Validity and responsiveness of GHC-index in patients with amalgam-attributed health complaints

Figures & data

Table 1. Cronbach’s alpha for reliability test and correlation coefficients for convergent validity.

	Baseline			Follow-up
	GBB-24	GHC-index	MAS	GBB-24	GHC-index	MAS
Internal consistency
Cronbach’s alpha (α)	0.93	0.87	0.92	0.92	0.86	0.89
Convergent validity (ρ)
BDSC	0.94	0.68	0.78	0.94	0.83	0.82
VAS	0.52	0.61	0.48	0.58	0.59	0.66
CL	0.59	0.60	0.54	0.67	0.62	0.67
Vitality	0.63	0.54	0.45	0.71	0.71	0.72

BDS_c: Bodily distress syndrome (continuous), CL: Cantril ladder of life satisfaction, GBB-24: 24-item Giessen subjective complaints list, GHC: General health complaint, MAS: Munch amalgam scale.

Table 2. Known group validity: Kruskal–Wallis statistics and relative efficiency of GHC-index and MAS against GBB-24.

	Kruskal–Wallis H test statistics ${\mathit{\chi }}_{(3)}^2$*	RE (95% CI)
BDS
GBB-24	14.66	1 (Reference)
GHC-index	10.04	0.69 (0.41–0.96)
MAS	8.65	0.59 (0.32–0.86)
SF-36 energy
GBB-24	16.24	1 (Reference)
GHC-index	13.40	0.83 (0.50–1.15)
MAS	15.37	0.95 (0.53–1.36)

${\mathit{\chi }}_{\left(3\right):}^2$ Chi-square statistic with 3 degrees of freedom, CI: 95% bootstrap confidence interval (with 1000 iterations), BDS: Bodily distress syndrome, GBB-24: 24-item Giessen subjective complaints list, GHC: General health complaint, MAS: Munch amalgam scale, SF-36 Energy: The Energy item of the subscale Vitality of the SF-36 Health Survey.

*p < .01.

Table 3. Predictive validity of GBB-24, GHC-index and MAS: logistic and ordinary least square regressions.

	β	se	p-value	CD/R^2	b^a
A. Logistic regression
Low self-reported health
GBB-24	0.086	0.038	.022	0.209	1.168
GHC-index	0.102	0.036	.005	0.255	1.464
MAS	0.096	0.036	.008	0.231	1.341
Moderate to severe BDS
GBB-24	0.503	0.212	.018	0.746	6.840
GHC-index	0.195	0.063	.002	0.380	2.803
MAS	0.125	0.060	.038	0.233	1.741
B. Ordinary least square regression
Vitality
GBB-24	1.151	0.243	<.001	0.505	0.711
GHC-index	1.080	0.239	<.001	0.496	0.705
MAS	1.134	0.196	<.001	0.514	0.717
BDSC
GBB-24	0.935	0.049	<.001	0.887	0.942
GHC-index	0.784	0.068	<.001	0.696	0.834
MAS	0.791	0.102	<.001	0.666	0.816

β: Unstandardized coefficient, b: Standardized coefficient, se: standard error of β, CD: Coefficient of discrimination (for logistic regression), R²: Coefficient of determination (commonly known as R-squared in linear regression), BDS_C: Bodily distress syndrome (continuous), GBB-24: 24-item Giessen subjective complaints list, GHC: General health complaint, MAS: Munch amalgam scale, Vitality: The Vitality subscale of the SF-36 Health Survey.

^aWhile only X is standardized in logistic regression, both X and Y are standardized in linear regression.

Table 4. Responsiveness of the symptom load measures: GHC-index, GBB-24 and MAS.

	MCS* (se)	SRM (se)	ES (se)	SEM	MDC90	%Δ > MDC90
GBB-24	7.1 (1.9)	0.66 (0.153)	0.51 (0.155)	3.88	6.36	43.8
GHC-index	12.8 (2.8)	0.81 (0.178)	0.72 (0.189)	6.57	10.81	56.3
MAS	10.3 (2.7)	0.67 (0.152)	0.58 (0.141)	5.83	9.60	46.9

MCS: Mean change score, GBB-24: 24-item Giessen subjective complaints list, GHC: General health complaint, MAS: Munch amalgam scale, SRM: Standard response mean, ES: Effect size, SEM: Standard error of measurement, MDC: Minimal detectable change, se: Standard error.

*p < .001 for paired t-test.

Data availability statement

The datasets generated and analyzed during the current study are not publicly available due to privacy concern, as relatively few patients participated in the study with implications for potential identification through personal characteristics.