Health care utilization and costs of patients with ME/CFS in Norway, a retrospective registry study, 2026, Anthun et al

[TiredMathematician]

Health care utilization and costs of patients with ME/CFS in Norway, a retrospective registry study

Anthun, Kjartan Sarheim; Halvorsen, Thomas; Melby, Line

Abstract
Background
Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex chronic illness without any known cause or treatment. Diagnostic processes are typically long, and with no established treatment option available, little is known about patients’ total health care utilization and costs. This study analyses health care utilization and costs among patients with ME/CFS in Norway and examines whether they differ from those observed in selected comparison groups.

Methods
We conducted a retrospective, population-based registry study using linked individual-level data from Norwegian national registries. Information on health care utilization, including hospital treatments, GP contacts and primary health care was collected and linked from national registries. Comparison groups included a random sample of the population and patients diagnosed with Lyme disease, breast cancer or fibromyalgia. Costs were measured as total hospital costs, while costs from general practitioners (GP) and primary care utilization were estimated based on GP tariff, estimated out of pocket payment to GP and national averages for each provided primary care service. Methods included descriptive statistics and regression of costs.

Results
Patients with ME/CFS utilize health care services less than other comparable diagnoses, but more than a random sample of the population. Health care costs among patients with ME/CFS were almost three times those observed in the random population sample, but were lower than the costs among patients with breast cancer.

Conclusions
Health care use and costs were particularly high for hospital treatments, while at the same time no clear treatment option is available. Greater attention should be paid to comorbidities among patients with ME/CFS. Further research is needed to determine whether some of their care could be provided more appropriately at other levels of the health care system rather than in hospitals.

Web | DOI | BMC Health Services Research

 

[Jonathan Edwards]

The conclusions do not seem to follow from the results.
I wonder whether this sort of exercise has any value.

 

[Midnattsol]

I need to reread this carefully as I just skimmed it, but how can they say anything about levels of care when the diagnostic code g93.3 is used only in hospitals? Is it the selection that is done based on at least one g93.3 code and then all healthcare costs of that patient used in the calculations? Do costs taper off as patients stop going back to their GP? I’m sure I cost more prior to diagnosis than after since there was no point going back except for yearly documentation for disability accommodations.

Data collection of g93.3 use in 2016-2018 is a bit unfortunate with the guidelines for diagnosis to be provided by GPs came in 2016, which means patients may not go to hospital and have g93.3 registered. I have no idea if I do as I was diagnosed after this change and most of my hospital visits and tests were before ME was on the table so I don’t think it would have been used as a preliminary diagnosis either because it was more “MR to check for MS” type things.

I want a registry study that just look at (potential) comorbidity and or change in diagnosis..

 

[Creekside]

Seems like nonsense. Patients with diagnoses of ME/CFS will utilize health services less because they'll be told: "We can't help you." PWME without diagnoses will cost extra, because they go from doctor to doctor hoping for some help.

Maybe the result will be that doctors will be encouraged to give ME/CFS diagnoses for any complaint that remotely fits, because according to this study, that will lower health care costs.

 

[Midnattsol]

Data for this study was provided by Statistics Norway, the Norwegian Patient Registry, Norwegian Control andPayment of Health Reimbursements Database and Norwegian Registry for Primary Health Care. These registers are national registers that are considered to be of good quality and completeness, well suited for use in research with linked data

While true we regard the registries as of high quality, some of the data can be.. questionable due to bad coding practices. For example you may find highly specific treatments being provided at hospitals or other institutions where they are not able to be performed. Or diagnostic codes are used wrong - I have no doubt someone could be coded as having chronic fatigue syndrome if they came for fatigue without actually having been diagnosed with CFS.

On calculating the cost there is another potential reason for it being lower than what it actually is: using the length of stay in hospital is fine, but I'm unsure about the use of reimbursements. For example units can be reimbursed for screening and diagnosing a patient with malnutrition, but we know this is seldom registered despite the reimbursement incentive. Another code that might be relevant for us would be rehabilitation codes, and they are also potentially lacking. Depending on the type of treatment patients in different group receives this could give biased estimates if one group receive more of treatments that are often not coded correctly.

They did include an analysis of cost depending on when the diagnostic code was first use, it does drop after the first year:

When we looked at costs separately for new ME/CFS patients compared to patients whose first diagnosis occurred before the observation period, we saw that the for the non-incident patients there was falling costs each year, indicating that the peak occurred in the years before 2016. For all incident patients, total costs were low in the year(s) prior to the first diagnosis, and lower in the years after, but not as low as the years prior to the first diagnosis. In other words, a large share of the increased health care costs for ME/CFS patients is related to the events around and following their diagnosis.