RV infection and HBGAs

Tables

Table 1: Rotavirus detection by enzyme immunoassay (EIA) according to …(n = 342)

Table 2: Distribution of RV status and RV genotypes according to sex (n = 342)

Table 3: Distribution of RV genotypes, Lewis profiles and ABO phenotypes according to RV status (n = 342)

Table 4: Association between ABO blood group, secretor status (secretor, non-secretor), and Lewis phenotypes with the infecting G/P rotavirus genotypes.

Contingency Tables:

RV vs Gender

RV vs Age_Interval_mo

RV vs Vaccine

RV vs fever

RV vs refusal_to_feed

RV vs ABO_pheno

RV vs H1_secretor

RV vs Le_ab_pheno

RV vs Combined amend

RV vs cough

RV vs source_of_water

RV vs toilet

RV vs heating

RV vs nursery

RV vs Site

Univariate models:

Association between Age Interval and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.12	0.06 – 0.22	<0.001
Age Interval mo [13 - 18]	1.60	0.62 – 4.15	0.331
Age Interval mo [19 - 24]	1.66	0.53 – 4.88	0.364
Age Interval mo [25 - 60]	0.73	0.16 – 2.57	0.651
Age Interval mo [7 - 12]	1.77	0.80 – 4.13	0.171
Observations	342
R2 Tjur	0.010

## We fitted a logistic model (estimated using ML) to predict RV with
## Age_Interval_mo (formula: RV ~ Age_Interval_mo). The model's explanatory power
## is very weak (Tjur's R2 = 1.00e-02). The model's intercept, corresponding to
## Age_Interval_mo = 0 - 6, is at -2.12 (95% CI [-2.84, -1.51], p < .001). Within
## this model:
## 
##   - The effect of Age Interval mo [13 - 18] is statistically non-significant and
## positive (beta = 0.47, 95% CI [-0.49, 1.42], p = 0.331; Std. beta = 0.47, 95%
## CI [-0.49, 1.42])
##   - The effect of Age Interval mo [19 - 24] is statistically non-significant and
## positive (beta = 0.51, 95% CI [-0.64, 1.58], p = 0.364; Std. beta = 0.51, 95%
## CI [-0.64, 1.58])
##   - The effect of Age Interval mo [25 - 60] is statistically non-significant and
## negative (beta = -0.31, 95% CI [-1.85, 0.94], p = 0.651; Std. beta = -0.31, 95%
## CI [-1.85, 0.94])
##   - The effect of Age Interval mo [7 - 12] is statistically non-significant and
## positive (beta = 0.57, 95% CI [-0.23, 1.42], p = 0.171; Std. beta = 0.57, 95%
## CI [-0.23, 1.42])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                             OR(95%CI)         P(Wald's test) P(LR-test)
## Age_Interval_mo: ref.=0 - 6                                  0.461     
##    13 - 18                  1.6 (0.62,4.1)    0.331                    
##    19 - 24                  1.66 (0.56,4.96)  0.364                    
##    25 - 60                  0.73 (0.19,2.83)  0.651                    
##    7 - 12                   1.77 (0.78,3.99)  0.171                    
##                                                                        
## Log-likelihood = -138.7083
## No. of observations = 342
## AIC value = 287.4166

Association between Gender and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.13	0.08 – 0.20	<0.001
Gender [Male]	1.59	0.86 – 2.99	0.142
Observations	342
R2 Tjur	0.006

## We fitted a logistic model (estimated using ML) to predict RV with Gender
## (formula: RV ~ Gender). The model's explanatory power is very weak (Tjur's R2 =
## 6.38e-03). The model's intercept, corresponding to Gender = Female, is at -2.05
## (95% CI [-2.56, -1.59], p < .001). Within this model:
## 
##   - The effect of Gender [Male] is statistically non-significant and positive
## (beta = 0.46, 95% CI [-0.15, 1.10], p = 0.142; Std. beta = 0.46, 95% CI [-0.15,
## 1.10])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                        OR(95%CI)         P(Wald's test) P(LR-test)
## Gender: Male vs Female 1.59 (0.86,2.95)  0.142          0.138     
##                                                                   
## Log-likelihood = -139.4147
## No. of observations = 342
## AIC value = 282.8295

Association between Vaccine and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.28	0.09 – 0.70	0.011
Vaccine [2 doses]	0.56	0.21 – 1.77	0.276
Vaccine [No RTHC]	0.83	0.22 – 3.28	0.785
Vaccine [Unvaccinated]	0.00	NA – 357687941411554125944610357248.00	0.984
Observations	342
R2 Tjur	0.007

## We fitted a logistic model (estimated using ML) to predict RV with Vaccine
## (formula: RV ~ Vaccine). The model's explanatory power is very weak (Tjur's R2
## = 7.00e-03). The model's intercept, corresponding to Vaccine = 1 dose, is at
## -1.28 (95% CI [-2.39, -0.36], p = 0.011). Within this model:
## 
##   - The effect of Vaccine [2 doses] is statistically non-significant and negative
## (beta = -0.58, 95% CI [-1.57, 0.57], p = 0.276; Std. beta = -0.58, 95% CI
## [-1.57, 0.57])
##   - The effect of Vaccine [No RTHC] is statistically non-significant and negative
## (beta = -0.19, 95% CI [-1.53, 1.19], p = 0.785; Std. beta = -0.19, 95% CI
## [-1.53, 1.19])
##   - The effect of Vaccine [Unvaccinated] is statistically non-significant and
## negative (beta = -14.29, 95% CI [, 68.05], p = 0.984; Std. beta = -14.29, 95%
## CI [, 68.05])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                      OR(95%CI)         P(Wald's test) P(LR-test)
## Vaccine: ref.=1 dose                                  0.422     
##    2 doses           0.56 (0.2,1.59)   0.276                    
##    No RTHC           0.83 (0.22,3.14)  0.785                    
##    Unvaccinated      0 (0,Inf)         0.984                    
##                                                                 
## Log-likelihood = -139.1099
## No. of observations = 342
## AIC value = 286.2198

