Multiple Correspondence Analysis is a Useful Tool to Visualize Complex Categorical Correlated Data
Bibliographic record
Abstract
ABSTRACT ObjectivesWe sought to identify the most expensive hospitalized individuals in the Canadian province of Saskatchewan in fiscal year 2012/13, and determine the primary cause of their high use of health services. Our aim was to identify health problems that can be prevented or better managed in a non-hospital health care setting. Comorbid conditions are an important and confounding covariate in this population and so we used multiple correspondence analysis (MCA) to investigate the association of these conditions with each other and the most responsible diagnosis for each hospitalization. MCA is a multivariable descriptive statistical technique that displays the relationship between categorical variables in 2-dimensional graphical form. ApproachWe identified the most expensive 5% of people hospitalized between 01APR2012 and 31MAR2013. Hospital costs accounted for the majority of costs, but physician, drug, long-term care, and home care costs were added. Comorbid conditions in any of the 25 hospital diagnostic fields were identified and grouped into categories based upon ICD-10-CA subcategories. For example, category 1 was ICD-10-CA codes F10-F19: Mental and behavioural disorders due to psychoactive drug use, while category 2 was ICD-10-CA codes F20-F29: Schizophrenia, schizotypal, and delusional disorders. SAS™ v9.3 was used to conduct MCA and generate graphs displaying the correlation between each comorbid condition category, where the distance of each dot from the other represents the strength of the association between the disease categories (i.e., diseases that are correlated cluster together.) The frequency of each category of comorbid condition was represented by the size of the dots on the graph (e.g., the more people with the disease, the larger the dot.) Categories of comorbid conditions were redefined based upon data findings and clinical expertise. ResultsThree patient groups emerged as being amenable to intervention and thus cost savings, specifically (1) individuals of advanced age who are no longer able to live at home and are hospitalized while waiting for a bed in a long-term care facility, (2) individuals with a mental health and/or addiction problem, and (3) individuals who experienced medical harm during their time in hospital. ConclusionMCA is a valuable graphical tool that is easy to learn and, in conjunction with other statistical techniques, can be used to elucidate the relationship between complex correlated categorical variables.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.024 | 0.097 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.012 | 0.011 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.081 | 0.009 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".