Preventing overuse of laboratory diagnostics: a case study into diagnosing anaemia in Dutch general practice

Table 2 Impact of laboratory tests on diagnose an underlying cause of anaemia by GPs. This table shows the impact of individual test results (within the complete set of tests) on the ability of the GP to diagnose an underlying cause of anaemia, regardless whether this diagnosis is correct

Tests with numerical results
Test		Underlying cause instead of unknown	Coefficient (log odds^a)	Exponentiated coefficient (odds)	p-value
ESR		ACD	0.031	1.032	< 0.001***
ESR		RA	−0.044	0.957	0.047*
ESR		Other	−0.060	0.941	0.022*
CRP		RA	−0.065	0.937	0.043*
Reticulocytes		Other	3.650	38.471	< 0.001***
eGFR		RA	−0.250	0.779	< 0.001***
Folic acid		Other	−0.091	0.913	0.023*
Serum iron		IDA	−0.158	0.854	0.024*
Tests with categorical results
Test	Shift in the result category	Underlying cause instead of unknown	Coefficient (log odds^a)	Exponentiated coefficient (odds)	p-value
Leukocytes	Normal to abnormal	Other	2.352	10.503	0.005**
Ferritin	Low normal to low	IDA	3.049	21.088	< 0.001***
Ferritin	Low normal to high normal	ACD	1.310	3.706	0.003**
Ferritin	Low normal to high normal	IDA	1.364	3.911	0.031*
Ferritin	Low normal to high	ACD	1.026	2.791	0.014*
Ferritin	Low normal to high	IDA	1.469	4.343	0.029*
Ferritin	Low normal to high	Other	−4.690	0.009	0.002**

ACD anaemia of chronic disease, CRP C-reactive protein, eGFR estimated glomerular filtration rate, ESR erythrocyte sedimentation rate, IDA iron deficiency anaemia, RA renal anaemia
Significance levels: *** = 0.001; ** = 0.01; * = 0.05
^aFor a more detailed explanation on the interpretation of the values reported in this table, see Additional file 1

ISSN: 1472-6947