Association between fever and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.12	0.08 – 0.18	<0.001
fever [Yes]	2.43	1.27 – 4.60	0.006
Observations	334
R2 Tjur	0.023

## We fitted a logistic model (estimated using ML) to predict RV with fever
## (formula: RV ~ fever). The model's explanatory power is weak (Tjur's R2 =
## 0.02). The model's intercept, corresponding to fever = None, is at -2.10 (95%
## CI [-2.52, -1.72], p < .001). Within this model:
## 
##   - The effect of fever [Yes] is statistically significant and positive (beta =
## 0.89, 95% CI [0.24, 1.53], p = 0.006; Std. beta = 0.89, 95% CI [0.24, 1.53])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                    OR(95%CI)         P(Wald's test) P(LR-test)
## fever: Yes vs None 2.43 (1.28,4.61)  0.006          0.008     
##                                                               
## Log-likelihood = -132.1382
## No. of observations = 334
## AIC value = 268.2764

Association between refusal to feed and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.14	0.09 – 0.20	<0.001
refusal to feed [Yes]	1.66	0.87 – 3.11	0.116
Observations	332
R2 Tjur	0.008

## We fitted a logistic model (estimated using ML) to predict RV with
## refusal_to_feed (formula: RV ~ refusal_to_feed). The model's explanatory power
## is very weak (Tjur's R2 = 7.55e-03). The model's intercept, corresponding to
## refusal_to_feed = None, is at -1.99 (95% CI [-2.41, -1.60], p < .001). Within
## this model:
## 
##   - The effect of refusal to feed [Yes] is statistically non-significant and
## positive (beta = 0.51, 95% CI [-0.13, 1.13], p = 0.116; Std. beta = 0.51, 95%
## CI [-0.13, 1.13])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                              OR(95%CI)         P(Wald's test) P(LR-test)
## refusal_to_feed: Yes vs None 1.66 (0.88,3.12)  0.116          0.12      
##                                                                         
## Log-likelihood = -134.1799
## No. of observations = 332
## AIC value = 272.3597

Association between H1 secretor and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.12	0.05 – 0.24	<0.001
H1 secretor [secretor]	1.51	0.68 – 3.83	0.340
Observations	342
R2 Tjur	0.003

## We fitted a logistic model (estimated using ML) to predict RV with H1_secretor
## (formula: RV ~ H1_secretor). The model's explanatory power is very weak (Tjur's
## R2 = 2.70e-03). The model's intercept, corresponding to H1_secretor =
## nonsecretor, is at -2.13 (95% CI [-3.01, -1.42], p < .001). Within this model:
## 
##   - The effect of H1 secretor [secretor] is statistically non-significant and
## positive (beta = 0.41, 95% CI [-0.38, 1.34], p = 0.340; Std. beta = 0.41, 95%
## CI [-0.38, 1.34])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                                      OR(95%CI)         P(Wald's test)
## H1_secretor: secretor vs nonsecretor 1.51 (0.65,3.54)  0.34          
##                                                                      
##                                      P(LR-test)
## H1_secretor: secretor vs nonsecretor 0.322     
##                                                
## Log-likelihood = -140.0245
## No. of observations = 342
## AIC value = 284.049

Association between ABO phenotypes and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.26	0.15 – 0.43	<0.001
ABO pheno [AB]	3.79	0.83 – 17.42	0.077
ABO pheno [B]	0.52	0.14 – 1.54	0.273
ABO pheno [O]	0.44	0.23 – 0.87	0.018
Observations	342
R2 Tjur	0.041

## We fitted a logistic model (estimated using ML) to predict RV with ABO_pheno
## (formula: RV ~ ABO_pheno). The model's explanatory power is weak (Tjur's R2 =
## 0.04). The model's intercept, corresponding to ABO_pheno = A, is at -1.33 (95%
## CI [-1.87, -0.85], p < .001). Within this model:
## 
##   - The effect of ABO pheno [AB] is statistically non-significant and positive
## (beta = 1.33, 95% CI [-0.19, 2.86], p = 0.077; Std. beta = 1.33, 95% CI [-0.19,
## 2.86])
##   - The effect of ABO pheno [B] is statistically non-significant and negative
## (beta = -0.65, 95% CI [-1.95, 0.43], p = 0.273; Std. beta = -0.65, 95% CI
## [-1.95, 0.43])
##   - The effect of ABO pheno [O] is statistically significant and negative (beta =
## -0.81, 95% CI [-1.49, -0.14], p = 0.018; Std. beta = -0.81, 95% CI [-1.49,
## -0.14])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                   OR(95%CI)          P(Wald's test) P(LR-test)
## ABO_pheno: ref.=A                                   0.01      
##    AB             3.79 (0.87,16.57)  0.077                    
##    B              0.52 (0.16,1.67)   0.273                    
##    O              0.44 (0.23,0.87)   0.018                    
##                                                               
## Log-likelihood = -134.7956
## No. of observations = 342
## AIC value = 277.5913

Association between Le ab phenotypes and rotavirus (RV) infection among children

model8 <- glm(RV ~ Le_ab_pheno,
    family = "binomial",
    data = African_infants)

tab_model(model8)

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.38	0.18 – 0.77	0.010
Le ab pheno [a-b+]	0.64	0.25 – 1.67	0.353
Le ab pheno [a+b-]	0.00	0.00 – 32009071.95	0.983
Le ab pheno [a+b+]	0.45	0.20 – 1.08	0.063
Observations	342
R2 Tjur	0.051

report::report(model8)

