Abstract
Background
Type 2 diabetes (T2D) causes a major burden on healthcare and is a leading cause of chronic kidney disease (CKD). While the healthcare service utilisation of individuals with CKD in speciality clinics and the high costs of renal replacement therapy have been widely studied, the burden of kidney function impairment on primary healthcare remains understudied. In this study, we aimed to examine the association of kidney function impairment with primary healthcare service utilisation in individuals with T2D.
Methods
This observational follow-up study consisted of 4773 individuals with T2D treated in the municipal Rovaniemi healthcare centre, Rovaniemi, Finland between 2011 and 2019 (mean age 70.9 years; 53.5% males). The mean follow-up period was 5.6 years. Substantial kidney function impairment was defined as a reduction of 25% or more in the estimated glomerular filtration rate (eGFR) over the follow-up and was contrasted to non-substantial kidney function impairment (eGFR reduction of less than 25%). Primary healthcare service utilisation included planned diabetes appointments, unplanned general practice [GP] appointments, other GP contacts (digital, phone or letter), unplanned nurse appointments, other nurse contacts (digital, phone or letter) and inpatient days in the healthcare centre that occurred during the follow-up. The association between kidney function impairment and primary healthcare service utilisation was studied with negative binomial regression models adjusted for potential confounding factors.
Results
Substantial kidney function impairment occurred in 12.9% of the study population. Compared to non-substantial kidney function impairment, substantial kidney function impairment associated with increased primary healthcare service utilization when measured with planned diabetes appointments (adjusted incidence rate ratio [aIRR] 1.18, 95% confidence interval [CI] 1.09–1.28), other GP contacts (aIRR 1.38, 95% CI 1.57–1.84), unplanned nurse appointments (aIRR 1.19, 95% CI 1.02–1.24), and inpatient days in the healthcare centre (aIRR 1.20, 95% CI 1.10–1.45).
Conclusions
Kidney function impairment in individuals with T2D was associated with an increase in primary healthcare service utilisation in several measured categories. This further underlines the importance of prevention and prompt treatment of CKD in individuals with T2D to reduce the burden on primary healthcare.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12913-025-13526-7.
Keywords: Type 2 diabetes, Chronic kidney disease, Primary health care, Health service utilisation
Background
Type 2 diabetes (T2D) is a global health crisis whose prevalence and incidence are rising across the globe [1]. The International Diabetes Federation reported a global overall 12.9% increase in diabetes prevalence between 2018 and 2021, with the disease now affecting over one in 10 adults globally [1]. Diabetes, along with hypertension, is the leading cause of chronic kidney disease (CKD) in the world [2, 3]. The rising prevalence of T2D entails a direct increase in individuals with CKD, since CKD is becoming increasingly common over time among individuals with T2D [3, 4]. However, there is concerning evidence indicating a disproportionate increase in CKD prevalence among T2D individuals compared to the increase in T2D itself [3]. Although a modest decline in the incidence and prevalence of renal replacement therapy has been observed locally in Finland during the last few years, globally the percentage of individuals requiring renal replacement therapy for end-stage renal disease (ESRD) is increasing at an alarming rate [3–5].
CKD is common in individuals with T2D, approximately half of whom develop CKD during their lives [3]. Multiple studies have estimated the prevalence of CKD to vary between 27.9% and 58.6% in individuals with T2D [6–11]. Due to the high prevalence of CKD, annual screening is generally recommended from the time of T2D diagnosis [12–14]. Early detection, renoprotective medication, and addressing modifiable risk factors such as high blood pressure, hyperglycemia, high LDL-cholesterol, poor nutrition, smoking and physical inactivity areimportant for slowing CKD’s progression and preventing further loss of health [13, 14]. Moreover, CKD can progress to ESRD, which significantly reduces individuals’ quality of life, increases all-cause mortality and has a substantial economic impact [3].
The economic impact of diabetes is also substantial, with individuals diagnosed with diabetes accounting for as much as 10% of all healthcare-related expenditures in the UK; due to the increasing prevalence of diabetes, this is projected to increase to around 17% in 2035 [15]. Of this cost, 80% will be attributable to treating complications caused by diabetes. Of all these complications, the costs related to the treatment of ESRD are the highest [16]. Active screening and treatment are effective in mitigating the progression of CKD to ESRD and the high costs related to dialysis and kidney transplantation [14, 17].
The costs of CKD treatment and healthcare service utilisation by individuals with CKD in speciality clinics have been widely studied, while the aforementioned high costs related to the treatment of ESRD are well known [16–18]. However, the management of T2D and early CKD is typically the responsibility of primary care providers, and to our knowledge, the association of CKD or kidney function impairment with primary healthcare service utilisation in individuals with T2D has not previously been studied. This longitudinal study aims to examine the association between substantial kidney function impairment (defined as ≥ 25% reduction in estimated glomerular filtration rate [eGFR]) and primary healthcare service utilisation among individuals with T2D. We hypothesise that an impairment in kidney function is associated with an increase in healthcare service utilisation within a primary care setting.
