Obstructive sleep apnea a synergistic health risk in type 2 diabetes mellitus: a cross-sectional analytical study using STOP-BANG questionnaire in a tertiary care teaching hospital of Eastern Odisha

Abstract

Background

Type 2 diabetes mellitus (T2DM) is frequently associated with vascular complications and sleep-disordered breathing, particularly obstructive sleep apnea (OSA), which adversely affects glycemic control. This study aimed to evaluate the risk of OSA among patients with T2DM and to promote awareness among physicians for early detection and improved management.

Methods

A hospital-based, cross-sectional analytical study was conducted among 448 T2DM patients attending the Endocrinology Department of IMS and SUM Hospital, a tertiary care centre in Eastern Odisha, from March to September 2022. OSA risk was assessed using the validated STOP-BANG questionnaire. Categorical variables were analysed using frequency distribution and Chi-square tests. Odds ratios (ORs) with 95% confidence intervals (CI) were calculated, and multivariate logistic regression identified predictors of high OSA risk. Data analysis was performed using SPSS version 24.0.

Results

Of the 448 participants, 286 (63.8%) were aged > 50 years, 249 (55.6%) were male, and 199 (44.4%) were female. The median STOP-BANG score was 3 (range 0–8), and 61.6% were categorized as high risk for OSA. Significant associations were observed between high OSA risk and obesity, hypertension, large neck and hip circumference, poor glycaemic control, and longer diabetes duration. Multivariate analysis confirmed these as independent predictors of OSA risk.

Conclusion

Over half of T2DM patients were at high risk for OSA. Routine OSA screening using simple tools such as STOP-BANG in diabetes clinics is recommended for early screening, better glycaemic control, and reduced cardiovascular morbidity. Hence, using the STOP-BANG questionnaire helps in screening for the risk of OSA in diabetes mellitus, though the gold standard for OSA diagnosis is polysomnography.

Introduction

Sleep-related breathing disorders such as obstructive sleep apnoea (OSA) are characterized by repetitive episodes of upper airway obstruction during sleep, resulting in intermittent oxygen desaturation, sleep fragmentation, and sympathetic overactivity [1]. OSA is strongly associated with cardiometabolic comorbidities, including hypertension, insulin resistance, hyperglycaemia, obesity, cardiovascular disease, and depression [2–4]. When OSA remains undiagnosed and untreated, patients often present with excessive daytime sleepiness, impaired cognitive performance, and an increased risk of road traffic accidents [5]. Polysomnography, though the gold standard for diagnosis, questionnaire-based screening tools that are validated are used to identify high-risk individuals in clinical settings. An apnoea–hypopnoea index (AHI) ≥ 5/hr is diagnostic of OSA [6, 7].

Obesity, a global epidemic, is a major shared risk factor for both OSA and type 2 diabetes mellitus (T2DM) [8]. The relationship between OSA and T2DM is bidirectional and it is mediated by several mechanisms: (i) intermittent hypoxia and sympathetic activation leading to insulin resistance; (ii) oxidative stress resulting from hypoxia–reoxygenation cycles that impair insulin signalling and endothelial function; and (iii) chronic systemic inflammation, with elevated cytokines such as IL-6 and TNF-α that disrupts glucose metabolism and worsen adipose tissue dysfunction. Furthermore, hypoxia may directly impair pancreatic β-cell function, liver glucose, and adipose tissue metabolism, thereby aggravating hyperglycaemia. Sleep fragmentation and deprivation contribute additional adverse effects on insulin sensitivity. Therefore, OSA-related hypoxemia stimulates the release of certain active proteins and oxidative stress that exacerbate insulin resistance, lipolysis, and cause an augmented prothrombotic and proinflammatory state, which can lead to premature death.

