Hot spot or not: a comparison of spatial statistical methods to predict prospective malaria infections

Mosha, Jacklin F; Sturrock, Hugh JW; Greenwood, Brian; Sutherland, Colin J; Gadalla, Nahla B; Atwal, Sharan; Hemelaar, Simon; Brown, Joelle M; Drakeley, Chris; Kibiki, Gibson; Bousema, Teun; Chandramohan, Daniel; Gosling, Roland D

doi:10.1186/1475-2875-13-53

Malaria Journal

Table 2 Odds of testing positive for P. falciparum infection during the follow-up survey: results from three geospatial models defined by baseline infection, anti-AMA-1 antibody prevalence, and anti MSP-1₁₉antibody prevalence adjusted for age

From: Hot spot or not: a comparison of spatial statistical methods to predict prospective malaria infections

Risk factor	Number tested	Malaria in second year n. %	OR 95% CI	Wald test P-value	Area under the ROC curve
PCR individual infection in baseline survey*
Neg	1,763	827 [46.9]	1	<0.001	0.560
Pos	905	521 [57.6]	1.58 [1.31-1.83]
PCR prevalence
Satscan exposure category	792	319 [40.3]	1	0.181	0.620
coldspot	1,728	864 [50.0]	1.35 [0.87-2.09]	<0.001	0.628
neither	726	500 [68.9]	4.54 [2.68-7.72]	0.966	0.597
hotspot	804	390 [48.5]	1	0.013
Kernel exposure quartiles	819	387 [47.2]	0.99 [0.60-1.64]	<0.001
<14.9	818	331 [40.5]	0.53 [0.32-0.88]	0.165
15-21.3	805	575 [71.4]	3.45 [2.06-5.75]	0.042
21.4-27.1	816	420 [51.5]	1	0.003
>27.1	794	344 [43.3]	0.69 [0.41-1.16]
Weighted exposure quartiles	807	372 [46.1]	0.58 [0.35-0.98]
<18.9	799	520 [65.1]	2.21 [1.31-3.73]
19-23.2
23.3-26.5
>26.5
AMA-1 individual prevalence
No	1,262	594 [47.1]	1	<0.001	0.554
Yes	1,071	593 [55.4]	1.45 [1.21-1.72]
AMA-1 prevalence
Satscan exposure category	904	310 [34.3]	1	<0.001	0.647
coldspot	1,092	554 [50.7]	2.65 [1.69-4.15]	<0.001	0.618
neither	1,250	819 [65.5]	5.84 [3.75-9.10]	0.002	0.609
hotspot	814	308 [37.8]	1	<0.001
Kernel exposure quartiles	813	414 [50.9]	2.26 [1.35-3.79]	<0.001
<27.9	812	425 [52.3]	2.62 [1.57-4.39]	0.154
28-38.9	807	536 [66.4]	5.16 [3.06-8.69]	<0.001
39-53.0	804	325 [40.4]	1	<0.001
>53.0	809	357 [44.1]	1.45 [0.86-2.44]
Weighted exposure quartiles	800	476 [59.5]	3.50 [2.07-5.91]
<18.9	803	498 [62.0]	3.33 [1.97-5.62]
19-23.9
24 -26.9
>26.9
MSP-1₁₉individual prevalence
No	1,730	924 [53.4]	1	0.196	0.541
Yes	681	341 [50.1]	0.88 [0.73-1.06]
MSP prevalence
Satscan exposure category	1,703	992 [58.2]	1	0.040	0.591
coldspot	967	493 [51.0]	0.64 [0.41-0.98]	<0.001	0.622
neither	576	198 [34.0]	0.21 [0.13-0.34]	0.773	0.625
hotspot	806	418 [51.9]	1	0.008	0.631
Kernel exposure quartiles	835	440 [52.7]	1.08 [0.65-1.78]	<0.001
<12.9	808	538 [66.6]	2.02 [1.21-3.38]	0.715
13-17.3	797	287 [36.0]	0.34 [0.20-0.55]	0.006
17.4-25.4	805	415 [51.6]	1	<0.001
>25.4	813	430 [52.9]	1.10 [0.66-1.81]
Weighted exposure quartiles	802	533 [66.5]	2.08 [1.23-3.51]
<16.5	796	278 [34.9]	0.35 [0.21-0.57]
16.6-18.3
18.4-22.7
>22.7
MSP-1₁₉&/ or AMA - 1 individual prevalence
No	986	253 [25.6]	1	0.986	0.530
Yes	1,237	466 [37.7]	1.00 [0.78-1.29]
MSP-1₁₉&/or AMA-1 prevalence
Satscan exposure category	-	400 [48.4]	-	-	-
coldspot	-	357 [44.6]	-	-	0.604
neither	-	386 [47.0]	-	-	0.530
hotspot	827	540 [67.8]	1	0.082
Kernel exposure quartiles	800	402 [49.3]	0.63 [0.37 - 1.07]	0.310
<44.5	822	380 [47.4]	0.77 [0.46-1.28]	0.001
44.6-51.4	797	372 [45.6]	2.44 [1.44-4.14]	0.507
51.5-59.3	816	502 [64.0]	1	0.063
>59.4	801		0.84 [0.49-1.42]	0.023
Weighted exposure quartiles	815		0.60 [0.36-1.03]
<16.5	784		1.86 [1.09-3.18]
16.6-18.3
18.4-22.7
>22.7

*Only individuals who were tested at both baseline and year 1.

Back to article page

ISSN: 1475-2875

Contact us

Submission enquiries: journalsubmissions@springernature.com