Table 2.
Costs per episode for influenza outpatients and associated risk factors in China, 2013 (US$)a
| Characteristic | Direct cost | Indirect costs | Total costs | Multiple linear regression | |
|---|---|---|---|---|---|
| Medical costs | Non-medical costs | ||||
| (95 % Confidence interval)b | |||||
| Total (n = 529) | 70 (69) | 26 (44) | 59 (59) | 155 (122) | |
| Gender | p = 0.141 | p = 0.904 | p = 0.333 | p = 0.814 | |
| Female (n = 248) | 64 (52) | 28 (49) | 62 (65) | 153 (116) | Reference |
| Male (n = 281) | 75 (81) | 25 (40) | 57 (52) | 157 (127) | −6 (−24,12) |
| Age group (years) | p = 0.062 | p < 0.001 | p = 0.006 | p = 0.003 | |
| < 5 (n = 122) | 85 (84) | 36 (53) | 75 (75) | 196 (152) | Reference |
| 5–14 (n = 232) | 63 (68) | 28 (43) | 61 (61) | 153 (121) | −52 (−89,−22)f |
| 15–59 (n = 160) | 68 (59) | 17 (39) | 45 (34) | 130 (91) | −81 (−119,−52)f |
| ≥ 60 (n = 15) | 61 (41) | 10 (13) | 58 (41) | 129 (62) | −104 (−151,−63)f |
| Risk statusc | p=0.006 | p=0.228 | p=0.163 | p=0.009 | |
| Low risk (n = 412) | 64 (57) | 26 (44) | 57 (54) | 146 (111) | Reference |
| High risk (n = 117) | 90 (98) | 28 (45) | 68 (71) | 186 (151) | 40 (16,76)f |
| Area | p = 0.197 | p = 0.406 | p < 0.001 | p < 0.001 | |
| Urban area (n = 438) | 69 (62) | 25 (42) | 67 (61) | 161 (119) | Reference |
| Rural area (n = 91) | 72 (96) | 33 (56) | 23 (23) | 128 (132) | −26 (−53,14) |
| Region | p = 0.013 | p = 0.403 | p < 0.001 | p = 0.004 | |
| East China (n = 292) | 72 (67) | 26 (48) | 71 (64) | 169 (124) | Reference |
| North China (n = 17)d | 82 (53) | 13 (20) | 38 (19) | 133 (78) | −45 (−95,−3)f |
| Central China (n = 52) | 69 (65) | 20 (27) | 40 (36) | 129 (95) | −64 (−98,−31)f |
| South China (n = 64) | 48 (45) | 27 (45) | 45 (37) | 120 (100) | −51 (−78,−22)f |
| Southwest China (n = 68) | 65 (48) | 26 (37) | 50 (67) | 140 (103) | −38 (−68,−8)f |
| Northwest China (n = 36) | 97 (132) | 38 (52) | 44 (46) | 179 (191) | −5 (−52,71) |
| Hospital | p < 0.001 | p = 0.745 | p = 0.002 | p = 0.002 | |
| Level 3 (n = 298) | 73 (60) | 28 (50) | 66 (67) | 167 (124) | Reference |
| Level 2 (n = 119) | 78 (100) | 22 (35) | 56 (45) | 156 (138) | 3 (−21,33) |
| Level 1 and lower (n = 112) | 51 (48) | 26 (38) | 45 (43) | 122 (88) | −34 (−61,−10)f |
| Virus type | p = 0.868 | p = 0.973 | p = 0.220 | p = 0.565 | |
| Untypede (n = 307) | 71 (79) | 25 (42) | 60 (62) | 157 (130) | Reference |
| Influenza A (n = 164) | 67 (53) | 27 (50) | 61 (56) | 156 (106) | −10 (−31,13) |
| Influenza B (n = 58) | 70 (57) | 27 (42) | 47 (41) | 144 (119) | 2 (−25,41) |
aMean (standard deviation) was presented to facilitate their use in economic evaluations despite the skew in sample distributions of costs. However, Rank-sum test was used for comparing two samples, and Kruskal-Wallis test was used for comparing three or more groups because the cost distribution is too right skewed
bCompared to the reference, absolute increase or decrease of the total cost (in US$). And we obtained the bias-corrected and accelerated (BCa) bootstrap percentile confidence interval using the R function “boot.ci”
cRisk status: high risk patients refer to those with underlying medical conditions including: chronic respiratory disease, asthma, chronic cardiovascular diseases, diabetes, chronic liver disease, and chronic renal disease, etc. Other patients without these underlying diseases are low risk patients
dNorth China: 2 patients from Northeast China were grouped into North China
eUntyped: Laboratory tests for influenza virus type identification were not conducted
f p < 0.05: significant differences