Table 3. Factors associated with a decrease in HICU.
Model 1. Simple probit with no selection equation (N = 1,483#) | Model 2. Probit with Heckman model to account for potntial selection bias (N = 1,636) | ||||
---|---|---|---|---|---|
Log-likelihood | -596.450 | ∅ | |||
Athrho (p value) | ∅ | -1.778 (0.001) | |||
Coefficient | P value | Coefficient | P value | ||
Gender and tumor | |||||
Male & any tumor site | -0.374 | 0.004 | -0.339 | 0.002 | |
Female & female-specific tumor site | -0.128 | 0.210 | -0.145 | 0.074 | |
Female & non-gender-specific tumor site | REF | REF | |||
Age at diagnosis | |||||
< 40 | -0.175 | 0.201 | -0.311 | 0.011 | |
> = 40 and < 50 | -0.083 | 0.429 | -0.164 | 0.071 | |
> = 50 | REF | REF | |||
Marital status at survey | |||||
In a relationship | -0.594 | < 0.001 | -0.524 | < 0.001 | |
Single | REF | REF | |||
Dependent children at survey | |||||
Yes | 0.309 | 0.002 | 0.309 | 0.001 | |
No | REF | REF | |||
Decrease in CU between diagnosis and survey | |||||
Yes | -0.198 | 0.060 | -0.174 | 0.038 | |
No | REF | REF | |||
Educational level at diagnosis | |||||
Secondary or less | 0.176 | 0.093 | 0.263 | 0.004 | |
Tertiary | REF | REF | |||
Housing tenure status | |||||
Homeowner | REF | REF | |||
Renter | 0.321 | 0.003 | 0.219 | 0.019 | |
Place of residence | |||||
Rural area | 0.228 | 0.017 | 0.166 | 0.033 | |
City | REF | REF | |||
Sector of activity at diagnosis | |||||
Public | 0.077 | 0.455 | -0.063 | 0.506 | |
Private | REF | REF | |||
Socio-professional category at diagnosis | |||||
Independent farmer, business owner | 0.206 | 0.445 | 0.302 | 0.132 | |
Craftsman, shopkeeper | 0.591 | < 0.001 | 0.479 | 0.001 | |
Manager (white collar) | REF | REF | |||
Intermediate occupation | -0.088 | 0.492 | -0.160 | 0.113 | |
Employee | 0.115 | 0.402 | -0.049 | 0.678 | |
Blue collar worker, farm worker | 0.256 | 0.121 | 0.104 | 0.445 | |
Working time at diagnosis | |||||
Part-time | 0.269 | 0.012 | 0.202 | 0.023 | |
Full-time | REF | REF | |||
HICU at diagnosis | |||||
High (> 3rd quartile) | REF | REF | |||
Intermediate (q1-q3) | -0.523 | < 0.001 | -0.430 | < 0.001 | |
Low (< 1st quartile) | -1.559 | < 0.001 | -1.261 | < 0.001 | |
Change of job and/or employer after diagnosis | |||||
Yes | 0.244 | 0.012 | 0.193 | 0.015 | |
No | REF | REF | |||
Working time reduction after diagnosis | |||||
Yes | 0.465 | < 0.001 | 0.337 | < 0.001 | |
No | REF | REF | |||
Duration of 1st sick leave (in months) | |||||
0.007 | 0.060 | 0.007 | 0.034 | ||
Chemotherapy at diagnosis | |||||
Yes | 0.215 | 0.021 | 0.160 | 0.038 | |
No | REF | REF | |||
Adverse cancer event between diagnosis and survey | |||||
Yes | -0.187 | 0.091 | -0.069 | 0.474 | |
No | REF | REF |
# Non-responses were excluded from model 1.
Note to the reader: All other factors being equal, and regardless of whether the tumor site was gender-specific or not, men were significantly less likely to experience a decrease in HICU than women.