Table 2 Variables, measures, hypotheses, and methods of analysis
Variable | Measure, parameterization | Variable name for syntax | Level | Hypothesis | Points in time | Methods of analysis: Stata command (version 14) |
|---|---|---|---|---|---|---|
1. Primary outcome | Â | Â | Â | Â | Â | Â |
a) Bone level | Radiographic bone levels around implants (mesial, distal, mm, continuous) | Bone level | Implant | Group A will show more bone loss than group B | (t0), t4, t6 | Mixed model for continuous responses: mixed |
2. Secondary outcome: Implant success | Modified criteria of Albrektsson (binary): composite variable on implant level | Success | Implant | Group A will show less success than group B | t4, t6 | Mixed model for binary responses: melogit |
a) Related to infection | Modified sulcus bleeding index (0–3 on 2 sites per implant) | SBI | Implant |  | t1, t2, t3, t4, t5, t6 | Mixed model for ordinal responses: meologit |
| Â | Bleeding on probing (binary on 4 sites per implant) | BOP | Implant | Â | t1, t2, t3, t4, t5, t6 | Mixed model for binary responses: melogit |
b) Clinical immobility | Clinical immobility of the implant (binary) | Mobility | Implant | Â | t1, t2, t3, t4, t5, t6 | Mixed model for binary responses: melogit |
c) Pain | Persistent pain or discomfort (binary) | Pain | Implant | Â | t1, t2, t3, t4, t5, t6 | Mixed model for binary responses: melogit |
d) Radiolucency | Evidence of peri-implant radiolucency (binary) | XLucency | Implant | Â | t4, t6 | Mixed model for binary responses: melogit |
e) Survival | Implant in situ (binary) | ImplantLoss | Implant | Â | Continuous time | Kaplan-Meier |
3. Tertiary outcome | Â | Â | Â | Â | Â | Â |
a) Periodontal and periimplant conditions | Probing depths (mm, continuous on 4 sites per implant) | ProbingDepth | Implant | Group A will show higher values than group B | t1, t2, t3, t4, t5, t6 | Mixed model for continuous responses: mixed |
| Â | Probing depths (mm, continuous on 4 sites per tooth) | ProbingDepth | Tooth | Overall improvement at teeth, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for continuous responses: mixed |
|  | Osstell (0–100, continuous on implant level) | Osstell | Implant | Group A will show lower values than group B until the fourth month, thereafter equalization between A and B | t1, t2, t3, t4, t5, t6 | Mixed model for continuous responses: mixed |
|  | Periotest (−8.0 - +50.0, continuous on tooth level) | Periotest | Tooth | Overall improvement at teeth, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for continuous responses: mixed |
|  | Periotest (−8.0 - +50.0, continuous on implant level) | Periotest | Implant | Group A will show higher values than group B until the fourth month, thereafter equalization between A and B | t1, t2, t3, t4, t5, t6 | Mixed model for continuous responses: mixed |
b) Oral health related quality of life | OHIP-G14 questionnnaire (0–56, continuous) | OHIP | Patient | Overall improvement, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for ordinal responses: meologit |
c) Patient’s satisfaction with the PRDP | Questionnnaire 8 items (Five-point Likert-scale, 8–40 continuous) | Satisfaction | Patient | Overall improvement, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for ordinal responses: meologit |
d) Nutrition of the patients | Food frequency questionnaire (1–7) | FFQ | Patient | Overall improvement, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for ordinal responses: meologit |
| Â | Food avoidance questionnaire (binary) | FAQ | Patient | Overall improvement, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for binary responses: melogit |
e) Chewing efficiency | Colour-mixing ability test with two coloured chewing gum (continuous) | Chewing | Patient | Overall improvement, improvement occurred faster in group A than in group B | t1, t2, t3, t4, t5, t6 | Mixed model for ordinal responses: meologit |
4. Exposure | Â | Â | Â | Â | Â | Â |
Group | 2 categories | Â | Jaw | Â | t0 | Â |
5. Time variables | Â | Â | Â | Â | Â | Â |
TimePoint | 0-6 for outcomes |  |  |  | 0-6 for t0 –t6 |  |
Week | Time [weeks] | Â | Patient | Â | Week | Â |
SqrtWeek | Square root of week | Â | Patient | Â | Root of week | Â |
Time | Â | Â | Patient | Â | Continuous | Â |
6. Confounder | Â | Â | Â | Â | Â | Â |
Age | Restricted cubic splines with 3 knots (2 coefficients) | Age | Patient |  | t−1 |  |
Gender | 2 categories (men; women) | Gender | Patient |  | t−1 |  |
Center | 4 categories | Center | Patient |  | t−1 |  |
Jaw class | 4 categories | JawClass | Jaw |  | t−1 |  |
Jaw | 2 categories (upper; lower) | Jaw | Jaw |  | t−1 |  |
Tooth | 1-16 within jaw | Tooth | Tooth |  | t−1 |  |
Site | Up to 4 sites | Site | Site | Â | Â | Â |
Smoking | 3 categories (never; ex; current) | Smoking | Patient | Â | Â | Â |
School education | 3 categories (<10, 10, >10 years) | Education | Patient |  | t−1 |  |
Probing depth | Restricted cubic splines with 3 knots (2 coefficients) | ProbingDepth0 | Tooth |  | t−1 |  |
Bone level | Before treatment in groups | BoneLevel0 | Â | Â | t0 | Â |
Periotest | Restricted cubic splines with 3 knots (2 coefficients) | Periotest0 | Tooth |  | t−1 |  |
OHIP-G14 questionnnaire (0–56, continuous) | Restricted cubic splines with 3 knots (2 coefficients) | OHIP0 | Patient |  | t−1 |  |
Questionnnaire 8 items (Five-point Likert-scale, 8–40 continuous) | Linear term only | Satisfaction0 | Patient |  | t−1 |  |
Food frequency questionnaire (1–7) | Linear term only | FFQ0 | Patient |  | t−1 |  |
Food avoidance questionnaire (binary) |  | FAQ0 | Patient |  | t−1 |  |
colour-mixing ability test with two coloured chewing gum (continuous) | Linear term only | Chewing0 | Patient |  | t−1 |  |
7. Subgroup analysis | Â | Â | Â | Â | Â | Â |
Jaw class 0 vs 1-3 | Secondary outcomes | JawClass |  | Improvement in group A is better than in group B | t−1 |  |
8. Additional analysis | Â | Â | Â | Â | Â | Â |
Maxilla vs. mandible | All outcomes |  |  | Maxilla will show less success and more bone loss than mandible; Implant stability (Periotest, Osstell) is lower in the maxilla than in the mandible; no differences in the improvement of other secondary outcomes | t−1 |  |