Insulin initiation in patients with type 2 diabetes is often delayed, but access to a diabetes nurse may help—insights from Norwegian general practice

Figures & data

Figure 1. Selection algorithm for identifying ROSA4 participants with type 2 diabetes (T2D) for the three categories of the primary outcome nominal variable.

Figure 2. Descriptive overview of key health care professionals and their roles and tasks in relation to diabetes follow-up in our Norwegian general practice study.

Table 1. Descriptive analysis of the characteristics of subjects with type 2 diabetes included into the three nominal study groups employed in primary outcome analysis.

Study groups Key descriptive factors	Timely basal insulin-initiation			Postponed basal insulin-initiation			NIAD only (reference group)
	Covariate	RRR (95%CI)	P-value	Covariate	RRR (95%CI)	P-value	Covariate
n (%)	294 (6.8)	–	–	219 (5.1)	–	–	3781 (88.0)
Demographics
Age (Mean ± SD)	63.6 ± 14.1	0.99 (0.98–1.00)	< 0.03	61.4 ± 12.1	0.98 (0.96–.099)	< 0.01	64.9 ± 12.4
Gender: n (%)
Males	167 (57.8)	Ref.	–	137 (63.5)	Ref.	–	2053 (54.9)
Females	124 (42.2)	0.89 (0.73–1.08)	0.37	80 (36.5)	0.73 (0.54–.098)	0.04	1685 (45.1)
Ethnicity: n (%)
European	254 (86.4)	Ref	–	173 (79.0)	Ref	–	3187 (84.3)
African	5 (1.7)	1.03 (0.40–2.63)	0.96	2 (0.9)	0.66 (0.16–2.77)	0.57	64 (1.7)
Asian	32 (10.9)	0.77 (0.51–1.15)	0.20	42 (19.1)	1.56 (1.06–2.30)	0.02	492 (13.0)
Other or unknown	3 (1.0)			2 (0.9)	–	–	38 (1.0)
Diabetes duration (years)	11.0 (6.0, 16.2)	1.07 (1.05–1.09)	< 0.01	10.0 (7.0, 15.0)	1.06 (1.04–1.08)	< 0.01	8.0 (5.0, 12.0)
Disease control efforts and status
Sum of NIADS	1.0 (0.0, 1.0)	0.31 (0.25, 0.37)	< 0.01	2.0 (2.0, 3.0)	3.60 (3.04, 4.29)	< 0.01	1.0 (1.0, 2.0)
HbA1c measurements per year*	3.67 (2.33; 4.67)	1.14 (1.11; 1.16)	<0.01	2.67 (2.00; 4.00)	1.01 (1.04; 1.07)	0.02	2.67 (1.67; 3.67)
HbA1c (% [mmol/mol])	8.8 (8.0, 10.2) [73 (64, 88)]	1.64 (1.55, 1.74)	< 0.01	10.0 (9.0,14.0) [86 (75, 130)]	1.91 (1.79, 2.03)	< 0.01	6.9 (6.3, 7.5)
BMI (kg/m^2)	31.0 (26.6, 35.6)	1.06 (1.03, 1.08)	< 0.01	30.7 (26.1, 33.5)	1.04 (1.00, 1.07)	0.02	29.1 (26.0, 32-8)
Systolic Blood Pressure (mmHg)	136 (125, 146)	1.00 (1.00, 1.01)	0.19	135 (125, 144)	1.00 (0.99, 1.01)	0.66	134 (125, 144)
Diastolic Blood Pressure (mmHg)	80 (70, 85)	1.00 (0.99, 1.02)	0.54	80 (75, 85)	1.02 (1.00, 1.04)	0.02	80 (71, 83)
Total cholesterol (mmol/l)	5.1 (4.3, 6.1)	1.05 (0.98, 1.12)	0.17	5.0 (4.2, 6.0)	1.01 (0.94, 1.10)	0.75	5.0 (4.3, 5.9)
LDL cholesterol (mmol/l)	3.0 (2.4, 3.9)	1.00 (0.92, 1.10)	0.94	3.0 (2.3, 3.8)	1.01 (0.91, 1.12)	0.91	3.1 (2.4, 3.8)
HDL cholesterol (mmol/l)	1.2 (1.0, 1.5)	0.79 (0.62, 0.99)	0.05	1.1 (1.0, 1.4)	0.84 (0.63, 1.10)	0.21	1.3 (1.1, 1.5)
Triglycerides (mmol/l)	2.5 (1.8, 3.9)	1.11 (1.07, 1.15)	< 0.01	2.4 (1.7, 3.4)	1.07 (1.02, 1.13)	< 0.01	2.0 (1.4, 2.8)
Creatinine (μmol/l)	91 (75, 125)	1.01 (1.01, 1.02)	< 0.01	83 (71, 94)	1.00 (1.00,1.01)	0.87	80 (68, 95)
Complications
Microvascular diseases, (n, %)
None	194 (66.0)	Ref	–	176 (80.4)	Ref	–	3354 (88.7)
One or more	100 (34.0)	4.09 (3.11–5.39)	<0.01	43 (19.6)	1.98 (1.36–2.88)	<0.01	427 (11.3)
Macrovascular diseases, (n, %)
None	182 (61.9)	Ref	–	161 (73.5)	Ref	–	2790 (73.8)
One or more	112 (38.1)	1.66 (1.29–2.14)	<0.01	58 (26.5)	0.97 (0.70–1.35)	0.88	991 (26.2)

