Managing Mental Health Disorders Resulting from Trauma through Yoga: A Review

Table 1

Summary of the twelve studies reviewed

Sl. Number	Category of trauma	Sample: (1) Age, (2) Gender, (3) 𝑛	Study design	Assessment tools and their reliability/validity	Statistics	Effect sizes	Limitations
(1)	Natural disaster (tsunami); (Location: the Andaman islands) [14]	(1) 28–50 years; (2) both genders; (3) 𝑛=47	Single group longitudinal design with before, after (7 day yoga program)	(i) Linear analog scales for fear, anxiety, disturbed sleep, and sadness (the reliability and validity has not been established),; (ii) Heart rate variability from the electrocardiogram),; (iii) breath rate and,; (iv) skin resistance; (ii), (iii), (iv) were recorded with a polygraph and were standardized	Paired t-tests	Not reported	(i) Absence of a control group, (ii) use of scales whose reliability and validity had not been established, and (iii) short duration of the intervention.
(2)	Natural disaster (tsunami) (Location: South-east coast of India) [15]	(1) 18–65 years; (2) Both genders; (3) 𝑛=183	Allocation (non random) to 3 groups (a) yoga breath intervention, (b) yoga breath intervention followed by trauma reduction exposure technique, and (c) a wait list control.; (d) Longitudinal assessments before, after (24 weeks) as well as intermittently	(i) The PTSD check list-17 (PCL-17).; (ii) Beck depression inventory (BDI-21) reliability and validity are established	Three factor ANOVA, with post hoc analysis using Newman Keuls tests	Not reported	Allocation to the 3 groups was not random
(3)	Natural disaster (Hurricane Katrina) (Location: New Orleans, U.S.A.) [16]	(1) 31–67 years; (2) Both genders; (3) 𝑛=20	(i) Single group; (ii) longitudinal assessments before, after (8 weeks of intervention)	(i) PTSD checklist-Specific version (PCL-S); (ii) Centre for (CES-D); (iii) State subscale of STAI (STAI-S); (iv) a follow up questionnaire (i), (ii), and (iii) are reliable and valid	Intention-to-treat regression analysis and one sample t-tests	PCL-S (𝑑=.38) CES-D (𝑑=.12) STATE-S (𝑑=.45)	(i) Small sample size and (ii) absence of a control group
(4)	Natural disaster (floods) (Location: Bihar, India) [17]	(1) 15–85 years; (2) Both genders; (3) 𝑛=1289	Cross sectional single group study evaluating risk for PTSD and depression in different age groups	Screening questionnaire for disaster mental health (SQD) with known reliability and validity	Two factor ANOVAs followed by post-hoc analyses	Not reported	Confounding variables which could influence susceptibility other than age and gender were not reported
(5)	Natural disaster (floods) (Location: Bihar, India) [18]	(1) 20–40 years; (2) Males; (3) 𝑛=22	(i) Randomized controlled study; (ii) Longitudinal with before, after assessment after 7 days of intervention	(i) Linear analog scales to assess fear, anxiety, disturbed sleep and sadness, (ii) heart rate variability based on electrocardiogram, and (iii) breath rate.; reliability and validity of linear analog scales are not established and (ii) and (iii) were recorded on a polygraph with standardized methods	Paired t-tests	Not reported	(i) Small sample size and; (ii) use of analog scales which are not valid
(6)	Exposure to combat and terrorism (Location: Kosovo) [19]	(1) 12 to 19 years; (2) Both genders; (3) 𝑛=139	(i) Single group.; (ii) Longitudinal with before, after (6 weeks) assessments	(i) PTSD reaction Index; With known reliability and validity	Paired t-tests RMANOVA for one group and the subset of another group	Effect size for group 1 (𝑑=0.5) and group II and II (𝑑=0.8)	Absence of a control group
(7)	Exposure to combat and terrorism (Location: Kosovo) [20]	(1) 12–19 years; (2) Both genders; (3) 𝑛=82	(i) Randomized control study; (ii) Longitudinal before, after (3 months)	Harvard trauma questionnaire (valid and reliable)	Repeated measures ANOVA	Not obtained possibly reported	NIL
(8)	Exposure to combat and terrorism (Location: Israel) [21]	(1) 8–12 years; (2) Both genders; (3) 𝑛=122	(i) Single group; (ii) Longitudinal before, after (4 months)	(i) WHO well being index; (ii) A satisfaction questionnaire. (iii) Connor abbreviated questionnaire (rated by the teacher).; (i) and (iii) are reliable and valid	(i) Wilcoxon paired signed ranked test; (ii) Kruskall Wallis test	Not reported	Absence of a control group
(9)	Exposure to combat and terrorism (Location: Bali) [22]	(1) 6–12 years; (2) Both genders; (3) 𝑛=226	(i) Single blind, randomized control design.; (ii) Longitudinal before, after (2 year follow up)	Not obtained	Not obtained	Not obtained	Numbers in the 2 groups were unequal (i.e., 48 in the SHAT group compared to 178 in the control) this is not usual in a randomized control design
(10)	Exposure to combat and terrorism as well as tsunami (Location: Sri Lanka) [23]	(1) 8–14 years; (2) Both genders; (3) 𝑛=71	Randomized to 2 interventions, before, after (1, 6 months)	(i) (UCLA PTSD index for DSM-IV.; (ii) 5 item scale for satisfaction.; (iii) 5 item scale related to the tsunami.; (i) and (ii) had established reliability and validity	Repeated measures ANOVA and chi-square tests	Effect Sizes (i) KIDNET 1.76 (one month post test), 1.96 (6 months post test) (ii) MED-RELAX 1.83 (one month post test), 2.20 (6 month post test)	(i) Absence of a no intervention control group.; (ii) The third questionnaire (related to the tsunami) was not standardized
(11)	Interpersonal violence (Location: Baltimore City, U.S.A.) [24]	(1) 9–11 years; (2) Both genders; (3) 𝑛=97	Randomized control before, after (12 weeks) assessments	(i) Response to stress questionnaire. (ii) The short mood and feelings questionnaire: Child version for depressive symptoms.; (iii) Emotional profile inventory; (iv) “People in my life self report measure”	(i) ANOVA for continuous variables.; (ii) chi-square tests for categorical variables	Effect sizes were calculated for Emotional profile positive affect (𝑑=.04), Emotional profile negative affect (𝑑=.13), SMFQ depression scores (𝑑=.13), PIML Trust in Friends (𝑑=.40), PIML Communications with Friends (𝑑=.17), PIML Teacher Affiliation (𝑑=.09), PIML Dissatisfaction with Teacher (𝑑=.08) and RSQ (𝑑=.83)	Recruitment may have involved highly motivated students with enthusiastic parents (the self-report measures may have been influenced by bias
(12)	Youth in a correctional facility (Location: Bangalore, India) [26]	(1) 12–16 years; (2) Females; (3) 𝑛=28	Matched pair allocation to intervention and control groups (pairs were matched for age and duration of stay in the community home), before, after (6 months) assessments	(i) Heart rate from the electrocardiogram.; (ii) Breath rate from a respirogram.; (iii) Skin resistance. The 3 variables were recorded with a polygraph using standard methods	Wilcoxon paired—sample tests	Not reported	(i) The small sample; size. (ii) absence of tests for self reported stress, and (ii) simplistic variables used are inadequate to assess physiological stress