## We fitted a logistic model (estimated using ML) to predict RV with Le_ab_pheno
## (formula: RV ~ Le_ab_pheno). The model's explanatory power is weak (Tjur's R2 =
## 0.05). The model's intercept, corresponding to Le_ab_pheno = a-b-, is at -0.96
## (95% CI [-1.73, -0.26], p = 0.010). Within this model:
## 
##   - The effect of Le ab pheno [a-b+] is statistically non-significant and
## negative (beta = -0.45, 95% CI [-1.40, 0.52], p = 0.353; Std. beta = -0.45, 95%
## CI [-1.40, 0.52])
##   - The effect of Le ab pheno [a+b-] is statistically non-significant and
## negative (beta = -17.61, 95% CI [-285.95, 17.28], p = 0.983; Std. beta =
## -17.61, 95% CI [-285.95, 17.28])
##   - The effect of Le ab pheno [a+b+] is statistically non-significant and
## negative (beta = -0.80, 95% CI [-1.62, 0.08], p = 0.063; Std. beta = -0.80, 95%
## CI [-1.62, 0.08])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

epiDisplay::logistic.display(model8)

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                        OR(95%CI)         P(Wald's test) P(LR-test)
## Le_ab_pheno: ref.=a-b-                                  < 0.001   
##    a-b+                0.64 (0.25,1.65)  0.353                    
##    a+b-                0 (0,Inf)         0.983                    
##    a+b+                0.45 (0.19,1.04)  0.063                    
##                                                                   
## Log-likelihood = -127.7171
## No. of observations = 342
## AIC value = 263.4342

Association between Combined and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.64	0.23 – 1.62	0.350
Combined amend [nonsec Lea+b-]	0.00	0.00 – 4247518100.04	0.985
Combined amend [Sec Lea-b-]	0.31	0.06 – 1.41	0.146
Combined amend [Sec Lea+b-]	0.00	0.00 – 27248842238187563646976.00	0.991
Combined amend [Sec Leb+]	0.30	0.11 – 0.86	0.020
Observations	342
R2 Tjur	0.059

## We fitted a logistic model (estimated using ML) to predict RV with
## Combined_amend (formula: RV ~ Combined_amend). The model's explanatory power is
## weak (Tjur's R2 = 0.06). The model's intercept, corresponding to Combined_amend
## = nonsec Lea-b-, is at -0.45 (95% CI [-1.45, 0.48], p = 0.350). Within this
## model:
## 
##   - The effect of Combined amend [nonsec Lea+b-] is statistically non-significant
## and negative (beta = -18.11, 95% CI [-328.00, 22.17], p = 0.985; Std. beta =
## -18.11, 95% CI [-328.00, 22.17])
##   - The effect of Combined amend [Sec Lea-b-] is statistically non-significant
## and negative (beta = -1.16, 95% CI [-2.86, 0.34], p = 0.146; Std. beta = -1.16,
## 95% CI [-2.86, 0.34])
##   - The effect of Combined amend [Sec Lea+b-] is statistically non-significant
## and negative (beta = -18.11, 95% CI [-554.85, 51.66], p = 0.991; Std. beta =
## -18.11, 95% CI [-554.85, 51.66])
##   - The effect of Combined amend [Sec Leb+] is statistically significant and
## negative (beta = -1.20, 95% CI [-2.19, -0.15], p = 0.020; Std. beta = -1.20,
## 95% CI [-2.19, -0.15])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                                    OR(95%CI)        P(Wald's test) P(LR-test)
## Combined_amend: ref.=nonsec Lea-b-                                 < 0.001   
##    nonsec Lea+b-                   0 (0,Inf)        0.985                    
##    Sec Lea-b-                      0.31 (0.07,1.5)  0.146                    
##    Sec Lea+b-                      0 (0,Inf)        0.991                    
##    Sec Leb+                        0.3 (0.11,0.83)  0.02                     
##                                                                              
## Log-likelihood = -127.0018
## No. of observations = 342
## AIC value = 264.0036

Association between cough and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.13	0.08 – 0.19	<0.001
cough [Yes]	1.97	1.04 – 3.68	0.035
Observations	332
R2 Tjur	0.014

## We fitted a logistic model (estimated using ML) to predict RV with cough
## (formula: RV ~ cough). The model's explanatory power is very weak (Tjur's R2 =
## 0.01). The model's intercept, corresponding to cough = None, is at -2.05 (95%
## CI [-2.48, -1.66], p < .001). Within this model:
## 
##   - The effect of cough [Yes] is statistically significant and positive (beta =
## 0.68, 95% CI [0.04, 1.30], p = 0.035; Std. beta = 0.68, 95% CI [0.04, 1.30])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                    OR(95%CI)         P(Wald's test) P(LR-test)
## cough: Yes vs None 1.97 (1.05,3.69)  0.035          0.038     
##                                                               
## Log-likelihood = -133.2281
## No. of observations = 332
## AIC value = 270.4563

Association between heating and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.14	0.10 – 0.19	<0.001
heating [paraffin]	3.24	1.32 – 7.47	0.007
heating [wood]	2.40	0.12 – 19.27	0.454
Observations	336
R2 Tjur	0.024

## We fitted a logistic model (estimated using ML) to predict RV with heating
## (formula: RV ~ heating). The model's explanatory power is weak (Tjur's R2 =
## 0.02). The model's intercept, corresponding to heating = electricity, is at
## -1.97 (95% CI [-2.33, -1.64], p < .001). Within this model:
## 
##   - The effect of heating [paraffin] is statistically significant and positive
## (beta = 1.17, 95% CI [0.27, 2.01], p = 0.007; Std. beta = 1.17, 95% CI [0.27,
## 2.01])
##   - The effect of heating [wood] is statistically non-significant and positive
## (beta = 0.87, 95% CI [-2.15, 2.96], p = 0.454; Std. beta = 0.87, 95% CI [-2.15,
## 2.96])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                           OR(95%CI)         P(Wald's test) P(LR-test)
## heating: ref.=electricity                                  0.036     
##    paraffin               3.24 (1.37,7.63)  0.007                    
##    wood                   2.4 (0.24,23.64)  0.454                    
##                                                                      
## Log-likelihood = -132.6583
## No. of observations = 336
## AIC value = 271.3166

