Fixed or loose? Dichotomy in RSA data for cemented cups

Figures & data

Table 1.  Breakdown of patients analyzed

Prosthesis	Cement	Department	Year 0–1	Year 0–2	Year 0–2
Lubinus eccentric	Cemex	Umeå	21	21	19
Lubinus eccentric	Palacos	Umeå	23	22	21
Crossfire Osteonics	Palacos	Umeå	10	9	10
Duration Rozing	Palacos	Umeå	17	19	17
Exeter	Palacos	Umeå	16	17	14
Lubinus eccentric	Palacos	Linköping	29	10	10
Flanged anti‐luxation (Link)	Palacos	Linköping	31	15	15
Total			147	113	106

Figure 1. Frequency distribution of the length of the migration vector, postoperatively to 2 years. Blue lines indicate observations. Green and red lines indicate the estimated distributions under the assumption that the material consists of one or more groups with log‐normal distribution. Red triangles represent the means of the 2 subgroups.

Table 2.  Conditional probabilities of a cup being of a certain type, belonging to a man, a woman, or a certain operating department, given that its migration (vector 0–2 year) is below or above the cut‐off point

	Below cut‐off point Probability	Above cut‐off point Probability
Cup
Lubinus Cemex	0.21	0
Lubinus Palacos	0.20	0.14
Crossfire	0.09	0
Duration	0.18	0.07
Exeter	0.09	0.50
Lubinus	0.07	0.21
Flanged anti‐luxation	0.15	0.07
Sex
Male	0.42	0.57
Female	0.58	0.43
Location
Linköping	0.22	0.29
Umeå	0.78	0.71

Figure 2. Frequency distribution of the length of the migration vector from 1 to 2 years. Blue lines indicate observations. Green and red lines indicate the estimated distributions under the assumption that the material consists of one or more groups with log‐normal distribution. Red triangles represent the means of the 2 subgroups.

Figure 3. Frequency distribution along the x‐axis from 1 to 2 years. Blue lines indicate observations. Green and red lines indicate the estimated distributions under the assumption that the material consists of one or more groups with log‐normal distribution. Red triangles represent the means of the 2 subgroups.

Figure 4. Frequency distribution along the y‐axis from 1 to 2 years. Blue lines indicate observations. Green and red lines indicate the estimated distributions under the assumption that the material consists of one or more groups with log‐normal distribution. Red triangles represent the means of the subgroups.

Figure 5. Frequency distribution along the z‐axis from 1 to 2 years. Blue lines indicate observations. Green and red lines indicate the estimated distributions under the assumption that the material consists of one or more groups with log‐normal distribution. Red triangles represent the means of the 2 subgroups.

Table 3.  Parameter estimates for mixture distributions. Component 1 corresponds to the distribution of the stable or less migrating population and component 2 to the distribution of the migrating population. The standard deviation for the estimate is given within parentheses

	Component 1			Component 2
Measurement	Proportion	Mean	SD	Proportion	Mean	SD
Vector 0–2 y	0.89 (0.04)	0.44 (0.06)	0.52 (0.13)	0.11 (0.04)	1.49 (0.10)	0.23 (0.07)
Vector 1 –2 y	0.89 (0.04)	0.15 (0.01)	0.07 (0.02)	0.11 (0.04)	0.54 (0.07)	0.16 (0.05)
x‐axis 1–2 y	0.96 (0.02)	0.00 (0.01)	0.10 (0.01)	0.04 (0.02)	−0.41 (0.10)	0.02(0.13)
y‐axis 1–2 y	0.78 (0.08)	0.02 (0.01)	0.06 (0.01)	0.22 (0.08)	−0.04 (0.06)	0.27 (0.05)
z‐axis 1–2 y	0.67 (0.15)	0.01 (0.01)	0.08 (0.01)	0.32 (0.15)	0.07 (0.05)	0.20 (0.04)

Table 4.  AIC values considered in the modeling process

Measurement	Distribution/distributions	AIC
Vector 0–2 y	Log‐normal	51.4
Vector 0–2 y	2 × Log‐normal	47.7
Vector 0–2 y	3 × Log‐normal	58.8
Vector 1 –2 y	Log‐normal	17.7
Vector 1 –2 y	2 × Log‐normal	16.6
Vector 1 –2 y	3 × Log‐normal	36.3
x‐axis 1–2 y	Normal	28.7
x‐axis 1–2 y	2 x Normal	13.0
x‐axis 1–2 y	3 × Normal	34.7
y‐axis 1–2 y	Normal	54.5
y‐axis 1–2 y	2 x Normal	16.7
y‐axis 1–2 y	3 × Normal	25.5
z‐axis 1–2 y	Normal	25.3
z‐axis 1–2 y	2 × Normal	19.1
z‐axis 1–2 y	3 × Normal	33.6