Individual, interpersonal, and organisational factors associated with discrimination in medical decisions affecting people with a migration background with mental health problems: the case of general practice

Figures & data

Figure 1. Patients played by actors in the video-vignettes in the survey.

Figure 2. Flow chart showing participation in the sample process.

Table 1. Sociodemographic description of the GP respondents according to the patient ethnicity in the video-vignette and statistical test on the random allocation of the video-vignette according to the patient ethnicity, Belgium, n = 797.

Respondents’ characteristic	Vignette of the patient without a migration background Mean (SD^a) or %	Vignette of the patient with a migration background Mean (SD) or %	Statistical test F-test or χ2 (p Value)
GPs, n (%)	400 (50.2)	397 (49.8)
Age (range = 24–79), years	38.5 (14.9)	38.0 (14.7)	0.21 (0.665)
Sex			0.08 (0.77)
Men	36.5	35.5
Women	63.5	64.5
Language			0.02 (0.88)
French-speaker	70.0	69.5
Dutch-speaker	30.0	30.5
GP’s ethnicity			0.10 (0.75)
Without a MB (i.e. Belgian)	73.0	74.8
With a MB (i.e. first or second-generation)	10.8	11.8
Licence status			2.04 (0.15)
Licensed GPs	52.8	57.4
GPs in training	47.2	42.6
Experience (years)	12.3 (14.2)	11.9 (13.8)	0.15 (0.70)
Working schedule			2.07 (0.56)
Full time	88.3	84.9
Half of normal working hours	3.7	5.3
< Half of normal working hours	0.8	1.0
Other	7.2	8.8
Perceived workload			0.78 (0.38)
Light or normal	48.5	51.6
High	51.5	48.4
Type of practice (>50% of their working schedule)			2.23 (0.13)
Solo	32.5	27.6
Group	67.5	72.4
Type of area			0.95 (0.62)
Unknown	.	0.7
Urban	11.2	12.3
Sub-urban	26.0	28.0
Rural	62.8	59.0
Frequency of contact with patients with migration background (1: never to 5: everyday)	3.4 (1.4)	3.7 (1.3)	9.36 (0.002)
Hudelson score (5, 35)	21.3 (5.1)	20.9 (5.0)	1.02 (0.31)
Cultural competence training No Yes	79.8 20.2	80.7 19.3	0.10 (0.74)
The patient exaggerates his distress (1: not likely at all to 5: very likely)	2.3 (0.6)	2.2 (0.6)	0.97 (0.32)

^a SD: Standard deviation.

Table 2. GPs’ diagnostic decisions, treatment, referral recommendations, and time spent on the vignette.

	Vignette of the patient without a migration background Mean (SD^a) or %	Vignette of the patient with a migration background Mean (SD) or %	Statistical test F-test or χ^2 (p Value)
Diagnosis
Depression, %	98.5	97.5	1.05 (0.31)
Anxiety, %	42.3	36.5	2.74 (0.10)
Post-traumatic Stress Disorder (PTSD), %	11.0	16.1	4.46 (0.04)
Severity of symptoms, /10	7.79 (1.03)	7.53 (1.18)	7.68 (0.006)
Treatment (/4)
Medical	2.95 (0.66)	2.85 (0.75)	4.09 (0.04)
Non-medical	3.65 (0.61)	3.71 (0.58)	2.29 (0.13)
A combination of both	3.60 (0.60)	3.44 (0.65)	11.55 (<0.001)
Referral (/4)
Likelihood of referral	3.26 (0.69)	3.31 (0.68)	0.95 (0.33)
MANOVA (F-test, p Value)^b	-	-	3.56 (<0.001)

^a SD: Standard deviation

^b MANOVA: Multivariate analysis of variance

Table 3. The effect of GPs’ individual, interpersonal, and organizational factors on ethnic differences in mental health diagnosis, treatment, and referral recommendations, estimated coefficients from logistic regression and ANOVA.

