Table 2:
Multivariable analyses of factors associated with malaria infections in pregnancy by type of infection and region, using individual participant data
| Available data, % | Any malaria infection vs no malaria | Microscopic infection vs no malaria | Submicroscopic infection vs no malaria | Risk of submicroscopic infections among women with NAAT-positive infections | |||||
|---|---|---|---|---|---|---|---|---|---|
| aOR (95% CI) | p value | aOR (95% CI) | p value | aOR (95% CI) | p value | aOR (95% CI) | p value | ||
| Africa, high transmission intensity* (base model, N=6351, 94·1% of available data, 8 sublocations†) | |||||||||
| Age, years | |||||||||
| <20 | 94·1% | 2·43 (1·96–3·00) | <0·0001 | 3·24 (2·53–4·15) | <0·0001 | 1·59 (1·21–2·10) | 0·0009 | 0·49 (0·37–0·66) | <0·0001 |
| 20–29 | 94·1% | 1·34 (1·14–1·57) | 0·0003 | 1·64 (1·35–2·01) | <0·0001 | 1·13 (0·94–1·36) | 0·20 | 0·69 (0·55–0·86) | 0·0008 |
| ≥30 | 94·1% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| Gravidity | |||||||||
| Primigravidae | 94·1% | 1.24 (1·03–1·49) | 0·026 | 1·84 (1·50–2·27) | <0·0001 | 0·65 (0·51–0·83) | 0·0007 | 0·35 (0·28–0·45) | <0·0001 |
| Secundigravidae | 94·1% | 0·95 (0·80–1·11) | 0·50 | 1·21 (1·00–1·46) | 0·044 | 0·73 (0·59–0·90) | 0·0035 | 0·60 (0·49–0·75) | <0·0001 |
| Multigravidae or more | 94·1% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| PfPR2–10 | 94·1% | 1·02 (1·00–1·03) | 0·0094 | 1·03 (1·02–1·04) | <0·0001 | 1·00 (0·99–1·02) | 0·64 | 0·98 (0·97–0·98) | <0·0001 |
| Rainy vs dry season | 94·1% | 1·18 (1·05–1·33) | 0·0072 | 1·30 (1·14–1·48) | 0·0001 | 0·98 (0·83–1·15) | 0·80 | 0·75 (0·64–0·89) | 0·0007 |
| First antenatal clinic visit‡ | 94·1% | 0·50 (0·26–0·98) | 0·045 | 0·50 (0·26–0·97) | 0·039 | 0·48 (0·22–1·04) | 0·06 | 0·96 (0·76–1·21) | 0·71 |
| Africa, high transmission intensity (base model with additional variables of interest with restricted sample size) | |||||||||
| Gestational age, weeks§ | 94·4% | 1·00 (0·99–1·01) | 0·88 | 1·00 (0·99–1·01) | 0·99 | 1·00 (0·99–1·01) | 0·75 | 1·00 (0·99–1·01) | 0·84 |
| HIV infection | 58·8% | 1·65 (1·13–2·41) | 0·010 | 1·82 (1·17–2·83) | 0·0075 | 1·47 (0·94–2·20) | 0·090 | 0·81 (0·51–1·28) | 0·37 |
| Rural setting§ | 24·3% | 1·18 (0·76–1·82) | 0·45 | 1·05 (0·76–1·45) | 0·76 | 1·37 (0·92–2·04) | 0·13 | 1·43 (0·91–2·24) | 0·12 |
| Antimalarial use§ | 20·0% | 0·32 (0·23–0·45) | <0·0001 | 0·27 (0·18–0·40) | <0·0001 | 0·56 (0·36–0·88) | 0·011 | 2·08 (1·42–3·05) | 0·0002 |
| ITN use | 695% | 0·99 (0·87–1·12) | 0·84 | 1·02 (0·89–1·18) | 0·76 | 0·91 (0·77–1·09) | 0·33 | 0·90 (0·75–1·08) | 0·25 |
| Any net use | 86·1% | 0·94 (0·84–1·06) | 0·33 | 0·97 (0·85–1·10) | 0·59 | 0·90 (0·77–1·04) | 0·16 | 0·96 (0·82–1·12) | 0·63 |
