Effectiveness of psycho-education on depression , hopelessness , suicidality , anxiety and substance use among basic diploma students at Kenya Medical Training College

41 SAJP June 2013 Vol. 19 No. 2 Substantial evidence supports the use of psychological therapies – particularly cognitive behavioural therapy (CBT) – through various methods of psycho-education in the prevention or treatment of mild to moderate depression, anxiety, moderate substance abuse and consequently, hopelessness and suicidality.[1] Researchers have devised several components of psycho-education geared towards symptom recognition and stress-coping skills, among other things, which mental health professionals have found to be effective in preventing and reducing the symptoms of these conditions. Psycho-education has also been found to promote self-referral to healthcare providers, by enhancing symptom recognition and reducing stigma.[2,3] An individual’s ability to understand the effects of psycho-stressors as risk factors can help reduce these conditions’ prevalence among patients or the general population, and individuals equipped with appropriate stress-coping strategies can deal with possible psycho-stressors before they become chronic.[2] The most common mental disorders diagnosed in primary-care settings are depression, anxiety and substance-related disorders, which can be present alone or comorbid with another mental disorder. [4] Hopelessness and suicidality are symptoms in a number of mental disorders, such as depression, schizophrenia, anxiety or substance abuse.[5-7] Common in the general population, these debilitating conditions affect general well-being, functioning, productivity and quality of life if left unmanaged, making it hard to do everyday tasks completely, competently and efficiently.[8-11] These conditions also have a psychosocial component that is closely related to negative social relationships, environmental challenges or an individual’s inadequate ability to cope effectively with stress. These challenges have been found to be predominant among medical students.[12-15] Because the diagnosis of depression, anxiety, hopelessness or suicidality by general practitioners is often poor, the proportion of sufferers who receive treatment is low. Among students entering the health professions, these conditions not only affect their lives negatively but may also have repercussions for patient care in the long run.[10,14] Rosenthal and Okie[5] noted that medical students are more prone to depression than their nonmedical peers. This could be because medical students constantly encounter very sick patients or even deaths among their patients. Psycho-education in recognising the symptoms of these conditions and understanding their causes, as well as skills for coping with stress, may facilitate their prevention, and possibly control of their occurrence. In a study involving 1stand 2nd-year medical students in a Californian University, aimed at investigating the use of mental health services and its barriers, Givens et al. found that about a quarter of Effectiveness of psycho-education on depression, hopelessness, suicidality, anxiety and substance use among basic diploma students at Kenya Medical Training College


Data collection instruments
The research instruments were 6 self-administered questionnaires: • A researcher-designed questionnaire for socio-demographic characteristics (SDQ): gender, age, year of study, marital status, place of residence while studying, religion and campus location.• The Beck's Depression Inventory (BDI), [19] a 21-item self-report inventory which measures the severity of depression in a general population.This widely used instrument discriminates subtypes of depression, differentiates depression from non-depressed patients has a high coefficient alpha of 0.93 (p=0.001) for college students, and a validity and reliability of 90%.The BDI cut-off points used in this study were: 0 -9 for minimal depression, 10 -18 for mild depression, 19 -29 for moderate depression and 30 -63 for severe depression.• The Beck's Hopelessness Scale (BHS) [20,21] is designed to reflect the respondent's negative expectancies in various psychopathological conditions.This scale has been administered in diverse samples of patients and among the general population to assess its psychometric properties, and has been found to have a high degree of internal consistency.It also shows a relatively high correlation with the clinical ratings of hopelessness and other self-administered measures of hopelessness.Scores provide a measure of the severity of reported hopelessness: 0 -3 for minimal, 4 -8 for mild, 9 -14 for moderate and 15 -20 for severe.• The Beck's Suicide Ideation Scale (BSIS) [21] measures the severity of risk of suicidality and constitutes a 21-item self-report questionnaire.
The instrument identifies those respondents with passive or active suicidal ideas, those with low, moderate or severe suicidal plans and those who have attempted suicide. [16] The Beck's Anxiety Inventory (BAI) [22] is a 21-question instrument designed to measure the severity of anxiety in a general population.It shows proven high interval consistency and test/re-test reliability over 1 week.In the general population, respondents must score ≥36 to be considered to have anxiety.The BDI and BAI have been validated against the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) (DSM-IV) respective diagnostic criteria in the western countries, where they have been used extensively for similar and other relevant surveys. [19,22]The cut-off points for the BAI used for this study were: 0 -7 for minimal, 8 -15 for mild, 16 -25 for moderate and 26 -63 for severe anxiety.• The World Health Organization (WHO) alcohol, smoking and substance involvement screening test (ASSIST). [23]The National Institute of Drug Abuse (NIDA) has adopted the WHO ASSIST version 6, used among the general population, which investigates the risk of psychoactive substance use/abuse in individuals who use a number of substances and exhibit substance use of a minimal, mild, moderate or severe degree. [24]The scores of all the other substances -including alcohol, tobacco, cannabis, cocaine, amphetamines, inhalants, sedatives, hallucinogens, opioids and others -are given. [24]he cut-off points for ASSIST were, (i) for alchohol: lower risk 0 -10, moderate risk 11 -26, high risk ≥27; (ii) for all other substances: lower risk 0 -3, moderate risk 4 -26, high risk ≥27.All 6 instruments were self-administered and took an average of 1.5 hours to complete.

