J R Lave1, R A Bankowitz, P Hughes-Cromwick
1University of Pittsburgh Graduate School of Public Health, PA, USA.
This study explored whether doctors' confidence in their initial diagnosis could predict how much hospital resources a patient would use. Researchers asked house staff to rate their certainty on a scale from 0 to 100%. They found that lower certainty ratings correlated with longer hospital stays, higher costs, and more consultations. Even though most patients had high certainty ratings, the relationship remained significant. The study suggests diagnostic confidence could be a useful indicator for managing hospital resources. However, the authors caution that more research is needed to confirm these findings.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Area of Science:
Background:
Healthcare systems seek ways to predict and manage inpatient resource use. Existing methods often rely on diagnostic codes or administrative data. This gap motivated researchers to explore whether subjective diagnostic certainty could predict resource consumption. Prior research has shown that diagnostic accuracy affects patient outcomes. However, no prior work had resolved whether diagnostic certainty itself correlates with resource use. The study aimed to address this uncertainty. It focused on general medicine inpatient settings. Researchers wanted to test if diagnostic certainty ratings could serve as a novel predictor. This approach could help identify patients at risk for higher resource use.
Purpose Of The Study:
The study aimed to assess if house staff's subjective diagnostic certainty at admission could predict inpatient resource use. Researchers wanted to determine if this subjective measure was independent of other known factors. They focused on general medicine wards at an academic medical center. The motivation was to find a novel predictor of resource consumption. Existing methods rely on administrative data or diagnostic codes. This study tested whether subjective certainty could add predictive value. The researchers sought to evaluate if diagnostic confidence influenced length of stay and consultations. They also wanted to assess if certainty ratings varied with other patient characteristics.
The study found diagnostic certainty ratings correlated with resource use in general medicine wards.
House staff used a visual analog scale with hash marks at 0, 25, 50, 75, and 100%.
General medicine wards handle diverse diagnoses, making them ideal for studying diagnostic certainty.
Administrative systems provided patient demographics and MedisGroups scores. Transaction databases tracked resource use.
77.2% of patients received a certainty rating of 75% or higher at admission.
Main Methods:
Researchers collected subjective diagnostic certainty ratings using a visual analog scale. House staff assigned ratings from 0 to 100% for each admission. The study included 1,778 patients admitted to general medicine wards. Data on prior hospitalizations and MedisGroups scores came from administrative systems. Length of stay and total charges were obtained from transaction databases. The team also recorded consultation numbers and pharmacy charges. Multivariate analysis tested the relationship between certainty ratings and resource use. They controlled for factors like MedisGroups scores and outlier status.
Main Results:
In univariate analysis, certainty ratings varied with prior hospitalizations and resource measures. They did not vary with MedisGroups scores or discharge status. Multivariate analysis showed certainty ratings predicted resource use. Length of stay increased with lower certainty ratings. Total costs and consultation numbers also rose with lower certainty. Adjusted pharmacy charges were an exception to this pattern. Only 22.8% of patients had certainty ratings below 75%. Despite this, the relationship between certainty and resource use remained significant. The findings suggest diagnostic uncertainty correlates with higher resource consumption.
Conclusions:
The authors found diagnostic certainty ratings predicted resource use in general medicine wards. These ratings remained significant even after adjusting for known factors. The relationship held for length of stay, costs, and consultations. However, it did not apply to adjusted pharmacy charges. The study did not claim certainty ratings are essential for resource prediction. Instead, it proposed they could serve as one of many indicators. The findings suggest diagnostic confidence may inform resource planning. The authors emphasized the need for further research on this relationship.
The authors propose diagnostic certainty could help predict resource use in inpatient settings.