WHAT IS ALREADY KNOWN ON THIS TOPIC

Non-diabetic hyperglycemia (NDH) represents a pre-diabetic state, with blood sugar higher than the normal range but not high enough to be formally termed type 2 diabetes mellitus (T2DM).

The prevalence of NDH has been shown to be increasing since the start of the millennium, with 2.7 million people in England identified to have NDH in 2021–2022 compared with 3.2 million identified to have T2DM.

Recent estimates suggest that, compared with 2015, 25% more adults will be diagnosed with T2DM in 2030, with an expected co-current rise in the number of individuals suffering from microvascular and macrovascular complications, both of which are directly correlated with suboptimal glycemic control.

WHAT THIS STUDY ADDS

Both NDH and T2DM were associated with significant comorbidities.

Of note were the high rates of hypertension, hypercholesterolemia, and obesity among individuals with NDH, with approximately 40% of individuals affected by at least one of the three, further highlighting the need for earlier detection of NDH and implementation of interventions to prevent progression to overt T2DM.

High service utilisation in primary and secondary care among individuals with both NDH and T2DM further highlights the need for early diagnosis and management to aid the prevention of associated chronic health conditions.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

This study highlights the significant morbidity associated with both NDH and T2DM, and the need for earlier identification of individuals with NDH to ensure early referral to the NHS Diabetes Prevention Program and the prevention/early recognition of multimorbidity associated with both NDH and T2DM.

Introduction

Non-diabetic hyperglycemia (NDH) represents a pre-diabetic state, with blood sugar higher than the normal range but not high enough to be formally termed type 2 diabetes mellitus (T2DM). The early recognition of NDH allows healthcare professionals and individuals alike with an opportunity to prevent the conversion from NDH to T2DM, a condition associated with increased morbidity and early mortality on account of its associated microvascular and macrovascular complications.1 In North West London (NWL), a lack of consistency in the care of individuals with NDH representing those at highest risk of future T2DM was recognised as early as 2010, with significant improvements observed after the formation of the North West London (NWL) Diabetes Transformation Program (NWL-DTP) in 2015 and the NHS Diabetes Prevention Program (NDPP).2 Similar improvements have been seen in other regions on account of an enhanced service for NDH detection and proactive management now increasingly common particularly after the development of the NDPP, local programs such as NWL-DTP, as well as the introduction of a NDH register within the Quality and Outcomes Framework (QOF) since 2020.3 Despite these improvements, there remains a lack of real-world data on the long-term impact of NDH on an individual’s health and the wider healthcare system in comparison to T2DM.4 We therefore used electronic health records (EHRs) from adults in NWL, available via the Discover-NOW dataset, to better appreciate the relationship between NDH/T2DM and disease states, healthcare usage, and clinical outcomes.

The prevalence of NDH has been shown to be increasing since the start of the millennium, with considerable variation seen across the UK. As per the NDH, 2021–2022, Diabetes Prevention Program, Detailed Analysis Report, 2.7 million people in England had NDH (defined according to National Institute for Health and Care Excellence (NICE) guidelines as a glycated haemoglobin (HbA1c) between 42 and 47 mmol/mol, 6.0–6.4%) between 2021 and 2022.5 With an increase of 776,790 individuals diagnosed with NDH between 2022 and 2023, most recent data show that there are 3.2 million people in England living with NDH, accounting for 6.4% of the adult population. A further 3.4 million (6.8%) people in England were recorded to have type 2 diabetes (defined as per NICE guidelines as a HbA1c >48 mmol/mol,6.5%) as opposed to 3 million people in 2017. The prevalence of NDH is estimated to be greater when using the American Diabetes Association (ADA) definition of NDH (39–47 mmol/mol, 5.7–6.4%), with estimates identified using Health Survey for England data showing an increase in the prevalence of NDH, as defined by ADA, in the UK from 11.6% in 2003 to 35.5% in 2011.6