Unadjusted relative risk ratios (RRR) with 95% CIs obtained from logistic regression. Relative risk ratios for the third nominal outcome, NIAD only, are not shown. Continuous variables are given as Medians, IQR. Differences between categorical variables, given as percentages. *Norwegian guidelines recommended (e.g. year 2012–2014) and still recommend (e.g. year 2023) at least one annual diabetes control for each patient, and eventually up to one to three additional follow-up controls based on an individual clinical assessment of each patient. All GPs and diabetes nurses are obliged to follow this recommendation.

Figure 3. HbA_1c development in NPH (basal insulin) initiators.

Table 2. Descriptive analysis of the characteristics of basal insulin (NPH) initiators who did or did not attain HbA_1c < 7% (53 mmol/mol) (e.g. from the study sample for secondary outcome analysis).

Study groups Key descriptive dimensions		Attaining HbA1c<7% (n = 100)	Not attaining HbA1c<7% (n = 158)	P-value
Demographics
Age in years (Mean ± SD)		63.8 ± 14.1	63.5 ± 13.7	0.86
Gender: Males/Females (n (%))		60 (60) / 40  (40)	88 (56) / 70 (44)	0.50
Ethnicity (n (%))	European	91 (91)	140 (89)	0.81
	African	0 (0)	3 (2)
	Asian	8 (8)	14 (9)
	Other or unknown	1 (1)	1 (1)
Diabetes duration (Median, IQR)		9.5 (5.5, 14.8)	11.2 (6.5, 16.3)	0.10
Non-insulin anti-diabetic therapy
NIAD usage (n (%))	Metformin	52 (52)	83 (53)	0.90
	SU	12 (12)	13 (8)	0.39
	DPP4i	14 (14)	25 (16)	0.73
	GLP-1-RA	7 (7)	12 (8)	0.86
	SGLT-2i	4 (4)	10 (6)	0.58
	TZD	0 (0)	2 (1)	0.52
Disease control status (No differences in total cholesterol, HDL cholesterol, triglycerides, blood pressure, data not shown)
HbA1c at baseline (Median, IQR)	%	8.6 (7.6, 9.7)	9.1 (8.1, 10.3)	0.02
	mmol/mol	70 (60, 83)	76 (65, 89)
BMI (Median, IQR) (kg/m^2)		31.7 (26.8, 37.1)	31.8 (26.9, 35.1)	0.61
LDL cholesterol (Median, IQR) (mmol/l)		3.1 (2.5, 4.1)	2.9 (2.2, 3.9)	0.21
Creatinine (Median, IQR]) (μmol/l)		96.0 (76.0, 132.3)	89.0 (75.0, 128.0)	0.60
Complications
Micro-vascular diseases (n, %)	None	72 (72)	96 (61)	0.07
	One or more	28 (28)	64 (39)
Macro-vascular diseases (n, %)	None	67 (67)	95 (60)	0.26
	One or more	33 (33)	63 (40)
Characterization of initiated insulin therapy
Share with shift or add-on with non-NPH insulin prior lowest HbA1c (%)		27	37	0.09
Time until change (Days)		30 (30, 173)	30 (30, 180)	0.34
Share who try fast acting insulin (%)		30	42	0.04
Share who try Pre-Mixed insulin (%)		12	15	0.55
Share who try basal insulin analogue (%)		4	3	0.50
Median number of NIADS at end of follow-up		1 (0, 1)	1 (0, 1)	0.93
Lowest attained HbA1c	%	6.5 [6.1, 6.9)	7.9 (7.5, 8.69	< 0.01
	mmol/mol	48 (42, 52)	63 (58, 70)
Change in HbA1c	%-point	−2.0 (-3.6, −1.1)	−0.9 (-2.1, −0.2)	< 0.01
	mmol/mol	−22 (-40, −12)	−10 (-23, −2)
Time to lowest HbA1c (Weeks)		35 (22, 53)	32 (21, 52)	0.81

Abbreviations: NIAD: Noninsulin anti-diabetic drug.

Differences between categorical variables, given as percentages, were assessed by Chi-square tests if expected cell counts were > 5, if not a by Fisher’s exact. Continuous variables are given as means ± SD or medians with interquartile range (IQR). Differences in mean and medians of continuous and normally distributed variables were explored using independent samples T-test. Non-normally distributed continuous variables were assessed by Mann–Whitney U-test.

Table 3. The relative risk ratios (RRR) and odds ratios (ORs) for the primary and secondary outcomes are shown for each key support staff factor (exposures).