| IRS | 46.4% | 0·98 (0·91–1·04) | 0·46 | 0·95 (0·87–1·03) | 0·21 | 1.00 (0·93–1·08) | 0·90 | 1·06 (0·97–1·16) | 0·21 |
| Africa, moderate-to-low transmission intensity* (base model, N=8199, 98·2% of available data, 24 sublocations†) | |||||||||
| Age, years | |||||||||
| <20 | 98·2% | 2·29 (1·87–2·81) | <0·0001 | 3·31 (2·37–4·62) | <0·0001 | 1·94 (1·54–2·45) | <0·0001 | 0·59 (0·41–0·85) | 0·0050 |
| 20–29 | 98·2% | 1·26 (1·08–1·47) | 0·0033 | 1·62 (1·23–2·14) | 0·0007 | 1·15 (0·97–1·36) | 0·11 | 0·71 (0·52–0·96) | 0·0268 |
| ≥30 | 98·2% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| Gravidity | |||||||||
| Primigravidae | 98·2% | 1·58 (1·34–1·88) | <0·0001 | 2·43 (1·88–3·14) | <0·0001 | 1·23 (1·01–1·51) | 0·039 | 0·51 (0·38–0·68) | <0·0001 |
| Secundigravidae | 98·2% | 1·23 (1·06–1·43) | 0·0078 | 1·42 (1·11–1·81) | 0·0045 | 1·15 (0·96–1·36) | 0·13 | 0·81 (0·61–1·06) | 0·12 |
| Multigravidae or more | 98·2% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| PfPR2–10 | 98·2% | 0·97 (0·95–0·98) | 0·0015 | 0·97 (0·94–1·00) | 0·031 | 0·97 (0·94–0·99) | 0·0042 | 0·99 (0·97–1·02) | 0·74 |
| Rainy vs dry season | 98·2% | 0·96 (0·86–1·07) | 0·42 | 0·95 (0·81–1·11) | 0·48 | 0·96 (0·85–1·09) | 0·52 | 1·02 (0·85–1·21) | 0·86 |
| First antenatal clinic visit} | 98·2% | 0·70 (0·37–1·30) | 0·25 | 0·61 (0·32–1·16) | 0·13 | 0·79 (0·39–1·58) | 0·50 | 1·30 (0·70–2·40) | 0·40 |
| Africa, moderate-to-low transmission intensity (base model with additional variables of interest with restricted sample size) | |||||||||
| Gestational age, weeks | 88.7% | 0·98 (0·96–0·99) | 0·0002 | 0·97 (0·96–0·99) | 0·0043 | 0·98 (0·96–0·99) | 0·0053 | 1·00 (0·98–1·03) | 0·65 |
| HIV infection | 90’6% | 1·19 (0·84–1·67) | 0·32 | 1·72 (1·04–2·86) | 0·036 | 0·95 (0·63–1·43) | 0·80 | 0·55 (0·30–1·00) | 0·049 |
| Rural setting | 37·0% | 1·41 (0·64–3·13) | 0·39 | 2·58 (1·06–6·29) | 0·037 | 1·13 (0·49–2·61) | 0·78 | 0·44 (0·22–0·87) | 0·018 |
| Antimalarial use | 17·7% | 0·61 (0·25–3·51) | 0·015 | 1·34 (0·80–2·26) | 0·27 | 0·40 (0·23–0·68) | 0·0007 | 0·39 (0·20–0·79) | 0·0085 |
| ITN use | 14·5% | 0·90 (0·63–1·29) | 0·57 | 0·85 (0·49–1·48) | 0·57 | 0·96 (0·62–1·47) | 0·85 | 1·14 (0·59–2·20) | 0·70 |
| Any net use | 66·3% | 0·88 (0·76–1·02) | 0·092 | 0·96 (0·78–1·19) | 0·72 | 0·84 (0·71–1·00) | 0·0447 | 0·87 (0·69–1·11) | 0·26 |
| IRS | 50·2% | 1·14 (0·82–1·58) | 0·43 | 1·13 (0·71–1·82) | 0·60 | 1·15 (0·79–1·68) | 0·46 | 1·02 (0·60–1·71) | 0·95 |
| The Americas and Asia (base model, N=10 068, 97·7% of available data, 22 sublocations † ) | |||||||||
| Age, years | |||||||||
| <20 | 97·7% | 1·58 (1·23–2·03) | 0·0003 | 2·76 (1·73–4·39) | <0·0001 | 1·31 (0·99–1·75) | 0·061 | 0·48 (0·28–0·80) | 0·0049 |
| 20·29 | 97·7% | 1·18 (0·99–1·42) | 0·071 | 1·66 (1·13–2·43) | 0·0093 | 1·09 (0·90–1·33) | 0·37 | 0·66 (0·44–0·99) | 0·046 |
