Skufca et al: Incidence rates of GBS, CFS/SEID and POTS prior to HPV vaccination in Finland

[mango]

Incidence rates of Guillain Barré (GBS), chronic fatigue/systemic exertion intolerance disease (CFS/SEID) and postural orthostatic tachycardia syndrome (POTS) prior to introduction of human papilloma virus (HPV) vaccination among adolescent girls in Finland, 2002–2012

J. Skufca a, b, , , J. Ollgren a, , E. Ruokokoski c, , O. Lyytikäinen a, , H. Nohynek c,

a Department of Infectious Diseases, National Institute for Health and Welfare (THL), Helsinki, Finland
b European Programme for Intervention Epidemiology Training (EPIET), European Centre for Disease Prevention and Control (ECDC), Stockholm, Sweden
c Department of Health Protection, National Institute for Health and Welfare (THL), Helsinki, Finland

Papillomavirus Research Volume 3, June 2017, Pages 91–96
Received 29 September 2016, Revised 9 March 2017, Accepted 9 March 2017, Available online 16 March 2017
http://dx.doi.org/10.1016/j.pvr.2017.03.001

Highlights

• CFS/SEID, GBS and POTS can be found in all distinct age- and sex-related diagnostic entities.
  
  
• Significant increase in CFS/SEID and POTS but not in GBS rates in years before HPV immunization.
  
  
• Rates based on register data should be interpreted with caution, especially for non-specific diagnostic entities.
  
  
• Factors contributing to clinicians setting the diagnoses and using different ICD-10 codes should be explored.

Abstract
Background
In Finland a vaccination programme against human papillomavirus (HPV) was introduced in November 2013 for girls aged 11–12 years with a catchup for girls 13–15 years. Allegations that HPV vaccine is causing Guillain Barré syndrome (GBS) and non-specific diagnostic entities, such as chronic fatigue syndrome/systemic exertion intolerance disease (CFS/SEID) and postural orthostatic tachycardia syndrome (POTS), continue to surface. We examined population register-based incidence rates of CFS/SEID, GBS and POTS to provide baseline data for future HPV vaccine safety evaluations.

Methods
First diagnosis of CFS/SEID, GBS and POTS in girls aged 11–15 years were obtained from the National Hospital Discharge Register during 2002–2012. We considered the following ICD-10 codes: G93.3 for CFS; G61.0 for GBS and G90.9, G90.8, G93.3, I49.8 for POTS. We calculated incidence rates per 100,000 person-years with 95% confidence intervals (CI).

Results
In total, 9 CFS/SEID, 19 GBS and 72 POTS cases were identified. The overall incidence rate was 0.53/100,000 (95% CI; 0.27–1.01) for CFS/SEID, 1.11 (95% CI; 0.71–1.74) for GBS and 4.21 (95%CI; 3.34–5.30) for POTS. Significant relative increase in annual incidence rate with a peak in 2012 was observed in CFS/SEID (33% (95% CI; 3.0–70.3: p=0.029) and POTS (16.5% (95% CI; 7.8–25.9: p<0.05), but not in GBS (5.4% (95% CI; −8.4–21.3: p=0.460).

Conclusions
Our findings provide baseline estimates of CFS/SEID, GBS and POTS incidences in Finland. However, rates based on register data should be interpreted with caution, especially for non-specific diagnostic entities for which internationally and even nationally agreed criteria are still being discussed. To assess the associations with HPV vaccine, methods using register linkage for cohort and self-controlled case series should be explored in addition to factors contributing to patients seeking care, treating physicians setting the diagnoses, and their preference of using of codes for these clinical entities.

Keywords
Papillomavirus vaccines; Vaccination adverse effects; Incidence rates; Guillain Barré syndrome; Chronic fatigue syndrome/Systemic exertion intolerance disease; Postural orthostatic tachycardia syndrome; Finland

http://www.sciencedirect.com/science/article/pii/S2405852116300696

 

[RogerBlack]

"National Hospital Discharge Register" - if this not a bad translation or misleadingly named, it would seem to be a massive problem. I can't imagine a significant number of people will be hospitalised normally. 1 in 200000 person-years (if average length of disease is 20 years, 0.01% of the population) seems really, really low compared with other estimates.

 

[Dolphin]

"National Hospital Discharge Register" - if this not a bad translation or misleadingly named, it would seem to be a massive problem. I can't imagine a significant number of people will be hospitalised normally. 1 in 200000 person-years (if average length of disease is 20 years, 0.01% of the population) seems really, really low compared with other estimates.

Yes, they appeared to have missed a lot of individuals either through diagnoses occurring elsewhere, different codes being used and/or a lack of awareness in Finland.

