i FACTORS ASSOCIATED WITH INTENTIONAL SELF- HARMING BEHAVIOURS AMONG PATIENTS ADMITTED TO AN EKURHULENI DISTRICT HOSPITAL Dr E MUGISHA Student Number: 300023 A submissible article as research report submitted to the Faculty of Health Sciences, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Medicine in Family Medicine (M. Med in Fam. Med) Supervisor: Prof. S. Moosa, MMed, MBA, PhD Department of Family Medicine and Primary Care, University of Witwatersrand Johannesburg, 2021 ii Declaration I, Dr Elsie MUGISHA, student number 300023, hereby declare that this research is my own unaided work, except where due acknowledgement for assistance received has been made. It is being submitted for the degree of Master of Family Medicine at the University of the Witwatersrand, Johannesburg. It has not been submitted previously for any other degree or examination at this or any other University. Signed: ____________________ (Signature of candidate) Dr E. MUGISHA Date: August 2021 iii Dedication To my Heavenly Father for giving me the strength and courage to complete this research report. To my wonderful husband Matthew Mugisha thank you for your endless love, comfort and counsel. In loving memory of my late grandmother Leah Kibuuka who never attended school but through her sacrifices inspired me to persevere. To my loving parents Prof Paul Kibuuka and Mrs Ruth Kibuuka, thank you, for your unconditional love, support and encouragement To my mother-in-law Dr Stella Mugisha, thank you for the constant care, prayers and reassurance. To my siblings Michelle Kibuuka and John-Paul Kibuuka, thank you for your friendship and thoughtfulness. Acknowledgements Prof H.S Schoeman: Statistical analysis. Prof S Moosa: Supervision, support and guidance for the completion of the research report Dr S Agbo: Supervision during the protocol development and data collection iv FACTORS ASSOCIATED WITH INTENTIONAL SELF-HARMING BEHAVIOURS AMONG PATIENTS ADMITTED TO AN EKURHULENI DISTRICT HOSPITAL Abstract Background: The single most important predictor of death by suicide is having engaged in one or more acts of self-harming behaviours. Annually, 7 582 people in South Africa die from self-harm; with as many as 151 646 suicide attempts. Aim: The aim of this study was to explore the factors associated with intentional self- harming behaviour among patients admitted to a district hospital. Methods: This was a descriptive cross-sectional study of all patients’ medical records admitted for intentional self-harm between June 2016 and June 2018 at Bertha Gxowa Hospital, a district hospital in Ekurhuleni Health District which provides level one health care service to Germiston. A modified version of the World Health Organization (WHO) Practice Manual data collection tool was used. The data recorded on collection sheets was then entered into Microsoft Excel and analysed using SAS|®. Results: One hundred and thirty-six patient records were reviewed. Patients prone to self- harming behaviours were mostly black single females aged 14-24 years who were unemployed but did not use alcohol. The method most commonly used was self-poisoning (93,4%). The commonest reason for attempt was interpersonal conflict (74,3%). Statistically significant factors associated with self-harming behaviours were employment status, alcohol use or not as well as history of attempts Conclusion: Health care workers working in primary and mental healthcare settings should have a higher index of suspicion for young black single female patients especially with a history of unemployment, no alcohol use and first self-harm attempt. Besides being a medical problem, this should be a societal problem. Keywords: Self-harm, Behaviour, Methods, Reason, Sociodemographics, Clinical profile, Associated factors v Table of Contents Declaration ii Dedication iii Acknowledgements iii Abstract iv Background iv Aim: iv Methods: iv Results: iv Conclusion: iv Keywords iv Table of Contents v Nomenclature vii List of figures i List of tables ii Introduction 1 Research methods 2 Study design 2 Setting 3 Study population 3 Data collection 3 Data analysis 4 Ethical considerations 4 Results 4 Discussion 12 Conclusions 14 Acknowledgements 14 Competing interests 15 Author contributions 15 Funding contributions 15 Data availability statement 15 Disclaimer 15 References 15 Appendices 19 Appendix I: Data collection tool 19 Appendix II: Ethical clearance 20 Appendix III: Permissions 22 vi Appendix IV: Turnitin report 25 Appendix V: Proofreading certificate 27 Appendix VII: Research proposal 28 References: 42 vii Nomenclature AIDS: Acquired immunodeficiency syndrome HIV: Human immunodeficiency virus MDD: Major depression disorder NSSI: Non-suicidal self-injury SSI: Suicidal self-injury WHO : World Health Organisation i List of figures Figure 1: Diagnosis by gender 17 Figure 2: History of attempts by gender 17 Figure 3: Psychiatric illness according to gender 19 Figure 4: Medical illness according to gender 20 ii List of tables Table 1: Sociodemographics of patients Table 2: Social habits of patients Table 3.: Methods used and documented reasons for attempt Table 4: Factors associated with self harming behaviour iii FACTORS ASSOCIATED WITH INTENTIONAL SELF-HARMING BEHAVIOURS AMONG PATIENTS ADMITTED TO AN EKURHULENI DISTRICT HOSPITAL An article submitted according to the style guide and requirements of the African Journal of Primary Health Care & Family Medicine (PHCFM). An appropriate title was selected and an application for title change will be submitted in conjunction with the examination process. 