	Diagnosis of PTSD (%Yes)^a		Severity of symptoms (/10)		A combination of both treatments (/4)^b
Covariates	B^c	CI95%	B	CI95%	B	CI95%
Main effect
Age (10 years)^d	−0.05**	(−0.09, −0.02)	0.09	(−0.03, 0.21)	0.04***	(−0.00, 0.08)
Women (ref = Men)	0.74*	(0.01,1.48)	−0.13	(−0.39, 0.14)	0.02	(−0.12, 0.15)
GP’s ethnicity (ref = Without a MB)	−0.63	(−1.60, 0.35)	0.21	(−0.19, 0.62)	−0.04	(−0.24, 0.17)
Perceived heavy workload (ref = Light)	−0.81*	(−1.46, −0.14)	0.12*	(−0.45, 0.39)	0.12**	(−0.01, 0.24)
Group practice (ref = Solo)	0.99*	(0.16, 1.84)	−0.08	(−0.34, 0.19)	−0.13*	(−0.26, 0.01)
Frequency of contact with patients with a migration background^e	0.24	(−0.12, 0.60)	0.02	(−0.12, 0.16)	−0.07***	(−0.14, −0.00)
Hudelson score (5-35)	0.01	(−0.05, 0.07)	0.01	(−0.02, 0.03)	−0.00	(−0.02, 0.01)
Cultural competence training (ref = No)	0.95**	(0.27, 1.62)	−0.43	(−0.86, −0.00)	−0.38***	(−0.55, −0.20)
The patient exaggerates his distress^f	−0.83**	(−1.43, −0.24)	−0.19**	(−0.40, 0.03)	0.10	(−0.02, 0.22)
Interaction between patient with a migration background (ref = without a migration background) and:
Age (10 years)^d	0.04*	(0.00, 0.08)	0.09	(−0.03, 0.21)	0.06*	(−0.00, 0.12)
Women (ref = Men)	−0.43	(−1.37, 0.51)	0.11	(−0.27, 0.49)	0.02	(−0.17,0.21)
GP’s ethnicity (ref = Without a MB)	−0.53	(−2.28, 1.23)	−0.26	(−0.08, 0.28)	0.38*	(0.08, 0.66)
Perceived heavy workload (ref = Light)	0.88*	(0.04, 1.73)	0.22	(−0.16, 0.59)	0.04	(−0.14, 0.22)
Group practice (ref = Solo)	−0.28	(−1.39, 0.82)	0.046	(−0.34, 0.43)	−0.01	(−0.21, 0.19)
Frequency of contact with patients with a migration background^e	0.06	(−0.43, 0.56)	−0.09	(−0.29, 0.11)	−0.05	(−0.15, 0.08)
Hudelson score (5-35)	−0.03	(−0.11, 0.05)	−0.03	(−0.06, 0.001)	0.01	(−0.01, 0.03)
Cultural competence training (ref = No)	0.07	(−0.83, 0.97)	0.38	(−0.27, 1.04)	−0.00	(−0.26, 0.25)
The patient exaggerates his distress^f	4.15**	(0.40, 1.90)	−0.11	(−0.42, 0.20)	−0.09	(−0.27, 0.09)

^a : These estimated coefficients come from a logistic regression, while the other come from an ANOVA. ^b: 4-point Likert scale from 1: not at all likely to 4: very likely

^c : bold coefficients have a significant p Value; * means p Value < 0.05; ** means p Value < 0.01; *** means p Value < 0.001

^d : Odds ratio associated with a 10-year increase in age

^e : 1 = less than monthly to 3 = at least every week

^f : 1 = not likely at all to 5 = extremely likely

Table

□	Schizophrenia and other psychotic disorders
□	Bipolar disorder
□	Depression
□	Anxiety
□	Obsessive Compulsive Disorder and related disorders
□	Post-Traumatic Stress Disorder
□	Somatization disorder and related disorders
□	Sleeping disorder
□	None of the above diagnoses applies to this patient.

Table

□	0	1	2	3	4	5	6	7	8	9	10
Vignette of the patient without a migration background (% of GPs)	0	0	0	0	0.7	1.4	7.1	26.1	40.4	21.8	2.5
Vignette of the patient with a migration background (% of GPs)	0	0.4	0	0	2.2	3.3	6.2	31.5	40.2	14.5	5

Table

	How likely are you to prescribe this treatment?
	Vignette of the patient without a migration background				Vignette of the patient with a migration background
	1 Very unlikely	2 Unlikely	3 Likely	4 Very likely	1 Very unlikely	2 Unlikely	3 Likely	4 Very likely
Medical treatment (%)	1.9	18.8	61.6	17.7	4.1	24.8	53.4	17.7
Non-medical treatment (%)	0.8	4.5	23.8	70.9	1.4	2.4	19.6	76.6
A combined medical and non-medical treatment (%)	0	6.1	28.3	65.6	0.3	8.2	38.9	52.6

Table

	How likely would you refer the patient in question?
	Vignette of the patient without a migration background				Vignette of the patient with a migration background
	1 Very unlikely	2 Unlikely	3 Likely	4 Very likely	1 Very unlikely	2 Unlikely	3 Likely	4 Very likely
Frequency (%)	0.8	11.5	48.5	39.2	1.0	9.6	46.8	42.6