OSA is more common among men, the elderly group, and those with overweight or morbid obesity [7]. Moreover, patients with common twin co-morbid conditions such as diabetes and hypertension are most commonly reported with sleep-disordered breathing [8]. Sleep apnoea, with partial or complete obstruction of the upper airways during sleep, leads to changes in intrathoracic pressure and hypoxia, which have an impact on quality of life impairs efficiency of doing work significantly [9]. Though a very common condition, OSA remains largely undiagnosed [10, 11]. It frequently coexists with hypertension, metabolic syndrome, polycystic ovary syndrome, chronic kidney disease, and T2DM [12–16]. Evidence suggests that OSA remains largely underdiagnosed despite its high prevalence and impact on quality of life and work productivity [9–11]. In a cross-sectional study, nearly 23% of patients with T2DM were found to have OSA [11]. Thus, identifying high-risk patients using validated tools such as the STOP-BANG questionnaire (cut-off ≥ 3) is clinically relevant, as it provides a simple, reliable, and cost-effective approach before referral for polysomnography [17].

Hence, sleep fragmentation and sleep deprivation associated with OSA may have an additive, adverse impact on insulin sensitivity [11]. Patients with OSA more often meet with accidents and have impaired social lives [18]. In urban areas of Western India, a hospital-based study revealed the prevalence of OSA in men aged between 35 and 65 years to be 19.5% [19].

So, from a public health perspective, OSA represents a growing concern in India. Meta-analysis estimates that tens of millions of Indian adults may have OSA, with prevalence varying by region, age, and comorbidity profiles. However, much of the evidence is based on studies from Western and Southern India, with limited epidemiological data from Eastern India. This gap is important because regional differences in craniofacial structure, body mass index distribution, lifestyle, and cardiometabolic risk factors may influence the burden and clinical presentation of OSA. Obstructive sleep apnoea-related hypoxemia stimulates the release of acute-phase proteins and reactive oxygen species that augment insulin resistance, lipolysis, prothrombotic and proinflammatory state, which can be fatal [20]. Intermittent hypoxia, sleep fragmentation, and sympathetic activation trigger the release of inflammatory markers like C-reactive protein (CRP) and interleukin-6 (IL-6), leading to dysfunctional adipose tissue and impaired insulin sensitivity [21].

Hence the biological plausibility of OSA contributing to poor glycaemic control and the lack of robust prevalence data from Eastern India, the present prospective study was undertaken to evaluate the utility of the STOP-BANG questionnaire for OSA screening among Indian patients with T2DM and to identify its associated risk factors.

Methods

Study design

This was a hospital-based, cross-sectional analytical study.

Study setting and study period

The study was conducted from March 2022 to September 2022, in the Outpatient Department (OPD) of Endocrinology, Institute of Medical Sciences, and SUM Hospital, a tertiary care teaching hospital located in Eastern Odisha. The hospital caters to both urban and semi-urban populations, with an average daily endocrinology OPD attendance of 80–120 patients. The centre is equipped with comprehensive diagnostic and treatment facilities for diabetes and related metabolic disorders, as well as access to sleep medicine evaluation services. (Sleep Lab). The catchment area of this tertiary care teaching hospital is large, with more than 2000 patients per day, and the socio-demographic background of the hospital is defined by its role as a key healthcare provider serving a diverse patient population, mostly from rural backgrounds and lower socio-economic strata. The consent of selected participants was taken after informing the benefits of the study. Then the responses were collected, and confidentiality was maintained throughout the study.

Study population and sampling

All adult patients (≥ 18 years) with a confirmed diagnosis of type 2 diabetes mellitus (T2DM) attending the endocrinology OPD during the study period were considered eligible. Patients with type 1 diabetes mellitus, secondary diabetes, pregnancy, or known sleep disorders already on treatment for OSA were excluded. A consecutive sampling technique was applied until the required sample size was achieved.

Sample size determination

Using Cochran’s formula and assuming an expected OSA prevalence of 25% among T2DM patients, a 95% confidence interval, and an absolute precision of 4%, the calculated sample size was 448.

