Can Too Much Sleep Signal Kidney Trouble in Type 2 Diabetes?
Andrew Kowalski, MD, FASN
Why This Study Caught My Attention
We spend a great deal of time telling patients that sleep is not just an important part of life, but that it is protective. Short sleep raises blood pressure, worsens insulin resistance, and fuels the cardiometabolic engine that drives kidney disease. So when a new study in the Journal of the Endocrine Society reported that the people with type 2 diabetes were most likely to develop or worsen diabetic kidney disease were the long sleepers, not the short sleepers, I did a double take and figured this deserves a careful look.¹
Diabetic nephropathy, now more often called diabetic kidney disease (DKD), affects roughly 30% to 40% of people with diabetes and is a leading path to dialysis and to cardiovascular events.¹ Anything simple and observable that flags who is at higher risk is worth understanding. As I will argue, and frequently do ;-), the most useful way to read this paper is not "sleep less" but ask why this happened and to "pay attention to the person who is sleeping a lot."
The Study at a Glance
Ikeda and colleagues at Juntendo University followed 678 Japanese outpatients with type 2 diabetes, aged 25 to 69, who were enrolled between 2013 and 2014 and tracked for a mean of 6.0 years.¹
Many key groups were excluded, such as people with prior heart attack, stroke, advanced heart failure, cirrhosis, active cancer, or dialysis were excluded, and anyone who already had macroalbuminuria at baseline was removed from the analysis.
No surprise here as they were looking at "healthier" patients and what happens.
Sleep was not captured once and forgotten. Participants reported their usual nightly sleep at baseline, year 2, and year 5, and the investigators averaged those reports up to the point of an event. The same approach was used for diet (a validated food questionnaire), physical activity, chronotype, sleep quality (Pittsburgh Sleep Quality Index), and depressive symptoms.
Feature | Detail |
Design | Prospective cohort, 3 Japanese centers |
Participants | 678 adults with type 2 diabetes, no macroalbuminuria at start |
Mean follow-up | 6.0 ± 2.7 years |
Exposure | Self-reported nightly sleep, averaged across up to 3 time points |
Outcome | New albuminuria (≥30 mg/g) or progression from micro- to macroalbuminuria (≥300 mg/g) |
Events | 215 participants (31.7%) |
Analysis | Multivariable Cox proportional hazards models |
The outcome was a grouping of either new-onset albuminuria in someone who started with a normal urine albumin, or progression from microalbuminuria to macroalbuminuria. Urine albumin was measured once a year.
Key Findings
The longer people slept, the more likely their kidneys were to show new or worsening albumin leak. Those averaging 8 or more hours a night had nearly twice the risk of the shortest sleepers!
Average nightly sleep | Participants | Reached endpoint | Adjusted hazard ratio (95% CI) |
Under 6 hours (reference) | 174 | 28.2% | 1.00 |
6 to under 7 hours | 273 | 27.8% | 1.08 (0.74 to 1.57) |
7 to under 8 hours | 170 | 37.1% | 1.56 (1.05 to 2.31) |
8 hours or more | 60 | 43.3% | 1.92 (1.15 to 3.19) |
The trend across categories was statistically significant (P for trend = .003), and it held after adjustment for age, sex, diabetes duration, HbA1c, blood pressure, triglycerides, uric acid, baseline urine albumin, diet, alcohol, and several medications.¹
Sleep was the only lifestyle factor (out of a long list that included salt, protein, calories, alcohol, and exercise) that stood out.
Three observations make this study more interesting than a simple "too much sleep is bad" headline.
Long sleepers looked metabolically healthier, not sicker. At baseline, the longer sleepers had a lower body mass index, lower LDL cholesterol, and HbA1c that was lowest in the 7 to 8 hour group. They also reported better sleep quality and a stronger morning preference. This is the opposite of what the large Western cohorts tend to show.
But they were older and their kidneys were already a step behind. Longer sleepers were older, more often male, had lower eGFR (71.8 vs 79.9 mL/min/1.73 m² in the longest vs shortest group), and more diabetic retinopathy. Over follow-up their eGFR stayed lower and their urine albumin rose faster.