The risk factors for T2DM and NDH are similar for individuals aged over 40, of South Asian ethnicity, overweight with a body mass index (BMI) greater than 25 kg/m2, and a history of hypertension are at greatest risk. While numerous diagnostic criteria exist, clinicians within the UK diagnose NDH based on a HbA1c between 42 and 47 mmol/mol (6.0–6.4%), as per NICE guidance published in 2012, following the WHO’s recommendation to use HbA1c to diagnose T2DM in 2011.7 8 With an increased appreciation for the existence of a pre-diabetic state, the recorded prevalence of NDH particularly since the formation of NDPP has improved. However, while NDH is now increasingly recognised, its association with cardiovascular disease (CVD) and other disease states traditionally associated after the onset of T2DM is poorly understood. This is of particular concern given the globally expected rise in NDH prevalence to approximately 470 million by 2030.9 10

Recent estimates suggest that 25% more adults will be diagnosed with T2DM in 2030, compared with 2015, with an expected co-current rise in the number of individuals suffering from microvascular and macrovascular complications, both of which are directly correlated with suboptimal glycemic control. This is likely to result in a profound reduction in quality and quantity of life among individuals with T2DM due, in many instances, to irreversible end-organ damage.11

While programs such as the local NWL-DTP and national NDPP have been successful in aiding individuals with NDH to live healthier lives and thus prevent its progression to T2DM, limited real-world data on the long-term complications of NDH and their impact on primary and secondary care services are available. This results in NDH being poorly understood and less well recognised as a clinical entity associated with significant morbidity, limiting monitoring, counseling, and referral to DM prevention programs. We therefore used the Discover London Secure Data Environment (SDE) dataset to determine the profile of individuals diagnosed with NDH in comparison with T2DM, in NWL, an area of high ethnic variability and prevalence of T2DM to better appreciate the relationship between NDH/T2DM and disease states, healthcare usage, and clinical outcomes.

Research design and methods

The Discover London SDE dataset is an anonymised linked dataset of over 2.8 million primary care registered individuals in NWL with linked data available from social, community, and mental healthcare services in addition to primary and secondary care from 2015 onward.12 While smaller in size when compared with the Clinical Practice Research Datalink, it is more ethnically diverse while being comparable in terms of overall age, gender distribution, and chronic disease prevalence in the UK.13

The Discover London SDE dataset was retrospectively analysed between January 2015 and December 2020. The dataset was accessed and analysed by data analysts (ZU-H, MS, TK, and BP). Using a retrospective cross-sectional study design, data were collected to understand the profiles of individuals with NDH/T2DM and the relationship between NDH/T2DM and disease states: hypertension, hypercholesterolemia, diabetic foot disease, diabetic eye disease, neuropathy, CVD, diabetic kidney disease (DKD), chronic kidney disease (CKD) stages 3–5, albuminuria >3 mg/mmol, obesity, and depression, as well as healthcare usage (primary care/accident and emergency care/inpatient care/outpatient care usage). First diagnosis recorded in the study period for each indication was used for each unique patient. Disease states were identified based on individuals being coded in primary or secondary care using International Classification of Diseases 10th Revision (ICD-10) or Read V.2 codes between January 2015 and December 2020 (Online supplemental file 1). The Discover London SDE is accessible via Discover-NOW Health Data Research Hub for Real World Evidence through their data scientist specialists and information governance committee-approved analysts, hosted by Imperial College Health Partners. Discover-NOW have secured Health Research Authority approval, with researchers not required to seek further ethical approval for use of the Discover London SDE dataset for research purposes for studies submitted to and approved by the NWL Data Access Committee (REC reference: 18/WM/0323; IRAS project ID: 253449).