Methods
Study population
This registry-based real-world study in primary healthcare is part of the Rovaniemi Primary Care T2D study. The initial study population consisted of all those 5104 individuals with a T2D diagnosis treated between November 1, 2011, and February 19, 2019, in Rovaniemi’s municipal healthcare centre in Finland. The international classification of diseases 10th revision (ICD-10) codes and the international classification of primary care codes (ICPC) were used to identify individuals with T2D. From this initial population, we included individuals with at least two laboratory eGFR or serum creatinine measurements at a minimum of 6-month intervals in this study, resulting in a study population of 4773 individuals. Individuals with only a single eGFR or serum creatinine measurement were excluded from the study. The study period varied between individuals, with the baseline starting from the individual’s first eGFR or serum creatinine measurement and the follow-up concluding with their final measurement.
The study data were taken from the local electronic health record system. All supplied data were anonymised and handled using individual IDs. The study protocol was approved by the Ethics Committee of Lapland Central Hospital, Rovaniemi, in May 2018. According to Finnish legislation, no written individual consent is required in a registry-based study.
eGFR reduction (exposure)
Each individual’s eGFR (ml/min/1.73 m2) was calculated from their serum creatinine level (µmol/l), sex and age using the Chronic Kidney Disease Epidemiology Collaboration equation if it was not automatically supplied by the laboratory. Substantial kidney function impairment was defined as an eGFR reduction of 25% or greater at (individual) baseline and follow-up, based on the 2012 guidelines of Kidney Disease: Improving Global Outcomes (KDIGO) for CKD evaluation and management [19]. In turn, non-substantial kidney function impairment was defined as an eGFR reduction of less than 25% (including no reduction or improvement) and was used as a reference in the analysis.
Primary healthcare utilisation (outcome)
We categorised individual visits to general practitioners (GP) and nurses in primary healthcare between baseline and follow-up by contact type as planned diabetes appointments, unplanned GP appointments, other GP contacts (digital, phone or letter), unplanned nurse appointments, other nurse contacts (digital, phone or letter) and inpatient days in the healthcare centre (18). Planned diabetes appointments were annual, biannual or otherwise recurring diabetes-related GP or nurse appointments scheduled on another visit by a GP or a nurse. Unplanned GP appointments accounted for all other unplanned face-to-face GP appointments, excluding emergency room visits. Unplanned nurse appointments accounted for all nurse appointments. Other contacts by GPs and nurses accounted for all other contact types to individuals by telemedicine (phone, digital records) or by letter. Inpatient days in the healthcare centre were inpatient days for any reason spent on a healthcare centre ward. In the Finnish healthcare system, healthcare centre wards are part of primary care services and provide a lower intensity of care compared to hospitals. These were used in cases when there were no severe problems in the individual’s oxygenation, ventilation or perfusion or the individual was in palliative care. Inpatient days were reported as an accumulated number of days, while all other appointments and contacts were reported as an accumulated number of events during the study period.
Covariates
Sex, age at follow-up, prescription of renoprotective medications using ATC codes in follow-up (ACE inhibitors, ARBs, SGLT2 inhibitors and GLP-1 receptor agonists), comorbidities and bioclinical variables at follow-up (systolic blood pressure [sBP], glycosylated haemoglobin [HbA1c] and low-density lipoprotein [LDL]) were identified as potential covariates. Blood pressure (BP) data consisted primarily of measurements taken by individuals at home and, when unavailable, measurements obtained by healthcare professionals during appointments. These variables play a critical role in both primary and secondary prevention of CKD, given their well-established prospective relationship with its development [13, 14]. Their selection as covariates was guided by international guidelines and the availability of corresponding measurements. The bioclinical variables sBP, HbA1c and LDL were used as dichotomised variables, following the T2D clinical practice guidelines at that time for each variable (sBP < 135 mmHg and ≥ 135 mmHg, HbA1c < 53 mmol/mol and HbA1c ≥ 53 mmol/mol, LDL < 2.5 mmol/l and ≥ 2.5 mmol/l). Individuals’ renoprotective medication data were collected from electronic health records using prescription and renewal data based on ATC codes (Supplement 3). Renoprotective medication data were divided from the study midpoint into baseline and follow-up categories, with baseline renoprotective medication data additionally including data collected 1 year before the study baseline.
Individuals with no additional chronic comorbidities besides T2D were classified as pure T2D. Those with other chronic comorbidities are split into three categories: diabetes-related comorbidities, non-diabetes-related comorbidities and combined comorbidities [20]. Conditions in the diabetes-related comorbidities category were considered pathophysiologically related to T2D, while non-diabetes-related comorbidities were not. Diabetes-related comorbidities included hypertension, atrial fibrillation, cardiovascular diseases, heart failure, hyperlipidemia and obesity. Non-diabetes-related comorbidities included musculoskeletal disorders, cancer, asthma, chronic obstructive pulmonary disease, prostate hyperplasia, hypothyroidism and dementia or Alzheimer’s disease. Combined comorbidities included individuals who had conditions from both previous categories. ICD-10 codes were used to identify comorbidities. The selection and categorisation of comorbidities were based on a previously published study from the same study population [20], except that we excluded CKD diagnoses due to the study design.