Primary outcome analysis Relative risk ratios (RRR) (Ref: NIAD only treated (N = 3,781), RRR = 1) with 95% CIs of postponed and timely basal insulin-initiation for each key exposure
Outcomes Exposures	Timely basal insulin initiation (N = 294)		Postponed basal insulin-initiation (N = 219)
	Crude RRR (95% CI)	Adjusted RRR (95% CI)	Crude RRR (95% CI)	Adjusted RRR (95% CI)
Support staff diabetes competence factors
Clinic diabetes nurse	2.42 (1.69, 3.46)***	2.10 (1.36, 3.24)^***a	1.06 (0.68, 1.65)	1.11 (0.65, 1.89)^a
Support staff diabetes course (e.g. share of sup. staff with course)	1.13 (0.80, 1.59)	0.88 (0.64, 1.22)^b	0.98 (0.67, 1.42)	0.93 (0.65, 1.32)^b
Support staff operational capacity factors
Support staff work-load (e.g. Support staff-to-GP FTE^# ratio)	0.96 (0.88, 1.06)	0.99 (0.91, 1.08)^c	1.01 (0.91, 1.11)	0.97 (0.89, 1.07)^c
Support staff size (e.g. Sum staff position sizes given in FTEs)	1.01 (1.00, 1.03)*	1.01 (1.00, 1.02)^*d	1.01 (0.99, 1.02)	1.01 (1.00, 1.02)^d
Secondary outcome analysis Odds ratios (ORs) with 95% CIs of attainment of HbA1c < 7% (53 mmol/mol)
Outcome Exposure	Do attain HbA1c<7% (53 mmol/mol)
	Crude OR (95% CI)		Adjusted OR (95% CI)
Support staff diabetes competence factors
Clinic diabetes nurse	1.17 (0.62, 2.20)		0.93 (0.46, 1.89)^a
Support staff diabetes course (e.g. share of sup. staff with course)	1.84 (1.01, 3.36)*		1.70 (0.90, 3.22)^b
Support staff operational capacity factors
Support staff work-load (e.g. Support staff-to-GP FTE ratio)	1.83 (0.60, 5.54)		1.41 (0.38, 5.18)^c
Support staff size (e.g. Sum of staff position sizes given in FTEs)	1.00 (0.99, 1.01)		1.00 (1.00, 1.01)^d

Primary outcomes: Relative risk ratios (RRR) with 95% CIs were obtained from the generalized structural equation model (GSEM) multinomial logistic regression with random effects showing key staff support factors that were associated with postponed and timely basal insulin-initiation.

Secondary outcomes: Table 3B. Odds ratios (ORs) and their 95% CIs were obtained from the binary logistic regression model with random effects showing factors that were associated with attainment of HbA_1c < 7% (53 mmol/mol).

Adjustments:.

^a Adjusted for: Support staff hands-on diabetes follow-up tasks (mediators), Support staff diabetes course (confounder), Support staff operational capacity factors (confounders);.

^b Adjusted for: Support staff hands-on diabetes follow-up tasks (mediators), Clinic diabetes nurse (confounder), Support staff operational capacity factors (confounders);.

^c Adjusted for: Support staff hands-on diabetes follow-up tasks (mediators), Clinic diabetes nurse (confounder), Support staff diabetes course (confounder), Total support staff FTEs (e.g. variable on support staff size) (confounder);.

^d Adjusted for: Support staff hands-on diabetes follow-up tasks (mediators), Clinic diabetes nurse (confounder), Support staff diabetes course (confounder), Support staff-to-GP-FTE ratio (e.g. variable on support staff work-load) (confounder).

All adjustments were based on directed acyclic graph analysis, and no other factors were identified as potential mediators or confounders.

About the variables: Support staff-hands on diabetes follow-up tasks: 1) staff patient support of BG measurements, a categorical variable (yes/no); 2) staff patient support of diet, a categorical variable (yes/no); 3) staff patient support of injection therapies (e.g. GLP-1-RAs and/or insulin), a categorical variable (yes/no); 4) staff usage of Noklus, a categorical variable (yes/no). Support staff diabetes competence factors: 1) Support staff diabetes course: a continuous variable on share of staff members with a formal diabetes course within the last 3 years; 2) Clinic diabetes nurse: indicates presence or absence of a diabetes nurse in the GP clinic, a categorical variable (yes/no). Support staff operational capacity factors: 1) Support staff-to-GP FTE ratio: a continuous variable expressing of staff work-load; 2) Total number of support staff FTEs: a continuous variable expressing staff size.

#Abbreviations: FTE: abbreviation for Full-Time-Equivalent. This is defined as number of total hours worked by an individual divided by the maximum number of compensable hours in a full-time schedule as defined by law in Norway. In Norway 1 FTE, also often referred to as one 100% employment position, equals an employment of 37.5 h per week;.

*** p < 0.001;.

** p < 0.01;.

* p < 0.05.