Individuals aged 18 years of age and over in NWL with a diagnosis of NDH or T2DM, for whom there were computerised records within the Discover London SDE dataset from January 1, 2015 were included. An unmatched control group of patients with thyroid disease without NDH or T2DM was included for comparison, as a group of individuals with an alternative chronic disease not recognised to be causally related to NDH or T2DM. The first diagnosis recorded within the study period for each disease state of interest was recorded for each unique individual. Those with an ever diagnosis of type 1 diabetes mellitus, gestational diabetes, or drug-induced diabetes were excluded. In addition, those diagnosed at >35 years of age, with continual insulin treatment from diagnosis (defined as record of insulin prescription within 3 months of diagnosis), and those diagnosed at <35 years of age, with commencement of continual insulin treatment within 6 months of diagnosis (defined as record of insulin prescription within 6 months of diagnosis), were also excluded due to likelihood of type 1 diabetes, as per the Royal College of General Practitioners 2010 UK Practical Classification Guidelines.13 14

For each of the study cohorts (control, NDH and T2DM), key demographic data, namely, age, ethnicity, gender, and Index of Multiple Deprivation (IMD) were collated. The activity of all subjects was subsequently evaluated by capturing the first diagnosis of thyroid disease, NDH, or T2DM in the study period as index date for each unique subject and mapping the development of morbidity during the study period, alongside the medications used to treat accumulated morbidity. Subject usage of primary, secondary care, and social care services, by means of the number of contacts per individual, were collated to determine healthcare resource utilisation among each of the study groups. Pearson’s χ2 test or Kruskal-Wallis rank-sum test was used to determine statistical significance between outcomes or the utilisation of healthcare services for individuals in the three cohorts.

ResultsStudy cohort characteristics

We identified 152,384 individuals (80,045 female; 71,883 male; 506 unknown) with NDH between January 1, 2015 and December 31, 2020, coded with NDH from 2015 to 2019, with a 61.2% decrease in 2020 compared with the year before, likely related to the COVID-19 pandemic resulting from a reduction in routine appointments and screening for DM. Comparatively, 124,190 individuals (54,579 female; 69,130 male; 480 unknown) were diagnosed with T2DM in the study period, with a 55.8% decrease in the number of newly diagnosed in the year January–December 2020 compared with the year before. The control group consisting of patients with thyroid disease without diabetes consisted of 11,626 individuals (9000 female; 2612 male; 14 unknown), with a 37.6% decrease in the number of individuals diagnosed in the year January–December 2020 compared with 2019. Median age of diagnosis was 42 (control cohort), 57 (NDH), and 60 (T2DM) years. The NDH and T2DM groups most commonly of Asian or Asian British ethnicity (42% and 48%, respectively), with the control group most commonly of white Caucasian ethnicity (51%). IMD for all three cohorts were broadly similar, with 51% (control group), 54% (NDH), and 52.7% (T2DM) of individuals in each cohort within IMD 3–7 (table 1).

Table 1

Demographic data for all three study cohorts: control, NDH, and T2DM

Morbidity associated with NDH and T2DM

Individuals with both NDH and T2DM had a high burden of disease, with only 13.1% of individuals with either NDH or T2DM not found to be suffering from at least one of the comorbidities of interest. The three most common comorbidities experienced by individuals with NDH were hypertension (41.4%), hypercholesterolemia (37.5%), and obesity (29.8%), compared with diabetic retinopathy (68.7%), hypertension (59.4%), and obesity (45.8%) in individuals with T2DM. Comparatively, the most common comorbidities in the control group were depression (30.8%), hypercholesterolemia (24.4%), and hypertension (17.1%). A high burden of CVD, hypercholesterolemia, and hypertension was noted in both cohorts with 14.7%, 37.5%, and 41.4% of individuals with NDH, respectively, compared with 23.5%, 42.0%, and 59.4% of individuals with T2DM. In addition, a high proportion of individuals with NDH or T2DM were identified as obese (29.8% and 45.8%, respectively, p<0.001)) versus 20.5% of individuals in the control group (p<0.001), further highlighting the association between NDH and T2DM with obesity and subsequent metabolic complications. A lower burden of CVD (6.3%), hypercholesterolaemia (24.4%), and hypertension (17.1%) was noted among individuals in the control group (table 2).

