1.2 The influential factors for help-seeking behavior
Crosstabs, an SPSS procedure, demonstrated that one’s family history of mental disorders, self-pay ratio and level of education were significantly related to help seeking-help- or- not behavior in individuals with SZ (X2 = 6.621,P = 0.010;X2 = 12.821, P = 0.005;Fisher exact P = 0.0225).
After pairwise comparison using the chi-square test or Fisher’s exact test, we found that the help-seeking behavior ratio is significantly higher in the 30% to 60% self-pay ratio(self-pay fee/total medical expenditure) group than in the 10% to30% self-pay ratio group (chi-square 8.40, p = 0.00375, respectively), with no significant difference in other self-pay groups. Nevertheless the help-seeking behavior ratio is significantly lower in the illiterate group than in the groups with primary education, junior high school education and senior high school education (the chi-square and p-value are 4.4557, 0.0348; 6.118, 0.0134; 11.019, 0.000902; respectively). No significant differences in help-seeking behavior exist between other self-pay and educational level groups. The patient group with a family history of mental disorders tend to have a higher help-seeking behavior ratio (chi-square and p-value are 6.621, 0.010, respectively). Furthermore, no significant differences in help seeking behavior(or non-help seeking behavior) exist contingent on the degree of self-pay insurance or educational level.
Similarly, help-seeking versus non-help-seeking behavior is not contingent upon gender, monthly family income, medical insurance, publicly disruptive behavior, self-injury, suicide attempts, risk assessment, or urban-rural residence (Table1).
Table 1: Comparison in help-seeking verses non help-seeking behavior based on influencing factors
Influencing factor
|
Number
|
Yes
|
No
|
Chi-square
|
P value
|
Logistic P
|
Gender
|
|
male
|
180
|
118
|
62
|
1.887
|
0.170
|
0.170
|
female
|
187
|
135
|
52
|
|
Family history of mental disorders
|
|
Yes
|
31
|
29
|
2
|
6.621
|
0.010
|
0.021
|
No
|
189
|
136
|
53
|
|
Monthly family income
|
|
< 1000 ¥
|
137
|
90
|
47
|
6.351
|
0.096
|
|
1000 – 3000 ¥
|
151
|
100
|
51
|
|
3000 – 5000 ¥
|
62
|
51
|
11
|
0.075
|
> 5000 ¥
|
17
|
12
|
5
|
|
Publicly funded healthcare
|
|
Yes
|
362
|
250
|
112
|
|
0.648
|
0.666
|
No
|
5
|
3
|
2
|
|
Self-pay ratio
|
|
< 10%
|
39
|
31
|
8
|
12.821
|
0.005
|
|
10 - 30%
|
231
|
144
|
87
|
0.004
|
30 - 60%
|
75
|
61
|
14
|
|
> 60%
|
22
|
17
|
5
|
|
Number of public disruptions
|
|
Yes
|
37
|
21
|
16
|
2.087
|
0.149
|
0.151
|
No
|
187
|
129
|
58
|
|
Self-injury
|
|
Yes
|
4
|
4
|
0
|
|
0.307
|
0.999
|
No
|
215
|
144
|
71
|
|
Suicide attempt
|
|
Yes
|
4
|
3
|
1
|
|
1.000
|
0.746
|
No
|
214
|
144
|
70
|
|
Risk assessment level
|
|
F
|
3
|
1
|
2
|
|
0.0787
|
|
E
|
11
|
5
|
6
|
|
D
|
40
|
22
|
18
|
|
C
|
64
|
43
|
21
|
0.098
|
B
|
96
|
72
|
24
|
|
A
|
152
|
109
|
43
|
|
Urban or rural residence
|
|
urban
|
170
|
118
|
52
|
0.033
|
0.855
|
0.855
|
rural
|
197
|
135
|
62
|
|
Educational level
|
|
illiterate
|
50
|
25
|
25
|
|
0.0225
|
|
primary
|
115
|
79
|
36
|
|
junior high school
|
134
|
95
|
39
|
|
senior high school
|
41
|
35
|
6
|
0.025
|
secondary school
|
14
|
10
|
4
|
|
post-secondary school
|
5
|
4
|
1
|
|
undergraduate or above
|
8
|
5
|
3
|
|
In consideration of potential interactions between various influencing factors, we used stepwise logistic regression analysis to study each factor’s impact on help-seeking behavior. We included all the p<0.1 (univariate binary logistic regression test) factors, which included family history of mental disorders, monthly family income, self-pay ratio, publicly disruptive behavior, risk assessment level and education level) in the model. The result shows that only a family history of mental disorders and education level are significantly related to whether or not the individual with SZ seeks help.
3 First choices of help-seeking behavior
3.1 Statistical description of the first choice of help
Two-hundred and fifty three out of 367 (68.9%) individuals with SZ in our study did seek help; for 64.6% (n = 163) of those, the first chose was “professional psychiatric service”, with only 41.1% (n = 104) of those admitted to inpatient care. Of the remainder of those seeking help, 30.8% choose non-medical options, in which 18.2% (n = 46) first asked relatives for help (Figure Legends 2).