Data collection tool and operational definition

The risk of OSA was assessed using the validated STOP-BANG questionnaire, both in English and the local language, Odia. A STOP-BANG score ≥ 3 was considered indicative of a high risk for OSA, and 0–2 or < 3 as low risk. Anthropometric measurements (BMI, neck and hip circumference) and clinical data (duration of diabetes, blood pressure, and glycaemic status) were recorded. Data were collected using the STOP-BANG questionnaire, which is explicitly used for screening and not for diagnosis or check the severity-grading [22]. H/O comorbid conditions such as hypertension, coronary artery disease, heart failure, stroke, or others were collected. H/O of medication and data of last fasting blood sugar (FBS) and post-prandial blood sugar (PPBS) were collected. The questionnaire is a tool that healthcare professionals can utilize to screen for obstructive sleep apnoea (OSA) in patients with type 2 diabetes mellitus (T2DM). The questionnaire includes eight dichotomous questions about sleep apnoea symptoms and demographic variables: STOP stands for Snoring, Tiredness, Observed Apnoea, and high blood pressure, and BANG stands for Body mass index (BMI), age, neck circumference, and gender. Each “yes” score is one point, and patients with a STOP-BANG score of ≥ 3 are considered at high risk of OSA. The questionnaire has shown high sensitivity to the extent of 93.9%. The overall screening process and identification is represent in Fig. 1.

Fig. 1.

PRISMA: flow diagram of the selection process, identification screening, and eligibility of OSA

Although the STOP-BANG questionnaire originally recommends a cut-off score of ≥ 5 to indicate “high risk” for obstructive sleep apnoea (OSA), several validation studies in different populations have demonstrated that a lower cut-off score of ≥ 3 offers higher sensitivity for identifying individuals at risk, particularly in resource-limited settings or in screening contexts where the priority is not to miss potential cases. Using the cut-off of three therefore maximizes the screening utility of the tool, enabling early identification and referral for confirmatory testing. Moreover, many clinical studies and guidelines have adopted ≥ 3 as the practical threshold for high OSA risk when polysomnography is not immediately feasible for all participants. This approach aligns with the tool’s intended use as a screening instrument rather than a diagnostic test. Therefore, in our study, we have used a score of three as the cut-off for categorizing participants into low-risk (0–2) and high-risk (≥ 3) groups [12, 17].

Data processing and statistical analysis

Data collected under the study were scrutinized, codified, and entered into the IBM SPSS Statistics 24.0 software for analysis [22]. The categorical variables like age group, gender, BMI & STOP-BANG questionnaire were done by using a frequency distribution procedure, and their association was studied with OSA by using the Chi-square test of independence. The continuous variables were subjected to the Shapiro-Wilk test of normality and found to be significantly different from the normal distribution. Hence, non-parametric tests were conducted in this analysis. Comparison of mean ± SD and median (IQR) of biometric measurements, blood glucose parameters, and duration of DM was analysed by using a non-parametric Mann-Whitney U test. A cut-off value ‘p’<0.05 was considered to indicate statistical significance. Multivariate binary logistic regression was undertaken with the dependent variable OSA with two levels: 0 = low risk and 1 = high risk, and independent variables like age, gender, BMI, neck size large, hip circumference in cm, fasting blood glucose, post-prandial blood glucose, and duration of DM in years.

Results

The study analyzed 448 sample cases of type 2 diabetic mellitus to get insight into the severity of obstructive sleep apnea (OSA) and the association with corresponding risk factors. Out of 448 cases, 162 belonged to the ≤ 50 years age group, 160 belonged to 51–60 years, and 126 belonged to > 60 years, thus 63.8% above 50 years of age. The overall mean ± SD and median (IQR) of age were 53.4 ± 11.1 years and 54(46–62) years, respectively. Among the study population, the male proportions were more than those of the females. Based upon the BMI classification, 4% were underweight, 45.3% normal, 37.5% overweight, and 13.2% obese. Overall mean ± SD and median (IQR) of BMI were 25.3 ± 4.4 kg/m² and 25.1(22.0–28.0) kg/m²,respectively (Table 1).

Table 1.

Demographic profile of cases (N = 448, endocrinology OPD)

	No.	%
Age group
≤ 50	162	36.2
51–60	160	35.7
> 60	126	28.1
Mean ± SD	53.4 ± 11.1
Median(IQR)	54(46–62)
Gender
Male	249	55.6
Female	199	44.4
BMI Group
Underweight (< 18.5)	18	4
Normal weight (18.5–24.9)	203	45.3
Overweight (25–29.9)	168	37.5
Obese (> 30)	59	13.2
Mean ± SD	25.3 ± 4.4
Median(IQR)	25.1(22.0–28.0)

The mean ± SD and median (IQR) of hip circumference were 100.5 ± 14.2 cm & 99.0(93.0–106.0.0.0) cm, respectively indicated higher hip circumference. The higher neck circumference with mean ± SD and median (IQR) 36.4 ± 6.4 cm & 37(33–39) cm, respectively, suggests potential increased cardio-metabolic risk and possible obstructive sleep apnoea risk.

