomes for analysis

What were the outcome measures?
In how many sites was a longer implant chosen after 3D imaging available?
In how many sites was a shorter implant chosen after 3D imaging available?
In how many sites was a wider implant chosen after 3D imaging available?
In how many sites was a narrower implant chosen after 3D imaging available?
In how many sites was bone augmentation prescribed after 3D imaging but not after conventional imaging?
In how many sites was bone augmentation prescribed after conventional imaging but not after 3D imaging?
Comments:

Definitions:

Implant siteThe position of a single implant

Bone augmentationSinus augmentation surgery or bone grafting of any type

Appendix C Data Extraction Formv2

(Adapted from Albon et al.)

Where the required information is not stated in the study report write NS

Study details

Author; year, trial name
Country(ies) and years of recruitment
Study design
Area of the mouth studied
Conventional imaging technique used
X sectional or 3D imaging technique used
Setting (practice, hospital etc)
Comments:

Patient characteristics

Population
Number of patients in study
Age (provide all information from study eg range, SD etc)
Gender – state percentage male (%)
State presenting problem if not partial or complete edentulism
Inclusion/exclusion criteria
Comments:

Evaluator characteristics

Number of evaluators
Types of evaluator (surgeon, radiologist etc)
Comments:

Interventions

Number of excluded patients
Number of patients after exclusions
Implant manufacturer(s)
Lengths and widths of implants available for selection
Number of implants placed
How many anterior mandible?
How many posterior mandible?
How many anterior maxilla?
How many posterior maxilla?
Did evaluators carry out a clinical examination?
If no clinical examination carried out, was other clinical information available to evaluators (eg. study casts, clinical findings etc)
Comments:

Outcomes for analysis

What were the outcome measures?
(complete the sections which apply below)

If outcome measure is simply selection of a different implant

In how many sites was a different implant selected after X sectional or 3D imaging evaluated?

If outcome specifies selection of a different length or width

In how many sites was a different length of implant selected after X sectional or 3D imaging evaluated?
In how many sites was a different width of implant selected after X sectional or 3D imaging evaluated?

If outcome specifies selection of a longer, shorter, wider or narrower implant

In how many sites was a longer implant chosen after X sectional or 3D imaging evaluated?
In how many sites was a shorter implant chosen after X sectional or 3D imaging evaluated?
In how many sites was a wider implant chosen after X sectional or 3D imaging evaluated?
In how many sites was a narrower implant chosen after X sectional or 3D imaging evaluated?

If outcome specifies prescription of additional procedures such as bone grafting, sinus lifting etc.

In how many sites was bone grafting prescribed after X sectional or 3D imaging but not after conventional imaging?
In how many sites was bone augmentation prescribed after conventional imaging but not after X sectional or 3D imaging?
In how many sites were other surgical procedures prescribed after conventional imaging but not after X sectional or 3D imaging?
In how many sites were other surgical procedures prescribed after conventional imaging but not after X sectional or 3D imaging?
Comments:
Can outcome data for anterior mandible only be analysed?

Appendix D – Final version of data extraction form as a Microsoft Excel file.