Table 4

Attendance of at least one intervention session in the National Health Service (NHS) Diabetes Prevention Programme after IA (N=33 544), mixed-effects logistic regression

	OR	95% lower CI	95% upper CI	P value
Pathway†
ICS Health and Well-being (DCA Vantage)
Living Well Taking Control (Afinion)	2.17	0.99	4.75	0.05
Reed Inhouse (A1c Now+)	1.67	0.72	3.86	0.23
Reed Lloyds (A1c Now+)	1.62	0.65	4.05	0.3
Sex
Male
Female	1.07	1.02	1.12	0.01
Indeterminate	0.76	0.34	1.67	0.49
Age (years)
<40
40–64	1.64	1.43	1.88	<0.001
65–74	2.37	2.06	2.73	<0.001
75+	1.9	1.64	2.2	<0.001
Ethnicity
White
Asian	1.09	1	1.18	0.05
Black	1.3	1.18	1.44	<0.001
Mixed	1.17	0.97	1.41	0.1
Other	1.28	0.95	1.71	0.1
Index of multiple deprivation quintile
1 (most deprived)
2	1.17	1.09	1.26	<0.001
3	1.29	1.2	1.4	<0.001
4	1.33	1.23	1.44	<0.001
5 (least deprived)	1.46	1.34	1.59	<0.001
BMI at IA
Underweight/healthy
Overweight	1.16	1.08	1.24	<0.001
Obese	1.27	1.18	1.36	<0.001
HbA1c measurement at IA
NDH (42.0–47.9 mmol/mol)
Normoglycemic (<42 mmol/mol)	0.71	0.67	0.75	<0.001
Diabetes (≥48 mmol/mol)*	0.32	0.28	0.35	<0.001
Random-effects parameters	Estimate	SE	95% lower CI	95% upper CI
Local referral area: identity
var(constant)	1.16	0.27	0.74	1.84

Likelihood-ratio test versus logistic model: chi-bar-squared (01)=1851.98, p=<0.001.
*Participants with a HbA1c>51 mmol/mol are taken off the program and put on to type 2 diabetes treatment pathways.
†For Ingeus, the IA was spread across the first intervention session and was not comparable to other providers; therefore, it was not included in the regression analysis.
BMI, body mass index; HbA1c, glycated hemoglobin; IA, initial assessment; NDH, non-diabetic hyperglycemia.