The average person on hemodialysis lives less than three years after starting treatment, a figure that has remained stubbornly stable for decades.1PubMed Central. Consequences of frequent hemodialysis: comparison to conventional hemodialysis and transplantation That number, though, obscures enormous variation. A relatively healthy 40-year-old starting dialysis with a functioning vascular access and no diabetes faces a profoundly different outlook than a frail 80-year-old with heart failure who begins treatment through a catheter in a hospital bed. Understanding what drives those differences matters far more than a single average.
The First Months Carry the Highest Risk
The period right after starting dialysis is disproportionately dangerous. In a study of over 1.1 million patients with end-stage kidney disease, roughly 8% died within the first 90 days of beginning treatment.2PubMed Central. A Machine Learning Model for Predicting Mortality within 90 Days of Dialysis Initiation Data from major registries in the United States, Canada, and Europe show a consistent pattern: deaths in the first three months account for about 35 to 36% of all deaths that occur in the entire first year.3Kidney International. Increased mortality early after dialysis initiation: a universal phenomenon This early spike is partly explained by the fact that many people start dialysis during a health crisis rather than through a planned transition. Beginning dialysis as an inpatient, for example, roughly doubles the risk of dying within the first year compared to starting as an outpatient.4JAMA Internal Medicine. One-Year Mortality After Dialysis Initiation Among Older Adults
For people who survive the initial adjustment, the risk drops but never returns to that of the general population. This is one reason the “less than three years” average can mislead: it is heavily weighted by patients who die early, while many others live considerably longer.
Age Changes Everything
No single factor predicts dialysis survival better than the age at which you start. In one study of 232 hemodialysis patients, the mortality rate among those under 65 was about 17%, compared with roughly 50% in those over 65. The gap widened with advancing age: mortality was about 45% in the 65-to-74 group, 55% in the 75-to-84 group, and 75% in patients 85 and older.5PubMed Central. Mortality in Hemodialysis Patients Over 65 Years of Age Being 85 or older at the start of dialysis was independently associated with nearly twice the risk of dying within the first year, even after accounting for other health conditions.4JAMA Internal Medicine. One-Year Mortality After Dialysis Initiation Among Older Adults
Younger patients who begin dialysis without significant complications can live for a decade or more. A 30-year-old starting dialysis is in a fundamentally different category from a 75-year-old, and blanket survival statistics rarely make that distinction visible.
Diabetes, Heart Failure, and the Weight of Comorbidities
Diabetes is the leading cause of kidney failure in most countries, and dialysis patients with diabetes consistently fare worse than those without it. A study from the Lombardy Registry found that age, type of diabetes, severe vascular disease, and cirrhosis all significantly shortened survival. Even diabetic patients who arrived at dialysis without any additional risk factors still had a poor prognosis and high rates of illness.6PubMed. Prognosis of diabetic patients on dialysis: analysis of Lombardy Registry data
The interplay between pre-dialysis health and survival on dialysis is striking. One study of diabetic patients on maintenance dialysis found that factors present before dialysis began, particularly older age, longer diabetes duration, lower blood pressure variability, body mass index, and existing cardiovascular disease, were more powerful predictors of death than anything measured after dialysis started.7Scientific Reports. Risk Factors Before Dialysis Predominate as Mortality Predictors in Diabetic Maintenance Dialysis patients To illustrate how dramatically risk can shift: a model estimating one-year mortality for diabetic dialysis patients showed that a 60-year-old nonsmoker with a history of heart attack and moderate functional ability had a 27% chance of dying within a year, while the same person at age 70 with poor functional ability had a 68% chance.8PLoS ONE. Predicting Mortality in Patients with Diabetes Starting Dialysis
Having four or more chronic conditions at dialysis initiation raised the risk of dying within one year by about 50%.4JAMA Internal Medicine. One-Year Mortality After Dialysis Initiation Among Older Adults Heart failure, in particular, interacts with dialysis in ways that compound risk, partly because the fluid shifts involved in treatment stress an already weakened heart.