1 Introduction Social value Intentional self-harm may be defined as injurious or harmful behaviour that is self-directed and purposely results in injury or the potential for injury to oneself, often as a reaction to feelings of extreme psychological distress or emotional pain1. It includes cutting, burning or hitting oneself; bingeing or starvation, abuse of drugs or alcohol; and overdosing on prescription medications2-4. This risky behaviour can have a fatal outcome5. Intentional self-harming behaviour encompasses parasuicide and attempted suicidal behaviours; terminology that are often used interchangeably in literature1. Parasuicide is referred to as non-suicidal self-injury (NSSI) and attempted suicide is suicidal self-injury (SSI)5 but often in a complex and overlapping relationship14,17. Scientific value The global burden of intentional self-harm has increased by 24% and it was found to be the 18th leading contributor worldwide to disease burden24. These numbers are predicted to increase over the coming decades6. With Africa being the largest and second populated continent in the world, there is scarcity of information on self-harming behaviours. A systematic review reported that data on self-harming behaviours were only available from 11 countries on the African continent6. Self-harming behaviour incidence rates varied from 0.7% in Nigeria to 6.0% in Liberia while South African estimates ranged from 2.9%-3.4%. Reported records indicate that self –harming behaviours were recorded among the top 10 causes of deaths in many countries9. It has been assessed that up to 7 582, people in South Africa die from self-harm; with as many as 151 646 suicide attempts annually24 An international metanalysis of the rates of suicide for psychiatric illnesses revealed an increased risk of suicide among patients with eating disorders. The risk of suicide is seven times higher for those with personality disorder and post-partum psychiatric disorders. The risk of suicide is increased especially during the first year and the lifetime risk of suicide for those individuals who have schizophrenia is 10%. Patients with cancer have twice the suicide rate compared to the general population and it is five times more common in those with epilepsy. Studies have also shown increased rates of suicidal ideation in patients with chronic pain and human immunodeficiency virus (HIV)/ acquired immunodeficiency syndrome 2 (AIDS)5. A study in Limpopo in 2004 revealed that there is a higher risk for suicide attempts amongst individuals who are younger, female and less educated13. This is in agreement with studies done in Africa where more females attempted suicide compared to males5. A study from Limpopo in 2014 suggests that the single most important predictor of death by suicide is having engaged in one or more acts of self-harming behaviours.22 Naidoo and Schlebusch9 showed that a wide variety of methods are used by the people committing suicide, the methods used are dependent on different factors such as, knowledge of lethality, availability and accessibility to the method by the patient. It showed that hanging and shooting are the preferred methods, trailed by self-poisoning with agents such as insecticides, drug overdoses, deliberately inhaling toxic gasses and setting fire to oneself.10 Conceptual framework A study on household factors associated with self-harm in Johannesburg in 2016 revealed potential risk factors or reasons for self- harming behaviours include experiencing judgement or feeling isolated in society, abuse or exposure to violence24. This is in correlation with the international metanalysis showing that the reasons identified included elevated numbers of negative life events, feelings of guilt or shame, feelings of loneliness sexual problems, childhood abuse/trauma, poor self-esteem parents with mental health problems and a family history suicidal behaviour5. Aims and objectives There has been no study within Ekurhuleni, a highly urbanised centre of South Africa. The aim of this study was to explore the factors associated with intentional self-harming behaviour among patients admitted to a district hospital. To fulfil this aim, the following were the objectives of the study: Explore the socio-demographic and clinical profile of patients admitted with intentional self-harming behaviour, as well as determine the methods used, the reasons documented and the associations between all these and intentional self-harming behaviour. Research methods Study design This was a descriptive cross-sectional record review of patients admitted for intentional self- harm between June 2016 and June 2018 at Bertha Gxowa Hospital. 