Intervention
All participants gave informed consent and confidentiality was maintained throughout the study.All respondents in both groups were advised that if they recognised the symptoms/conditions being investigated, they were free to make self-referrals.All respondents were given information on resources available for help.
In the experimental group, baseline characteristics were obtained.This was followed by 4 psycho-education sessions.These 2-hour sessions comprised lectures, simulations, group discussions and role-plays.They incorporated predisposing/precipitating causes of depression, hopelessness, suicidality, anxiety and substance abuse, as well as symptom recognition and stress-coping skills/ strategies.Reassessment was carried out 3 months after the initial intervention to investigate its effects, followed immediately by another 8 hours of psycho-education in 2-hour sessions.A final reassessment was carried out 3 months after the second assessment (6 months after the baseline assessment).All participants were free to contact the principal investigator if they felt that they had a psychological problem or related to the symptoms featured in the questionnaires.
In the control group, no interventions were provided.(However, respondents were free to self-refer to a healthcare provider.Reassessment was done at 3 and 6 months.Participants were provided with a 24-hour helpline in case they developed suicidal tendencies or needed help of any kind.

Psycho-education intervention and reassessments
The initial 4 sessions of psycho-education given to the experimental group included a 2-hour session on definitions of terms, causes of the conditions and symptom recognition.Another 2 sessions involved theoretical lectures and simulations on stress-coping strategies/skills.These included scheduling/time management, communication skills, decision-making techniques, problem solving, assertiveness training, improving self esteem, sleep hygiene, breathing techniques, controlled breathing/de-arousal, anger-management techniques, relaxation exercises, progressive muscle relaxation, general exercise activities and adherence training.A clinical psychologist supervised small group discussions.
After 3 months, assessment was followed immediately by a repeat of the same process of psycho-education, in addition to addressing the challenges that the respondents had experienced after the first round.
Each psycho-education programme in all the departments in the experimental group lasted between 1.5 months and 2 months.

Data management
The collected data were double-entered by 2 separate groups of data-entry clerks, and cleaned and analysed using SPSS (version 16), utilising descriptive and inferential statistics.Results were presented in the form of tables and narratives.

Results
The socio-demographic characteristics of the study sample across the three assessments are summarised in Table 1.
At assessment 1 (baseline), there was no difference in the prevalence of depression between the control and the experimental groups: minimal 20.6% v. 20.8%; mild 12.6% v. 13.1%; moderate 18.4% v. 20.2%; and severe 48.5% v. 45.8% respectively.During assessment 2, 3 months after the first psycho-education session, there was no difference in the prevalence of depression.
At assessment 3, 3 months after assessment 2, there was a reduction in the prevalence of severe and moderate depression in the experimental group compared with the control group: moderate 5.2% v. 10.4%; and severe 3.4% v. 5.0%, respectively.The reduction in depression severity after psycho-education was significant, (p<0.0001) after 6 months.
There was no difference in the prevalence of hopelessness at assessment 1 between the groups.At assessment 2, there was a significant difference (p<0.0001):moderate depression 3.2% v. 8.6% in the experimental and control groups, respectively.The difference was also significant at assessment 3 (p=0.001).
There was no difference between the groups with regards to suicidality ideas, passive or active, at all the assessments (p=0.830,p=0.288, p=0.059, chronologically).However, there was a difference in the prevalence reduction rate which was not significant.At assessment 1, passive suicide ideation was 98.3% v. 99.0%, at assessment 2 was 99.7% v. 99.8% and at assessment 3 was 99.2% v. 98.4% in the experimental and control groups, respectively.Respectively, the prevalence of active suicidality ideas was 1.7% v. 1.0% at assessment 1, 0.3% v. 0.2% at assessment 2 and 0.8% v. 1.6% at assessment 3.
At assessment 1, there was a significant difference between the groups in relation to suicidal plan (p=0.005) and suicidal attempts (p<0.0001)(Table 2).However, after 6 months of psycho-education there was no difference between the groups (p=0.146 and p=0.615, respectively).
At assessment 1 and 2, there was no difference in the percentage prevalence of anxiety in the 2 groups (Table 2).However, after months of psycho-education, there was a difference in reduction rates of anxiety prevalence between the 2 groups.Moderate anxiety was more prevalent in the experimental group at 14.1%, compared with the control (10.5%).Severe anxiety was more prevalent in the control group at 7.8%, compared with the experimental group (at 5.4%).Overall, the difference in the severity of anxiety in the groups was significant after 6 months of psycho-education (p<0.0001)(Table 2).