Other variables
Additional bioclinical variables collected to characterise the study population included body mass index (BMI, kg/m2), diastolic BP (mmHg) and haemoglobin (Hb, g/l). For the bioclinical variables, the baseline value was defined as the first available measurement and the follow-up value as the last available one during the study period. We gathered data on cardiovascular and diabetes medications using ATC codes. A comprehensive list of preventive medications can be found in Supplement 3. BMI was calculated by dividing the individual’s weight in kilograms by the square of their height in meters and was treated as a continuous variable. Stages of CKD according to eGFR (G1-G5) following KDIGO classification were used for descriptive purposes [19].
Statistical methods
The descriptive statistics of the study population are presented as means and SDs or medians and interquartile ranges (IQR) for continuous and count variables and as proportions and numbers for categorical variables. Annual healthcare service utilisation for eGFR reduction categories was also calculated. The Mann-Whitney U test and Pearson’s chi-square test were used to estimate statistical significance in comparative analyses between eGFR categories when appropriate.
As primary healthcare service utilisation data were overdispersed, we used a negative binomial model to analyse the association between eGFR reduction and healthcare service utilisation. We estimated the models separately for all outcomes: planned diabetes appointments, unplanned GP appointments, other GP contacts (digital, phone or letter) unplanned nurse appointments, other nurse contacts (digital, phone or letter) and inpatient days in the healthcare centre. We reported the partly adjusted (adjusted for the study period) and fully adjusted incidence rate ratios (aIRRs) with 95% confidence intervals. Fully adjusted models included sex, age, study period, comorbidities, achievement of the treatment targets for HbA1c, LDL and the use of ACEi/ARB, SGLT2i and GLP-1 RAs medication as covariates. The level of statistical significance was set at a p-value of less than 0.05. The statistical analyses were conducted using R software (version 4.3.1., http://www.r-project.org).
Results
The total study population consisted of 4773 individuals with T2D; the mean age was 70.9 years, and 53.5% of the individuals were male. The total study population characteristics can be found in Supplement 1. At baseline, the mean eGFR was 80.5 ml/min/1.73 m², which reduced to 75.5 ml/min/1.73 m² at follow-up during an average study period of 5.6 years. At baseline, 85.1% of individuals were classified in CKD classes G1-G2, but this percentage decreased to 77.0% at follow-up. At the same time, the proportion of individuals in CKD classes G3a-G3b increased from 13.8% to 18.8%. Similarly, the prevalence of CKD stages G4-G5 rose from 1.2% at baseline to 4.2% at follow-up. The median interval of serum creatinine or eGFR measurements was 371 (IQR 326–459) days. Of the study sample, 79.9% of the individuals had chronic comorbidities and 20.1% had only T2D. Medication usage was prevalent, with 82.4% of individuals using hypertensive medications and 76.8% using diabetes medications.
As shown in Table 1, compared to those in the non-substantial kidney function impairment group, individuals in the substantial kidney function impairment group were older (77.9 years vs. 69.9 years; p < 0.001) and had a lower mean eGFR (40.7 ml/min/1.73 m² vs. 80.8 ml/min/1.73 m²; p < 0.001). Other statistically significant differences in the distribution of the studied variables between these groups were noted; for example, individuals in the substantial kidney function impairment group had lower systolic BP (135.9 vs. 139.4; p < 0.001) and higher HbA1c (53.8 vs. 48.9; p < 0.001) than those in the non-substantial kidney function impairment group. Individuals in the substantial kidney function impairment group also used hypertensive medication more often than those in the non-substantial kidney function impairment group (89.2% vs. 81.3%; p < 0.001). In particular, the usage of ACEi/ARB was more frequent (71.4% in the substantial kidney function impairment group vs. 64.8% in the non-substantial kidney function impairment group, respectively; p < 0.001). Other renoprotective medication usage did not follow the same trend, with SGLT2i usage lower in the substantial kidney function impairment group than in the non-substantial kidney function impairment group (8.3% vs. 15.0%, respectively; p < 0.001), and no significant difference was observed in the usage of GLP-1 RA (3.0% vs. 3.7%, respectively; p = 0.400).
Table 1.
