Disability weights based on patient-reported data from a multinational injury cohort

Belinda J. Gabbe, Ronan A Lyons, Pamela Simpson, Frederick P. Rivara, Shanthi Ameratunga, Suzanne Polinder, Sarah Derrett, James HarrisonBulletin 2016 · DOI

TL;DR

This paper introduces a new method for calculating "disability weights" – a measure of how much an injury impacts a person's life – by using patient self-reported data instead of expert panels. These new weights are generally higher than previous estimates, suggesting that the burden of injuries is underestimated, and could lead to more accurate assessments of public health impacts.

What Problem Does This Paper Address?

Estimating the burden of disease, particularly for injuries, is crucial for allocating resources and developing effective public health policies. A key component of this estimation is the "disability weight," which quantifies the severity of a health condition. Traditionally, these weights have been derived from expert panels or general public surveys (panel-based weights). However, these methods have limitations: they may not generalize well across different cultural contexts, struggle to capture the complexity of varied injuries, and often overlook the long-term, chronic nature of many injuries. This study aims to address these limitations by developing disability weights directly from the experiences of individuals who have sustained injuries.

Methods

The researchers utilized data from the "Validating and Improving Injury Burden Estimates Study" (Injury-VIBES), a multinational cohort study involving over 29,000 participants from Australia, the Netherlands, New Zealand, the UK, and the US. They employed the EQ-5D, a standardized questionnaire that measures health status across five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Participants reported their health status at multiple time points post-injury (3, 6, and 12 months).

The core of their method involved:

  1. Calculating EQ-5D Preference Scores: Based on patient responses, a numerical score was generated for each individual's health state.
  2. Deriving Disability Weights: These scores were then used to calculate disability weights. For each injury, the patient's EQ-5D score was compared to age- and sex-specific population norms. The difference represented the degree of disability.
  3. Time-Averaging: Weights were calculated at three, six, and 12 months post-injury and then "time-averaged" to represent the overall impact over a period. The 12-month weights were assumed to reflect both immediate and expected lifelong disability.
  4. Categorization: Weights were calculated for various injury classifications, including those used by the Global Burden of Disease (GBD) 2013 study, EUROCOST groups, and specific ICD-10 diagnosis codes.

Datasets

The study used pooled data from six longitudinal injury-outcome studies that form the Injury-VIBES cohort. These included:

  • DIPS (Dutch Injury Patient Survey): Netherlands
  • NSCOT (National Study on Costs and Outcomes of Trauma): United States of America
  • POIS (Prospective Outcomes of Injury Study): New Zealand
  • VOTOR (Victorian Orthopaedic Trauma Outcomes Registry): Australia
  • VSTR (Victorian State Trauma Registry): Australia
  • UKBOIS (United Kingdom Burden of Injury Study): United Kingdom

A total of 29,770 injury cases with at least one EQ-5D score were analyzed.

Key Insights and Ideas

  • Patient Experience Matters: The study strongly advocates for using patient-reported data, arguing it offers a more direct and potentially more accurate reflection of injury burden than expert opinions or general population perceptions.
  • Injuries are Chronic: A recurring theme is that injuries are not always acute events with full recovery. Many result in long-term, persistent disability, a fact often underestimated by traditional methods.
  • Higher Weights, Higher Burden: The newly derived disability weights are generally higher than those used in the GBD 2013 study, particularly for common injuries like fractures. This suggests that previous estimates of injury burden may have been too low.
  • Granularity is Key: By calculating weights for specific ICD-10 codes, the study highlights significant heterogeneity in disability even within seemingly similar injury types (e.g., different types of fractures).

Key Results

  • Increased Disability Estimates: The new patient-reported disability weights were consistently higher than the GBD 2013 panel-based weights for many injury categories. For example, for hospitalized cases, the new weights were 1.1 to 22.2 times higher for 22 injury categories.
  • Long-Term Disability is Common: The study found that long-term disability was prevalent across most injury categories, reinforcing the idea that injuries are often chronic conditions.
  • Hospitalized vs. Emergency Cases: Weights were higher for patients admitted to the hospital compared to those treated solely in emergency departments, as expected.
  • Specific ICD-10 Codes Reveal Nuance: For fractures, the new weights indicated significant differences in disability based on the specific bone or region affected, a level of detail not always captured by broader injury classifications.

Limitations and Future Work

The authors acknowledge several limitations:

  • Coding Accuracy: The accuracy of injury diagnoses, especially for emergency department cases, cannot be guaranteed.
  • Sample Size for Some Categories: Disability weights for some specific injury categories were based on relatively small numbers of cases.
  • Data Set Differences: Variations in follow-up rates and EQ-5D data availability across the six contributing studies could introduce bias.
  • Generalizability to Low- and Middle-Income Countries: The study was conducted in high-income countries, and it's unclear if these weights would be directly applicable elsewhere.
  • Focus on Single Injuries: The analysis primarily considered single injuries, while many patients sustain multiple injuries, which can exacerbate disability.
  • Adults Only: The study focused on adult participants, and recovery trajectories may differ for children.
  • Comorbidity Not Explicitly Considered: While age-specific population preferences in the EQ-5D might indirectly account for some comorbidity, it wasn't a direct factor.

Future work could involve:

  • Multi-Injury Analysis: Incorporating the impact of multiple injuries on disability.
  • Cross-Cultural Validation: Extending this methodology to low- and middle-income countries to assess generalizability.
  • Pediatric Cohorts: Developing disability weights for children.
  • Application to Other Conditions: Adapting this patient-reported approach to estimate disability weights for non-injury-related health conditions.

Summary updated

Abstract

Injury is often a chronic disorder and burden of disease estimates should reflect this. Application of the new weights to burden studies would substantially increase estimates of disability-adjusted life-years and provide a more accurate reflection of the impact of injuries on peoples' lives.
BibTeX
@article{gabbe2016disability,
  author = {Belinda J. Gabbe and Ronan A Lyons and Pamela Simpson and Frederick P. Rivara and Shanthi Ameratunga and Suzanne Polinder and Sarah Derrett and James Harrison},
  title = {{Disability weights based on patient-reported data from a multinational injury cohort}},
  journal = {Bulletin},
  year = {2016},
  doi = {10.2471/blt.16.172155},
}
Figure