Association between study site and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.40	0.26 – 0.61	<0.001
Site [OPHC]	0.22	0.12 – 0.42	<0.001
Observations	342
R2 Tjur	0.071

## We fitted a logistic model (estimated using ML) to predict RV with Site
## (formula: RV ~ Site). The model's explanatory power is weak (Tjur's R2 = 0.07).
## The model's intercept, corresponding to Site = DGMAH, is at -0.91 (95% CI
## [-1.36, -0.49], p < .001). Within this model:
## 
##   - The effect of Site [OPHC] is statistically significant and negative (beta =
## -1.49, 95% CI [-2.13, -0.87], p < .001; Std. beta = -1.49, 95% CI [-2.13,
## -0.87])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                     OR(95%CI)         P(Wald's test) P(LR-test)
## Site: OPHC vs DGMAH 0.22 (0.12,0.42)  < 0.001        < 0.001   
##                                                                
## Log-likelihood = -129.4831
## No. of observations = 342
## AIC value = 262.9662

Association between source of water and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.50	0.02 – 5.22	0.571
source of water [indoor tap]	0.45	0.04 – 9.95	0.523
source of water [outdoor tap]	0.24	0.02 – 5.21	0.247
source of water [river water]	4236359.95	0.00 – NA	0.986
Observations	337
R2 Tjur	0.033

## We fitted a logistic model (estimated using ML) to predict RV with
## source_of_water (formula: RV ~ source_of_water). The model's explanatory power
## is weak (Tjur's R2 = 0.03). The model's intercept, corresponding to
## source_of_water = borehole, is at -0.69 (95% CI [-3.76, 1.65], p = 0.571).
## Within this model:
## 
##   - The effect of source of water [indoor tap] is statistically non-significant
## and negative (beta = -0.80, 95% CI [-3.18, 2.30], p = 0.523; Std. beta = -0.80,
## 95% CI [-3.18, 2.30])
##   - The effect of source of water [outdoor tap] is statistically non-significant
## and negative (beta = -1.44, 95% CI [-3.83, 1.65], p = 0.247; Std. beta = -1.44,
## 95% CI [-3.83, 1.65])
##   - The effect of source of water [river water] is statistically non-significant
## and positive (beta = 15.26, 95% CI [-163.70, ], p = 0.986; Std. beta = 15.26,
## 95% CI [-163.70, ])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                                OR(95%CI)           P(Wald's test) P(LR-test)
## source_of_water: ref.=borehole                                    0.034     
##    indoor tap                  0.45 (0.04,5.19)    0.523                    
##    outdoor tap                 0.24 (0.02,2.72)    0.247                    
##    river water                 4236359.95 (0,Inf)  0.986                    
##                                                                             
## Log-likelihood = -131.8013
## No. of observations = 337
## AIC value = 271.6025

Association between toilet and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.50	0.07 – 2.56	0.423
toilet [flush]	0.31	0.06 – 2.29	0.185
toilet [none-outdoors]	0.00	NA – 25002402671041197162688852485778799889884632842240.00	0.988
toilet [pit latrine]	0.32	0.06 – 2.46	0.213
Observations	333
R2 Tjur	0.007

## We fitted a logistic model (estimated using ML) to predict RV with toilet
## (formula: RV ~ toilet). The model's explanatory power is very weak (Tjur's R2 =
## 6.87e-03). The model's intercept, corresponding to toilet = bucket, is at -0.69
## (95% CI [-2.67, 0.94], p = 0.423). Within this model:
## 
##   - The effect of toilet [flush] is statistically non-significant and negative
## (beta = -1.18, 95% CI [-2.87, 0.83], p = 0.185; Std. beta = -1.18, 95% CI
## [-2.87, 0.83])
##   - The effect of toilet [none-outdoors] is statistically non-significant and
## negative (beta = -14.87, 95% CI [, 113.74], p = 0.988; Std. beta = -14.87, 95%
## CI [, 113.74])
##   - The effect of toilet [pit latrine] is statistically non-significant and
## negative (beta = -1.13, 95% CI [-2.84, 0.90], p = 0.213; Std. beta = -1.13, 95%
## CI [-2.84, 0.90])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                     OR(95%CI)         P(Wald's test) P(LR-test)
## toilet: ref.=bucket                                  0.55      
##    flush            0.31 (0.05,1.76)  0.185                    
##    none-outdoors    0 (0,Inf)         0.988                    
##    pit latrine      0.32 (0.05,1.91)  0.213                    
##                                                                
## Log-likelihood = -132.6672
## No. of observations = 333
## AIC value = 273.3344

Association between toilet and rotavirus (RV) infection among children

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.14	0.10 – 0.20	<0.001
nursery [Yes]	1.95	0.82 – 4.28	0.108
Observations	335
R2 Tjur	0.008

## We fitted a logistic model (estimated using ML) to predict RV with nursery
## (formula: RV ~ nursery). The model's explanatory power is very weak (Tjur's R2
## = 7.95e-03). The model's intercept, corresponding to nursery = None, is at
## -1.94 (95% CI [-2.30, -1.61], p < .001). Within this model:
## 
##   - The effect of nursery [Yes] is statistically non-significant and positive
## (beta = 0.67, 95% CI [-0.20, 1.45], p = 0.108; Std. beta = 0.67, 95% CI [-0.20,
## 1.45])
## 
## Standardized parameters were obtained by fitting the model on a standardized
## version of the dataset. 95% Confidence Intervals (CIs) and p-values were
## computed using a Wald z-distribution approximation.

