Article Figures & Data
Figures
- Figure 1. Flow chart of inclusion and exclusion criteria
- Figure 2. Receiver operating characteristic curves for the derivation and validation samples.
AUC = area under curve.
Tables
Points Age, years 18–45 0.00 46–59 1.25 60–70 1.50 BMI <30 0.00 ≥30 1.50 Sex Female 0.00 Male 1.50 BMI = body mass index.
Derivation (n = 352),
n (%)Validation (n = 260),
n (%)P value Primary care data complete 350 (99.4) 249 (95.8) 0.002 Lack of sleep 74 (21.0) 27 (10.4) 0.001 Male 122 (34.7) 67 (25.8) 0.019 Mean age, SD 48.84 (11.15) 49.01 (11.55) 0.852 Mean weight, SD 81.11 (16.20) 86.66 (17.34) <0.001 Mean height, SD 169.45 (9.70) 169.03 (8.96) 0.588 Mean BMI, SD 28.18 (4.88) 30.26 (5.24) <0.001 Frequent snoring 346 (98.3) 257 (98.8) 0.740 Breathing pauses 134 (38.1) 182 (70.0) <0.001 Daytime hypersomnia 149 (42.3) 163 (62.7) <0.001 History of accidents 12 (3.4) 22 (8.5) 0.007 Morning tiredness 151 (42.9) 157 (60.4) <0.001 Waking with sensation of asphyxia 68 (19.3) 96 (36.9) <0.001 Mean neck circumference, SD 39.08 (4.27) 40.09 (4.09) 0.003 AHT 79 (22.4) 89 (34.2) 0.001 Heart failure 3 (0.9) 7 (2.7) 0.106 DM 20 (5.7) 33 (12.7) 0.002 Smoking habit Non-smoker 131 (37.2%) 74 (28.5%) 0.027 Smoker 106 (30.1%) 76 (29.2%) Ex-smoker 115 (32.7%) 110 (42.3%) Drinks alcohol 126 (35.8%) 111 (42.7%) 0.083 Mean Epworth score, SD 9.24 (4.14) 10.04 (4.90) 0.034 Diagnostic test CPSG 11 (3.8) 159 (68.8) <0.001 RP 272 (92.8) 67 (29.0) None 6 (2.0) 5 (2.2) Both RP and CPSG 4 (1.4) 0 (0.0) Therapeutic decision CPAP 79 (28.5) 118 (50.6) <0.001 No CPAP 192 (69.3) 111 (47.6) Mandibular advancement device 4 (1.4) 4 (1.7) Postural device 2 (0.7) 0 (0.0) AHT = arterial hypertension. BMI = body mass index. CPAP = continuous positive airway pressure. CPSG = conventional polysomnography. DM = diabetes mellitus. RP = respiratory polygraphy. SD = standard deviation.
- Table 3. Derivation sample. Univariate logistic regression for the primary variable 'therapeutic decision'. Selection of variables for the final model.
Variable OR 95% CI P value Male 3.15 1.74 to 5.99 <0.001 Age 18–45; 46–59; 60–70 (1) 2.56 1.37 to 4.95 0.002 Age 18–45; 46–59; 60–70 (2) 3.15 1.52 to 6.68 BMI <30; ≥30 3.61 2.11 to 6.24 <0.001 Frequent snoring 0.66 0.11 to 5.07 0.656 Breathing pauses 1.92 1.14 to 3.23 0.013 Daytime hypersomnia 1.06 0.63 to 1.77 0.835 History of accidents 0.31 0.02 to 1.81 0.220 Morning tiredness 0.62 0.36 to 1.04 0.071 Waking with sensation of asphyxia 1.02 0.54 to 1.89 0.948 Neck circumference ≤38; 38.1–42; >42 (1) 2.04 1.07 to 3.99 <0.001 Neck circumference ≤38; 38.1–42; >42 (2) 7.20 3.57 to 15.08 AHT 4.00 2.26 to 7.17 <0.001 DM 3.36 1.04 to 11.64 0.043 Smoking status (1) 1.08 0.56 to 2.09 0.122 Smoking status (2) 1.80 0.99 to 3.32 Alcohol status 1.07 0.62 to 1.81 0.816 Epworth score <9; ≥9 1.67 0.99 to 2.86 0.056 AHT = arterial hypertension. BMI = body mass index. CI = confidence intervals. DM = diabetes mellitus. OR = odds ratio.
- Table 4. Derivation sample. Final logistic regression and scores for the clinical prediction rule.
B SE Significance Exp(B) 95% CI for Exp(B) βi/ βmin Score AHT 0.9150 0.3304 0.00562 2.497 1.306 to 4.787 1 1.00 Age 46–59 1.1955 0.3746 0.00141 3.305 1.614 to 7.055 1.30657 1.25 Age 60–70 1.4440 0.4421 0.00109 4.238 1.804 to 10.281 1.57814 1.50 BMI ≥30 1.4490 0.3192 5.65e–06 4.259 2.300 to 8.074 1.58360 1.50 Male 1.3740 0.3488 8.18e–05 3.951 2.040 to 8.059 1.50164 1.50 Constant −2.0823 0.3435 1.34e–09 0.125 0.061 to 0.236 Hosmer–Lemeshow test P = 0.903.
AHT = arterial hypertension. BMI = body mass index. CI = confidence interval. SE = standard error.
Cut-off point Treated
(N = 85)Not treated
(N = 192)Sensitivity, %
(95% CI)Specificity, %
(95% CI)PPV, %
(95% CI)NPV, %
(95% CI)Derivation (n = 277) ≥1.50 80 154 94.1
(89.1 to 99.1)19.8
(14.2 to 25.4)34.2
(28.1 to 40.3)88.4
(78.8 to 98.0)≥2.25 73 94 85.9
(78.5 to 93.3)51.0
(44.0 to 58.1)43.7
(36.2 to 51.2)89.1
(83.3 to 94.9)≥2.50 73 88 85.9
(78.5to 93.3)54.2
(47.1 to 61.2)45.3
(37.7 to 53.0)89.7
(84.1 to 95.2)≥2.75 71 80 83.5
(75.6 to 91.4)58.3
(51.4 to 65.3)47.0
(39.1 to 55.0)88.9
(83.4 to 94.4)Cut-off point Treated
(N = 122)Not treated
(N = 111)Sensitivity, %
(95% CI)Specificity, %
(95% CI)PPV, %
(95% CI)NPV, %
(95% CI)Validation (n = 233) ≥1.50 119 90 97.5
(94.8 to 100.3)18.9
(11.6 to 26.2)56.9
(50.2 to 63.7)87.5
(74.3 to 100.7)≥2.25 100 64 82.0
(75.1 to 88.8)42.3
(33.2 to 51.5)61.0
(53.5 to 68.4)68.1
(57.1 to 79.1)≥2.50 99 63 81.1
(74.2 to 88.1)43.2
(34.0 to 52.5)61.1
(53.6 to 68.6)67.6
(56.7 to 78.5)≥2.75 93 57 76.2
(68.7 to 83.8)48.6
(39.4 to 57.9)62.0
(54.2 to 69.8)65.1
(54.8 to 75.3)
In this issue
Derivation and validation of a clinical prediction rule for sleep apnoea syndrome for use in primary care
Jump to section
More in this TOC Section
Related Articles
Cited By...
Intended for Healthcare Professionals
