Digital Kidney Monitoring Passes the Clinical Test but Fails to Cut Costs, Danish Trial Finds

Remote monitoring has long been pitched as the fix for overburdened hospitals: let patients report their symptoms and measurements from home, and clinics can reserve face-to-face appointments for those who genuinely need them. For people living with chronic kidney disease, a condition that demands lifelong surveillance of kidney function, blood pressure and quality of life, the promise is especially appealing. But a rigorous new economic evaluation from Denmark suggests that the financial case for digital follow-up is far less clear-cut than the technology industry might hope. In the PROKID randomised trial, published in BMC Health Services Research, researchers found that patient-reported outcome based follow-up was clinically non-inferior to routine outpatient care, yet it delivered no cost savings and no statistically significant improvement in quality-adjusted survival over eighteen months.

The trial, led by Birgith Engelst Grove of AmbuFlex, Centre for Patient-reported Outcomes at Gødstrup Hospital and colleagues at Aarhus University and the Danish Centre for Health Economics, enrolled 152 adults with chronic kidney disease. Participants were randomly allocated in a three-arm design. Forty-seven patients continued with usual outpatient follow-up, serving as the control group. Fifty-four received follow-up built around patient-reported outcomes delivered through telephone consultations, a model the investigators call PRO-tele. The remaining fifty-one were assigned to PRO-based remote monitoring, or PRO-remote, in which patients submitted structured digital questionnaires from home and clinicians responded based on the answers. The trial was registered with ClinicalTrials.gov under identifier NCT03847766, retrospectively registered on 23 January 2019.

The clinical logic of the intervention rests on a concept known as the patient-reported outcome, or PRO. Rather than relying solely on laboratory values such as estimated glomerular filtration rate, the standard measure of kidney filtering capacity, PRO-based care systematically collects patients’ own accounts of symptoms, wellbeing and treatment burden at regular intervals. Algorithms or clinicians then flag responses that suggest deterioration, triggering contact or an earlier appointment. In theory, this creates a safety net that catches problems between scheduled visits while allowing stable patients to skip unnecessary trips to hospital. The PROKID investigators had already established that both digital models were non-inferior to usual care on kidney function; the new study asked the harder question of whether they also made economic sense.

To answer it, the team conducted a full cost-effectiveness analysis over an eighteen-month horizon, following the CHEERS standards for health economic evaluations. Individual-level costs were drawn from Danish national registers, an unusually rich data source that captures nearly every contact patients have with the health system. The cost categories included outpatient and inpatient care, primary care services, medication, and productivity losses. Effects were expressed in two ways: quality-adjusted life-years, or QALYs, which combine survival with health-related quality of life as measured by the EuroQol-5L questionnaire, and the number of patients who met the predefined non-inferiority margin for estimated glomerular filtration rate. Incremental cost-effectiveness ratios were calculated for each intervention against usual care and for the two digital models against each other, with uncertainty quantified through 500 bootstrap replications and summarised in cost-effectiveness acceptability curves.

The headline numbers tell a story of near-equivalence rather than transformation. Mean total costs during the intervention period were €16,664 for the control group, €19,300 for PRO-tele and €18,486 for PRO-remote. Notably, the increases in median costs from the pre-intervention period were similar and statistically non-significant across all three arms: €1,254 for usual care, €1,104 for PRO-tele and €1,009 for PRO-remote. In other words, none of the groups experienced a cost trajectory that meaningfully diverged from the others once baseline spending was taken into account. Chronic kidney disease is expensive to manage regardless of how follow-up is organised, and the digital models did not bend that curve.

On the effectiveness side, PRO-tele produced a small, non-significant QALY gain of 0.08 compared with usual care, with a 95 percent confidence interval running from −0.03 to 0.19, at additional costs of €1,572. That combination yields an incremental cost-effectiveness ratio of €19,552 per QALY gained. PRO-remote showed even smaller differences: €347 in extra costs and a QALY difference of 0.01, with a confidence interval from −0.11 to 0.14, producing an ICER of €31,835 per QALY. Both interventions achieved kidney function progression comparable to the control group, confirming the earlier non-inferiority finding. The confidence intervals, however, straddle zero, which means the data cannot rule out either a modest benefit or a modest harm in quality of life for either digital model.

The cost-effectiveness acceptability curves add a probabilistic layer to these point estimates. At conventional willingness-to-pay thresholds of €30,000 and €50,000 per QALY gained, the probability that either intervention was cost-effective ranged from roughly 40 to 70 percent. Those figures sit in an ambiguous middle zone. They are far from the near-certainty that health systems typically want before scaling a programme nationally, yet they are also not evidence of poor value. The uncertainty stems largely from the modest sample size of 152 patients and the relatively short eighteen-month follow-up, during which few patients with chronic kidney disease die or progress to end-stage renal disease. Longer horizons might reveal differences that this trial could not detect.

The authors’ conclusion is refreshingly candid for a field often accused of overselling digital health. PRO-based follow-up interventions resulted in comparable QALYs and costs relative to usual outpatient follow-up, they report, and the absence of cost savings or significant QALY gains means implementation decisions should also weigh non-economic factors, including patient preferences and workflow integration. That framing matters because cost-effectiveness is only one dimension of the decision. Remote follow-up may still be worth adopting if patients prefer fewer hospital visits, if it frees clinic capacity for other patients, or if it fits smoothly into clinicians’ routines. Conversely, a model that costs the same as usual care but adds digital friction for patients and staff offers little reason to switch.

For the broader debate about telehealth, PROKID offers a sobering data point. The pandemic era produced a wave of enthusiasm for remote care, often based on feasibility studies rather than randomised economic evidence. This trial is one of the few to subject a PRO-based kidney follow-up model to a full economic evaluation alongside a non-inferiority assessment, using register-based costs that avoid the recall bias and incomplete accounting that plague many telehealth analyses. Its message is that clinical equivalence is achievable, but equivalence is not the same as efficiency. Digital follow-up did not reduce total healthcare spending, likely because any savings from fewer outpatient visits were offset by other contacts, intervention infrastructure and the underlying disease burden.

What happens next will depend on how health systems interpret ambiguous economics. A decision-maker applying a €30,000 per QALY threshold would see PRO-tele as a coin flip with a slight edge, while PRO-remote hovers closer to the threshold boundary. The Danish team’s work, funded through the HEALTH CONNECT research programme of Central Denmark Region, provides the evidentiary foundation for that judgement rather than dictating the answer. As chronic disease prevalence climbs and outpatient capacity strains, the pressure to adopt remote monitoring will only grow. PROKID’s contribution is to replace hype with numbers: the technology works clinically, it does not save money in the short term, and the choice to deploy it should rest on patient experience, workflow realities and how much a health system is willing to pay for the possibility of a small quality-of-life gain.

Subject of Research: Cost-effectiveness of patient-reported outcome based digital home monitoring versus usual outpatient follow-up in chronic kidney disease

Article Title: Digital home monitoring versus usual outpatient follow-up in chronic kidney disease: cost-effectiveness of the PROKID randomised non-inferiority trial

Article References: Grove, B. E., Skovsgaard, C. V., Wiben, A., Schougaard, L. M. V., Ivarsen, P., & De Thurah, A. (2026). Digital home monitoring versus usual outpatient follow-up in chronic kidney disease: cost-effectiveness of the PROKID randomised non-inferiority trial. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15795-2

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15795-2

Keywords: chronic kidney disease, remote patient monitoring, patient-reported outcomes, cost-effectiveness analysis, QALY, telehealth, randomised controlled trial, health economics, eGFR, outpatient follow-up, PROKID trial, Denmark