Skip to main content

Table 2 Associations between PHQ-9 scores and annual healthcare utilization and OOP expenditure among adults in Chitwan district, Nepal, 2013

From: Healthcare utilization and out-of-pocket expenditures associated with depression in adults: a cross-sectional analysis in Nepal

 Utilization (Number of visits)OOP Expenditure (US dollars)
IRR (95% CI)P valueβ (95% CI)P value
Inpatient Admissions N = 479Univariate1.08 (0.94–1.25)0.2803.63 (−0.99–8.26)0.124
Multivariate1.07 (0.94–1.22)0.3144.00 (−1.84–9.84)0.179
Age1.00 (0.94–1.22)0.9814.37 (−3.63–13.49)0.284
Female0.70 (0.32–1.54)0.378−83.84 (−211.31–43.62)0.197
Relative wealth index0.96 (0.71–1.30)0.79426.92 (−19.08–72.92)0.251
Outpatient Consultations N = 479Univariate1.14 (1.06–1.23)< 0.00015.52 (2.36–8.67)0.001
Multivariate1.14 (1.06–1.23)< 0.00015.36 (2.26–8.47)0.001
Age1.00 (0.98–1.01)0.5490.62 (−0.22–1.45)0.150
Female1.07 (0.66–1.71)0.78617.26 (−7.58–42.09)0.173
Relative wealth index1.00 (0.66–1.71)0.9866.49 (0.18–13.11))0.044
Total Healthcare N = 479Univariate1.14 (1.07–1.22)< 0.00019.15 (3.12–15.18)0.003
Multivariate1.14 (1.07, 1.22)< 0.00019.37 (2.15–16.58)0.011
Age1.00 (0.98–1.01)0.5634.99 (−3.52–13.49)0.251
Female1.05 (0.67–1.65)0.817−66.59 (−204.87–71.70)0.345
Relative wealth index1.00 (0.88–1.13)0.98233.57 (−16.17–83.31)0.186
  1. Healthcare utilization and OOP Expenditure were estimated using negative binomial regression and bootstrapped linear regression to model incidence rate ratios (IRR), and β coefficients respectively along with and 95% confidence intervals
  2. Both IRR and β coefficients were adjusted by survey-sampling weights and multivariate analyses were further adjusted for sex, gender and relative wealth index
  3. Estimates reflect changes associated with a one unit increase in PHQ-9 score after adjustment for age, gender and relative wealth index