Indian Journal of Psychological Medicine
  Home | About Us | Editorial Board | Search | Ahead of print | Current Issue | Archives | Instructions | Contact | Advertise | Submission | Login 
Users Online: 821 
Wide layoutNarrow layoutFull screen layoutHome Print this page Email this page Small font sizeDefault font sizeIncrease font size


 
 Table of Contents    
LEARNING CURVE
Year : 2018  |  Volume : 40  |  Issue : 4  |  Page : 395-397  

Prenatal depression and infant health: The importance of inadequately measured, unmeasured, and unknown confounds


Department of Psychopharmacology, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka, India

Date of Web Publication16-Jul-2018

Correspondence Address:
Chittaranjan Andrade
Department of Psychopharmacology, National Institute of Mental Health and Neurosciences, Bengaluru, Karnataka
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/IJPSYM.IJPSYM_232_18

Rights and Permissions
   Abstract 


A recent study found that maternal antenatal depressive symptoms were associated with adverse infant general health outcomes and that gestational age, birth weight, and breastfeeding did not mediate the observed relationship. The authors suggested that antenatal depression can have a harmful effect on infant health through disturbed fetal programming driven by maternal symptoms and behaviors that influence the maternal and hence the fetal internal environment. The authors implied that interventions to diagnose and treat maternal depression can have a protective effect against disturbances in infant health. However, because of the observational nature of the study, cause–effect relationships cannot be conclusively stated. This is especially so because there were many confounds that the authors did not consider. The present article provides examples and explanations of how inadequately measured, unmeasured, and unknown confounds can explain observed relationships between explanatory and outcome variables, thereby negating cause–effect interpretations of study findings. It is important to minimize confounding when conducting observational studies, and this can only be done by comprehensively listing and efficiently measuring potential confounders in advance.

Keywords: Antenatal depression, bias, cause–effect relationship, confounding variables, infant health outcomes, observational research


How to cite this article:
Andrade C. Prenatal depression and infant health: The importance of inadequately measured, unmeasured, and unknown confounds. Indian J Psychol Med 2018;40:395-7

How to cite this URL:
Andrade C. Prenatal depression and infant health: The importance of inadequately measured, unmeasured, and unknown confounds. Indian J Psychol Med [serial online] 2018 [cited 2019 Jul 22];40:395-7. Available from: http://www.ijpm.info/text.asp?2018/40/4/395/236648




   Introduction Top


Depression is associated with biological changes that range from disturbances in neuroendocrine regulation to disturbances in sleep and appetite. Depression is also associated with self-injurious changes in behavior; patients may smoke, drink, use illicit substances, disregard advice related to current medical conditions, and attempt deliberate self-harm. Thus, the internal environment of a depressed patient is perturbed by the illness itself and by behaviors related to the illness.

During pregnancy, the growth and development of the fetus are influenced by maternal health, maternal behaviors, and the maternal internal environment. It is, therefore, reasonable to speculate that depression complicating pregnancy can influence fetal, infant, and child health outcomes, and perhaps outcomes in adult life as well. This, in fact, is a matter that has been considerably researched. One recent investigation was that of Coburn et al.[1]

The study by Coburn et al.[1]

The authors [1] studied 322 low-income Mexican-American mothers and their infants. The mothers were assessed at a mean of 35 weeks of gestation and, again, 12 weeks after birth. At both time points, maternal depressive symptoms, if any, were quantified using the Edinburgh Postpartum Depression Scale (EPDS). At the 12-week assessment, mothers also completed a Baby Health Questionnaire that recorded whether or not eight common health disturbances such as fever, rash, and diarrhea had occurred during infancy.

Regression analyses were conducted to assess the relationship between antenatal depression ratings (EPDS scores) and infant health disturbances. Maternal age and the presence of a romantic partner were included as covariates in the regressions. Education, household income, immigrant status, and the number of other children were not adjusted for in the regressions because they were found to have no influence on infant health disturbances.

