Srinivasan 1995.
| Methods | Randomised controlled field trial in rural south India | |
| Participants | 4 primary health centres were selected from each of 3 subcentres: therefore 12 subcentres in Karur health unit district, Tamil Nadu. 3 packages (2 intervention and 1 control) were implemented and included 294 pregnant women in the high‐risk package, 242 in the uniform package recommended by the Tamil Nadu Government, and 335 in the control package | |
| Interventions | Intervention: all packages were implemented by trained female auxiliary nurse midwives (ANMs). (1) The high‐risk package involved ANMs detecting pregnancies between the 12th and 18th weeks of gestation, registering them, and collecting baseline data. Screenings in the high‐risk package included at registration and at subsequent visits (20, 28, 34, 38 weeks), where the ANM undertook a clinical examination and distributed folic acid (dose dependent on haemoglobin). All high‐risk women were advised to give birth in a hospital. Postnatally, ANMs were expected to visit each new mother 3 times (3, 10, 40 postnatal days), when they were expected to detect and treat infections in mother and neonate, and to refer to the hospital those she could not diagnose/treat. (2) The Tamil Nadu Government (TNG) uniform package involved a set of routine antenatal care services recommended by the local government. These services were similar to the high‐risk package with regard to registration and visits by the ANM; however there was no high‐risk approach. Therefore all women were treated uniformly. Five visits postnatally (1, 3, 7, 15, 30) were to be made by the ANMs to all women who gave birth Control: implementation of the TNG package was the responsibility of general health services; therefore no special input was provided by project staff (e.g. ANMs). Data from pregnant women in this group represent outcomes under routine programme conditions |
|
| Outcomes | Preventable maternal morbidity, preventable perinatal morbidity, preventable neonatal morbidity | |
| Notes | Funding: no funding was acknowledged | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | High risk | Quote: "from each Primary Health Centre, three sub centres were selected at random from among those beyond a radius of 10 km from the PHC: one each was randomly allocated to the high‐risk package and the Tamil Nadu Government package and the third served as the control" Comment: this is a randomised controlled trial, the the unit of random allocation was the subcentre; however the method of randomisation has not been disclosed |
| Allocation concealment (selection bias) | Unclear risk | Comment: there is no mention of allocation concealment, although this is a randomised controlled field trial; information is insufficient to support a judgement |
| Blinding of participants and personnel (performance bias) All outcomes | Unclear risk | Comment: insufficient information to permit judgement |
| Blinding of outcome assessment (detection bias) All outcomes | Unclear risk | Comment: insufficient information to permit judgement |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Quote: "...69% in the HR series, 72% in the TNG series, and 77% in the control series could be considered for analysis..." Comment: reasons were given (migrated, abortion, and refusal), resulting in an attrition rate of 21% |
| Selective reporting (reporting bias) | Unclear risk | Comment: this is not a registered trial, but outcomes mentioned in the methods are reported in the results |
| Other bias | Low risk | Study seems to be free from other biases |