| ≥30 | 97·7% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| Gravidity | |||||||||
| Primigravidae | 97·7% | 0·95 (0·80–1·14) | 0·61 | 1·20 (0·87–1·64) | 0·27 | 0·87 (0·71–1·07) | 0·18 | 0·73 (0·51–1·04) | 0·077 |
| Secundigravidae | 97·7% | 0·99 (0·83–1·19 | 0·95 | 1·03 (0·73–1·45) | 0·86 | 0·99 (0·81–1·20) | 0·90 | 0·96 (0·66–1·39) | 0·81 |
| Multigravidae or more | 97·7% | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. | 1 (ref) | .. |
| PfPR2–10 | 97·7% | 1·04 (1·02–1·07) | 0·0014 | 1·00 (0·96–1·04) | 0·90 | 1·06 (1·03–1·09) | 0·0004 | 1·06 (1·01–1·12) | 0·021 |
| Rainy vs dry season | 97·7% | 1·05 (0·90–1·21) | 0·56 | 1·24 (0·96–1·61) | 0·10 | 0·98 (0·83–1·16) | 0·83 | 0·79 (0·59–1·06) | 0·11 |
| First antenatal clinic visit} | 97·7% | 2·73 (0·86–8·68) | 0·089 | 11·15 (2·26–54·97) | 0·0031 | 1·93 (0·49–7·55) | 0·35 | 0·15 (0·02–0·86) | 0·034 |
| The Americas vs Asia | 97·7% | 0·87 (0·32–2·32) | 0·78 | 1·57 (0·39–6·31) | 0·53 | 0·71 (0·22–2·29) | 0·57 | 0·38 (0·08–1·84) | 0·23 |
| The Americas and Asia (base model with additional variables of interest with restricted sample size ¶ ) | |||||||||
| Gestational age, weeks | 96·1% | 1·00 (0·99–1·01) | 0·51 | 0·99 (0·97–1·01) | 0·17 | 1·00 (0·99–1·01) | 0·98 | 1·01 (0·99–1·04) | 0·22 |
| Rural setting | 80·7% | 1·46 (1·20–1·76) | <0·001 | 1·76 (1·25–2·46) | 0·0010 | 1·34 (1·07–1·67) | 0·012 | 0·77 (0·52–1·13) | 0·18 |
| Antimalarial use | 52·5% | 1·24 (0·90–1·70) | 0·18 | 1·35 (0·88–2·09) | 0·17 | 1·17 (0·80–1·72) | 0·41 | 0·87 (0·52–1·45) | 0·59 |
| ITN use | 53·6% | 1·08 (0·86–1·37) | 0·50 | 1·10 (0·78–1·55) | 0·58 | 1·09 (0·83–1·44) | 0·52 | 0·98 (0·65–1·48) | 0·94 |
| Any net use | 88·2% | 1·12 (0·96–1·31) | 0·16 | 1·16 (0·88–1·54) | 0·29 | 1·10 (0·92–1·31 | 0·29 | 0·95 (0·69–1·29) | 0·73 |
| IRS | 36·6% | 1·03 (0·76–1·39) | 0·86 | 0·86 (0·39–1·91) | 0·71 | 1·04 (0·75–1·44) | 0·80 | 1·21 (0·52–2·83) | 0·66 |
Multigravidae excluded secundigravidae. Available data refer to proportion of participants with an outcome on submicroscopic, microscopic, and no malaria. Studies in Asia and the Pacific were included under Asia; studies in central or South America were included under the Americas. In the multinomial model, the parameter estimates are relative to the referent group; for a unit change in the covariate the logit of the outcome is expected to change by its respective parameter estimate given the variables in the model are held constant (eg, for the probability of having microscopic malaria during pregnancy in high transmission areas in Africa, one increase in unit of PfPR2–10 resulted in an increase of 0·029 logit of microscopic malaria, which translates to an OR of 1·3 [exponent of 0·029]; this OR means that for 1 