Background
Methods
Systems assessed
Outcome measures
-
System: Any approach to classifying causes of neonatal deaths and/or stillbirths that was described by authors of included papers as a “system” or “approach”, and/or that included a clearly delineated list of causes separated from the data.
-
Modified system: Any system that was created as a result of making changes to an existing system, where:
-
the system presented was described by the authors as a modification of an existing system, or
-
it was apparent that the system had been modified, despite the authors stating that the system was unchanged from its original form (e.g. different number of levels, number of categories at the top level, meaning of categories, etc.).
-
-
New system: Any system that was created without modifying an existing system.
-
Used system: A system that was used for any purpose (e.g. clinical, research) other than purely developmental (e.g. testing for reliability).
-
Global system: Any system used to classify or estimate causes of stillbirths and neonatal deaths in all countries for which data is available.
-
National system:
-
∘ used by a national government for annual reporting of causes for the majority (>50%) of SB and/or NND nationwide, or
-
∘ used by any research group (e.g. the United States Agency for International Development, USAID, or the United Nations Children’s Fund, UNICEF) to classify causes of death
-
▪ as reported by Demographic and Health Surveys (DHS) in at least one year, where DHS data is assumed to be nationally representative, or
-
▪ of the majority (>50%) of SB and/or NND that occur in a country in at least one year, or
-
-
∘ otherwise stated to be a system developed on purpose for national government use.
-
-
Widely used system: any system used to classify 1000+ deaths and/or in 2+ countries between 2009 and 2014.
-
Level: Some systems may have a single “level” of causes and other systems may have several levels of causes, with the top level listing more general causes and each lower level listing sub-categories within a given general cause. For example, classifying the cause of a SB or NND in a system with multiple levels would mean that a set of causes, from most general (taken from the top level) to most specific (taken from the lowest level), would be selected, e.g. “congenital anomaly” from the top level and then more detail on that cause via assignation of a sub-category at the next level down, e.g. “trisomy 13”.
Data collection and analysis
Characteristics | Weight | Variables used to assess alignment | Aligned if | Judgment of variable accuracy as a measure of alignment | |
---|---|---|---|---|---|
Structural characteristics | |||||
1 | A global system must use rules to ensure valid assignment of cause of death categories | .98 | Rules available? | Yes | Strong |
2 | A global system must be able to work with all levels of data (from both low-income and high-income countries), including minimal levels | .98 | Yes for all three variables | ||
Used in both HIC and LMIC? | Strong | ||||
Used with verbal autopsy? | Strong | ||||
Used in >1 LMIC? | Weak | ||||
3 | A global system must ensure cause of death categories are relevant in all settings | .96 | Used in both HIC and LMIC? | Yes | Weak |
4 | A global system must require associated factors to be recorded and clearly distinguished from causes of death | .94 | Yes for both variables below | ||
Associated factors included? | Strong | ||||
Distinguishes associated factors from causes? | Strong | ||||
5 | A global system must distinguish between antepartum and intrapartum conditions | .90 | Distinguishes IP from AP? | Yes | Strong |
