Comorbidities treated in primary care in children with chronic fatigue syndrome / myalgic encephalomyelitis: A nationwide registry linkage study from Norway

Chronic fatigue syndrome (CFS), or myalgic encephalomyelitis (ME), is a debilitating disorder, characterized by severe fatigue of unknown cause and often accompanied by other symptoms such as muscle pain and headache [1, 2]. The terms CFS and ME are often used interchangeably, and Norwegian health authorities recommend using the combined term CFS/ME [3]. In addition to fatigue, CFS/ME is associated with symptoms such as post-exertional malaise, pain, unrefreshing sleep, and cognitive impairment [1]. We have previously shown that CFS/ME is threefold more common among women than among men, with highest incidence rates among children and adolescents aged 10 to 19 years [4].

In Norway, children with possible CFS/ME are routinely referred by their general practitioner to hospital for diagnostic workup, while follow-up after diagnosis is the responsibility of the general practitioner. In the present population-based registry study we have used data from Norwegian primary and specialist health care registries with national coverage to further investigate the burden of CFS/ME in children. The study had two main objectives. Our first aim was to describe comorbidities diagnosed in primary care in children diagnosed with CFS/ME in specialist health care with children in the general population as control group. As treatment-seeking study-samples may be different from the general population in terms of registered comorbidities, we also used a comparison group of children with a chronic disease, type 1 diabetes mellitus (T1DM). Our second aim was to describe the timing of the diagnoses from primary care in relation to the timing of the CFS/ME diagnosis.

The mean age at CFS/ME diagnosis among the 1670 children registered at least once in specialist health care with this diagnosis was 14.8 years (standard deviation [SD] 2.5 years) and 66.7 % were girls. Mean age at first diagnosis among the 4937 children in the T1DM control group was 10.9 years (SD 4.2 years) and 46.5 % were girls. The general child population control group consisted of 1 337 508 children (48.7 % girls).

In this large, population-based study, children and adolescents with CFS/ME were significantly more often diagnosed with several conditions in primary care than children with T1DM and children in the general population. These diagnoses not only included conditions closely related to CFS/ME but also depression, anxiety, infections, and migraine. When restricting to primary care diagnoses prior to the specialist health care diagnoses, even stronger differences between children with CFS/ME and children with T1DM were observed. Three out of four children with CFS/ME in specialist health care were registered with a prior diagnosis of weakness / general tiredness in primary care. However, as we had access to specialist health care diagnoses for the period 2008–2014 and primary care diagnoses only from 2006 to 2014, it cannot be ruled out that the proportion with this diagnosis in primary care was even higher. For many children the time span from the first primary care diagnosis of weakness / general tiredness to the first CFS/ME diagnosis was 1 year or longer.

CFS/ME is a rare disease and previous studies have typically included small number of patients. The main strength of this study is the utilization of registry data covering the entire population over a long study period. Children with a serious condition of prolonged nature (in this case CFS/ME) are likely to be followed up more closely in primary care than children without such a condition. By including a control group with another chronic condition (T1DM) we were able to make comparisons of children with CFS/ME both to the general child population and to a chronic condition group.

As in other studies based on observational, routinely collected data, a major limitation is the lack of validity testing of the diagnoses. While ICD-10 code G93.3 is to be used for CFS/ME in specialist health care, there is no specific ICPC-2 code for the condition that is to be used in primary care. The observation that children diagnosed with CFS/ME in specialist health care were far more often registered with the ICPC-2 code for weakness / general tiredness in primary care than children with T1DM and children in the general child population is reassuring with regard to the reliability of the CFS/ME diagnosis from specialist health care. Similarly, ICPC-2 codes for diabetes were registered in primary care for most children with the ICD-10 code for T1DM in specialist health care, but only for a small proportion of the children in the two other groups. The Norwegian Directorate of Health recommends use of the ICD-10 code G93.3 for CFS/ME [3]. Still, some children fulfilling criteria for a diagnosis of CFS/ME might have been registered with other ICD-10 codes. In a recent study, paediatricians generally reported that they follow the guidelines of the Norwegian Paediatric Association [6].

