Fig 1.
*Correspondence: Short letters to journal editors and similar forms of communication
Table 1.
Studies reporting the PPV of an MND code in routinely-collected datasets, separated by dataset evaluated.
Table 2.
Studies reporting the sensitivity of an MND code in routinely collected datasets, separated by dataset evaluated.
Table 3.
International Classification of Diseases (ICD) codes for motor neurone disease and its subtypes.
Fig 2.
PPV of an MND code, stratified by type of routinely-collected dataset.
Boxes weighted by study size. PPV: Positive predictive value. Study size: Number of cases with MND codes that were assessed. Routine dataset: The routinely collected source of coded datasets P: primary care data, H: hospital data, D: death certificate data. Heterogeneity measures: I2 = 97%, Chi-squared = 321.6 (df = 12) p<0.0001.
Fig 3.
Sensitivity of an MND code, stratified by type of routinely-collected dataset.
Boxes weighted by study size. Study size: Number of known MND cases for which a code was sought. Routine dataset: The routinely collected source of coded datasets. P: primary care data, H: hospital data, D: death certificate data. Heterogeneity measures: I2 = 83%, Chi-squared = 23.0 (df = 4) p = 0.0001.
Table 4.
Within-study sub-group analyses.