Risk of alcohol and drug abuse
At baseline, there was a significant difference in the risk for alcohol abuse between the 2 groups (p=0.016), with no difference in the risk of abuse of tobacco, cannabis, cocaine, amphetamine, inhalants, sedatives, opioids and hallucinogens.After 6 months of psychoeducation, there was no significant difference in risk except for cocaine (p=0.034) and cannabis (p=0.026)(Table 3).
Analysis showed that more respondents in the experimental group had self-referred to a mental health facility/professional by the second and third assessment (n= 200 v. n=239, respectively) than in the control group (n=144 v. n=137, respectively).The self-referral means were higher in the experimental group than the control group in the second (0.1787 v. 0.2495, respectively) and third assessments (0.0919 v. 0.0835, respectively).ANOVA testing showed a statistically significant difference between the means in the 2 study groups in assessments 2 and 3 (p=0.001and p<0.0001, respectively).
Analysis of variance was used to test for differences in the means of the prevalence of conditions between the 2 groups in the 3 assessments, to show the effects on outcomes of the intervention used.Analysis of the means of depression, hopelessness, suicidality, anxiety and the risk of substance abuse showed progressive reduction of means across the assessments between the 2 study groups, with a greater reduction among the experimental group in the third assessment (Table 4).In assessment 1, the conditions which showed significant differences between the 2 groups were suicidal attempts, anxiety, tobacco and sedatives.In assessment 2, only hopelessness had a significant difference (Table 5) while in assessment 3, depression, hopelessness, suicidal ideas, suicidal plans, cannabis and cocaine showed significant differences.
There was significant difference between the means of some conditions in each group, with more effectiveness within the experimental group.Even where differences were not significant, they were larger in the experimental group than in the control group, with the highest difference between the first and third assessments.

Population studies
The high return rate was attributed to the fact that, before the data collection period, the researchers had worked to sensitise both the respondents and campus administrators.Previous studies by Ndetei et al. [12,13] found 100% response rates in educational institutions.
There were more male than female respondents, which the researchers infer is due to the imbalances between males and females in the various campuses, since there were higher numbers of males in the experimental group.Most respondents were aged <24 years and single, having been admitted to KMTC after high school.The majority of students aged ≥24 years, and who were married, separated, divorced or widowed, had come for upgrading from a certificate to a diploma.The researchers inferred that the age difference was due to the high number of first year students, who were more often under 24 than 2nd year students.More than half of the respondents in both groups were protestants which meant there was a skew which contributed to differential representation of the religions in the sociodemographic characteristics of the 2 age groups.