Characteristics of the study population, stratified by substantial and non-substantial kidney function impairment
| Population categorised by kidney function impairment | P-value | ||
|---|---|---|---|
| eGFR change < 25% | eGFR change ≥ 25% | ||
| General descriptors | |||
| n | 4144 (86.8%) | 629 (13.2%) | |
| Male, n (%) | 2228 (53.8%) | 325 (51.7%) | 0.326 |
| Female, n (%) | 1916 (46.2%) | 304 (48.3%) | |
| Age, years | 69.9 (12.3) | 77.9 (10.2) | < 0.001 |
| Study period, years | 5.5 (2.1) | 5.8 (2.0) | 0.001 |
| Bioclinical variables | |||
| eGFR in baseline, ml/min/1.73 m² | 81.8 (19.0) | 71.6 (20.1) | < 0.001 |
| eGFR in end, ml/min/1.73 m² | 80.8 (18.4) | 40.7 (16.5) | < 0.001 |
| Systolic BP, mmHg | 139.4 (20.6) | 135.9 (22.2) | < 0.001 |
| Diastolic BP, mmHg | 79.7 (11.7) | 75.8 (12.9) | < 0.001 |
| HbA1c, mmol/mol | 48.9 (13.6) | 53.8 (15.1) | < 0.001 |
| Hemoglobin, g/l | 139 (19.2) | 124 (21.7) | < 0.001 |
| BMI, kg/m² | 29.9 (5.9) | 29.6 (6.1) | 0.151 |
| LDL-cholesterol, mmol/l | 2.6 (1.0) | 2.3 (1.0) | < 0.001 |
| Comorbidities | |||
| Pure type 2 diabetes, n (%) | 854 (20.6%) | 107 (17.0%) | < 0.001 |
| Diabetes-related comorbidities, n (%) | 770 (18.6%) | 83 (13.2%) | |
| Non-diabetes-related comorbidities, n (%) | 301 (7.3%) | 36 (5.7%) | |
| Combined comorbidities, n (%) | 2219 (53.5%) | 403 (64.1%) | |
| Chronic kidney disease, n (%) | 317 (7.6%) | 157 (25.0%) | < 0.001 |
| Medications | |||
| Any diabetes medication, n (%) | 3258 (78.6%) | 504 (80.1%) | 0.389 |
| Any hypertensive medication, n (%) | 3370 (81.3%) | 561 (89.2%) | < 0.001 |
| ACEi/ARB, n (%) | 2685 (64.8%) | 449 (71.4%) | < 0.001 |
| MRAs, n (%) | 196 (4.7%) | 112 (17.8%) | < 0.001 |
| Ca-channel blockers, n (%) | 1436 (34.7%) | 281 (44.7%) | < 0.001 |
| Beta-blockers, n (%) | 1695 (40.9%) | 369 (58.7%) | < 0.001 |
| Diuretics, n (%) | 1135 (27.4%) | 415 (66.0%) | < 0.001 |
| Metformin, n (%) | 2516 (60.7%) | 303 (48.2%) | < 0.001 |
| SGLT2i, n (%) | 622 (15.0%) | 52 (8.3%) | < 0.001 |
| GLP-1 RA, n (%) | 153 (3.7%) | 19 (3.0%) | 0.400 |
| Gliptin, n (%) | 1207 (29.1%) | 295 (46.9%) | < 0.001 |
| Glinide, n (%) | 16 (0.4%) | < 5 (< 0.7%) | 0.736 |
| Glitazone, n (%) | 14 (0.3%) | < 5 (< 0.7%) | 0.936 |
| Sulphonylureas, n (%) | 77 (1.9%) | 16 (2.5%) | 0.246 |
| Any long-acting insulin, n (%) | 844 (20.4%) | 232 (36.9%) | < 0.001 |
| Multiple daily injections insulin therapy, n (%) | 464 (11.2%) | 166 (26.4%) | < 0.001 |
| Statin, n (%) | 2691 (64.9%) | 394 (62.6%) | 0.261 |
| Ezetimibe, n (%) | 261 (6.3%) | 35 (5.6%) | 0.477 |
| Fibrates, n (%) | 16 (0.4%) | 0 (0.0%) | 0.119 |
Data are presented in follow-up unless otherwise indicated. Data are presented as mean (SD), except for sex, eGFR, comorbidities and medication data, which are given as numbers and percentages. Note: p-values are calculated for the differences between the groups using the Mann-Whitney U test or χ2 test. eGFR = estimated glomerular filtration rate, BP = blood pressure, HbA1c = glycated haemoglobin, BMI = body mass index, LDL = low-density lipoprotein, ACEi = angiotensin-converting enzyme inhibitor, ARB = angiotensin II receptor blocker, SGLT2i = sodiumglucose cotransporter 2 inhibitor, GLP-1 RA = glucagon-like peptide 1 receptor agonist
Table 2 shows the eGFR reduction groups’ annual primary healthcare service utilisation. Compared to individuals in the non-substantial kidney function impairment group, individuals in the substantial kidney function impairment group had more planned diabetes appointments (mean [SD] 1.3 [1.1] vs. 1.1 [1.2]; p < 0.001), unplanned nurse appointments (1.9 [3.0] vs. 1.6 [2.4]; p = 0.007) and more digital, phone or letter contacts by GPs (3.0 [3.1] vs. 1.8 [2.2]; p < 0.001). Moreover, individuals in the impaired kidney function group also had, on average, more inpatient days on the healthcare centre ward (9.3 [15.8] vs. 8.6 [22.8], respectively; p < 0.001). Digital, phone or letter contacts by nurses occurred more frequently in the non-substantial kidney function impairment group than in the substantial kidney function impairment group (1.1 [1.1] vs. 0.9 [1.2]; p < 0.001). No significant differences were noted between the groups for unplanned GP appointments.
Table 2.
