Impact of Medical Certification of Cause of Death (MCCOD) training interventions in improving the quality of MCCOD | ||||||
---|---|---|---|---|---|---|
Patient or population: Physicians or prospective physicians Setting: Global Intervention: Generic academic training in training curricula Comparison: Pre-intervention parameters of MCCOD quality | ||||||
Outcomes | Anticipated absolute effects* (95% CI) | Risk difference (95% CI) | № of certificates assessed (studies) | Certainty of the evidence (GRADE) | Comments regarding similar studies that did not meet the meta-analysis inclusion criteria | |
Risk with pre-intervention | Risk with post-intervention | |||||
No time interval | 832 per 1000 | 275 per 1000 (250 to 300) | 0.33 (0.30 to 0.36) | 3596 (3 observational studies) | ⨁⨁⨁◯ Moderatea | In one study in Canada, 83 and 146 death certificates were assessed with 69.2% and 75.9% error percentages. In one Indian study, the related percentages were 29.2% and 27.5%. In another two Indian studies with just 75 and 80 death certificate assessments, the percentages were 100% versus 22.6%, and 85% versus 0.0%, respectively |
Presence of abbreviations | 328 per 1000 | 53 per 1000 (43 to 59) | 0.16 (0.13 to 0.18) | 3596 (3 observational studies) | ⨁⨁⨁◯ Moderatea | In the above Canadian study, the error percentages were 19.9% and 18.1%. In the three Indian studies, the related percentages were 21.9.% versus 33.3%; 86.3% versus 29.3%; and 22.5% versus 0.0%, respectively |
Improper sequence | 349 per 1000 | 63 per 1000 (52 to 70) | 0.18 (0.15 to 0.20) | 4335 (3 observational studies) | ⨁⨁⨁◯ Moderatea | In the above Canadian study, the error percentages were 15.8% and 6%. In the three Indian studies, the related percentages were 25% versus 59%; 89.3% versus 36%; and 60% versus 3.75%, respectively |
Multiple causes | 265 per 1000 | 40 per 1000 (34 to 45) | 0.15 (0.13 to 0.17) | 4204 (3 observational studies) | ⨁⨁⨁◯ Moderate a | In one study in Papua New Guinea, the respective percentages were 16.3% and 7.9% |
Ill-defined underlying cause of death | 363 per 1000 | 55 per 1000 (44 to 62) | 0.15 (0.12 to 0.17) | 4455 (3 observational studies) | ⨁⨁◯◯ Lowa,b | In one Sri Lankan study, ill-defined underlying cause of death was observed to be higher post-intervention (10.6% versus 4.4%) |