The study population was the complete population (average population of 5,300,000) with permanent residence in Finland with a focus on the target group, i.e. females of age 11–15 years between 2002 and 2012 registered in the Finnish Population Registry. We obtained the information on the CFS/SEID, GBS and POTS diagnoses from the National Hospital Discharge Register (HILMO), which since 1996 contains comprehensive individual level healthcare records on inpatients and since 1998 also outpatient visits provided by all hospitals in Finland. For each hospitalization and hospital outpatient visit, reporting of data to HILMO is mandatory. The basic variables in the register include personal identity number, area of residence, admission and discharge dates, outpatient visit dates and diagnoses of patients.

Since 1996, the discharge diagnoses are reported as International Classification of Diseases version 10 (ICD-10) codes. In the analysis we used the first registered ICD-10 code episode for each patient and we considered the following codes: G93.3 for CFS/SEID cases; G61.0 for GBS cases and G90.9, G90.8, G93.3, I49.8 for POTS cases. We calculated sex- and age-specific incidence rates per 100,000 personyears with 95% confidence intervals (CI) by dividing the number of CFS/SEID, GBS and POTS cases in each sex/age category by the total number of person years at risk in the same category. We estimated a relative increase in incidence rate per year by Poisson regression. For the data analysis we used the Stata software package, Version 14 (USA: StataCorp LP).

 

[Dolphin]

These are the figures for all ages. Extremely low.

total of 58,335 738 person-years for females and males in all agegroups in overall population of Finland, 3490 972 person-years for females and males of age 11–15 years and 1710 376 person-years for females of age 11–15 years were included in the analysis in the study period years 2002–2012. A total of 327 CFS/SEID, 1511 GBS and 3788 POTS cases were identified overall. The average annual incidence rate was 0.56/100,000 (95% CI; 0.50–0.62) for CFS/SEID, 2.59 (95% CI; 2.46–2.72) for GBS and 6.49 (95%CI; 6.29–6.70) for POTS.

 

[Dolphin]

Our findings on the CFS/SEID slightly differed from a Norwegian study [24], which is to date the largest register-based study on CFS/ SEID and the only study describing the distribution by both, sex and age. Contrary to our study, they found a fairly stable overall incidence rates over the years. There have been several attempts in the past to update case definition and diagnostic criteria for CFS/SEID [25]. It may be that the increased awareness among clinicians about CFS/SEID has led to increasing ascertainment and coding of these syndromes in Finland. While we observed an increasing trend in CFS/SEID during the years of this study (2002–2012), other diagnostic entities with overlapping symptoms, such as unspecified rheumatism, burn-out, and neurasthenia, had a decreasing trend during the same time period (data not shown). The increase in CFS/SEID and decrease of other codes describing similar symptoms during the same time period might be coincidental, but it is also possible that the CFS/SEID coding became increasingly used among physicians replacing the other previously used codes. The Norwegian study also found a much higher overall rate compared to our study (25.8 vs. 0.56 respectively), although their ICD-10 code used for the CFS/SEID case-search was the same as in our study (G93.3) and only first registered diagnosis was used in the analysis [24]. However, they also included the diagnoses assigned in the outpatient clinics, which might be a major reason for the differences with our rates, as HILMO does not capture information on the outpatient clinic visits done outside hospital. A higher incidence in women than in men is consistent in both studies, although the Norwegians found a peak of all cases in the age group 10–19 years and 30–39 years vs. the peak in age group 26–30 years and not observed among the 11–15 years old girls in our study. Their estimated incidence rate among women of age 10–14 years was approximately 60 compared to 0.53 among 10–15 years old girls in our study [24]. CFS/SEID has been associated with preceding infections [26] and vaccines, such as the pandemic influenza [26] and the HPV vaccination, but the association between the CFS/SEID with HPV vaccines has not been confirmed [27].

 

[Dolphin]

The limitation of this hospital register-based study is that the number CFS/SEID and POTS cases are likely to be under-reported as these two syndromes are often not diagnosed or misdiagnosed. The diagnoses in our study were also not validated by the review of the patient records, but will be reviewed in the framework of a future HPV vaccine safety study. Our study includes all information on inpatients and outpatient visits in hospitals, thus one of the limitations is that HILMO does not include the outpatient primary care visits, nor the visits in private health care.

 

[Dolphin]

We might have missed the less severe GBS, CFS/SEID and POTS from the primary care, but due to severity of these entities, which can mostly occur in the inpatient setting [18], and due to very few private health-care providers in Finland, we believe that we captured majority of severe cases from HILMO.

I am not convinced. I think a lot of severely affected people would avoid being inpatients: I know I have.

 

[RogerBlack]

I am not convinced. I think a lot of severely affected people would avoid being inpatients: I know I have.

I think this is a different set of 'severe' patients - it's probably those who for one reason or another - suspected abuse, someone reported they aren't coping, accidents/fires/falls who get into hospital for some other reason than simply a static case of CFS.
It's not people who are 'managing'.
By this definition, they may be capturing 'severe' cases - but while this set of patients is interesting, it's a tiny, tiny fraction of those severely affected enough that CFS has a devastating impact on their lives.