Responses

Download PDFPDF

Early childhood vaccination and subsequent mortality or morbidity: are observational studies hampered by residual confounding? A Danish register-based cohort study
Compose Response

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g. higgs-boson@gmail.com
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Statement of Competing Interests

PLEASE NOTE:

  • Responses are moderated before posting and publication is at the absolute discretion of BMJ, however they are not peer-reviewed
  • Once published, you will not have the right to remove or edit your response. Removal or editing of responses is at BMJ's absolute discretion
  • If patients could recognise themselves, or anyone else could recognise a patient from your description, please obtain the patient's written consent to publication and send them to the editorial office before submitting your response [Patient consent forms]
  • By submitting this response you are agreeing to our full [Response terms and requirements]

Vertical Tabs

Other responses

  • Published on:
    Comments to the response “Response to comment regarding “Early childhood vaccination and subsequent mortality or morbidity: are observational studies hampered by residual confounding? A Danish register-based cohort study”” by Lise Gehrt, Peter Aaby, Chris
    • Andreas Jensen, MSc in Statistics Department of Paediatrics and Adolescent Medicine, Rigshospitalet, Copenhagen University Hospital, DK-2100 Copenhagen Ø, Denmark
    • Other Contributors:
      • Lone Graff Stensballe, Professor
      • Per Kragh Andersen, Professor

    Andreas Jensen (0000-0003-4302-2982), Per Kragh Andersen (0000-0003-0490-0885), Lone Graff Stensballe (0000-0003-1569-153X)

    Department of Paediatrics and Adolescent Medicine, Rigshospitalet, Copenhagen University Hospital, DK-2100 Copenhagen Ø, Denmark. Andreas Jensen, MSc in Statistics
    Department of Paediatrics and Adolescent Medicine, Rigshospitalet, Copenhagen University Hospital. Lone Graff Stensballe, professor
    Section of Biostatistics, Department of Public Health, University of Copenhagen. Per Kragh Andersen, professor

    We appreciate the response to our study.
    First, considering the latest response, we further restricted our study population to children who reached 16 months of age before 1 January 2007, where PCV was introduced, and received 3 DTP vaccines before that age i.e. omitting the 2DTP+MMR-group. Here we present the corresponding estimates (Table 2 - https://blogs.bmj.com/bmjopen/files/2020/01/table-2.png).

    Second, the interpretation and properties of a ratio between a hazard ratio for one outcome and a hazard ratio for another outcome is not obvious.

    Finally, we note that Table 1 of our original response has been revised (please find it here: https://blogs.bmj.com/bmjopen/files/2020/01/table-1.png) . The first version of Table 1 presented the estimates from the crude analy...

    Show More
    Conflict of Interest:
    None declared.
  • Published on:
    Response to comment regarding “Early childhood vaccination and subsequent mortality or morbidity: are observational studies hampered by residual confounding? A Danish register-based cohort study”
    • Lise Gehrt, Ph.D. Stipendiate CVIVA, Bandim Health Project, Statens Serum Institut, Denmark, AND OPEN, University of Southern Denmark, Denmark
    • Other Contributors:
      • Peter Aaby, Professor
      • Christine Stabell Benn, Professor
      • Signe Sørup, Postdoc

    In our previous response we listed a number of differences between the study by Jensen et al.[1] and the studies by Sørup et al.[2, 3]. We thank Jensen et al. for providing the estimates for a restricted cohort of children who received 2 doses of DTP before 11 months of age (Table 1: https://bmjopen.bmj.com/pages/wp-content/uploads/sites/7/2019/11/table-1...). However,
    restriction of the cohort to children with 2 DTP vaccines before 11 months of age only removes a limited part of the differences we mentioned; hence, the results in Table 1 are still incomparable to the previous studies by Sørup et al.
    Jensen et al. included the 2DTP+MMR group in the modelling of effects (which Sørup et al. did not), and they did not analyse the data by type of infection, by sequence of vaccinations, and by the many confounders we had used. Since we specifically limited our studies to the period before PCV was part of the routine immunization programme, it should be noted that Jensen et al include many years where PCV was used (2007-2016). The introduction of PCV, a vaccine against respiratory infections, may have eliminated the need for some of the beneficial non-specific effects of MMR.
    Rather than testing something else, finding something else, and concluding that previous studies testing and finding different things were flawed, the fruitful way forward would be to investigate the...