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                      OR(95%CI)         P(Wald's test) P(LR-test)
## nursery: Yes vs None 1.95 (0.86,4.42)  0.108          0.124     
##                                                                 
## Log-likelihood = -132.8338
## No. of observations = 335
## AIC value = 269.6676

Showing bivariate models

	RV			RV			RV			RV			RV			RV
Predictors	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p
(Intercept)	0.12	0.06 – 0.22	<0.001	0.13	0.08 – 0.20	<0.001	0.28	0.09 – 0.70	0.011	0.12	0.08 – 0.18	<0.001	0.14	0.09 – 0.20	<0.001	0.12	0.05 – 0.24	<0.001
Age Interval mo [13 - 18]	1.60	0.62 – 4.15	0.331
Age Interval mo [19 - 24]	1.66	0.53 – 4.88	0.364
Age Interval mo [25 - 60]	0.73	0.16 – 2.57	0.651
Age Interval mo [7 - 12]	1.77	0.80 – 4.13	0.171
Gender [Male]				1.59	0.86 – 2.99	0.142
Vaccine [2 doses]							0.56	0.21 – 1.77	0.276
Vaccine [No RTHC]							0.83	0.22 – 3.28	0.785
Vaccine [Unvaccinated]							0.00	NA – 357687941411554125944610357248.00	0.984
fever [Yes]										2.43	1.27 – 4.60	0.006
refusal to feed [Yes]													1.66	0.87 – 3.11	0.116
H1 secretor [secretor]																1.51	0.68 – 3.83	0.340
Observations	342			342			342			334			332			342
R2 Tjur	0.010			0.006			0.007			0.023			0.008			0.003

Multivariate models:

Models0

mo <- glm(RV ~ Site + Le_ab_pheno + ABO_pheno + heating + fever,
    family = "binomial",
    data = African_infants)

summary(mo)

## 
## Call:
## glm(formula = RV ~ Site + Le_ab_pheno + ABO_pheno + heating + 
##     fever, family = "binomial", data = African_infants)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.3331  -0.5736  -0.3547  -0.0001   2.4074  
## 
## Coefficients:
##                  Estimate Std. Error z value Pr(>|z|)   
## (Intercept)       0.17878    0.59016   0.303  0.76194   
## SiteOPHC         -1.27835    0.43811  -2.918  0.00352 **
## Le_ab_phenoa-b+  -0.83213    0.54047  -1.540  0.12365   
## Le_ab_phenoa+b- -17.28175  791.02720  -0.022  0.98257   
## Le_ab_phenoa+b+  -0.99606    0.47175  -2.111  0.03473 * 
## ABO_phenoAB       1.79938    0.98310   1.830  0.06720 . 
## ABO_phenoB       -0.62193    0.64497  -0.964  0.33491   
## ABO_phenoO       -0.72874    0.37975  -1.919  0.05499 . 
## heatingparaffin   1.19290    0.52142   2.288  0.02215 * 
## heatingwood       0.93419    1.24962   0.748  0.45471   
## feverYes         -0.01676    0.43836  -0.038  0.96951   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 271.08  on 332  degrees of freedom
## Residual deviance: 215.04  on 322  degrees of freedom
##   (9 observations deleted due to missingness)
## AIC: 237.04
## 
## Number of Fisher Scoring iterations: 17

tab_model(mo)

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	1.20	0.37 – 3.80	0.762
Site [OPHC]	0.28	0.12 – 0.65	0.004
Le ab pheno [a-b+]	0.44	0.15 – 1.26	0.124
Le ab pheno [a+b-]	0.00	0.00 – 4096769.01	0.983
Le ab pheno [a+b+]	0.37	0.15 – 0.95	0.035
ABO pheno [AB]	6.05	0.92 – 51.72	0.067
ABO pheno [B]	0.54	0.13 – 1.75	0.335
ABO pheno [O]	0.48	0.23 – 1.02	0.055
heating [paraffin]	3.30	1.15 – 9.08	0.022
heating [wood]	2.55	0.11 – 24.15	0.455
fever [Yes]	0.98	0.41 – 2.31	0.970
Observations	333
R2 Tjur	0.182

epiDisplay::logistic.display(mo) 

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                           crude OR(95%CI)    adj. OR(95%CI)     P(Wald's test)
## Site: OPHC vs DGMAH       0.22 (0.12,0.42)   0.28 (0.12,0.66)   0.004         
##                                                                               
## Le_ab_pheno: ref.=a-b-                                                        
##    a-b+                   0.6 (0.23,1.57)    0.44 (0.15,1.26)   0.124         
##    a+b-                   0 (0,Inf)          0 (0,Inf)          0.983         
##    a+b+                   0.45 (0.19,1.04)   0.37 (0.15,0.93)   0.035         
##                                                                               
## ABO_pheno: ref.=A                                                             
##    AB                     4.98 (1.03,24.2)   6.05 (0.88,41.52)  0.067         
##    B                      0.55 (0.17,1.78)   0.54 (0.15,1.9)    0.335         
##    O                      0.4 (0.2,0.8)      0.48 (0.23,1.02)   0.055         
##                                                                               
## heating: ref.=electricity                                                     
##    paraffin               3.38 (1.42,8.02)   3.3 (1.19,9.16)    0.022         
##    wood                   2.38 (0.24,23.47)  2.55 (0.22,29.47)  0.455         
##                                                                               
## fever: Yes vs None        2.42 (1.28,4.59)   0.98 (0.42,2.32)   0.97          
##                                                                               
##                           P(LR-test)
## Site: OPHC vs DGMAH       0.003     
##                                     
## Le_ab_pheno: ref.=a-b-    < 0.001   
##    a-b+                             
##    a+b-                             
##    a+b+                             
##                                     
## ABO_pheno: ref.=A         0.02      
##    AB                               
##    B                                
##    O                                
##                                     
## heating: ref.=electricity 0.07      
##    paraffin                         
##    wood                             
##                                     
## fever: Yes vs None        0.97      
##                                     
## Log-likelihood = -107.5213
## No. of observations = 333
## AIC value = 237.0425