The respective mean ± SD and median (IQR) of FBS were 168.8 ± 65.7 mg/dl & 155.5(122.0–200.0.0.0) mg/dl. The mean ± SD and median (IQR) of PPB were 250.6 ± 92.8 mg/dl & 236.0 (189.0–295.0.0.0) mg/dl. The mean duration of DM was 7.6 ± 6.1 years, and the median duration, along with IQR, was 6.0(3.0–10.0) days (Table 2). On average, the sample displays characteristics indicative of metabolic risk factors and poorly controlled diabetes, particularly regarding glucose levels and body composition.

Table 2.

Descriptive statistics of selected biometric measurements and blood glucose level (Sample size N = 448)

Variables	Descriptive statistics
	Mean ± SD	Median (IQR)
HIP Circumference in cm	100.5 ± 14.2	99.0(93.0–106.0.0.0)
Neck Circumference in cm	36.4 ± 6.4	37(33–39)
Fasting Blood Glucose	168.8 ± 65.7	155.5(122.0–200.0.0.0)
PP Blood Glucose Level	250.6 ± 92.8	236.0(189.0–295.0.0.0)
Duration of DM in year	7.6 ± 6.1	6.0(3.0–10.0)

STOP-BANG questionnaire comprised of ‘S’ for snoring, ‘T’ for tiredness, ‘O’ for anybody who observed the patient snoring, ‘P’ for the presence of Blood pressure, ‘B’ for BMI more than 35 Kg/m2, ‘A’ for age more than 50 years, and ‘G’ for gender equal to male. Out of 448 cases, 40.6% had snoring, 65% were experiencing tiredness, 23% were observed to be snoring, and 43.5% had high blood pressure, 2.5% had a BMI more than 35 Kg/m2, 63.8% with age of more than 50 years, 65.6% neck size, and 55.6% males.

OSA score has been categorized into 2 groups, i.e., low-risk OSA between 0 and 2 (< 3) and high-risk OSA (≥ 3) which comprises of intermediate (3–4) and high risk OSA (5–8), with shares of 38.4% and 61.6%, respectively. The overall mean ± SD of OSA was 3.0 ± 1.3, and the median (IQR) was 3(2–4) (Table 3).

Table 3.

Distributions of cases by STOP-BANG questionnaire and obstructive sleep apnea (OSA) risk (N = 448)

Variables	No.	%
Snoring (Loud and Frequent Snoring)	182	40.6
Tired (Frequent Daytime Tiredness/Fatigue)	291	65
Observed (Reported Observed Apneas)	103	23
Pressure (Diagnosis of High Blood Pressure)	195	43.5
BMI > 35 Kg./m2 (Body Mass Index)	11	2.5
Age > 50 years	286	63.8
Neck size large (> 17 in. for males; >16 in. for females)	294	65.6
Male Gender	249	55.6
OSA Risk Category
Low-risk (< 3)	172	38.4
High-risk (≥ 3)	276	61.6
Mean ± SD of OSA score	3.0 ± 1.3
Median(IQR) of OSA score	3(2–4)

The association of demographic variables with OSA revealed that out of 162 cases of ≤ 50 years age, 102 had low-risk OSA and 60 had high-risk OSA. Among 160 cases of 51–60 years, 41 had low-risk OSA, and 119 had high-risk OSA. Corresponding proportions in > 60 years were 29 & 97, respectively (Fig. 2). The proportions of high risk of OSA were significantly higher among the 51–60 & > 60 years than the ≤ 50 years age group, with p = < 0.001. The gender and BMI group did not present a significant association with the risk level of OSA (p > 0.05). However, BMI more than 35 Kg/M², age older than 50 years, and large neck size have significant associations with high-risk OSA (p < 0.05). The OR (95%CI) for gender (Male/Female) = {0.912(0.621–1.339)}. The odds of the outcome (High risk OSA) for males are 0.912 times the odds for females (the reference group). Therefore, the difference in the odds of the outcome between males and females is not statistically significant. Being older than 50 years is a very strong risk factor for the outcome, with individuals in this age group having 5.246 times the odds of being high-risk OSA compared to those aged 50 or younger. Having a large neck size (compared to a non-large neck size) is a significant risk factor for high OSA, with an OR of 1.928 times higher odds (Table 4).

