Understanding Heterogeneity in Clinical Cohorts Using Normative Models: Beyond Case-Control Studies

Abstract

BackgroundDespite many successes, the case-control approach is problematic in biomedical science. It introduces an artificial symmetry whereby all clinical groups (e.g., patients and control subjects) are assumed to be well defined, when biologically they are often highly heterogeneous. By definition, it also precludes inference over the validity of the diagnostic labels. In response, the National Institute of Mental Health Research Domain Criteria proposes to map relationships between symptom dimensions and broad behavioral and biological domains, cutting across diagnostic categories. However, to date, Research Domain Criteria have prompted few methods to meaningfully stratify clinical cohorts.MethodsWe introduce normative modeling for parsing heterogeneity in clinical cohorts, while allowing predictions at an individual subject level. This approach aims to map variation within the cohort and is distinct from, and complementary to, existing approaches that address heterogeneity by employing clustering techniques to fractionate cohorts. To demonstrate this approach, we mapped the relationship between trait impulsivity and reward-related brain activity in a large healthy cohort (N = 491).ResultsWe identify participants who are outliers within this distribution and show that the degree of deviation (outlier magnitude) relates to specific attention-deficit/hyperactivity disorder symptoms (hyperactivity, but not inattention) on the basis of individualized patterns of abnormality.ConclusionsNormative modeling provides a natural framework to study disorders at the individual participant level without dichotomizing the cohort. Instead, disease can be considered as an extreme of the normal range or as—possibly idiosyncratic—deviation from normal functioning. It also enables inferences over the degree to which behavioral variables, including diagnostic labels, map onto biology.

Nonreplication of functional connectivity differences in autism spectrum disorder across multiple sites and denoising strategies

“…Using case-control comparisons and searching for common alterations exhibited across the group can easily obscure nonshared, heterogeneous patterns of differences across individuals. In other words, it is possible that certain types of statistics, which largely rest on the assumption of group-level similarity, are inappropriate and may mask important differences that can be better resolved by applying individually-sensitive analytic approaches (Byrge et al, 2015;Dubois & Adolphs, 2016;Marquand et al, 2019;Marquand, Rezek, Buitelaar, & Beckmann, 2016).…”

Multimodal fusion of structural and functional brain imaging in depression using linked independent component analysis

“…However, the robustness and generalizability of such studies have been brought into question (Dinga et al, ), which may be partly due to substantial brain heterogeneity within groups, which has been illustrated in terms of morphometry in schizophrenia (Alnæs et al, ). Alternatively, dimensional measures such as brain age prediction (Kaufmann et al, In press) and normative modeling (Marquand et al, ; Marquand, Rezek, Buitelaar, & Beckmann, ) have shown promising results in elucidating brain heterogeneity in mental disorders such as schizophrenia (Wolfers et al, ) and attention deficit/hyperactivity disorder (Wolfers et al, ). An important consideration is the multicausal nature of depression, which means that it is likely that including other measures such as psychosocial, cognitive, and genetic factors would increase the explained variance, beyond brain imaging by itself.…”

Neuroimaging Studies Illustrate the Commonalities Between Ageing and Brain Diseases

“…The alternative is to not merely focus on where groups differ, but to model where individuals differ. The illustration that many of the tools we use for measuring aspects of biology provide little or no diagnostic specificity and the evidence of commonalities across diagnostic categories reinforces the value of considering the individual instead of the group . Furthermore, given the similarities between brain diseases and ageing I described earlier, then it is even more important to consider the individual in the context of what would be expected given their age.…”