Poor Sleep Plagues Over a Third of Ghanaians With Type 2 Diabetes, Study Finds

More than one in three adults living with type 2 diabetes in Ghana’s Mampong Municipality are sleeping badly, and the strongest predictors of their restless nights turn out to be psychological rather than metabolic, according to a new cross-sectional study published in BMC Endocrine Disorders. The research, led by Godfred Darko of Akenten Appiah-Menka University of Skills Training and Entrepreneurial Development, surveyed 401 adults with type 2 diabetes mellitus across three health facilities between January and March 2026, and found that 37.4 percent qualified as poor sleepers on a widely used clinical measure of sleep quality.

The study used the Pittsburgh Sleep Quality Index, or PSQI, a validated instrument that scores sleep across seven components including duration, latency, disturbances, efficiency, and daytime dysfunction. Participants recorded a median PSQI score of 7, with an interquartile range stretching from 6 to 20, a distribution indicating that while many patients hovered near the threshold of disturbed sleep, a substantial minority experienced severe and multifaceted sleep problems. A global score above 5 conventionally signals poor sleep quality, and the 150 participants who crossed that line represent a burden that clinicians in the municipality had not previously quantified.

What elevates the study beyond a simple prevalence count is its parallel measurement of psychological and social variables. Each participant also completed the Perceived Stress Scale, a ten-item questionnaire capturing how unpredictable, uncontrollable, and overloaded respondents find their lives, and the Multidimensional Scale of Perceived Social Support, which assesses support from family, friends, and significant others. The statistical relationships that emerged were striking. Perceived stress correlated positively with PSQI scores with a Spearman coefficient of 0.668, while social support correlated inversely at minus 0.645, both highly significant. In epidemiological terms, these are strong associations, suggesting that the emotional and social context of living with a chronic disease may matter as much as the disease itself when it comes to sleep.

To test whether these correlations held up under scrutiny, the researchers turned to multivariable logistic regression, a technique that estimates the independent effect of each factor while adjusting for all the others. The results were consistent with the correlation analysis. Every five-point increase in perceived stress nearly quadrupled the odds of poor sleep quality, with an adjusted odds ratio of 3.67 and a 95 percent confidence interval of 1.99 to 6.75. Conversely, every five-point increase in perceived social support cut the odds by nearly half, with an adjusted odds ratio of 0.57 and a confidence interval of 0.41 to 0.79. In plain terms, stressed patients slept poorly, and supported patients slept better, regardless of their other characteristics.

Age and treatment regimen also left clear fingerprints on sleep outcomes. Compared with adults aged 30 to 44, participants aged 45 to 60 and those older than 60 faced significantly higher odds of poor sleep, a pattern consistent with the known age-related erosion of sleep architecture, in which deep slow-wave sleep diminishes and nocturnal awakenings become more frequent. Treatment type mattered as well: patients using insulin, or combining insulin with oral agents, were more likely to be poor sleepers than those on oral medication alone. The authors suggest several plausible mechanisms, including nocturnal hypoglycemia, the burden of injection routines, and the fact that insulin prescription typically marks longer disease duration and greater metabolic complexity.

One finding stands out for its sheer magnitude. Participants reporting high levels of physical activity had dramatically lower odds of poor sleep quality, with an adjusted odds ratio of 0.11 and a confidence interval of 0.02 to 0.55. Although the wide interval reflects a small number of highly active participants and warrants caution, an odds reduction of nearly 90 percent is difficult to ignore. The result aligns with a substantial international literature linking regular exercise to improved sleep onset, sleep continuity, and sleep depth, and it hints at a practical, low-cost intervention available even in resource-constrained primary care settings.

The biological plausibility of the stress-sleep connection in diabetes is well established. Chronic hyperglycemia triggers osmotic diuresis, causing nocturia that fragments sleep, while sleep deprivation in turn impairs insulin sensitivity and glucose tolerance, creating a feedback loop in which poor sleep worsens glycemic control and poor glycemic control worsens sleep. Add the psychological load of managing a lifelong condition, often with financial strain and limited family resources, and the high prevalence of disturbed sleep in this population becomes less surprising. The strong inverse association with social support suggests that this loop may be buffered by the people around the patient, a modifiable factor that few diabetes programs currently measure.

The study’s setting gives its findings particular weight. Evidence on sleep and diabetes from sub-Saharan African primary care has been sparse, and most of what is known comes from high-income countries with different health systems, diets, and social structures. By recruiting across the Mampong Municipal Government Hospital and the Kofiase and Adudwan health centres, the researchers captured a multicentre sample that reflects routine care rather than specialist clinics. The cross-sectional design, however, imposes real limits: because stress, support, and sleep were measured at the same moment, the data cannot establish which factor drives which. It is entirely possible that poor sleep amplifies perceived stress just as stress disrupts sleep, and only longitudinal or interventional studies can untangle the direction of causation.

Even so, the practical implications are immediate. The authors argue that sleep quality and psychosocial assessment should be woven into routine diabetes care, a recommendation that requires no new drugs or expensive equipment. A short PSQI administered at clinic visits could flag the more than one-third of patients whose sleep is quietly undermining their metabolic health, while screening for high perceived stress and low social support could identify those most at risk. Given that sleep disturbance is associated with worse glycemic control, greater diabetes distress, and reduced quality of life, treating sleep as a vital sign in diabetes management is an idea whose time may have come, in Mampong and well beyond it.

The study, which received no external funding and was approved by the Ghana Health Service Ethics Review Committee, adds Ghana to the growing map of countries documenting the hidden burden of sleep disorders among people with chronic disease. As diabetes prevalence continues to climb across Africa, the message from Mampong is clear: the fight against type 2 diabetes is not only fought with medication and diet, but also, increasingly, in the quality of patients’ sleep and the strength of the social ties that surround them.

Subject of Research: Sleep quality and its psychosocial and clinical determinants among adults with type 2 diabetes in Ghana

Article Title: Sleep quality and associated factors among adults living with type 2 diabetes mellitus in Mampong municipality, Ghana: a multicentre cross-sectional study

Article References: Sleep quality and associated factors among adults living with type 2 diabetes mellitus in Mampong municipality, Ghana: a multicentre cross-sectional study. (n.d.). https://doi.org/10.1186/s12902-026-02585-4

Image Credits: AI Generated

DOI: 10.1186/s12902-026-02585-4

Keywords: type 2 diabetes, sleep quality, Pittsburgh Sleep Quality Index, perceived stress, social support, physical activity, Ghana, insulin therapy, cross-sectional study, glycemic control, primary care, BMC Endocrine Disorders