Model A

	RV
Predictors	Odds Ratios	CI	p
(Intercept)	0.74	0.23 – 2.28	0.607
fever [Yes]	2.03	1.01 – 4.06	0.045
ABO pheno [AB]	7.10	1.16 – 57.83	0.039
ABO pheno [B]	0.50	0.13 – 1.59	0.275
ABO pheno [O]	0.50	0.24 – 1.05	0.066
heating [paraffin]	3.94	1.43 – 10.47	0.006
heating [wood]	2.01	0.09 – 18.20	0.570
Combined amend [nonsec Lea+b-]	0.00	0.00 – 4174784215.02	0.984
Combined amend [Sec Lea-b-]	0.29	0.05 – 1.41	0.138
Combined amend [Sec Lea+b-]	0.00	0.00 – 25059605166731082858496.00	0.991
Combined amend [Sec Leb+]	0.23	0.08 – 0.69	0.007
Observations	333
R2 Tjur	0.157

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                                    crude OR(95%CI)    adj. OR(95%CI)    
## fever: Yes vs None                 2.42 (1.28,4.59)   2.03 (1.02,4.06)  
##                                                                         
## ABO_pheno: ref.=A                                                       
##    AB                              4.98 (1.03,24.2)   7.1 (1.1,45.82)   
##    B                               0.55 (0.17,1.78)   0.5 (0.15,1.73)   
##    O                               0.4 (0.2,0.8)      0.5 (0.24,1.05)   
##                                                                         
## heating: ref.=electricity                                               
##    paraffin                        3.38 (1.42,8.02)   3.94 (1.47,10.57) 
##    wood                            2.38 (0.24,23.47)  2.01 (0.18,22.12) 
##                                                                         
## Combined_amend: ref.=nonsec Lea-b-                                      
##    nonsec Lea+b-                   0 (0,Inf)          0 (0,Inf)         
##    Sec Lea-b-                      0.31 (0.07,1.5)    0.29 (0.05,1.5)   
##    Sec Lea+b-                      0 (0,Inf)          0 (0,Inf)         
##    Sec Leb+                        0.29 (0.11,0.81)   0.23 (0.08,0.67)  
##                                                                         
##                                    P(Wald's test) P(LR-test)
## fever: Yes vs None                 0.045          0.048     
##                                                             
## ABO_pheno: ref.=A                                 0.013     
##    AB                              0.039                    
##    B                               0.275                    
##    O                               0.066                    
##                                                             
## heating: ref.=electricity                         0.031     
##    paraffin                        0.006                    
##    wood                            0.57                     
##                                                             
## Combined_amend: ref.=nonsec Lea-b-                < 0.001   
##    nonsec Lea+b-                   0.984                    
##    Sec Lea-b-                      0.138                    
##    Sec Lea+b-                      0.991                    
##    Sec Leb+                        0.007                    
##                                                             
## Log-likelihood = -110.8197
## No. of observations = 333
## AIC value = 243.6394

Model B

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                              crude OR(95%CI)    adj. OR(95%CI)    
## fever: Yes vs None           2.4 (1.27,4.55)    2.18 (1.09,4.37)  
##                                                                   
## ABO_pheno: ref.=A                                                 
##    AB                        4.98 (1.03,24.2)   5.51 (1.02,29.75) 
##    B                         0.55 (0.17,1.78)   0.48 (0.14,1.64)  
##    O                         0.41 (0.21,0.81)   0.43 (0.21,0.88)  
##                                                                   
## heating: ref.=electricity                                         
##    paraffin                  3.35 (1.41,7.96)   3.51 (1.36,9.05)  
##    wood                      2.36 (0.24,23.29)  2.07 (0.18,24.27) 
##                                                                   
## Gender: Male vs Female       1.66 (0.88,3.12)   1.78 (0.9,3.5)    
##                                                                   
## refusal_to_feed: Yes vs None 1.65 (0.88,3.1)    1.4 (0.7,2.82)    
##                                                                   
##                              P(Wald's test) P(LR-test)
## fever: Yes vs None           0.028          0.03      
##                                                       
## ABO_pheno: ref.=A                           0.005     
##    AB                        0.047                    
##    B                         0.244                    
##    O                         0.02                     
##                                                       
## heating: ref.=electricity                   0.042     
##    paraffin                  0.01                     
##    wood                      0.561                    
##                                                       
## Gender: Male vs Female       0.095          0.091     
##                                                       
## refusal_to_feed: Yes vs None 0.346          0.349     
##                                                       
## Log-likelihood = -120.3031
## No. of observations = 331
## AIC value = 258.6063

Model C

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                              crude OR(95%CI)    adj. OR(95%CI)    
## fever: Yes vs None           2.4 (1.27,4.55)    1.94 (0.95,3.94)  
##                                                                   
## ABO_pheno: ref.=A                                                 
##    AB                        4.98 (1.03,24.2)   7.08 (1.08,46.58) 
##    B                         0.55 (0.17,1.78)   0.43 (0.12,1.53)  
##    O                         0.41 (0.21,0.81)   0.53 (0.26,1.11)  
##                                                                   
## heating: ref.=electricity                                         
##    paraffin                  3.35 (1.41,7.96)   4.2 (1.54,11.5)   
##    wood                      2.36 (0.24,23.29)  1.47 (0.11,18.81) 
##                                                                   
## Gender: Male vs Female       1.66 (0.88,3.12)   1.55 (0.78,3.1)   
##                                                                   
## refusal_to_feed: Yes vs None 1.65 (0.88,3.1)    1.36 (0.66,2.78)  
##                                                                   
## Le_ab_pheno: ref.=a-b-                                            
##    a-b+                      0.6 (0.23,1.57)    0.47 (0.16,1.33)  
##    a+b-                      0 (0,Inf)          0 (0,Inf)         
##    a+b+                      0.45 (0.19,1.04)   0.36 (0.15,0.89)  
##                                                                   
##                              P(Wald's test) P(LR-test)
## fever: Yes vs None           0.067          0.069     
##                                                       
## ABO_pheno: ref.=A                           0.016     
##    AB                        0.042                    
##    B                         0.195                    
##    O                         0.093                    
##                                                       
## heating: ref.=electricity                   0.027     
##    paraffin                  0.005                    
##    wood                      0.767                    
##                                                       
## Gender: Male vs Female       0.214          0.21      
##                                                       
## refusal_to_feed: Yes vs None 0.405          0.407     
##                                                       
## Le_ab_pheno: ref.=a-b-                      < 0.001   
##    a-b+                      0.152                    
##    a+b-                      0.983                    
##    a+b+                      0.027                    
##                                                       
## Log-likelihood = -110.5603
## No. of observations = 331
## AIC value = 245.1205