Table 2

Patient comorbidities for each three of the study cohorts: control, NDH, and T2DM

CKD was observed among both individuals with NDH and T2DM, with 4970 unique individuals (3435 coded for CKD 3 to CKD 5, 1535 uncoded with an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2; in total, representing 3.3% of total NDH cohort) with NDH and 5659 unique individuals (4369 coded for CKD 3 to CKD 5, 1290 uncoded with an eGFR <60 mL/min/1.73 m2; in total, representing 4.6% of total T2DM cohort). This was associated with 8418 (5.5%) individuals with NDH and 23,743 (19.1%) individuals with T2DM noted to have an urine albumin-creatinine ratio (uACR) >3 mg/mmol, signifying, as already recognised, albuminuria as an early indicator of CKD on an account of eGFR preservation secondary to glomerular hyperfiltration. However, for the NDH and T2DM cohorts, this did not translate into individuals being coded with DKD, with only 71 (0.05%) and 1884 (1.5%) individuals coded to have DKD. Comparatively, 257 (2.2%) individuals in the control group had an uACR >3 mg/mmol (table 2).

Diabetic retinopathy and diabetic foot disease were more commonly associated with T2DM (68.7% and 26.7%, respectively) compared with 8.8% and 2.1%, respectively, in individuals with NDH. Comparatively, 1.4% and 0.9% of individuals within the control group were coded for diabetic retinopathy and diabetic foot disease, respectively (table 2).

A statistically significant difference was noted for 10 out of 11 investigated disease states when comparing the control group and NDH cohorts, with no significant difference noted in the degree of DKD between the two cohorts (p=0.5). A difference was noted in the degree of CKD stages 3–5 and albuminuria, suggesting for the NDH cohort to have undocumented/under-recognised DKD or CKD secondary to an underlying disease state other than T2DM. Comparatively, a statistically significant difference between each of the 11 investigated disease states was noted when comparing the control cohort versus T2DM and NDH versus T2DM cohorts (table 3).

Table 3

Patient comorbidities comparing control cohort vs NDH cohort; control cohort vs T2DM cohort; NDH vs T2DM cohort

Primary care utilisation

Individuals aged 60–69 years were the highest service users. Overall, over the 5-year study period, 1,640,186 (control group) vs 9,221,379 (NDH) vs 9,891,479 (T2DM) primary care appointments were identified, representing 28 (control group) vs 12 (NDH) vs 16 (T2DM) contacts per individual per year (table 4). Furthermore, 577,043 (control group) vs 5,599,019 (NDH) vs 7,087,084 (T2DM) individual prescriptions were made over the 5-year study period, accounting for 10 (control group) vs 8 (NDH) vs 11 (T2DM) prescriptions per individual per year. Further, 77,952 (control group) vs 829,589 (NDH) vs 894,172 (T2DM) outward referrals were made from primary care, accounting for 7 (control group) vs 5 (NDH) vs 7 (T2DM) referrals per individual over the 5-year study period, with hospital admission, referral to community musculoskeletal service, referral to radiology services, referral to district nurse, and referral to diabetes structured education program representing the top five outward referrals among the two study groups. Among all referrals, only 11% and 54% of individuals with NDH and T2DM, respectively, were referred to a diabetes’s structured education program.

Table 4

Five-year primary care, outpatient, inpatient, and accident and emergency unit utilisation for each of the three study cohorts: control, NDH, and T2DM

Outpatient service utilisation

Outpatient service utilisation for both NDH and T2DM cohorts was high, with 141,302 (NDH) vs 161,680 (T2DM) contacts over the 5-year period compared with 19,627 contacts over the 5-year period for the control cohort (table 4). The most common cause of outpatient visits was attendance to ophthalmology, radiology, and cardiology services for all three study cohorts.

