Els van Meijel

89 Parental posttraumatic stress | Chapter 5 Table 5.1 The univariable and multivariable associations between parent and child factors and the severity of parental PTSS (IES-R scores) at T1 (n = 135) Univariable model Multivariable model Beta a 95% confidence interval p -value Beta a 95% confidence interval p -value Parent characteristics Prior trauma history 1.35 0.91 - 2.00 0.13 - - - Acute stress, irritable 2.07 1.44 - 2.98 0.000 1.60 1.11 - 2.29 0.01 Acute stress, physical 2.26 1.52 - 3.35 0.000 2.10 1.40 - 3.06 0.000 Witnessing accident 0.62 0.34 - 1.15 0.13 - - - Child characteristics Hospital > 1 day 1.69 1.16 - 2.48 0.007 1.67 1.17 - 2.38 0.005 Severe pain 0.99 0.66 - 1.47 0.94 - - - T1 = 3 months after the accident. PTSS = clinically significant posttraumatic stress symptoms. IES-R = Impact of Event Scale Revised. a Due to the skewed distribution of the parental PTSS data, we performed linear regression analysis on log10 transformed data. To aid interpretation of the results, we present back transformed regression parameter estimates and corresponding upper and lower limits of confidence intervals. Table 5.2 The univariable and multivariable associations between parent and child factors and the severity of parental PTSS (IES-R scores) at T2 (n = 69) Univariable model Multivariable model Beta a 95% confidence interval p -value Beta a 95% confidence interval p -value Parent characteristics Prior trauma history 0.93 0.54 - 1.61 0.80 - - - Acute stress, irritable 1.78 1.05 - 3.01 0.03 1.68 1.00 - 2.81 0.048 Acute stress, physical b 1.71 0.95 - 3.07 0.07 - - - Witnessing accident 0.59 0.28 - 1.27 0.17 - - - Child characteristics Hospital > 1 day 1.49 0.86 - 2.57 0.15 - - - Severe pain 0.77 0.45 - 1.34 0.36 - - - Permanent physical impairment 2.51 1.44 - 4.40 0.002 2.16 1.23 - 3.81 0.008 T2 = 2–4 years after the accident. PTSS = clinically significant posttraumatic stress symptoms. IES-R = Impact of Event Scale Revised. a Due to the skewed distribution of the parental PTSS data, we performed linear regression analysis on log10 transformed data. To aid interpretation of the results, we present back transformed regression parameter estimates and corresponding upper and lower limits of confidence intervals. b Did not positively contribute to the multivariable level and was therefore left out.