Fig. 2.

OSA risk categories among patients with T2DM

Table 4.

Association of demographic variables with OSA score

Variables	OSA				Total		[χ2, p] {OR(95%CI)}
	Low-risk (< 3)		High-risk (≥ 3)
Age group	No.	%	No.	%	No.	%
≤ 50	102	63	60	37	162	100	χ2 = 64.971 p = < 0.001
51–60	41	25.6	119	74.4	160	100
> 60	29	23	97	77	126	100
Gender
Male	98	39.4	151	60.6	249	100	[χ2 = 0.221 p = 0.639] {0.912(0.621–1.339)}
Female	74	37.2	125	62.8	199	100
BMI group
Underweight (< 18.5)	5	27.8	13	72.2	18	100	χ2 = 4.859 p = 0.182
Normal weight (18.5–24.9)	75	36.9	128	63.1	203	100
Overweight (25–29.9)	74	44	94	56	168	100
Obese (> 30)	18	30.5	41	69.5	59	100
Body Mass Index More Than 35 Kg/M2
Yes	0	0	11	100	11	100	χ2 = 7.028 p = 0.008
No	172	39.4	265	60.6	437	100
Age older than 50
Yes	70	24.5	216	75.5	286	100	[χ2 = 64.768 p = < 0.001] {5.246(3.455–7.964)}
No	102	63	60	37	162	100
Neck size large
Yes	97	33	197	67	294	100	χ2 = 10.543 p = 0.001 {1.928(1.294–2.873)}
No	75	48.7	79	51.3	154	100
Total	172	38.4	276	61.6	448	100

The median (IQR) of neck circumference was found to be significantly higher in high-risk OSA, with a respective 37.0 cm, than the corresponding values of low risk of OSA (p < 0.001). HIP circumference, Fasting, and PP blood glucose level didn’t have significant variations between low and high-risk of OSA (p > 0.05). The mean and median duration of diabetic mellitus between high-risk OSA class [8.0(4.0–12.0) years] was significantly higher than low-risk OSA [5.0(2.0–9.0)] class, with p < 0.001 (Table 5.).

Table 5.

Comparison of biometric measurements, blood glucose parameters, and duration of DM by OSA

Variables	OSA		Mann-Whitney U 'p' value
	Low-risk (<3) (N=172)	High-risk (≥3) (N=276)
	Median(IQR)	Median(IQR)
HIP Circumference in cm	99.0(95.0-104.0)	99.0(92.0-106.0)	0.918
Neck Circumference in cm	35.0(31.0-38.0)	37.0(33.0-40.0)	<0.001
Fasting Blood Glucose	160.0(119.0-198.8)	153.0(124.3-200.0)	0.654
PP Blood Glucose Level	229.0(182.5-287.8)	240.0(190.0-297.8)	0.312
Duration of DM years	5.0(2.0-9.0)	8.0(4.0-12.0)	<0.001

Binary logistic regression analysis with dependent variable OSA with two levels 0 = low risk and 1 = high risk of OSA, and independent variables were Age, gender, BMI, neck size large, hip circumference in cm, fasting blood glucose, post-prandial blood glucose, and duration of DM in years was undertaken following Backward likelihood regression analysis. The iteration continued for 4 steps. The significant variables selected in the 4th & final step were age older than 50, large neck size, and duration of DM. The percentage of correct prediction by the model was 72.5%, which was good, and the model coefficient, as per the omnibus test, was significant (p < 0.001). However R² value was 0.247, which implied only 24.7% of the variation in OSA was explained by the regression model. The age older than 50 years had Exp (B), i.e., odds ratio equals 5.723 with a 95% confidence interval (CI) (3.627–9.031) in comparison to age 50 or lower (p < 0/001). That means those aged more than 50 have significantly higher odds of developing severe OSA. Large neck size had Exp (B) equal to 2.892 with 95% CI 1.826–4.581, indicating significantly high odds of severe OSA for large neck size (p < 0.001). Duration of DM had Exp (B) 1.004 with CI 1.043–1.083. This implied a significant contribution of the duration of DM to the severe OSA. Thus, the multivariate analysis implied three significant variables, i.e., age older than 50 years, large neck size, and duration of DM (Table 6).