Inpatient admission utilisation

Inpatient admission utilisation of both NDH and T2DM cohorts was high, with 282,371 and 314,880 inpatient admissions for the NDH and T2DM cohorts, respectively, compared with 27,881 inpatient admissions for the control cohort (table 4). This was associated with a length of stay of 53,129 (control cohort) vs 531,708 (NDH) vs 803,365 (T2DM) days, resulting in 1.9 (control cohort) vs 1.9 (NDH) vs 2.6 (T2DM) days per admission (table 5). The most common cause of admission for all three study cohorts were lobar pneumonia, urinary tract infection, and congestive heart failure.

Table 5

Prescribing of medications for diabetes for each of the three study cohorts: control, NDH, and T2DM

Accident and emergency utilisation

Accident and emergency (A&E) attendances for both NDH and T2DM cohorts was high, with 358,234 and 342,006 attendances, respectively. This compared with 33,201 attendances among the control cohort (table 4). The most common cause for A&E attendance for both NDH and T2DM cohorts after “no diagnosis” were gastrointestinal, cardiac, and respiratory conditions as opposed to gastrointestinal, respiratory, and ophthalmological conditions for the control cohort.

Prescription of medications for diabetes

Prescription of drugs used to treat diabetes in individuals with NDH and T2DM was 27,772 (18.2%) and 109,361 (88.1%), respectively. This compared with <5 individuals from the control group (table 5).

Discussion

In our study, a higher number of individuals were diagnosed with NDH as opposed to T2DM over the 5-year study period. NDH was seen to be more common in women, with T2DM being more common in men. The number of individuals diagnosed with NDH increased year on year, from 2015 to 2019, with an expected decrease in new diagnoses in 2020 likely due to the COVID-19 pandemic and a change in the way routine non-urgent care was managed in both primary and secondary care settings. While the rate of conversion was not specifically measured within this study, 27,722 unique individuals with NDH were prescribed medications for the management of diabetes mellitus (DM) over the 5-year study period. This crudely accounts for 18.2% of individuals diagnosed with NDH during the study period subsequently developing T2DM. This is in accordance with Ravindrarajah et al who suggested a 5-year conversion rate of 23%.4 This high rate of conversion from NDH to T2DM, alongside estimates suggesting 10% of adults within the UK to be diabetic by 2030, highlights the need for earlier identification of individuals with NDH to not only reduce the rate of conversion of NDH to T2DM but also allow for earlier interventions aimed at delaying/preventing the onset of complications associated with impaired glycemic control.15

In NWL, the NWL-DTP and NDPP have been highly successful in improving NDH detection and allowing for proactive management in primary care.2 This has been achieved through strategies aimed at (1) patient empowerment, (2) clinician education, (3) engaging the wider multidisciplinary team, and (4) optimising information technology (IT) systems across NWL. Between 2015 and 2018, innovations implemented via the NWL-DTP resulted in the NDH register increasing by 62,491 individuals from 9071 in 2015 to 71,562 in 2018. Furthermore, significant improvements in HbA1c, blood pressure, and hyperlipidemia management were achieved as were the number of referrals to the NDPP, highlighting the significant improvements in care capable of being achieved through clinical innovations and collaborative working across disciplines.2

Both NDH and T2DM were associated with significant comorbidities. Of note were the high rates of hypertension, hypercholesterolemia, and obesity among individuals with NDH, with approximately 40% of individuals affected by at least one of the three, further highlighting the need for earlier detection of NDH and assessment of disease profiles classically associated with T2DM to ensure the implementation of interventions to prevent progression of NDH to overt T2DM as well as the development of multimorbidity. This was particularly evident by the increase in the proportion of individuals developing organ-specific diseases, including CVD, diabetic eye complications, and diabetic foot disease among individuals with both NDH and T2DM. The control cohort, while displaying a lower degree of morbidity compared with the NDH and T2DM cohorts, had high rates of depression (31%), hypercholesterolemia (24%), and obesity (20%), while also having an average length of stay per inpatient admission (days) comparable to the NDH cohort. The control cohort therefore needs further investigation to gauge the demographics most commonly associated with the NDH and T2DM cohorts.