Annual primary care contacts categorised by kidney function impairment (eGFR impairment < 25% or ≥ 25%)
| Annual primary care service utilisation categorised by kidney function impairment | p-value | ||||
|---|---|---|---|---|---|
| eGFR change < 25% | eGFR change ≥ 25% | ||||
| Mean (SD) | Median (IQR) | Mean (SD) | Median (IQR) | ||
| Planned diabetes appointments | 1.1 (1.2) | 0.8 (0.3–1.5) | 1.3 (1.1) | 1.0 (0.4–1.8) | < 0.001 |
| Unplanned GP appointments | 1.1 (1.0) | 1.0 (0.3–1.7) | 1.0 (0.9) | 1.0 (0.3–1.6) | 0.504 |
| Unplanned nurse appointments | 1.6 (2.4) | 0.9 (0.4–1.8) | 1.9 (3.0) | 0.8 (0.0-2.2) | 0.007 |
| Other GP contacts (digital, phone or letter) | 1.8 (2.2) | 1.2 (0.6–2.3) | 3.0 (3.1) | 2.0 (0.9-4.0) | < 0.001 |
| Other nurse contacts (digital, phone or letter) | 1.1 (1.1) | 0.8 (0.3–1.5) | 0.9 (1.2) | 0.6 (0.0-1.3) | < 0.001 |
| Inpatient days in healthcare centre | 8.6 (22.8) | 2.4 (0.8–7.5) | 9.3 (15.8) | 4.1 (1.4–9.7) | < 0.001 |
Data are presented as the number of visits per year, except for inpatient days in the healthcare centre, which are represented as days per year. Note: p-values are calculated for the differences between the groups using the Mann-Whitney U test. Statistically significant values are marked in bold
A reduction of 25% or more in eGFR was associated with increased primary healthcare service utilisation in several outcome variables after adjusting for all potential covariates, as shown in Table 3. After full adjustment, a substantial reduction in kidney function was associated with increases in planned diabetes appointments (adjusted incidence rate ratio [aIRR] 1.18, 95% CI 1.09–1.28), other GP contacts (aIRR 1.38, 95% CI 1.57–1.84), unplanned nurse appointments (aIRR 1.19, 95% CI 1.02–1.24) and inpatient days in the healthcare centre (aIRR 1.20, 95% CI 1.10–1.45). No statistically significant associations between unplanned GP appointments and other nurse contacts were found.
Table 3.
Associations of substantial kidney function impairment with primary healthcare service utilization
| Associations of kidney function impairment with primary healthcare service utilisation by type | ||||
|---|---|---|---|---|
| Incidence rate ratio (study period only) |
95% CI | Incidence rate ratio (all covariates) |
95% CI | |
| Planned diabetes appointments | 1.18 | 1.09–1.28 | 1.18 | 1.09–1.28 |
| Unplanned GP appointments | 0.94 | 0.87–1.02 | 0.99 | 0.92–1.07 |
| Unplanned nurse appointments | 1.12 | 1.02–1.24 | 1.19 | 1.08–1.32 |
| Other GP contacts (digital, phone or letter) | 1.70 | 1.57–1.84 | 1.38 | 1.27–1.50 |
| Other nurse contacts (digital, phone or letter) | 0.90 | 0.83–0.97 | 0.99 | 0.92–1.08 |
| Inpatient days in healthcare centre | 1.26 | 1.10–1.45 | 1.20 | 1.02–1.40 |
Incidence rate ratio (all covariates) is adjusted for the study period, sex, age, comorbidities, HbA1c, LDL, ACEi/ARB, SGLT2i, and GLP-1 RAs usage. Incidence rate ratio (study period only) is adjusted only for the study period. Statistically significant incidence rate ratios are marked in bold. Partly adjusted and fully adjusted associations are based on negative binomial regression analysis
As a sensitivity analysis, we adjusted the models for sBP, which slightly attenuated the effect estimates but did not otherwise alter the overall results (see Supplement 2).
Discussion
Main findings
In this study, based on a mean follow-up time of 5.6 years, a kidney function impairment of 25% or more was associated with increased primary healthcare service utilisation in terms of planned diabetes appointments, unplanned nurse appointments, other GP contacts and inpatient days in the healthcare centre. The association remained significant after adjusting for follow-up time, sex, age, bioclinical variables, comorbidities and renoprotective medication usage. Additional adjustment for sBP did not change the results, even though the effect estimates were slightly attenuated.
The increase in planned diabetes appointments and other GP contacts can be attributed to increased treatment and monitoring of kidney function by GPs, in line with national and international guidelines. However, the noted increase in unplanned nurse appointments and inpatient days was most likely not directly related to kidney function monitoring or treatment per se. This suggests that the comprehensive increase in healthcare service utilisation in the substantial kidney function impairment group cannot be fully explained by the expected measures taken by healthcare professionals. The increases in unplanned nurse appointments and inpatient days may be attributable to multiple factors, such as increases in related complications associated with kidney function impairment and the exacerbation of comorbid conditions. It is noteworthy that among individuals in the substantial kidney function impairment group, approximately 25% progressed to CKD stages 4–5 (data not shown). This decline may have led to a transition from primary health care to specialised care, such as nephrology. As our dataset includes only visits recorded in primary health care, this shift could partly explain why no increase in unplanned GP visits was observed. Consequently, the total number of health care contacts may be underestimated in our study. Overall, these findings aim to raise awareness of the broader implications that kidney function decline has on healthcare systems, particularly in light of the current economic pressures facing health services in many developed countries, including Finland.