Table 6.

Binary logistic regression statistics of OSA with risk factors (a)

Variables	B	S.E.	Wald	df	Sig.	Exp (B) (95% CI	% Correct	Omnibus Tests of Model Coefficients	Nagelkerke R2
Age older than 50 (Yes)	1.744	0.233	56.186	1	< 0.001	5.723 (3.62–9.03)	72.5	< 0.001	0.247
Neck size large (Yes)	1.062	0.235	20.49	1	< 0.001	2.892 (1.83–4.58)
Duration of DM	0.042	0.019	4.813	1	0.028	1.043 (1.04–1.08)
Constant	−1.565	0.280	31.334	1	< 0.001	0.209

Clinical interpretation of effect sizes

Binary logistic regression analysis showed that age > 50 years was the strongest predictor of high-risk OSA, with individuals in this age group having 5.7-fold higher odds compared with those ≤ 50 years (OR = 5.72; 95% CI: 3.62–9.03; p < 0.001). A large neck circumference was associated with nearly threefold increased odds of high-risk OSA (OR = 2.89; 95% CI: 1.83–4.58; p < 0.001). Duration of diabetes mellitus demonstrated a smaller but significant effect, with each additional year increasing the odds of high-risk OSA by approximately 4% (OR = 1.04; 95% CI: 1.04–1.08; p = 0.028). The overall model was statistically significant (Omnibus test, p < 0.001) and explained 24.7% of the variance in high-risk OSA (Nagelkerke R² = 0.247).

Discussion

The present hospital-based study recruited 448 patients with Type 2 diabetes attending the Endocrinology Department of IMS & SUM Hospital and screened them for sleep-disordered breathing using a validated questionnaire. Overall, we observed a high prevalence of OSA risk features in our sample, with 61.6% screening positive (score ≥ 3), a rate consistent with several recent clinic-based reports that show markedly elevated OSA risk among people with T2DM compared with the general population [23].

A male predominance was observed, with 249 males and 199 females among the 448 participants. Notably, 63.8% of the study population were above 50 years of age and had higher body weight. Similar findings were reported by Alghafli et al. [24], who also used a questionnaire-based screening method for OSA and found a mean age of 53 years, with more than 50% males, 32% with a BMI > 35 kg/m², and 76.7% being hypertensive. Male sex (249/448), older age (63.8% > 50 years), and higher body weight were prominent in our cohort; these demographic patterns mirror other clinic-based series where older age, male sex, and adiposity cluster with increased OSA risk in diabetes clinics. Recent systematic and clinical studies confirm that patients with T2DM commonly present with these same risk profiles [25, 26].

Adiposity and anthropometry were strongly associated with high OSA risk in our sample. A higher BMI and longer duration of diabetes were both linked to increased screening positivity findings that align with contemporary analyses showing that obesity is a major driver of Sleep-disordered breathing in people with diabetes and that metabolic burden compounds OSA risk. In particular, neck circumference > 40 cm in our patients was a significant predictor of high-risk OSA; multiple recent studies (including cross-sectional and cohort analyses) have demonstrated that neck circumference and related upper-body fat metrics correlate more tightly with OSA risk than BMI alone, presumably because localized para-pharyngeal adipose deposition directly alters upper-airway mechanics [27–30].

Our findings also highlighted that a higher BMI and longer duration of diabetes were associated with increased OSA risk, consistent with studies by Algeffari et al., Alkhodaidi ST et al., Bamanikar A et al., and Obaseki DO et al. [31–35]. A neck circumference greater than 40 cm was significantly associated with a high risk of OSA, in agreement with previous studies [36]. This may be explained by localized adipose tissue deposition around the neck, which can alter upper airway dynamics and contribute to airway obstruction. Obesity leads to para-pharyngeal fat deposition, narrowing the airway and predisposing individuals to desaturation events and the need for continuous positive airway pressure ventilation [37].