What Dialysis Patients Actually Die From
The causes of death on dialysis reflect the profound cardiovascular toll of kidney failure. In a Chinese study of hemodialysis patients, the leading causes of death were cardiovascular disease at about 30%, cerebrovascular events (strokes) at roughly 20%, and infection at about 17%.9PubMed Central. Study on causes of death and influencing factors in hemodialysis patients with End-Stage renal disease A large Brazilian cohort told a somewhat different story, with infection as the leading killer at 41%, followed by cardiovascular disease at 30%.10PubMed. Analysis of adjudicated causes of death in a large cohort of Brazilian patients undergoing dialysis The difference between regions likely reflects variation in access to care, infection control practices, and population health profiles. Regardless of setting, cardiovascular disease and infection consistently dominate. Cancer, malnutrition, and dialysis withdrawal make up smaller but meaningful shares.
Vascular Access and Why It Matters So Much
For people on hemodialysis, how the blood is accessed for treatment turns out to be a surprisingly powerful predictor of survival. Patients who start dialysis through a surgically created fistula in the arm have the best outcomes, those with a synthetic graft do somewhat worse, and those dialyzed through a catheter fare the worst. A meta-analysis found that catheters were associated with about 72% higher risk of death compared with fistulas, while grafts carried about 26% higher risk.11PubMed Central. Mortality outcomes associated with vascular access types in hemodialysis for ESRD: a systematic review and meta-analysis Another study showed that infection-related death was roughly three times more common among catheter users than among those with a fistula.12Kidney International. Vascular access and increased risk of death among hemodialysis patients
The picture is more complicated than “fistulas save lives,” though. A study comparing patients who successfully started dialysis through a fistula, patients who ended up on a catheter after a failed fistula attempt, and patients who simply started with a catheter found that even the failed-fistula group had much lower mortality than the catheter-only group. This suggests that the type of patient who gets referred for fistula creation early tends to be healthier overall, and that a large share of the fistula’s apparent survival advantage actually reflects patient factors rather than the access type itself.13PubMed Central. The Survival Benefit of “Fistula First, Catheter Last” in Hemodialysis Is Primarily Due to Patient Factors Still, catheter infections are a real and direct threat, so minimizing catheter time remains a practical goal.
Peritoneal Dialysis Versus Hemodialysis
Peritoneal dialysis, which uses the lining of the abdomen rather than an external machine to filter blood, has been compared with hemodialysis in many studies. The general pattern is that peritoneal dialysis is associated with similar or better survival in the first year or two, though longer-term comparisons are more mixed and vary across patient subgroups.14PubMed. Mortality studies comparing peritoneal dialysis and hemodialysis: what do they tell us? One study using methods designed to account for the fact that patients often switch between modalities found that peritoneal dialysis was associated with roughly half the risk of death compared with hemodialysis during the first two years.15PubMed Central. Comparing mortality of peritoneal and hemodialysis patients in the first 2 years of dialysis therapy: a marginal structural model analysis
These comparisons come with a significant caveat. People who choose home-based peritoneal dialysis tend to be younger, more independent, and have fewer additional health problems than those who end up on in-center hemodialysis. Several observational studies report a survival benefit for home dialysis, but it remains unclear whether the advantage comes from the treatment setting or from the characteristics of the people who choose it.16PubMed Central. Controversies in targeting a “home-dialysis first” renal replacement therapy policy No large randomized trial has settled the question.
Residual Kidney Function and Nutritional Health
Even after kidney failure has progressed to the point of needing dialysis, most people retain some residual function in their native kidneys. That leftover capacity, even a small amount, consistently predicts better survival for both hemodialysis and peritoneal dialysis patients.17American Journal of Kidney Diseases. The Importance of Residual Kidney Function in Dialysis Patients Residual function helps clear toxins that dialysis machines handle poorly, and preserving it is considered a key clinical goal.18PubMed Central. Advances in Understanding and Management of Residual Renal Function in Patients with Chronic Kidney Disease Certain medications, adequate blood pressure control, and avoiding kidney-toxic drugs can slow the loss of this residual function.