The observed eGFR decline in the substantial kidney function impairment group is likely associated with older age, more combined comorbidities, being treated more often with long-acting or multiple daily injections of insulin, as well as having a higher average HbA1c, suggesting a more advanced T2D compared to those in the non-substantial kidney function impairment group. Surprisingly, individuals in the non-substantial kidney function impairment group experienced, on average, a lower than expected decline in eGFR (0.18 ml/min/1.73 m²/year), despite having higher blood pressure and more diabetes-related comorbidities than the substantial kidney function impairment group. Individuals in this group were more frequently treated with SGLT2i, but reciprocally less frequently with ACEi/ARB, both of which have been shown to slow down the progression of CKD [12, 13]. We speculate that this observed effect could be attributed to factors not captured in our data, such as duration and phenotype of T2D, younger age of the patients and the increased usage of SGLT2i in the group.
Previous studies
To the best of our knowledge, there are no previous studies on kidney function impairment and primary healthcare service utilisation. Most studies on costs or healthcare service utilisation related to CKD have been conducted mainly on the treatment of advanced CKD and ESRD in speciality clinics, or have explored healthcare service utilisation in different CKD stages [15–18, 21]. One recent cohort study examining the relationship of comorbidities in older individuals with T2D with primary healthcare service utilisation showed an increase in the utilisation of GP and nurse appointments with an increase in comorbidities, which included advanced CKD [22]. Another large cohort study in the UK showed that comorbidities (including CKD) and polypharmacy increased healthcare service utilisation [23]. Two studies found that T2D increased healthcare service utilisation compared to individuals without T2D, but they did not analyse the effects of CKD [24, 25]. A systematic review and meta-analysis done in 2020 found that more advanced CKD stages are significantly associated with increased healthcare costs, longer hospital stays and increased cardiovascular morbidity and mortality [26]. One large U.S. claims database study consisting of individuals with T2D and newly recognized CKD found increased health care service utilisation in several categories at more advanced CKD stages [21]. These results are in line with ours; however, the present study contributes additional insight by examining the association of primary health care service utilisation with kidney function impairment irrespective of prior kidney function status or CKD diagnosis, thereby capturing a broader and clinically relevant patient population in a real-world setting.
Primary and secondary prevention of CKD
Given the high prevalence of CKD in individuals with T2D, managing modifiable risk factors is fundamental for effective primary prevention. If unsuccessful, secondary prevention by further addressing modifiable risk factors and initiating renoprotective medication can significantly slow the progression of the disease and delay the development of ESDR by several years [12]. As demonstrated by our findings, preventing kidney function impairment through primary prevention or early intervention not only slows the progression of CKD but also reduces the burden of primary healthcare. The time point at which the data were collected (between November 1, 2011, and February 19, 2019) most likely explains the low levels of SGLT2i and GLP1 RA prescriptions because the effect of SGLT2i and GLP-1 RAs in the treatment of CKD was not yet fully acknowledged at the time, and the guidelines even discouraged the usage of the aforementioned medications in individuals with impaired kidney function [27]. This has since changed, with international guidelines recommending SGLT2i as the first-in-line therapy for CKD, along with ACEi/ATRb, with GLP-1 RAs the second-in-line therapy [12].
Primary healthcare service utilisation
When assessing healthcare service utilisation from a primary care standpoint, it should be noted that primary healthcare operates differently depending on the country in question. In Finland, GPs work in publicly funded healthcare centres, while in the UK, France, Norway and Denmark, GPs are self-employed [28]. This leads to differences in the management of individuals. In the Finnish healthcare system, the main responsibility for the screening, diagnosis and treatment of T2D and its complications lies in primary care. Individuals with advanced CKD (eGFR < 30 ml/min/1.73 m2) are referred to nephrology, while the treatment of T2D and other comorbidities, including earlier stage CKD, are managed in primary care. Finnish primary care treats individuals with multimorbidities and polypharmacy and individuals on multiple daily injections of insulin therapy, while in other countries, these individuals may be referred to speciality clinics. Finnish nurses manage appointments independently and consult GPs if necessary. Furthermore, having primary care inpatient wards is a special feature of the Finnish healthcare system and may correspond to hospital care in other countries. These differences in individual management should be taken into account when evaluating the applicability of these results to primary healthcare in other countries.
Strengths and weaknesses
The main strengths of this registry-based study are its inclusion of a large majority of T2D individuals, well-representing T2D individuals in the region, as well as the longitudinal follow-up design in a real-world primary care setting. The study data were obtained from the electronic health registry, which encompasses all available information on individuals in primary care, including all contacts by the individual to primary care. The samples from individuals were processed by the same laboratory, allowing the results to be compared without the risk of possible discrepancies. Data on individual smoking status or history, physical activity, socioeconomic status, medication purchase information (medication data was based on prescriptions) and mortality were not available for study, and these can be considered study limitations. Additionally, insufficient data on urine albumin measurements restricted their inclusion in this study. Despite the longitudinal study design, causality cannot be confirmed.
Conclusions
Our novel finding regarding the association between kidney function impairment and increased primary healthcare service utilisation reinforces the importance of primary prevention, regular screening, timely identification and prompt treatment of CKD in individuals with T2D to prevent CKD progression and reduce the burden on primary healthcare. Future studies should address the cost-effectiveness and the role of new renoprotective medications in relation to primary health care utilisation among individuals with T2D.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Acknowledgements
Not applicable.