6 | A global system should record the level of data available to assign the cause of death (e.g. verbal autopsy only, placental histology, autopsy, etc.) | .86 | Records type of data used? | Yes | Strong |
7 | A global system must have multiple levels of causes of death, with a small number of main categories | .82 | As below | ||
Number of causes | ≤10 | Strong | |||
Number of levels | 2+ | Strong | |||
8 | A global system must include a sufficiently comprehensive list of categories to result in a low proportion of deaths classified as “other” | .80 | % “other” | Max <20 % | Weak |
Functional characteristics | |||||
9 | A global system must be easy to use, and produce data that are easily understood and valued by users | 1 | As below | ||
# deaths classified/# countries of use | 500+ cases and/or 2+ countries | Weak | |||
Definitions available? | Yes | Weak | |||
Rules available? | Yes | Weak | |||
National? | Yes | Weak | |||
10 | A global system must have clear guidelines for use and definitions for all terms used | 1 | Yes for both variables below | ||
Definitions available? | Strong | ||||
Rules available? | Strong | ||||
11 | A global system must produce data that can be used to inform strategies to prevent perinatal deaths | .96 | As below | ||
IP vs AP? | Yes | Weak | |||
% “other” | Max <20 % | Weak | |||
National? | Yes | Weak | |||
12 | A global system must require neonatal deaths to be clearly distinguished from stillbirths | .94 | Yes for both variables below | ||
Distinguishes SB and NND? | Strong | ||||
Separate categories for SB and NND? | Strong | ||||
13 | A global system must have high inter- and intra-rater reliability | .94 | Reliability testing? | Yes; min ≥0.60 | Strong |
14 | A global system must be available in different formats including inexpensive ehealth and mhealth options, and in multiple languages | .92 | Yes for both variables below | ||
E-format? | Strong | ||||
>1 language? | Weak | ||||
15 | A global system must allow easy access to the data by the end-users | .92 | Accessible data? | Yes | Weak |
16 | A global system must incorporate both stillbirths and neonatal deaths | .86 | Both SB and NND? | Yes | Strong |
17 | A global system must require the single most important factor leading to the death to be recorded | .86 | As below | ||
Hierarchical? | No or partially | Weak | |||
Only 1 cause allowed? | Yes | Strong | |||
Includes FGR/IUGR/SGA? | No | Strong |
Results
Overall alignment
Score using all variables | Score using “strong” variables only | |||
---|---|---|---|---|
Unweighted | Weighted | Unweighted | Weighted | |
Maximum possible score | 17 | 15.64 | 12 | 11.00 |
Froen 2009-Codac [9] | 9 | 7.94 | 8 | 7.14 |
Korteweg 2006-Tulip [10] | 7 | 6.20 | 6 | 5.40 |
Black 2010-CHERG [11] | 6 | 5.50 | 3 | 2.82 |
Cole 1986 [12] | 6 | 5.48 | 5 | 4.52 |
Flenady 2009-PSANZ-PDC [13] | 6 | 5.50 | 5 | 4.54 |
Kotecha 2014-Wales [14] | 6 | 5.42 | 4 | 3.70 |
Ujwala 2012 [15] | 6 | 5.18 | 5 | 4.38 |
Chan 2004-PSANZ-NDC [23] | 5 | 4.46 | 4 | 3.66 |
Kidanto 2009 [24] | 5 | 4.38 | 5 | 4.38 |
Lawn 2006-CHERG [25] | 5 | 4.72 | 4 | 3.82 |
Manning 2013-maternal & fetal-Ireland [26] | 5 | 4.52 | 4 | 3.60 |
Pattinson 1989 [27] | 5 | 4.58 | 4 | 3.78 |
Schmiegelow 2012 [28] | 5 | 4.42 | 5 | 4.42 |
Varli 2008-Stockholm [29] | 5 | 4.58 | 4 | 3.78 |
Wigglesworth 1980 [30] | 5 | 4.46 | 3 | 2.70 |
Abdellatif 2013 [31] | 4 | 3.34 | 3 | 2.54 |
CMACE 2010-maternal & fetal [32] | 4 | 3.60 | 4 | 3.60 |
CMACE 2010-neonatal [32] | 4 | 3.46 | 3 | 2.66 |
de Galan-Roosen 2002 [16] | 4 | 3.48 | 4 | 3.48 |