In the present study we found a high frequency of being diagnosed with depression among children with CFS/ME. In a recent case-control study from Norway, children (12–18 years of age) with CFS/ME according to strict clinical criteria had a strong reduction in health-related quality of life compared to healthy peers [7]. That study also reported that children with CFS/ME had eight times higher risk of depressive symptoms, but that this difference could not explain their poor quality-of-life scores. High levels of depression (29 %) have also been reported from a large UK cross-sectional study among children aged 12–18 years with CFS/ME [8]. We found higher frequencies of anxiety diagnoses among children with CFS/ME than in our two control groups. Another study from the UK included children with CFS/ME 7–17 years of age and compared anxiety scores for this group to normal population scores [9].

In a previous study using data from the NPR 2008–2012, we have shown that there are two distinct age peaks in the incidence of CFS/ME (age groups 10 to 19 years and 30 to 39 years) [4]. We hypothesized that this pattern could be explained by primary exposure to infectious agent(s) in adolescents (first age peak) and subsequent reactivation of latent infection (second age peak). In the present study, we have shown that the frequency of infections diagnosed in primary care is higher prior to a diagnosis of CFS/ME than prior to a diagnosis of T1DM, indicating that infections may indeed be among the factors that could cause CFS/ME. In a recent review, it was concluded that the severe fatigue that can afflict patients with a disorders such as multiple sclerosis, Parkinson’s disease, and systemic lupus erythematosus, may be caused by peripheral inflammation and immune activation, and that similar mechanisms may also play a role in the development of CFS/ME [10]. In a cohort of 301 adolescents with infectious mononucleosis from Chicago, Illinois, 13 % met the criteria for chronic fatigue syndrome after 6 months [11]. Also, in an adult UK cohort consisting of patients with infectious mononucleosis, influenza or tonsillitis (but no control group without infection), the odds of clinically diagnosed fatigue was found to be four times higher for infectious mononucleosis when compared to influenza and nearly seven times higher for infectious mononucleosis when compared to tonsillitis [12]. These results fit well with results from the present study, where we observed that the greatest difference between the CFS/ME group and either control group was for infectious mononucleosis. However, influenza, acute tonsillitis, strep throat, and pneumonia were also significantly more often observed in the CFS/ME group, indicating that different infections might trigger the development of CFS/ME. In a previous study we found a strong, positive association between pandemic influenza A (H1N1) and CFS/ME, but no changes in risk of CFS/ME after pandemic influenza vaccination [13]. Similarly, an earlier Norwegian study showed no relation between vaccination against meningococcal disease and CFS/ME, while a tenfold increased risk for CFS/ME was observed after mononucleosis with clear symptoms [14].

We found a somewhat higher frequency of asthma among children with CFS/ME than in the T1DM and general child population control groups. In US adults, high rates of asthma have been described prior to the onset of CFS [15], and from Taiwan, the risk has been reported to increase following atopy [16]. We also observed that one in ten patients with CFS/ME had been diagnosed with migraine in primary care, which was four times as frequent as in patients with T1DM. The link between CFS/ME and migraine seems rather well established. For instance, a 1.5-fold increased risk of CFS was described in adults with migraine in another study from Taiwan [17]. Also, in a study including two different adult groups of patients with CFS, migraines were often reported (39–69 %) [18]. Similar to our findings, high frequencies of headaches and muscle pain were observed among children (mean age 15.2 years) with CFS/ME in Dutch paediatric departments [19].

The findings in the present study confirm results from previous studies in that children with CFS/ME show higher levels of depression, anxiety, migraine, and muscle pain than their peers. Furthermore, children with CFS/ME are often diagnosed with infections, in particular prior to the CFS/ME diagnosis. The long time spans observed from the first diagnosis of weakness / general tiredness in primary care to a specialist health care diagnosis of CFS/ME might indicate that the treatment of these patients is sometimes not optimal.