Abbreviations
- aIRR
Adjusted incidence rate ratio
- ACEi
Angiotensin-converting enzyme inhibitor
- ARB
Angiotensin II receptor blocker
- MRA
Mineralocorticoid receptor antagonist
- BMI
Body mass index
- CKD
Chronic kidney disease
- CKD-EPI
Chronic Kidney Disease Epidemiology Collaboration
- eGFR
Estimated glomerular filtration rate
- ESRD
End-stage renal disease
- GLP-1 RA
glucagon-like peptide 1 receptor agonist
- GP
General practitioner
- HbA1c
Glycosylated hemoglobin
- ICD-10
The International Classification of Diseases 10th Revision
- ICPC
International Classification of Primary Care
- KDIGO
Kidney Disease: Improving Global Outcomes
- LDL
Low-density lipoprotein
- sBP
Systolic blood pressure
- SGLT2i
Sodium-glucose cotransporter 2 inhibitor
- T2D
Type 2 diabetes
Author contributions
HS, MH, JJ, IM, IP and EH contributed to the concept and design of the study. All authors contributed to the acquisition, analysis, or interpretation of the data. HS prepared the first draft. MH, JR, IM, IP and EH critically revised the manuscript. JJ performed the statistical analysis. All authors read and approved the final manuscript.
Funding
Open Access funding provided by University of Oulu (including Oulu University Hospital). This study has received funding from the Sakari Alhopuro Foundation, the Lapland Regional Fund of the Finnish Cultural Foundation and the Finnish Kidney Foundation.
Data availability
Restrictions apply to the availability of the data, which were used under license for the current study, and are not publicly available. Data are, however, available from the corresponding author upon reasonable request and with permission from the wellbeing services county of Lapland.
Declarations
Ethics approval and consent to participate
The study was approved by the Ethics Committee of the Lapland Central Hospital, Rovaniemi, Finland (Reg.no. 05/2018). Informed consent to participate from individuals is not required in a registry-based study according to Finnish legislation (Personal Data Act 523/1999). All methods were carried out following relevant guidelines and regulations. The access to the data used in the present study was authorised by the administrative senior physician of the city of Rovaniemi, Lapland, Finland. Only pseudonymised data were used.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, et al. IDF diabetes atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183. 10.1016/j.diabres.2021.109119 [DOI] [PMC free article] [PubMed]
- 2.International Diabetes Federation. IDF Diabetes Atlas. 9th edn. Brussels. Belgium: 2019. International Diabetes Federation; 2019. Available from: https://www.diabetesatlas.org
- 3.Thomas MC, Cooper ME, Zimmet P. Changing epidemiology of type 2 diabetes mellitus and associated chronic kidney disease. Nat Rev Nephrol. 2016;12:73–81. 10.1038/nrneph.2015.173. [DOI] [PubMed] [Google Scholar]
- 4.Finnish Registry for Kidney Diseases. (national quality register) 2012–2023 report, the national institute for health and welfare, Finland. Available from: https://www.thl.fi/kansallisten-laaturekisterien-raportit/munuaistautirekisteri/
- 5.Harding JL, Pavkov ME, Magliano DJ, Shaw JE, Gregg EW. Global trends in diabetes complications: a review of current evidence. Diabetologia. 2019;62:3–16. 10.1007/s00125-018-4711-2. [DOI] [PubMed] [Google Scholar]
- 6.Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. Lancet. 2020;396:1204–22. 10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Metsärinne K, Bröijersen A, Kantola I, Niskanen L, Rissanen A, Appelroth, Tina, et al. High prevalence of chronic kidney disease in Finnish patients with type 2 diabetes treated in primary care. Prim Care Diabetes. 2015;9:31–8. 10.1016/j.pcd.2014.06.001. [DOI] [PubMed] [Google Scholar]
- 8.Koro CE, Lee BH, Bowlin SJ. Antidiabetic medication use and prevalence of chronic kidney disease among patients with type 2 diabetes mellitus in the united States. Clin Ther. 2009;31:2608–17. 10.1016/j.clinthera.2009.10.020. [DOI] [PubMed] [Google Scholar]
- 9.Cid Ruzafa J, Paczkowski R, Boye KS, Di Tanna GL, Sheetz MJ, Donaldson R, et al. Estimated glomerular filtration rate progression in UK primary care patients with type 2 diabetes and diabetic kidney disease: a retrospective cohort study. Int J Clin Pract. 2015;69:871–82. 10.1111/ijcp.12640. [DOI] [PubMed] [Google Scholar]
- 10.Hagnäs M, Sundqvist H, Jokelainen J, Tumminia A, Vinciguerra F, Loreto C, et al. The prevalence of chronic kidney disease and screening of renal function in type 2 diabetic patients in Finnish primary healthcare. Prim Care Diabetes. 2020;14:639–44. 10.1016/j.pcd.2020.05.005. [DOI] [PubMed] [Google Scholar]