Nutritional status is another powerful marker. Serum albumin, a protein in the blood that reflects both nutrition and inflammation, is one of the strongest predictors of death on dialysis. In hemodialysis patients with heart failure, lower albumin at the start of treatment independently predicted death within the first year.19PubMed Central. Hypoalbuminaemia and One-Year Mortality in Haemodialysis Patients with Heart Failure: A Cohort Analysis Across both hemodialysis and peritoneal dialysis patients, each small drop in plasma albumin was consistently associated with more than a 20% increase in the risk of death.20Scientific Reports. Dynamic mortality predictions from serum albumin in dialysis patients using robust joint models with competing risks Low albumin can reflect poor protein intake, chronic inflammation, or both, so it serves as a red flag for deteriorating health even when other numbers look stable.
More Frequent Dialysis and Improving Trends
Standard hemodialysis in most countries involves three sessions per week, each lasting about four hours. A randomized trial tested what would happen if patients came in six times a week instead. Frequent dialysis was associated with significantly lower risk of death or worsening heart function, and lower risk of death or declining physical health, compared with the standard three-times-a-week schedule.21PubMed Central. In-center hemodialysis six times per week versus three times per week The tradeoff is the practical burden: twice the number of clinic visits per week is a major lifestyle commitment, and access to daily dialysis remains limited in many areas.
There is evidence that outcomes have improved modestly over time. Swedish registry data covering 2006 to 2015 showed that one-year mortality for hemodialysis patients dropped from about 13% to about 11% over that decade. The improvement disappeared after accounting for changes in treatment practices like better blood-pressure management and use of newer medications, suggesting that the gains came from evolving clinical care rather than from any shift in the patient population.22PubMed Central. Association Between Implementation Of Novel Therapies And Improved Survival In Patients Starting Hemodialysis: The Swedish Renal Registry 2006-2015
How Dialysis Compares to a Kidney Transplant
A kidney transplant is the only treatment for kidney failure that consistently offers a large survival advantage over dialysis. In a study that followed transplant-eligible patients on the waiting list, those who received a transplant gained about 2.4 extra years of life over a 10-year follow-up compared with those who stayed on dialysis.23JAMA Network Open. Survival Benefit of First Single-Organ Deceased Donor Kidney Transplantation Compared With Long-term Dialysis Across Ages in Transplant-Eligible Patients With Kidney Failure Even among people over 70, transplant recipients had about 38% lower mortality than matched dialysis patients over a median follow-up of less than two years, and five-year survival was roughly 80% for transplant recipients versus 53% for those on dialysis.24PubMed. Survival after kidney transplantation compared with ongoing dialysis for people over 70 years of age: A matched-pair analysis
This benefit isn’t immediate. Transplant surgery itself carries risk, and the survival curves typically don’t separate until several months afterward. But among those who clear the early post-surgical period, the long-term advantage is substantial. A more recent analysis showed that even kidneys from higher-risk donors conferred a survival gain of several months over staying on the waitlist, while kidneys from lower-risk donors offered larger gains.25PubMed Central. Survival Benefits of Deceased Donor Kidney Transplant vs Waitlisting The practical limitation is organ availability: most people who could benefit from a transplant spend years on a waitlist, and many never receive one.