Basically, those that were longer sleepers already had chronic disease!
The signal concentrated in vulnerable subgroups. The association was strongest in people aged 65 and older (P for interaction = .046), in those who already had microalbuminuria (P for interaction < .001), and in those with eGFR below 60. In adults under 65 there was no meaningful association.
Blood pressure did not differ across sleep groups at baseline or over time, which argues against a simple hemodynamic explanation.
What Is Probably Going On Long Sleep as a Messenger, Not the Culprit
To simplify the bottom line, longer sleep was more of a warning sign among these individuals that they will get worse. The authors are NOT saying that sleeping longer will give you more disease.
Their lead author put it plainly, people should not take this as advice to sleep less.
So why would a leaner, well-controlled, older person with type 2 diabetes need nine hours of sleep? Several threads of evidence point in the same direction.
Inflammation. A major topic for me as this is the start of all disease (CRIMS model). A meta-analysis of 72 studies and more than 50,000 adults found that long sleep (more than 8 hours) was associated with higher C-reactive protein and interleukin-6, while short sleep was not consistently linked to IL-6. Chronic low-grade inflammation is a well-recognized accelerant of diabetic kidney disease.
Muscle loss and frailty. Another major topic of mine. Muscle loss will 100% accelerate chronic disease! Another study found that sleeping more than 8 hours was associated with about 52% higher odds of sarcopenia (muscle wasting) (OR 1.52; 95% CI, 1.23 to 1.88), with men affected mainly by long rather than short sleep. And another study of more than 41,000 people linked long sleep to frailty as well (OR 1.21; 95% CI, 1.10 to 1.32). Notice that the long sleepers in this study were leaner and mostly male, exactly the profile in which low body weight can reflect lost muscle rather than good metabolic health.
Hidden illness and reduced reserve. Long sleep is one of the most consistent epidemiologic markers of all-cause mortality, which suggests that it often shows up with worsening disease, rather than causing it. Unrecognized sleep disorders, low-level heart failure, depression, and medication effects can all lengthen time in bed without making sleep restorative.
Through the CRIMS Lens
For readers who follow my CRIMS framework (Cardio-Renal-Inflammatory-Metabolic-Sarcopenia), this paper is a near-perfect illustration of why the five domains cannot be read in isolation.
On the Metabolic axis these long sleepers looked fine: lower BMI, lower LDL, decent HbA1c. Yet the Renal axis was already weaker (lower eGFR, more retinopathy as a sign of microvascular injury), and the likely drivers sit on the Inflammatory and Sarcopenia axes, which a standard diabetes visit rarely measures or even notices. A favorable lab panel can hide a vulnerable patient.
One thought to keep in mind: Long habitual sleep in an older adult with diabetes may be a low-cost clinical nudge/suggestion to look for sarcopenia, frailty, inflammation, and sleep-disordered breathing.
But remember this is a thought and observed correlation made by looking at the data, not a proven screening strategy.
What the Study Does Well (Pros)
• A solid design with long follow-up. Six years on average, with roughly a decade of planned observation, gives time for kidney outcomes to emerge.
• Repeated sleep measurements. Averaging sleep across baseline, year 2, and year 5 captures habitual behavior far better than a single questionnaire, which is the weakness of most prior studies.
• Laboratory-defined outcomes. Albuminuria categories were based on measured urine albumin rather than billing codes.
• Unusually thorough lifestyle adjustment. Diet (including protein source and salt), alcohol, physical activity, chronotype, sleep quality, and depressive symptoms were all measured and modeled.
• Longitudinal trajectories. The authors showed that eGFR stayed lower and urine albumin rose faster over time in long sleepers, not just that the endpoint was reached more often.
• Sensitivity checks. Results held when sleep was regrouped into three categories and after adjusting for diagnosed sleep apnea, sleeping-pill use, and RAS blockade.
• Honest interpretation. The authors explicitly frame long sleep as a marker, not a target.
Where Caution Is Needed (Cons and Limitations)
• It was an observational design. No real claim can be made, it might just be a coincidence (a very interesting coincidence). Reverse causation (early illness causing longer sleep) is entirely plausible, but needs to be investigated to make that claim.