While the expected prevalence of DKD in individuals with T2DM ranges between 30% and 50%, this is most frequently only observed clinically after prolonged periods of DM, with the UK Prospective Diabetes Study (UKPDS) showing clinically identifiable DKD in the form of albuminuria and an eGFR <60 mL/min/1.73 m2 at a median of 15 years after the onset of T2DM.15–18 While the total duration of T2DM is unknown for the 52% of individuals within the T2DM cohort, who were diagnosed with T2DM prior to 2015, with 48% of individuals within the T2DM study cohort having T2DM for less than 5 years, the low proportion of formally coded CKD/DKD is likely skewed by the cohort consisting of individuals with T2DM of varied durations. However, we identified a lower proportion of individuals with formally coded DKD in both the NDH and T2DM cohorts compared with the number of individuals with either: CKD 3–5 (coded or uncoded within EHRs) or albuminuria with an uACR >3 mg/mmol. While 4970 (NDH) and 5659 (T2DM) individuals were identified to have CKD stages 3–5 through coding for CKD 3–5 or the identification of eGFR values <60 mL/min/1.73 m2, 8418(NDH) and 23,743 (T2DM) individuals had a uACR >3 mg/mmol, signifying nephropathy, and a greater number of individuals in NWL living with DKD than formally identified. This further highlights the need to (1) improve the screening of CKD by assessment of both eGFR and albuminuria in individuals with NDH and T2DM, (2) widen the recognition of albuminuria as an early indicator of CKD particularly in individuals with eGFR preservation secondary to glomerular hyperfiltration, and (3) improve the formal coding of CKD within EHRs.

Identification of a higher prevalence of diabetic retinopathy in the NDH cohort as opposed to the control group suggests an association between NDH and microvascular disease. 8.8% of the NDH cohort was identified to have diabetic retinopathy as opposed to 1.4% within the control cohort. In a systematic review by Kirthi et al, the prevalence of retinopathy among individuals with NDH was shown to be 6.6% (IQR 1.9–9.8%) compared with individuals with normoglycemia which had a prevalence of 3.2% (IQR 0.3–7.3%).19 In this study, diabetic retinopathy included all forms of diabetic retinopathy including background, pre-proliferative, proliferative retinopathy, and maculopathy. As a result, we are unable to comment on the severity of eye disease associated with NDH and T2DM and will aim to assess the association between non-proliferative and proliferative retinopathy with NDH/T2DM in a subsequent follow-up study.

The comorbidity among both NDH and T2DM cohorts is further highlighted by the extensive use of primary care services by both cohorts. With primary care services under extensive pressure, with an exacerbated problem of declining GP numbers, a lack of trained allied health professionals, and increased demand, individuals suffering from preconditions such as NDH are likely to receive suboptimal care, with a resultant increase in the proportion of individuals developing T2DM and its associated healthcare complications. With inpatient admissions increasing annually for both cohorts between 2015 and 2019, the need for earlier intervention in individuals diagnosed with NDH is paramount to prevent the potential development of T2DM and multimorbidity as highlighted in this study. Improved utilisation of the NDPP and innovations implemented via the NWL-DTP particularly those relating to digital systems support and closer multi-disciplinary team (MDT) working allowed for processes to be streamlined, resulting in improved efficiency within the primary care healthcare system and improved patient care.

With T2DM capable of causing irreversible end-organ damage, intervention at earlier stages of DM including NDH is essential. This includes the early referral of individuals to diabetes management/prevention programs and ensuring appropriate treatment of hyperglycemia and other medical conditions with prognostically beneficial medications. Outward referrals to a diabetes structured education program in our cohort were low, with only 11% and 54% of individuals with NDH and T2DM, respectively, being referred during the study period. This poses a significant cause for concern, with many individuals being denied the opportunity to gain an understanding of the importance of dietary and lifestyle measures, as well as medication compliance to prevent the development and progression of T2DM and its associated long-term health complications. This is despite the NDPP having been shown to be associated with a reduction in weight and HbA1c in high-risk individuals.20 Further interventions such as those implemented via the NWL-DTP are thus needed to improve the uptake of referrals to education programs in NWL.