    Show More
    Conflict of Interest:
    None declared.
  • Published on:
    Comments to the response “Regarding “Early childhood vaccination and subsequent mortality or morbidity: are observational studies hampered by residual confounding? A Danish register-based cohort study”” by Lise Gehrt et al
    • Andreas Jensen, MSc in Statistics Department of Paediatrics and Adolescent Medicine, Rigshospitalet, Copenhagen University Hospital
    • Other Contributors:
      • Per Kragh Andersen, Professor
      • Lone Graff Stensballe, Professor

    We appreciate the response to our study.
    The response assumes that a restriction in the population under study also limited the bias in two previous studies (1;2). In the two previous studies only individuals, who had two diphtheria–tetanus–pertussis–polio–H. influenzae type b-vaccines at 11 months of age were included.
    Therefore, we found it relevant to apply the same restriction to our study population and present the corresponding estimates adjusted for the confounders included in our study (3) (Table 1 - https://blogs.bmj.com/bmjopen/files/2019/11/Jenson-et-al-table.jpg).
    In Table 1 it can be seen that the restriction of the analysis to include only individuals with two diphtheria–tetanus–pertussis–polio–H. influenzae type b-vaccines at 11 months of age had little impact on the estimates. Importantly, the associations showing a reduced risk of hospitalisation for accidents among children with two or three diphtheria–tetanus–pertussis–polio–H. influenzae type b-vaccines and the measles-mumps-rubella vaccine were essentially unchanged when we restricted the analysis to include individuals with two diphtheria–tetanus–pertussis–polio–H. influenzae type b-vaccines at 11 months of age only.

    Reference List

    (1) Sorup S, Benn CS, Poulsen A, Krause TG, Aaby P, Ravn H. Live vaccine against measles, mumps, and rubella and the risk of hospital admissions for no...

    Show More
    Conflict of Interest:
    None declared.
  • Published on:
    Regarding “Early childhood vaccination and subsequent mortality or morbidity: are observational studies hampered by residual confounding? A Danish register-based cohort study”
    • Lise Gehrt, Ph.D Stipendiate CVIVA, Bandim Health Project, Statens Serum Institut, Denmark, AND OPEN, University of Southern Denmark, Denmark
    • Other Contributors:
      • Peter Aaby, Professor
      • Christine Stabell Benn, Professor
      • Signe Sørup, Post. Doc

    The study by Jensen et al. aimed “to examine the degree of residual confounding” in the association between childhood vaccinations and subsequent morbidity and mortality [1]. Jensen et al. conclude that “our study also suggests that residual confounding may have been present in previous Danish register-based studies investigating the impact of MMR [Measles-mumps-rubella vaccine] on the morbidity outcomes respiratory syncytial virus and hospitalisation for infection until 2 years of age” [1].

    Our previous Danish register-based studies referred to in this conclusion examined the effect of having MMR (recommended at 15 months of age) compared with the 3rd dose of diphtheria– tetanus–pertussis–polio–H. influenzae type b (DTP, recommended at 3,5 and 12 months of age) as the most recent vaccine on subsequent infectious disease hospitalisations (adjusted incidence rate ratios [IRR], 0.86; 95% confidence interval [CI], 0.84-0.88) [2] and respiratory syncytial virus infections (adjusted IRR, 0.78; 95% CI, 0.66-0.93)[3].

    There are fundamental methodological differences between our previous studies and the study by Jensen et al. Therefore, the results from Jensen et al. cannot be generalized to these previous studies.

    First, to enhance exchangeability, we restricted the study population to children, who had received the first two DTP doses before 11 months of age [2, 3], to exclude children who did not receive vaccines in a timely fashion, which could be relate...

    Show More
    Conflict of Interest:
    None declared.