• Small long-sleep group. Only 60 people slept 8 or more hours, which is why the confidence interval for that group is wide (1.15 to 3.19). More patients = more data = something to really look at.
• Single annual urine albumin. This was a mistake on the ones designing this study. Most guidelines call for confirming albuminuria on repeated samples over 3 to 6 months. One random sample per year is likely a simple fluke and some participants were inappropriately grouped.
• Self-reported sleep. As we all know, anything that we self report will likely be higher or lower, if it is to our benefit. There was no actigraphy or sleep study. People often overestimate time asleep, and "time in bed" is not the same as sleep.
• Incomplete sleep-disorder data. Obstructive and central sleep apnea were not distinguished, and insomnia diagnoses were not captured.
• Borderline primary signal. As a continuous variable, sleep duration did not reach significance (HR 1.13; P = .082). The two groups that were followed were exploratory, with no correction for multiple testing.
• Pre-modern therapy era. Enrollment was in 2013 to 2014. No one was on an SGLT2 inhibitor at baseline, GLP-1 receptor agonist use was low, finerenone did not yet exist, and the importance of being on a ACE inhibitor or Angiotensin receptor blocker was still not as aggressively pushed as they are today.
• Generalizability. A single ethnic group (Japanese), relatively lean (mean BMI about 24.5), with good baseline kidney function. Results may not translate to populations with higher obesity rates or highly Westernized cultures.
• No inflammation or muscle data. The proposed mechanisms (inflammation, sarcopenia, frailty) were not directly measured in this cohort.
What supports the authors. Every diabetes-specific cohort I found links long sleep to DKD. The Chinese cohort is especially aligned: long sleep raised DKD risk by about 47%, and the combination of long sleep with weight gain doubled it. The large UK Biobank analyses reached similar conclusions over 13 years of follow-up.
What challenges them. Most studies describe a U-shaped curve in which both extremes are harmful, with the lowest risk near 7 hours. The absence of a short-sleep signal in this Japanese cohort may reflect its small size, the choice of <6 hours (rather than 7 hours) as the reference group, or the leaner population.
The mechanism debate. Western cohorts tend to attribute long-sleep risk to obesity and metabolic excess. In the UK Biobank, BMI, CRP, liver enzymes, and lipids explained about 20% of the association. The Japanese long sleepers were leaner, which points instead toward frailty, sarcopenia, and inflammation. Both pathways probably exist, and which one dominates likely depends on the patient in front of you.
What This Means for You
If you have type 2 diabetes, please do not cut your sleep to protect your kidneys. Nothing in this study shows that sleeping less helps, and short sleep has its own well-documented harms. Most adults do best with roughly 7 to 8 hours of restful sleep.
If you find you routinely need 9 or more hours, or you sleep long and still wake tired, that is worth mentioning at your next visit. It may be a clue to something else, such as sleep apnea, low muscle mass, low thyroid, depression, anemia, a medication effect, or early heart or kidney strain.
The kidney basics still matter most. Every Kkidney patient should know your UACR and eGFR, keep blood pressure and glucose on target, stay active with both walking and resistance training to protect muscle, and ask whether you are a candidate for the medications that are now proven to protect diabetic kidneys.
For clinicians: Asking "How many hours do you usually sleep?" takes ten seconds. In an older patient with diabetes, especially one with microalbuminuria or eGFR below 60, a habitual answer of 8 hours or more is a reasonable prompt to:
1. Confirm albuminuria with repeated UACR testing and recheck kidney function on schedule.
2. Screen for sleep-disordered breathing (STOP-BANG or referral for a sleep study).
3. Screen for sarcopenia and frailty (grip strength, chair-stand test, gait speed, recent weight or muscle loss).
4. Review sedating medications and screen for depression and hypothyroidism.
5. Make sure the four pillars of DKD therapy (RAS blockade, SGLT2 inhibitor, nonsteroidal MRA where indicated, GLP-1 receptor agonist) are optimized.
None of these steps is new, and all are already recommended. Long sleep simply gives us one more reason to make sure they happen.