Hypertension showed a strong association with high OSA risk in our cohort (approximately two-fold increased risk). This relationship is well supported in recent literature: mechanistic and clinical reviews published in 2023–2024 emphasize that recurrent intermittent hypoxia and sleep fragmentation in OSA trigger sustained sympathetic activation, endothelial dysfunction, and oxidative stress — pathophysiologic processes that promote chronic blood pressure elevation and cardiovascular risk. These mechanisms plausibly explain the higher prevalence of hypertension we observed among high-risk participants [38–40].

Clinically relevant implications follow from these convergent findings. Given the high yield of questionnaire screening in diabetes populations (and consistent external validations of STOP-BANG and similar tools in recent years), routine screening for OSA risk in diabetes clinics can help identify patients who would benefit from confirmatory sleep testing and targeted interventions (weight management, CPAP indicated, and BP optimization). Several studies stress that early detection in high-risk medical populations (including T2DM) is an actionable strategy to reduce cardiometabolic morbidity.

Limitations and recommendations: as reinforced by recent methodological reviews, questionnaire-based screening is useful for case-finding but cannot replace objective polysomnography for diagnostic classification or severity grading; future work in our setting should include home or lab sleep testing to quantify apnoea-hypopnea index and to examine whether OSA treatment improves glycaemic indices and BP control in this population. As it is a cross-sectional study design, it was not possible to establish a causal relationship, and being a questionnaire-based study chance of response bias is possible.

There is potential selection bias due to hospital based study design because patients seeking care are different from the general population, leading to non-representative samples, especially regarding severity of illness, comorbidities, demographics (age, sex, socio-economic status), and health care seeking behaviour, which can skew results on treatment effectiveness. Hospital based sample does not reflect the community.

Conclusion

This study highlights a strong association between type 2 diabetes mellitus (T2DM) and obstructive sleep apnoea (OSA), revealing a significantly higher prevalence of OSA risk among diabetic patients. The increased risk was closely linked to obesity (diabesity), larger neck circumference, hypertension, poor glycaemic control, and longer duration of diabetes. The use of the STOP-BANG questionnaire proved to be a practical and efficient tool for early OSA risk detection in outpatient settings. Identifying high-risk individuals through simple screening can help prioritize patients for polysomnography confirmation. Early diagnosis and management of OSA are crucial because untreated cases can lead to worsening cardiovascular morbidity, increased daytime fatigue, sleepiness, poorer diabetes control, and diminished quality of life. These findings underscore the need to integrate routine OSA screening into standard diabetes care protocols. By doing so, healthcare systems can facilitate early intervention, improve clinical outcomes, and reduce long-term complications. The study also emphasizes the importance of multidisciplinary management involving endocrinology, sleep medicine, and cardiology. Although limited by a modest sample size and reliance on a questionnaire, the research provides strong evidence supporting structured screening programs. Implementing such strategies can help mitigate disease burden, enhance patient well-being, and reduce healthcare costs. However, the impact of OSA on quality of life in type 2 DM and future research directions, such as interventional or longitudinal studies for greater public health, indicate the need for prompt screening. in T2DM patients to ensure timely diagnosis, comprehensive management, and improved patient prognosis.

Author contributions

All authors contributed to the conceptualization of the paper and had an opportunity to review and edit the paper. P.G, M.B, S.M, and T.M. wrote the original draft and review and edited the manuscripts. T.M, MB, D.M.,S.M. and SM contributed to the investigation, methodology, data curation and formal analyses. All authors contributed to the manuscript review and editing.

Funding

Open access funding provided by Siksha 'O' Anusandhan (Deemed To Be University). This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

None.

Declarations

Ethics approval and consent to participate

The participants were assured that their responses to the Stop-Bang questionnaire would remain anonymous. This study was approved by the Ethical Committee of IMS and SUM Hospital: Ref No IMS/SRC/294/2023 dated 21/08/2023.

Consent for publication

All participants agreed to publish.

Competing interests

The authors declare no competing interests.