Engmann 2012 [33] | 4 | 3.46 | 3 | 2.66 |
Flenady 2009-PSANZ-NDC [13] | 4 | 3.58 | 3 | 2.78 |
Gardosi 2014-MAINa
| 4 | 3.52 | 4 | 3.52 |
Gordijn 2009 [18] | 4 | 3.68 | 4 | 3.68 |
Khanum 2009 [34] | 4 | 3.42 | 3 | 2.62 |
Kidron 2009 [35] | 4 | 3.78 | 4 | 3.78 |
McClure 2015 [42]b
| 4 | 3.74 | 5 | 4.60 |
Mo-Suwan 2009 [36] | 4 | 3.52 | 4 | 3.52 |
MRC 2002-PPIP-South Africa [37] | 4 | 3.52 | 3 | 2.72 |
National Services Scotland 2013-neonatal [38] | 4 | 3.40 | 2 | 1.68 |
NIPORT 2005-Bangladesh [39] | 4 | 3.70 | 2 | 1.98 |
Shah 2011 [40] | 4 | 3.60 | 4 | 3.66 |
Van Diem 2010 [41] | 4 | 3.46 | 3 | 2.66 |
VanderWielen 2011-WiSSP [42] | 4 | 3.60 | 4 | 3.60 |
Wood 2012 [43] | 4 | 3.44 | 4 | 3.44 |
Basys 2014-Lithuania [44] | 3 | 2.58 | 3 | 2.58 |
Chan 2004-PSANZ-PDC [23] | 3 | 2.84 | 3 | 2.84 |
CMACE 2011-maternal & fetal [45] | 3 | 2.62 | 3 | 2.62 |
Cole 1989-ICE [46] | 3 | 2.84 | 3 | 2.84 |
De Reu 2009-Tulip mod. [47] | 3 | 2.52 | 2 | 1.72 |
Froen 2009-simplified Codac [9] | 3 | 2.54 | 3 | 2.54 |
Gardosi 2005-ReCoDe [48] | 3 | 2.76 | 2 | 1.80 |
Glinianaia 2010 [49] | 3 | 2.52 | 2 | 1.72 |
Hey 1986 [50] | 3 | 2.84 | 4 | 3.70 |
Hinderaker 2003 [51] | 3 | 2.52 | 2 | 1.72 |
Manandhar 2010 [52] | 3 | 2.70 | 4 | 3.56 |
National Services Scotland 2013-obstetric [38] | 3 | 2.64 | 2 | 1.72 |
Nausheen 2013 [53] | 3 | 2.88 | 4 | 3.74 |
Nga 2012 [54] | 3 | 2.60 | 3 | 2.66 |
SCRN WG 2011 [55] | 3 | 2.70 | 3 | 2.70 |
Simpson 2010 [56] | 3 | 2.48 | 2 | 1.68 |
Abha 2011 [57] | 2 | 1.68 | 2 | 1.68 |
Aggarwal 2011 [58] | 2 | 1.78 | 1 | 0.98 |
Aggarwal 2013 [59] | 2 | 1.74 | 1 | 0.94 |
Black-2010-CHERGc [11] | 2 | 1.68 | 2 | 1.68 |
De Reu 2009-Cole mod. [47] | 2 | 1.66 | 2 | 1.72 |
De Reu 2009-Wigglesworth mod. [47] | 2 | 1.66 | 2 | 1.72 |
Dias e Silva 2013 [60] | 2 | 1.68 | 2 | 1.68 |
Dudley 2010-INCODE [61] | 2 | 1.80 | 3 | 2.72 |
Hey 1986-short form [50] | 2 | 1.66 | 2 | 1.72 |
Khanal 2011 [62] | 2 | 1.66 | 1 | 0.86 |
Lawn 2009-consistent classification for causes of stillbirth [63] | 2 | 1.88 | 2 | 1.88 |
Lawn 2010 [36] | 2 | 1.82 | 1 | 0.86 |
Lawn 2012 [64] | 2 | 1.82 | 1 | 0.86 |
National Services Scotland 2013-FIGO [38] | 2 | 1.72 | 2 | 1.72 |
Olamijulo 2011 [65] | 2 | 1.72 | 2 | 1.72 |
Seaton 2012 [66] | 2 | 1.66 | 1 | 0.86 |
Serena 2013-Wigglesworth mod. [67] | 2 | 1.66 | 1 | 0.86 |
Winbo 1998-NICE [68] | 2 | 1.66 | 2 | 1.72 |
Winter 2013-Rwanda [69] | 2 | 1.66 | 1 | 0.86 |
Cunninghamd 1997 [70] | 1 | 0.82 | 1 | 0.82 |
Freitas 2012 [71] | 1 | 0.86 | 1 | 0.86 |
Gupta 2012-Bhutan [72] | 1 | 0.80 | 0 | 0.00 |
Hama Diallo 2012 [73] | 1 | 0.82 | 1 | 0.82 |
Jehan 2009e [74] | 1 | 0.80 | 0 | 0.00 |
Kruse 2014 [75] | 1 | 0.80 | 1 | 0.86 |
Nabeel 2012 [76] | 1 | 0.82 | 2 | 1.68 |
Public Health Agency of Canada 2008 [77] | 1 | 0.86 | 1 | 0.86 |
Rocha 2011 78
| 1 | 0.82 | 1 | 0.82 |
Smith 2010 [79] | 1 | 0.80 | 1 | 0.86 |
Serena 2013-ReCoDe mod. [67] | 0 | 0.00 | 0 | 0.00 |
Wou 2014 [80] | 0 | 0.00 | 0 | 0.00 |
Characteristics with greatest and least alignment
Characteristics | % consensus | % systems in alignment with each characteristic | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
All (81) | Widely useda (27) | Less used (54) | Used in HIC only (36) | Used in LMIC only (32) | SB-only systems (15) | NND-only systems (26) | Combined systems (NND and SB) (40) | |||
Structural | ||||||||||
1 | A global system must use rules to ensure valid assignment of cause of death categories. | 98 % | 41 % | 52 % | 35 % | 44 % | 28 % | 53 % | 35 % | 40 % |
2 | A global system must be able to work with all levels of data (from both low-income and high-income countries), including minimal levels. | 98 % | 3 % | 7 % | 0 % | 0 % | 0 % | 0 % | 8 % | 0 % |