Model D

## 
## Logistic regression predicting RV : Positive vs Negative 
##  
##                              crude OR(95%CI)    adj. OR(95%CI)    
## fever: Yes vs None           2.4 (1.27,4.55)    0.97 (0.41,2.31)  
##                                                                   
## ABO_pheno: ref.=A                                                 
##    AB                        4.98 (1.03,24.2)   6.42 (0.91,45.25) 
##    B                         0.55 (0.17,1.78)   0.54 (0.15,1.94)  
##    O                         0.41 (0.21,0.81)   0.5 (0.24,1.06)   
##                                                                   
## heating: ref.=electricity                                         
##    paraffin                  3.35 (1.41,7.96)   3.56 (1.26,10.01) 
##    wood                      2.36 (0.24,23.29)  2.32 (0.18,29.64) 
##                                                                   
## Gender: Male vs Female       1.66 (0.88,3.12)   1.36 (0.67,2.76)  
##                                                                   
## refusal_to_feed: Yes vs None 1.65 (0.88,3.1)    1.33 (0.65,2.76)  
##                                                                   
## Le_ab_pheno: ref.=a-b-                                            
##    a-b+                      0.6 (0.23,1.57)    0.44 (0.15,1.27)  
##    a+b-                      0 (0,Inf)          0 (0,Inf)         
##    a+b+                      0.45 (0.19,1.04)   0.37 (0.14,0.93)  
##                                                                   
## Site: OPHC vs DGMAH          0.22 (0.12,0.42)   0.3 (0.13,0.71)   
##                                                                   
##                              P(Wald's test) P(LR-test)
## fever: Yes vs None           0.953          0.953     
##                                                       
## ABO_pheno: ref.=A                           0.024     
##    AB                        0.062                    
##    B                         0.346                    
##    O                         0.072                    
##                                                       
## heating: ref.=electricity                   0.056     
##    paraffin                  0.016                    
##    wood                      0.516                    
##                                                       
## Gender: Male vs Female       0.396          0.394     
##                                                       
## refusal_to_feed: Yes vs None 0.435          0.437     
##                                                       
## Le_ab_pheno: ref.=a-b-                      0.001     
##    a-b+                      0.127                    
##    a+b-                      0.983                    
##    a+b+                      0.035                    
##                                                       
## Site: OPHC vs DGMAH          0.006          0.006     
##                                                       
## Log-likelihood = -106.79
## No. of observations = 331
## AIC value = 239.58

Showing Multivariate models

	RV			RV			RV			RV
Predictors	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p	Odds Ratios	CI	p
(Intercept)	0.74	0.23 – 2.28	0.607	0.11	0.05 – 0.23	<0.001	0.30	0.10 – 0.83	0.024	0.84	0.23 – 3.11	0.796
fever [Yes]	2.03	1.01 – 4.06	0.045	2.18	1.08 – 4.37	0.028	1.94	0.95 – 3.94	0.067	0.97	0.41 – 2.30	0.953
ABO pheno [AB]	7.10	1.16 – 57.83	0.039	5.51	1.02 – 32.98	0.047	7.08	1.13 – 58.59	0.042	6.42	0.96 – 56.35	0.062
ABO pheno [B]	0.50	0.13 – 1.59	0.275	0.48	0.12 – 1.52	0.244	0.43	0.11 – 1.41	0.195	0.54	0.13 – 1.79	0.346
ABO pheno [O]	0.50	0.24 – 1.05	0.066	0.43	0.21 – 0.88	0.020	0.53	0.26 – 1.12	0.093	0.50	0.24 – 1.07	0.072
heating [paraffin]	3.94	1.43 – 10.47	0.006	3.51	1.31 – 8.90	0.010	4.20	1.49 – 11.42	0.005	3.56	1.23 – 9.94	0.016
heating [wood]	2.01	0.09 – 18.20	0.570	2.07	0.09 – 20.47	0.561	1.47	0.06 – 15.96	0.767	2.32	0.10 – 24.30	0.516
Combined amend [nonsec Lea+b-]	0.00	0.00 – 4174784215.02	0.984
Combined amend [Sec Lea-b-]	0.29	0.05 – 1.41	0.138
Combined amend [Sec Lea+b-]	0.00	0.00 – 25059605166731082858496.00	0.991
Combined amend [Sec Leb+]	0.23	0.08 – 0.69	0.007
Gender [Male]				1.78	0.91 – 3.56	0.095	1.55	0.78 – 3.15	0.214	1.36	0.67 – 2.80	0.396
refusal to feed [Yes]				1.40	0.69 – 2.81	0.346	1.36	0.66 – 2.78	0.405	1.33	0.64 – 2.75	0.435
Le ab pheno [a-b+]							0.47	0.16 – 1.34	0.152	0.44	0.15 – 1.27	0.127
Le ab pheno [a+b-]							0.00	0.00 – 10299545.78	0.983	0.00	0.00 – 6851241.20	0.983
Le ab pheno [a+b+]							0.36	0.15 – 0.91	0.027	0.37	0.15 – 0.95	0.035
Site [OPHC]										0.30	0.13 – 0.71	0.006
Observations	333			331			331			331
R2 Tjur	0.157			0.103			0.150			0.182

Measures of Association by epiR:

Fever

##          Outcome
## Predictor Positive Negative
##      Yes        20       67
##      None       27      220
## 
##              Outcome +    Outcome -      Total        Inc risk *        Odds
## Exposed +           20           67         87              23.0       0.299
## Exposed -           27          220        247              10.9       0.123
## Total               47          287        334              14.1       0.164
## 
## Point estimates and 95% CIs:
## -------------------------------------------------------------------
## Inc risk ratio                                 2.10 (1.25, 3.55)
## Odds ratio                                     2.43 (1.28, 4.61)
## Attrib risk in the exposed *                   12.06 (2.40, 21.72)
## Attrib fraction in the exposed (%)            52.45 (19.69, 71.84)
## Attrib risk in the population *                3.14 (-2.25, 8.53)
## Attrib fraction in the population (%)         22.32 (2.56, 38.07)
## -------------------------------------------------------------------
## Uncorrected chi2 test that OR = 1: chi2(1) = 7.735 Pr>chi2 = 0.005
## Fisher exact test that OR = 1: Pr>chi2 = 0.007
##  Wald confidence limits
##  CI: confidence interval
##  * Outcomes per 100 population units 
## 
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  dat
## X-squared = 6.7704, df = 1, p-value = 0.009268

By cough

##          Outcome
## Predictor Positive Negative
##      Yes        21       83
##      None       26      202
## 
##              Outcome +    Outcome -      Total        Inc risk *        Odds
## Exposed +           21           83        104              20.2       0.253
## Exposed -           26          202        228              11.4       0.129
## Total               47          285        332              14.2       0.165
## 
## Point estimates and 95% CIs:
## -------------------------------------------------------------------
## Inc risk ratio                                 1.77 (1.05, 3.00)
## Odds ratio                                     1.97 (1.05, 3.69)
## Attrib risk in the exposed *                   8.79 (0.04, 17.54)
## Attrib fraction in the exposed (%)            43.53 (4.42, 66.63)
## Attrib risk in the population *                2.75 (-2.82, 8.33)
## Attrib fraction in the population (%)         19.45 (-2.11, 36.45)
## -------------------------------------------------------------------
## Uncorrected chi2 test that OR = 1: chi2(1) = 4.540 Pr>chi2 = 0.033
## Fisher exact test that OR = 1: Pr>chi2 = 0.041
##  Wald confidence limits
##  CI: confidence interval
##  * Outcomes per 100 population units 
## 
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  dat
## X-squared = 3.8452, df = 1, p-value = 0.04989

By Site

##          Outcome
## Predictor Positive Negative
##     OPHC        20      221
##     DGMAH       29       72
## 
##              Outcome +    Outcome -      Total        Inc risk *        Odds
## Exposed +           20          221        241               8.3      0.0905
## Exposed -           29           72        101              28.7      0.4028
## Total               49          293        342              14.3      0.1672
## 
## Point estimates and 95% CIs:
## -------------------------------------------------------------------
## Inc risk ratio                                 0.29 (0.17, 0.49)
## Odds ratio                                     0.22 (0.12, 0.42)
## Attrib risk in the exposed *                   -20.41 (-29.90, -10.93)
## Attrib fraction in the exposed (%)            -245.99 (-482.06, -105.67)
## Attrib risk in the population *                -14.39 (-23.96, -4.81)
## Attrib fraction in the population (%)         -100.40 (-150.52, -60.31)
## -------------------------------------------------------------------
## Uncorrected chi2 test that OR = 1: chi2(1) = 24.164 Pr>chi2 = <0.001
## Fisher exact test that OR = 1: Pr>chi2 = <0.001
##  Wald confidence limits
##  CI: confidence interval
##  * Outcomes per 100 population units 
## 
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  dat
## X-squared = 22.529, df = 1, p-value = 2.07e-06

By refusal to feed

##          Outcome
## Predictor Positive Negative
##      Yes        20       88
##      None       27      197
## 
##              Outcome +    Outcome -      Total        Inc risk *        Odds
## Exposed +           20           88        108              18.5       0.227
## Exposed -           27          197        224              12.1       0.137
## Total               47          285        332              14.2       0.165
## 
## Point estimates and 95% CIs:
## -------------------------------------------------------------------
## Inc risk ratio                                 1.54 (0.90, 2.61)
## Odds ratio                                     1.66 (0.88, 3.12)
## Attrib risk in the exposed *                   6.46 (-2.01, 14.94)
## Attrib fraction in the exposed (%)            34.91 (-10.66, 61.71)
## Attrib risk in the population *                2.10 (-3.58, 7.78)
## Attrib fraction in the population (%)         14.86 (-6.84, 32.15)
## -------------------------------------------------------------------
## Uncorrected chi2 test that OR = 1: chi2(1) = 2.506 Pr>chi2 = 0.113
## Fisher exact test that OR = 1: Pr>chi2 = 0.131
##  Wald confidence limits
##  CI: confidence interval
##  * Outcomes per 100 population units 
## 
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  dat
## X-squared = 2.0023, df = 1, p-value = 0.1571

By nursery

##          Outcome
## Predictor Positive Negative
##      Yes         9       32
##      None       37      257
## 
##              Outcome +    Outcome -      Total        Inc risk *        Odds
## Exposed +            9           32         41              22.0       0.281
## Exposed -           37          257        294              12.6       0.144
## Total               46          289        335              13.7       0.159
## 
## Point estimates and 95% CIs:
## -------------------------------------------------------------------
## Inc risk ratio                                 1.74 (0.91, 3.34)
## Odds ratio                                     1.95 (0.86, 4.42)
## Attrib risk in the exposed *                   9.37 (-3.86, 22.59)
## Attrib fraction in the exposed (%)            42.67 (-9.94, 70.10)
## Attrib risk in the population *                1.15 (-4.14, 6.43)
## Attrib fraction in the population (%)         8.35 (-4.34, 19.49)
## -------------------------------------------------------------------
## Uncorrected chi2 test that OR = 1: chi2(1) = 2.665 Pr>chi2 = 0.103
## Fisher exact test that OR = 1: Pr>chi2 = 0.142
##  Wald confidence limits
##  CI: confidence interval
##  * Outcomes per 100 population units 
## 
## 
##  Pearson's Chi-squared test with Yates' continuity correction
## 
## data:  dat
## X-squared = 1.9327, df = 1, p-value = 0.1645