Frailty as a Hidden Predictor
Frailty, a state of reduced physical reserve marked by weakness, slow walking speed, exhaustion, and low activity, affects a disproportionate number of dialysis patients across all ages. Frail dialysis patients face higher rates of hospitalization, infection, cardiovascular events, and death.26PubMed. Frailty in patients on dialysis In a study of patients initiating dialysis, those who were frail at baseline had about 57% higher mortality after adjusting for age, other conditions, and laboratory values. Frailty was a stronger predictor than the level of kidney function at which dialysis was started.27PubMed Central. Frailty, Dialysis Initiation, and Mortality in End-Stage Renal Disease
This finding has practical implications for older patients and their families weighing whether to start dialysis. A prospective study found that starting dialysis was associated with a significant reduction in the risk of death for older people with advanced kidney disease, regardless of their frailty level.28PubMed Central. A prospective, observational study of frailty, quality of life and dialysis in older people with advanced chronic kidney disease But frailty also predicts worse quality of life and more treatment complications, which is why the decision to start dialysis in the very elderly and very frail is increasingly treated as a shared decision between patient and physician rather than an automatic one. Depression, which is common in dialysis patients, compounds the problem by reducing treatment adherence and quality of life, and is independently associated with higher mortality.29PubMed Central. Depression in Chronic Kidney Disease and End-Stage Renal Disease: Similarities and Differences in Diagnosis, Epidemiology, and Management
The Racial Survival Paradox
One of the most puzzling findings in dialysis research is that Black Americans, who face higher rates of kidney failure and worse health outcomes in the general population, tend to survive longer on dialysis than white Americans. In one study, five-year mortality was 34% for Black patients versus 56% for white patients, and the difference persisted after adjusting for a long list of health factors including diabetes, heart disease, and markers of inflammation.30PubMed Central. Inflammation and the paradox of racial differences in dialysis survival
The picture becomes more complicated with age. A large JAMA study found that the overall survival advantage for Black dialysis patients was driven almost entirely by those over 50. Among younger patients aged 18 to 50, Black patients actually had significantly higher mortality than white patients. At ages 18 to 30, for instance, mortality was roughly 28% for Black patients versus 14% for white patients. The survival advantage only appeared and widened in older age groups.31PubMed Central. Association of race and age with survival among patients undergoing dialysis Researchers suspect this age-dependent reversal reflects complex interactions between transplant access (younger Black patients face longer waits), health conditions at dialysis initiation, and biological factors that are not yet fully understood.
Global Disparities in Dialysis Survival
Where you live matters enormously. A global study found that in low-income and lower-middle-income countries, dialysis withdrawal due to cost of care was reported as a cause of death, something that essentially never appears in wealthier nations.32PubMed Central. Dialysis Outcomes Across Countries and Regions: A Global Perspective From the International Society of Nephrology Global Kidney Health Atlas Study The median annual cost of dialysis worldwide is roughly $19,000 per patient, and government coverage varies dramatically by country wealth. In high-income countries, about half require no copayment for peritoneal dialysis; in low-income countries, none do. When people cannot afford treatment, they stop treatment, and stopping dialysis is essentially a terminal decision.
When Patients Choose to Stop
Dialysis withdrawal is a recognized end-of-life pathway. For patients who discontinue treatment and enter hospice, time is extremely short. In a study of nearly 2,000 patients who stopped dialysis and enrolled in hospice, average survival afterward was just 7.4 days, with a range of zero to 40 days. By comparison, hospice patients with other terminal diagnoses survived an average of about 54 days.33PubMed Central. Survival after dialysis discontinuation and hospice enrollment for ESRD The brevity of this window matters for planning: families who are considering stopping dialysis need to understand that the transition from treatment to death is typically measured in days, not weeks, leaving very little time for the extended hospice support that many people associate with end-of-life care.
This rapid timeline also highlights the degree to which dialysis is doing life-sustaining work, even in patients who feel that their quality of life is poor. The decision to stop is deeply personal and increasingly supported by palliative care teams, particularly for older and frailer patients who may feel the burdens of treatment outweigh the benefits. Having that conversation early, while the patient can still participate meaningfully, is something clinicians are pushing for more than they used to.