Skip to main content

Table 3 Results of the multilevel multinomial regression of consistent use of bed nets on selected sociodemographic, ideational, household and community variables, Nigeria, 2015

From: Factors associated with caregivers’ consistency of use of bed nets in Nigeria: a multilevel multinomial analysis of survey data

Predictor

Using every night vs. using rarely/never

Using most or some nights vs. using rarely/never

Relative risk ratio

95% confidence interval

Relative risk ratio

95% confidence interval

Sociodemographic and media exposure variables

 Age in years

0.971***

0.957, 0.986

0.987

0.973, 1.002

 Female gender (RC = male)

1.810***

1.385, 2.367

1.488**

1.156, 1.916

 Education (RC = none)

  Primary

0.740

0.520, 1.054

0.690*

0.460, 0.972

  Secondary and higher

0.670*

0.471, 0.953

0.638**

0.453, 0.898

 Regularly listened to the radio (RC = no)

0.878

0.669, 1.153

0.873

0.671, 1.135

 Regularly watched the television (RC = no)

0.562***

0.416, 0.759

0.757

0.569, 1.006

 Religion (RC = christian)

  Muslim

0.442***

0.274, 0.711

0.741

0.473, 1.160

  Others

0.290*

0.095, 0.886

0.541

0.182, 1.609

 Exposed to malaria prevention message on the media in last 12 months (RC = not exposed)

1.665***

1.279, 2.168

1.939***

1.509, 2.492

Ideational characteristics

 Perceived severity of malaria (RC = did not perceive)

0.675**

0.529, 0.862

0.888

0.706, 1.117

 Perceived susceptibility to malaria (RC = did not perceive)

0.936

0.748, 1.171

1.009

0.815, 1.249

 Knew that malaria is caused by mosquito bite (RC = did not know)

1.342

0.950, 1.896

1.652**

1.182, 2.309

 Reported perceived self-efficacy to prevent malaria (RC = did not report)

1.283

0.971, 1.695

1.048

0.807, 1.360

 Reported perceived self-efficacy to detect severe malaria (RC = did not report)

1.040

0.792, 1.366

1.050

0.811, 1.360

 Reported perceived response-efficacy of bed nets (RC = did not report)

1.468***

1.163, 1.854

0.897

0.719, 1.119

 Reported perceived self-efficacy to use nets (RC = did not report)

1.802***

1.397, 2.324

1.340*

1.052, 1.706

 Discussed malaria with others (RC = did not discuss)

1.450*

1.047,2.009

2.477***

1.775, 3.458

 Discussed nets with others (RC = did not discuss)

1.382*

1.069, 1.787

1.283*

1.003, 1.640

 Higher score for positive attitudes towards net use (RC = lower score)

1.547***

1.230, 1.944

1.551***

1.245, 1.929

 Perceived net use to be the norm in the community (RC = did not perceive)

2.079***

1.657, 2.609

1.999***

1.608, 2.483

 Knew where to buy nets in community

2.514***

1.841, 3.434

2.080***

1.528, 2.831

 Willing to pay for nets (RC = not willing to pay)

1.655***

1.299, 2.109

1.824***

1.443, 2.307

Household/community variables

 Household size

0.862***

0.822, 0.904

0.904***

0.863, 0.946

 Number of nets in household

1.867***

1.656, 2.104

1.491***

1.327, 1.675

 Urban residence (RC = rural)

0.854

0.590, 1.237

0.772

0.539, 1.105

 State of residence (RC = Akwa Ibom)

  Kebbi

17.328***

7.924, 37.893

4.107***

1.934, 8.721

  Nasarawa

3.215***

1.861, 5.553

4.010***

2.379, 6.760

 Household wealth quintile (RC = lowest)

  Second

0.767

0.484, 1.213

0.707

0.441, 1.132

  Middle

0.591*

0.362, 0.964

0.720

0.439, 1.180

  Fourth

0.545*

0.324, 0.918

0.832

0.495, 1.401

  Highest

0.600

0.343, 1.049

0.979

0.563, 1.702

 Community advantage index (RC = low)

  Medium

1.225

0.720, 2.084

0.944

0.563, 1.582

  High

1.513

0.788, 2.905

1.444

0.770, 2.708

Random effects

 Cluster random effects/(SE)a

0.542*** (0.128)

0.480*** (0.121)

  1. RC reference category
  2. p < 0.1; * p < 0.05; ** p < 0.01; *** p < 0.001
  3. aSignificance of community-level random effects is assessed using log-likelihood ratio tests (comparing models with and without random effects) with one-sided p-values because the null value is on the border of the parameter space