- 11.Rodriguez-Poncelas A, Garre-Olmo J, Franch-Nadal J, Diez-Espino J, Mundet-Tuduri X. Prevalence of chronic kidney disease in patients with type 2 diabetes in spain: PERCEDIME2 study. BMC Nephrol. 2013;14:8. 10.1186/1471-2369-14-46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rossing P, Caramori ML, Chan JCN, Heerspink HJL, Hurst C, Khunti K, et al. KDIGO 2022 clinical practice guideline for diabetes management in chronic kidney disease. Kidney Int. 2022;102:S1–127. 10.1016/j.kint.2022.06.008. [DOI] [PubMed] [Google Scholar]
- 13.ElSayed NA, Aleppo G, Aroda VR, Bannuru RR, Brown FM, Bruemmer D, et al. 11. Chronic kidney disease and risk management: standards of care in Diabetes—2023. Diabetes Care. 2023;46:S191–202. 10.2337/dc23-S011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sundqvist H, Heikkala E, Jokelainen J, Russo G, Mikkola I, Hagnäs M. Association of renal function screening frequency with renal function decline in patients with type 2 diabetes: a real-world study in primary health care. BMC Nephrol. 2022;23:356. 10.1186/s12882-022-02979-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hex N, Bartlett C, Wright D, Taylor M, Varley D. Estimating the current and future costs of type 1 and type 2 diabetes in the UK, including direct health costs and indirect societal and productivity costs. Diabet Med. 2012;29:855–62. 10.1111/j.1464-5491.2012.03698.x. [DOI] [PubMed] [Google Scholar]
- 16.Li R, Bilik D, Brown MB, Zhang P, Ettner SL, Ackermann RT, et al. Medical costs associated type 2 diabetes complications comorbidities. 2015;16. [PMC free article] [PubMed]
- 17.McEwan P, Hafner M, Jha V, Correa-Rotter R, Chernin G, De Nicola L, et al. Translating the efficacy of Dapagliflozin in chronic kidney disease to lower healthcare resource utilization and costs: a medical care cost offset analysis. J Med Econ. 2023;26:1407–16. 10.1080/13696998.2023.2264715. [DOI] [PubMed] [Google Scholar]
- 18.Parker ED, Lin J, Mahoney T, Ume N, Yang G, Gabbay RA, et al. Economic costs of diabetes in the U.S. in 2022. Diabetes Care. 2023;dci230085. 10.2337/dci23-0085. [DOI] [PubMed]
- 19.KDIGO Guideline. Kidney Disease: Improving Global Outcomes (KDIGO) CKD work Group. KDIGO 2012 clinical practice guideline for theevaluation and management of chronic kidney disease. Kidney inter. Off J Int Soc Nephrol. 2013;3(Suppl):1–150. 10.1038/kisup.2012.73.
- 20.Heikkala E, Mikkola I, Jokelainen J, Timonen M, Hagnäs M. Multimorbidity and achievement of treatment goals among patients with type 2 diabetes: a primary care, real-world study. BMC Health Serv Res. 2021;21(1):964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Folkerts K, Petruski-Ivleva N, Kelly A, Fried L, Blankenburg M, Gay A, et al. Annual health care resource utilization and cost among type 2 diabetes patients with newly recognized chronic kidney disease within a large U.S. Administrative claims database. JMCP. 2020;26(12):1506–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Mellanen E, Kauppila T, Kautiainen H, Lehto M, Rahkonen O, Pitkälä K, et al. Use of primary health care services and mortality in older patients with type 2 diabetes with or without comorbidities. Scand J Prim Health Care. 2023;41:392–9. 10.1080/02813432.2023.2255062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hodgson S, Morgan-Harrisskitt J, Hounkpatin H, Stuart B, Dambha-Miller H. Primary care service utilisation and outcomes in type 2 diabetes: a longitudinal cohort analysis. BMJ Open. 2022;12:e054654. 10.1136/bmjopen-2021-054654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.O’Neill KN, McHugh SM, Tracey ML, Fitzgerald AP, Kearney PM. Health service utilization and related costs attributable to diabetes. Diabet Med. 2018;35:1727–34. 10.1111/dme.13806. [DOI] [PubMed] [Google Scholar]
- 25.Aro A-K, Karjalainen M, Tiihonen M, Kautiainen H, Saltevo J, Haanpää M, et al. Use of primary health care services among older patients with and without diabetes. BMC Prim Care. 2022;23:233. 10.1186/s12875-022-01844-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Darlington O, Dickerson C, Evans M, McEwan P, Sörstadius E, Sugrue D, et al. Costs and healthcare resource use associated with risk of cardiovascular morbidity in patients with chronic kidney disease: evidence from a systematic literature review. Adv Ther. 2021;38(2):994–1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.American Diabetes Association. Standards of medical care in Diabetes—2018 abridged for primary care providers. Clin Diabetes. 2018;36:14–37. 10.2337/cd17-0119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kringos DS, Boerma WGW, Hutchinson A, Saltman RB, editors. Building primary care in a changing Europe. Copenhagen (Denmark): European Observatory on Health Systems and Policies; 2015. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Restrictions apply to the availability of the data, which were used under license for the current study, and are not publicly available. Data are, however, available from the corresponding author upon reasonable request and with permission from the wellbeing services county of Lapland.