The study team searched eight databases for studies with data on frequency of maternal peripartum infection. Specific outcomes comprised chorioamnionitis in labour, puerperal endometritis, wound infection following caesarean section or perineal trauma, and sepsis occurring from onset of labour until 42 days postpartum. Extracted data included study characteristics; geographical location, "study extent" to reflect how nationally representative the study population might be, data collection methods, diagnostic methods, follow-up time and low-risk groups. These variables were used in meta-regression to explore their effect on incidence of infection.
| variable name | description | values | description |
| id | numeric id created by EppiReviewer software | ||
| title | author name and date | ||
| start | date that study data collection started | ||
| end | date that study data collection ended | ||
| country | country where data collection took place | ||
| study extent | measure of how nationally representative the study population is | 1 | National level (total country population or representative sample) |
| 2 | State/regional level | ||
| 3 | Network of health facilities e.g. surveillance network or insurance scheme | ||
| 4 | >2 facilities or field sites | ||
| 5 | Single facility or field site | ||
| data collection | data collection method | 1 | Routine healthcare data |
| 2 | Data collected specifically for the study | ||
| 3 | Unclear | ||
| follow-up | day at which measurement of infection outcome ended | dc/0 | hospital discharge |
| nk/88 | not known/not clear | ||
| 1 to 42 | days postpartum | ||
| diagnosis | method of diagnosing infection | 1 | Clinical diagnosis |
| 2 | Self-reported symptoms | ||
| 3 | Unclear | ||
| diagnosis of chorioamnionitis | method of diagnosing infection | 1 | ICD9/10 codes |
| 2 | Fever plus other signs e.g. tachycardiac | ||
| 3 | Fever alone | ||
| 4 | Unclear | ||
| livebirth | study population restricted to women with a live birth | 0 | No - both live and stillbirth included |
| 1 | Yes | ||
| vaginal delivery | study population restricted to women with a vaginal delivery | 0 | No - all modes of delivery included |
| 1 | Yes | ||
| singleton pregnancy | study population restricted to women with a singleton pregnancy | 0 | No - women with one or more foetuses included |
| 1 | Yes | ||
| term delivery | study population restricted to women with a term delivery | 0 | No - both pre-term and term delivery included |
| 1 | Yes | ||
| lowrisk preganncy | study population restricted to women with a low-risk pregnancy | 0 | No - both low and high-risk pregnancies included |
| 1 | Yes | ||
| severity (sepsis only) | Sepsis severity | 0 | Blood stream infection only - bacteraemia/septicaemia |
| 1 | Systemic inflammatory response syndrome (SIRS) - may also include more severe forms of disease and/or measures of blood stream infection | ||
| 2 | Severe sepsis with organ dysfunction and/or septic shock | ||
| quality | critical appraisal of study quality | 0 to 5 | |
| infection | number of women with the infection | ||
| denominator | total number of women in the study | ||
| percentage | percentage of women with the infection |