Important findings in regression analyses were as follows:

  1. Higher antenatal EPDS scores were associated with higher postnatal EPDS scores
  2. Higher antenatal and postnatal EPDS scores were independently associated with more infant health disturbances
  3. Antenatal EPDS scores did not predict gestational age, birth weight, or breastfeeding, and these three variables did not predict infant health concerns.


Coburn et al.[1] concluded that maternal antenatal depressive symptoms may be associated with adverse infant health outcomes and that gestational age, birth weight, and breastfeeding do not mediate the observed relationship. They suggested the need for interventions for antenatal depression to reduce the potentially harmful effects of antenatal depression on fetal and infant health.

Critical comments

Coburn et al.[1] demonstrated an association between the severity of antenatal depression and the magnitude of infant health concerns at 12 weeks after birth. They suggested that antenatal depression can have a harmful effect on infant health through disturbed fetal programming driven by maternal symptoms and behaviors that influence the maternal and hence fetal internal environment. They implied that interventions to diagnose and treat maternal depression can have a protective effect against infant health concerns.

However, a cause–effect relationship between antenatal depression and infant health cannot be determined from an observational study such as this because inadequately measured, unmeasured, and unknown confounds could have explained both antenatal depression and poor infant health, as the following examples show.

Inadequately measured confounds

The authors [1] recorded the annual household income, but this single variable would not have been able to capture the range of environmental adversities that could have affected maternal and infant health. For example, poor living conditions, poor access to nutritious food, and poor access to quality medical care could have been responsible for both antenatal depression and poor infant health. Thus, environmental adversity is an example of an inadequately measured confound because it is not satisfactorily described by the annual household income, alone.

Unmeasured confounds

Maternal physical illness, maternal use of medications, maternal use of alcohol, cigarettes, and illicit drugs, and other variables could have increased the risk of both maternal depression and poor infant care, with the latter resulting in infant health disturbances. Information related to such variables was not recorded. Thus, these are examples of unmeasured confounds. The failure to record whether the women were receiving antidepressant medication is a particularly important unmeasured confound, given that medication exposure during pregnancy can moderate maternal depression severity and worsen fetal and infant health outcomes through direct effects, or improve these outcomes through attenuation of depression.

Unknown confounds

The infant shares some of its mother's genes. It is possible that genetic influences that are responsible for maternal depression may also influence infant vulnerability to health disturbances. We do not know what these genetic influences are and so cannot measure them. These would be examples of unknown confounds.

Implications

Although this is a promising study,[1] it merely indicates that the severity of antenatal depression, as measured by the EPDS, is a potential marker for infant general health. There is insufficient evidence to posit a cause–effect relationship between antenatal depression and infant general health. Whereas it is important to diagnose and treat antenatal depression, it is too early to conclude that doing so will result in improved infant general health outcomes.

Parting notes

If researchers do not efficiently collect data on variables that influence the outcomes that they plan to study, they will not be able to determine whether or not these variables bias their findings. The reader is referred to specific articles that explain the concept of confounding and bias in research.[2],[3],[4] The possibility of confounding is an important consideration in observational research, such as studies that examine gestational and infant outcomes after depression or antidepressant exposure during pregnancy.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
   References Top

1.
Coburn SS, Luecken LJ, Rystad IA, Lin B, Crnic KA, Gonzales NA. Prenatal maternal depressive symptoms predict early infant health concerns. Matern Child Health J 2018;22:786-93.  Back to cited text no. 1
    
2.
Rhodes AE, Lin E, Streiner DL. Confronting the confounders: The meaning, detection, and treatment of confounders in research. Can J Psychiatry 1999;44:175-9.  Back to cited text no. 2
    
3.
Grimes DA, Schulz KF. Bias and causal associations in observational research. Lancet 2002;359:248-52.  Back to cited text no. 3
    
4.
Andrade C. Confounding. Indian J Psychiatry 2007;49:129-31.  Back to cited text no. 4
[PUBMED]  [Full text]  




 

Top
 
 
  Search
 
  
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
    Abstract
   Introduction
    References

 Article Access Statistics
    Viewed398    
    Printed1    
    Emailed0    
    PDF Downloaded25    
    Comments [Add]    

Recommend this journal