The use of prognostically beneficial medication, particularly in the case of cardiovascular and renal complications of diabetes, is essential. While the renoprotective effects of renin angiotensin aldosterone inhibitors are well known with an associated reduction in albuminuria, CKD progression, and death, it is the early use of such agents after a diagnosis with T2DM that is of most importance in reducing the rate of renal disease progression and cardiovascular mortality.21–23 Improved interventions for both individuals with NDH and T2DM are thus needed to allow for the detection of albuminuria, even in individuals with a seemingly preserved eGFR >60 mL/min/1.73 m2. It is through this early identification of albuminuria that will allow for the early introduction of prognostically beneficial medication. Furthermore, with the introduction of sodium glucose cotransport-2 inhibitors (SGLT2i) into clinical guidelines, work is needed to ensure primary care are implementing updated guidelines to allow for patients to reap the benefits of early prescribing of SGLT2i.

This study has many strengths, the biggest of which is the Discover-NOW health research platform, on account of the linkage of anonymised patient data from across social, community, and mental healthcare services in addition to primary and secondary care from 2015 onward, making it among the largest patient datasets available, capturing approximately one-third of London’s population. However, this study has several limitations. The study period of 5 years is short and offers only a 0-year to 4-year follow-up period for almost half of the T2DM study cohort. While a current limitation, we aim to conduct a 10-year and 20-year follow-up of the study cohort in due course. As discussed, individuals for the various comorbidities of interest were identified utilising ICD-10 and READ V.2 codes, with the quality of data capture thus based on the quality of the codes included within the study. Due to the high number of READ V.2 codes associated with any one condition, there is a risk that codes could have been omitted in error despite best efforts made to ensure coding accuracy and inclusion. The cohorts were not matched by age, ethnicity, gender, and IMD, thereby introducing significant variability with regard to how these factors may affect the development of multimorbidity. While matching may significantly reduce the sample size and thus reduce power, we aim to use regression analysis to investigate the effects of these variables among the three study cohorts undertaken in our subsequent study reviewing 10-year follow-up of the study cohort. While the financial incentives associated with the QOF have, to a large degree, ensured accurate and high degrees of recording for T2DM, annual monitoring of NDH was only included in the QOF from 2020 onward, the final year included in this study. With the extensive experience gained via the NWL-DTP, it is expected for the inclusion of NDH monitoring within the QOF to result in improved recording and management of NDH. The true impact of NDH inclusion within the QOF, however, will be evaluated in a follow-up study.

Conclusions

This study highlights the significant morbidity associated with both NDH and T2DM, and the need for earlier identification of individuals with NDH/T2DM to ensure early management and prevent/delay the development of multimorbidity. With an increase in the prevalence of NDH and T2DM expected internationally, a greater emphasis is needed to ensure individuals are educated with regard to the long-term implications of both NDH and T2DM and how best to prevent its occurrence. Similarly, it is paramount that clinicians ensure optimal glycemic and weight management, as well the early prescription of prognostically beneficial medications, to prevent or halt the progression of associated health complications in individuals with both NDH and T2DM. The success of the NDPP and NWL-DTP within NWL has shown the potential positive impact such widescale transformation programs can have on improving the detection and management of individuals with NDH. With an improved understanding of the multimorbidity associated with both NDH and T2DM, clinicians and patients alike are now better informed in the battle of limiting the impact burden of both NDH and T2DM on patients and the healthcare system alike.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplemental information.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

Not applicable.

Acknowledgments

We acknowledge the patients of North West London who have consented to their clinical data being available for research purposes via the Discover-NOW Health data research hub for real-world evidence. FWKT is supported by the Diamond Fund from Imperial College Charity, and the Ken and Mary Minton Chair of Renal Medicine. RD is supported by Imperial College Institute of Global Health Innovation, Clinical Research Fellowship.