percentage point of increase in transmission there is a 3% increase in the odds of having microscopic malaria among pregnant women). aOR=adjusted OR. ITN=insecticide treated net. IRS=indoor residual spraying. LAMP=loop-mediated isothermal amplification. NAAT=nucleic acid amplification test (PCR or LAMP). OR=odds ratio. PfPR2–10=Plasmodium falciparum prevalence among children aged 2–10 years at the year of study visit, as estimated by the Malaria Atlas Project.
Africa, high transmission: PfPR2–10 ≥35%; Africa, moderate-to-low transmission: PfPR2–10 <35% (only 1 study [two sublocations] had PfPR2–10 <10%). In the model for Asia and the Americas, one study in Indonesia (9% of data), had PfPR2–10 of 25%, all other studies had a PfPR2–10 in the range of 0–5%, with 81% of studies having PfPR2–10 of <2%. PfPR2–10 was added as a continuous variable in all models.
Available data (data with information on microscopic infection, submicroscopic infection, and no malaria infection) in Africa, in high transmission areas: N=6746, nine sublocations (microscopic infection n=2455 [36·3%], submicroscopic infection n=1406 [20·8%], and no malaria infection n=2885 [42·8%]); in Africa, in moderate-to-low transmission areas: N=8350, 25 sublocations (microscopic infection n=949 [11·4%], submicroscopic infection n=1858 [22·3%], and no malaria infection n=5543 [66·4%]); and in the Americas and Asia: N=10 305, 23 sublocations (microscopic infection n=373 [3·6%], submicroscopic infection n=919 [8·9%], and no malaria infection n=9013 [87·5%]).
Comparison group for first antenatal clinic visit was not first antenatal clinic visit or unknown whether first visit or not. For Africa, in high transmission areas, the protective effect of first antenatal clinic visit for microscopic malaria comes from one study that only included women in their first or second pregnancy compared with six studies in which visits were not first antenatal clinic visit, or this information was unknown. Exclusion of first antenatal clinic visits from this model did not result in meaningful changes for the other covariates (data not shown).
In Africa, in high transmission areas, models for gestational age were run without first antenatal clinic visit because of non-convergence; PfPR2–10 was not included in the model for submicroscopic vs microscopic malaria. In Africa, in high transmission areas: models for setting and antimalarial use did not include first antenatal clinic visit because all involved studies included women at any antenatal clinic visit, or it was unknown. PfPR2–10 was not included in the models of antimalarial use, rural setting, and any net use because of non-convergence.
There was insufficient information on HIV infection in studies in the Americas and Asia.