Key Points
• In 678 Japanese adults with type 2 diabetes followed for about 6 years, those sleeping 8 or more hours had nearly twice the risk (HR 1.92) of new or worsening albuminuria compared with those sleeping under 6 hours.
• Sleep was the only lifestyle factor, out of many measured, that independently tracked with kidney outcomes.
• The risk was concentrated in adults 65 and older and in those who already had microalbuminuria or eGFR below 60.
• Long sleepers were leaner with lower LDL, which suggests long sleep reflects frailty, muscle loss, and inflammation rather than metabolic excess.
• Other diabetes cohorts agree that long sleep is a DKD risk marker, but most also find short sleep harmful; the safest range remains about 7 to 8 hours.
• The study is observational, used single annual urine samples, relied on self-reported sleep, and predates SGLT2 inhibitors and finerenone.
• Do not shorten sleep to protect your kidneys. Treat habitual long sleep as a cue to check for sleep apnea, sarcopenia, frailty, depression, and medication effects, and to optimize proven kidney-protective care.
References
1. Ikeda F, Mita T, Osonoi Y, et al. Longer sleep duration and the incidence and progression of diabetic nephropathy: a prospective cohort study. J Endocr Soc. 2026;10(11):bvag201. doi:10.1210/jendso/bvag201
2. Irwin MR, Olmstead R, Carroll JE. Sleep disturbance, sleep duration, and inflammation: a systematic review and meta-analysis of cohort studies and experimental sleep deprivation. Biol Psychiatry. 2016;80(1):40-52. doi:10.1016/j.biopsych.2015.05.014
3. Pourmotabbed A, Ghaedi E, Babaei A, et al. Sleep duration and sarcopenia risk: a systematic review and dose-response meta-analysis. Sleep Breath. 2020;24(4):1267-1278. doi:10.1007/s11325-019-01965-6
4. Pourmotabbed A, Boozari B, Babaei A, et al. Sleep and frailty risk: a systematic review and meta-analysis. Sleep Breath. 2020;24(3):1187-1197. doi:10.1007/s11325-020-02061-w
5. Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep. 2010;33(5):585-592. doi:10.1093/sleep/33.5.585
6. Liu C, Zhang J, Wei X, et al. Effects of sleep duration and changes in body mass index on diabetic kidney disease: a prospective cohort study. Front Endocrinol (Lausanne). 2023;14:1278665. doi:10.3389/fendo.2023.1278665
7. Niu X, Liu H, Wang Y, et al. Sleep duration, mediating biomarkers, and risk of microvascular complications among individuals with type 2 diabetes: a prospective cohort study. Diabetes Res Clin Pract. Published online February 8, 2025. PMID:39929336
8. Xiao ZH, Zhang X, Ji L, et al. Sleep patterns and risk of microvascular complications in type 2 diabetes (UK Biobank) [confirm exact title]. Sleep Health. Published online July 10, 2025.
9. Ohkuma T, Fujii H, Iwase M, et al. Association between sleep duration and urinary albumin excretion in patients with type 2 diabetes: the Fukuoka Diabetes Registry. PLoS One. 2013;8(11):e78968. doi:10.1371/journal.pone.0078968
10. Tan NYQ, Chan J, Cheng CY, Wong TY, Sabanayagam C. Sleep duration and diabetic kidney disease. Front Endocrinol (Lausanne). 2019;9:808. doi:10.3389/fendo.2018.00808
11. Hao Q, Xie M, Zhu L, et al. Association of sleep duration with chronic kidney disease and proteinuria in adults: a systematic review and dose-response meta-analysis. Int Urol Nephrol. 2020;52(7):1305-1320. doi:10.1007/s11255-020-02488-w
12. Park S, Lee S, Kim Y, et al. Short or long sleep duration and CKD: a Mendelian randomization study. J Am Soc Nephrol. 2020;31(12). doi:10.1681/ASN.2020050666
13. Mazidi M, Shekoohi N, Katsiki N, Banach M. Longer sleep duration may negatively affect renal function. Int Urol Nephrol. Published online September 24, 2020. doi:10.1007/s11255-020-02624-6