Prevalence of conditions across the assessments
At assessment 1, there were generally similar overall prevalence levels of all conditions among the experimental and control groups.It was inferred that the factors precipitating the respondents to develop the conditions may have been similar among the 2 study groups, that  The high levels of severe depression and anxiety among respondents in the first assessment could be explained by the stressors facing these students, such as new lifestyles and cultures, new friends and exposure to alternative ways of thinking.As they progressed in their training and adjusted to new lifestyles and social environments, they were required to develop new coping strategies.Those who could not cope effectively were more susceptible to anxiety or depression symptoms, which may have progressed to full-blown conditions and consequently increased hopelessness and suicidality, or to higher risks of substance abuse. [3,25] similar study investigating the prevalence of depression, anxiety and their associated factors among medical students in Karachi, Pakistan, [26] reported a 70% rate of anxiety and depression among the respondents.Precipitating causes were not exclusively academic but included loss, relationship problems, residence and substance abuse.
In this study, the prevalence of active suicidal ideations found among both groups in the first assessment can be attributed to the comorbidity of hopelessness among the respondents, or other factors which were not investigated.In the experimental group, the low prevalence of suicidal plans v. suicide attempts suggests that those who had attempted suicide may have been influenced by other circumstances than hopelessness, like substance abuse or causes not investigated.It can also be inferred that not all who attempt suicides display prior suicidal ideas or plans.
These results were much lower than the findings of a study of prevalence of suicide ideation among college students, where 43% had experienced some level of suicidal ideations, 14.9% had suicidal plans and 5.5% had made suicide attempts suicide which were attributed to inability to cope with psycho-stressors. [27]he high risk of alcohol abuse in the 2 groups at the first assessment can be explained by alcohol's ready availability and cheapness, making it the substance of choice.The high risk of alcohol abuse was followed closely by the risk of tobacco abuse, amphetamines and sedatives.Furthermore, new lifestyles, experiences, financial challenges, psycho-stressors and academic challenges may precipitate involvement in substance abuse.Although Kenya's legal drinking age is 18 years, alcohol and tobacco are readily available even to underage buyers from small-scale, often illicit producers and sellers. [28] similar study [26] listed reasons for substance use among college students: to relax (62.2%), to relieve stress (60.8%), desire to experiment (41.9%), peer pressure (38.9%), and to cope with problems (38.9%).In another study among Kenyan universities, the prevalence rate of cigarette use was 42.8%, cannabis 2% and cocaine 0.6%. [29]Among those who admitted to using substances, 75.1% were introduced to substance use by a friend and 23.5% were introduced by a relative other than a member of the nuclear family.The majority wanted to relax (62.2%) or relieve stress (60.8%).

Effectiveness of psycho-education interventions
The prevalence of all the studied conditions and the risk of substance abuse had fallen by assessment 2 and was even lower by assessment 3 among both study groups, with higher progressive reduction among the experimental group.This indicates that the psycho-education given to the experimental respondents was effective.The reduction among the control group can be attributed to repeat assessment with the same tools, which may have awakened the respondents to their respective symptoms and inspired them to deal with their conditions.However, the reductions were not consistent.These results are consistent with previous studies. [1,16,30]he higher rate of reduction of all conditions among the experimental respondents across the assessment can be related to the effectiveness of the psycho-education intervention.Experimental respondents were enabled to cope with psycho stressors and the ability to recognise symptoms led them to seek professional help early.The highest effect occurred after the 2nd psycho-education intervention, 3 months after the 1st.Outcomes at 6 months were more positive and consistent across the assessments among the experimental respondents, with more respondents self-referring.
A similar study involving 400 patients diagnosed with depression concluded that, if offered to adults, problem solving treatment was effective in reducing cases and improving subjective function. [31]34][35][36] In the current study, addressing symptom recognition, conditions' predisposing and precipitating causes and comprehensive stresscoping strategies was found effective in cost-effective case reduction. 9]

Study limitations
The study did not investigate whether respondents had suffered depression or anxiety previously, or indentify other possible psychostressors which may have precipitated depression or anxiety.These unknown factors could have contributed to the reduction in the conditions.It is also a limitation that all the participants in the experimental group were recruited from the same site.
Acknowledgements.This work formed part of the requirements of a PhD at the University of Nairobi by SKM.This PhD was supervised by NDM, the lead supervisor, Dr M Mathai (Nairobi University) and Prof. B Sebit (Muhimbili University, Dar Es Salaam, Tanzania).We thank all KMTC student respondents, lecturers and principals who participated in this study; as well as the Director of KMTC, for facilitating the logistics of the study; our professional colleagues J Nga'nga, E Kariuki and V Mutiso for their invaluable assistance during psycho-education; and G Mutevu of Africa Mental Health Foundation for her prompt typographical assistance.
Ethics.The study was approved by the Kenyatta National Hospital/ University of Nairobi ethical and research committee.The Director of the KMTC authorised research on the selected campuses.
general risk factors above specific or peculiar campus factors, and that there were risk factors other than academic stressors which predisposed the respondents to psychological morbidity.

Table 6 . Correlation co-efficient test on mean difference of effect of psycho-education between assessments 1 and 3 within each study group
Significance: *p<0.001,† p<0.05, ‡ p<0.01.