Skip to main content

Table 4 Regression difference-in-differences (DiD) estimates of the impact of CHAs on modern contraceptive use adjusting for the effects of socio-demographic covariates

From: The impact of adding community-based distribution of oral contraceptives and condoms to a cluster randomized primary health care intervention in rural Tanzania

Covariates

Odds Ratio

P > t

[95% Conf. Interval]

DiD (Ref: Baseline)

 Post (Endline)

1.026

0.779

.857

1.229

Village (Ref: Control Village)

 Treatment Village

1.000

0.999

.839

1.191

 DiD Interaction (Post*Treatment)

1.026

0.822

.819

1.286

 Age

1.456

0.000**

1.387

1.528

 Age squared

.994

0.000**

.993

.995

Education (ref: no education)

 Less than secondary education

1.438

0.000**

1.202

1.719

 Secondary education or more

1.480

0.002**

1.154

1.898

Religion (Ref: Christian)

 

 Muslim

1.000

0.992

.879

1.140

 Traditional or other

.072

0.000**

.035

.149

Marital Status (Ref: Married)

 Living with partner

1.154

0.159

.946

1.409

 Divorced

.739

0.015*

.580

.942

 Widowed

.567

0.003**

.388

.828

 Single

.847

0.073

.707

1.016

Wealth Quintile (Ref: Poorest)

 Poor

1.013

0.909

.806

1.274

 Better

1.299

0.004**

1.086

1.553

 Less Poor

.940

0.591

.751

1.177

 Least Poor

1.101

0.422

.871

1.392

 Parity

1.075

0.000**

1.035

1.118

 _cons

.002

0.000**

.001

.003

  1. Summary statistics *p < 0.05 **p < 0.01
  2. Observations: 5469
  3. Wald chi2 (18) 522.36
  4. Prob > chi2 0.000
  5. Pseudo R2: 0.0718