3 | A global system must ensure cause of death categories are relevant in all settings. | 96 % | 10 % | 30 % | 0 % | 0 % | 0 % | 7 % | 15 % | 8 % |
4 | A global system must require associated factors to be recorded and clearly distinguished from causes of death. | 94 % | 14 % | 19 % | 11 % | 17 % | 13 % | 7 % | 8 % | 20 % |
5 | A global system must distinguish between antepartum and intrapartum conditions. | 90 % | 20 % | 19 % | 20 % | 22 % | 16 % | 20 % | 0 % | 33 % |
6 | A global system should record the level of data available to assign the cause of death (e.g. verbal autopsy only, placental histology, autopsy, etc.). | 86 % | 9 % | 19 % | 4 % | 19 % | 0 % | 7 % | 4 % | 13 % |
7 | A global system must have multiple levels of causes of death, with a small number of main categories. | 82 % | 40 % | 33 % | 43 % | 33 % | 44 % | 33 % | 42 % | 40 % |
8 | A global system must include a sufficiently comprehensive list of categories to result in a low proportion of deaths classified as “other”. | 80 % | 48 % | 52 % | 46 % | 53 % | 53 % | 27 % | 65 % | 45 % |
Functional | ||||||||||
9 | A global system must be easy to use, and produce data that are easily understood and valued by users. | 100 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % |
10 | A global system must have clear guidelines for use and definitions for all terms used. | 100 % | 17 % | 15 % | 19 % | 17 % | 16 % | 20 % | 19 % | 15 % |
11 | A global system must produce data that can be used to inform strategies to prevent perinatal deaths. | 96 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % |
12 | A global system must require neonatal deaths to be clearly distinguished from stillbirths. | 94 % | 5 % | 7 % | 4 % | 0 % | 9 % | 0 % | 0 % | 10 % |
13 | A global system must have high inter- and intra-rater reliability. | 94 % | 7 % | 11 % | 6 % | 8 % | 6 % | 7 % | 0 % | 13 % |
14 | A global system must be available in different formats including inexpensive ehealth and mhealth options, and in multiple languages. | 92 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % | 0 % |
15 | A global system must allow easy access to the data by the end-users. | 92 % | 10 % | 11 % | 9 % | 14 % | 6 % | 0 % | 12 % | 13 % |
16 | A global system must incorporate both stillbirths and neonatal deaths. | 86 % | 49 % | 48 % | 50 % | 56 % | 44 % | 0 % | 0 % | 100 % |
17 | A global system must require the single most important factor leading to the death to be recorded. | 86 % | 47 % | 52 % | 44 % | 50 % | 41 % | 33 % | 50 % | 50 % |
Subgroup analyses
Alignment according to type of death classified
Alignment of widely used systems
Alignment by region of use
Alignment by type of characteristic
Sensitivity analysis
Discussion
-
A global system must be easy to use, and produce data that are easily understood and valued by users (agreed by 100 % of panellists)
-
A global system must have clear guidelines for use and definitions for all terms used (agreed by 100 % of panellists)
-
A global system must use rules to ensure valid assignment of cause of death categories (agreed by 98 % of panellists)
-
A global system must be able to work with all levels of data (from both low-income and high-income countries), including minimal levels (agreed by 98 % of panellists)
-
A global system must ensure cause of death categories are relevant in all settings (agreed by 96 % of panellists)
-
A global system must produce data that can be used to inform strategies to prevent perinatal deaths (agreed by 96 % of panellists)