# Investigating Data Interventions for Subgroup Fairness: An ICU Case Study

arXiv: https://arxiv.org/abs/2604.03478 · HTML: https://mlsanity.com/paper/2604.03478/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 44 references that are in the map.
- First posted on arXiv 3 Apr 2026.

## Builds on (most cited first, 19 of 19)

- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Fair prediction with disparate impact: A study of bias in recidivism prediction instruments](https://mlsanity.com/paper/1610.07524/) (2016)
- [A Reductions Approach to Fair Classification](https://mlsanity.com/paper/1803.02453/) (2018)
- [Inherent Trade-Offs in the Fair Determination of Risk Scores](https://mlsanity.com/paper/1609.05807/) (2016)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Simple data balancing achieves competitive worst-group-accuracy](https://mlsanity.com/paper/2110.14503/) (2021)
- [CheXclusion: Fairness gaps in deep chest X-ray classifiers](https://mlsanity.com/paper/2003.00827/) (2020)
- [Inherent Tradeoffs in Learning Fair Representations](https://mlsanity.com/paper/1906.08386/) (2019)
- ['It's Reducing a Human Being to a Percentage'; Perceptions of Justice in Algorithmic Decisions](https://mlsanity.com/paper/1801.10408/) (2018)
- [Distributionally Robust Optimization: A Review](https://mlsanity.com/paper/1908.05659/) (2019)
- [Fairness by Learning Orthogonal Disentangled Representations](https://mlsanity.com/paper/2003.05707/) (2020)
- [On the relation between accuracy and fairness in binary classification](https://mlsanity.com/paper/1505.05723/) (2015)
- [Data augmentation for learning predictive models on EEG: a systematic comparison](https://mlsanity.com/paper/2206.14483/) (2022)
- [Representation Matters: Assessing the Importance of Subgroup Allocations in Training Data](https://mlsanity.com/paper/2103.03399/) (2021)
- [Yet Another ICU Benchmark: A Flexible Multi-Center Framework for Clinical ML](https://mlsanity.com/paper/2306.05109/) (2023)
- [Statistical Inference for Fairness Auditing](https://mlsanity.com/paper/2305.03712/) (2023)
- [The Data Addition Dilemma](https://mlsanity.com/paper/2408.04154/) (2024)
- [How Robust is Your Fairness? Evaluating and Sustaining Fairness under Unseen Distribution Shifts](https://mlsanity.com/paper/2207.01168/) (2022)
- [On the Origins of Sampling Bias: Implications on Fairness Measurement and Mitigation](https://mlsanity.com/paper/2503.17956/) (2025)

## Abstract (arXiv)

In high-stakes settings where machine learning models are used to automate decision-making about individuals, the presence of algorithmic bias can exacerbate systemic harm to certain subgroups of people. These biases often stem from the underlying training data. In practice, interventions to "fix the data" depend on the actual additional data sources available -- where many are less than ideal. In these cases, the effects of data scaling on subgroup performance become volatile, as the improvements from increased sample size are counteracted by the introduction of distribution shifts in the training set. In this paper, we investigate the limitations of combining data sources to improve subgroup performance within the context of healthcare. Clinical models are commonly trained on datasets comprised of patient electronic health record (EHR) data from different hospitals or admission departments. Across two such datasets, the eICU Collaborative Research Database and the MIMIC-IV dataset, we find that data addition can both help and hurt model fairness and performance, and many intuitive strategies for data selection are unreliable. We compare model-based post-hoc calibration and data-centric addition strategies to find that the combination of both is important to improve subgroup performance. Our work questions the traditional dogma of "better data" for overcoming fairness challenges by comparing and combining data- and model-based approaches.