3 Setting The Bertha Gxowa Hospital is the only district hospital in the Ekurhuleni Health District in Gauteng Province. The hospital has 230 approved beds. It provides level one health care service to Germiston - a town with a population of 394,2533 and surrounding communities. Ekurhuleni Health District is one of the five health districts in Gauteng Province. There is currently no documented data with regards to intentional self-harming behaviours within the local area. Study population The study population included all patients who were admitted with intentional self-harming behaviour to the Bertha Gxowa Hospital during June 2016 and June 2018. The details of patients were identified from the hospital admission record archives with the assistance of the hospital records clerk. A total of 207 patients admitted to the adult medical ward of the hospital were identified with the diagnostic label of parasuicide or attempted suicide. However, only 136 patient’s folders were available in the in-patient filing cabinets of the hospital and included in the study. The data of 136 patient files from the admissions register were examined to capture socio-demographics attributes, history of attempts, social habits, clinical profile, methods used and the documented reason for the attempt for each patient. The patient folders with inadequate data and those with completely illegible records were excluded from the study as decided by the researcher and reviewed by the supervisor to avoid bias. Data collection Data collection was done using a modified data collection tool from the World Health Organization (WHO) Practice Manual, a tool used for establishing and maintaining surveillance systems for suicide attempts and self-harm6. The modified tool collected the following data sets: diagnosis, socio-demographic aspects, clinical profile, self-harming methods, and documented reasons. The researcher recorded data on data collection sheets. Confidentiality was maintained with a study number allocated to each patient record in a separate coding sheet for the patients file number. The data was then entered into Microsoft Excel software, saved, password protected and used for analysis. 4 Data analysis Descriptive and inferential statistical analysis was done using Statistical Analysis Software®. Fisher’s exact test was utilised to perform associations between intentional self harming behaviours among patients admitted to an Ekurhuleni district hospital. A p-value of < 0.05 was considered to be statistically significant. Ethical considerations Permission to perform the research was granted by the management at the Bertha Gxowa Hospital and Ekurhuleni Health District (GP_201903_028). Ethical clearance was given from the Health Research and Ethics Committee of the University of the Witwatersrand, Johannesburg (M190104). Confidentiality was preserved by coding each patient record with a special number. Results Results are presented as socio-demographics attributes, history of attempts, social habits, clinical profile, methods used and the documented reason for the attempt for each patient. Sociodemographic attributes of patients Out of 136 patient records, 50.7% of the study sample were female while 49.3% were male. Black patients comprised the majority (75%). This was trailed by white, coloured and Indian patients. In terms of gender and race, black women (79,7%) and black men (70,2%) dominated. With regards to age and gender - a majority of patients were women in the age range of 14-24 years (63,9%). Majority (93,4%) of the patients were single, with 5.2% of the patients married, and the rest widowed or divorced. Furthermore, only 20,6% of the patients were employed. Data on education and religion was incomplete and had to be left out of the study. Detailed socio-demographic data is laid out in Table 1. 5 Table 1: Socio-demographic data of patients Men n=67(%) Women n=69(%) Total n=136(%) Age groups n (%) 14–24 25 (37,3) 44 (63,9) 69 (50,7) 25–34 23 (34,3) 15 (21,7) 38 (27,9) 35–44 14 (20,9) 5 (7,3) 19 (14,0) >45 5 (7,5) 5 (7,3) 10 (7,4) Mean 30 25.6 27.7 Median 27 24 24 Mode 22 24 24 Race Black 47 (70,2) 55 (79,7) 102 (75,0) White 18 (26,8) 10 (14,5) 28 (20,6) Coloured 2 (3,0) 3 (4,4) 5 (3,7) Indian 0 (0,0) 1 (1,4) 1 (0,7) Marital status Single 63 (94,0) 64 (92,7) 127 (93,4) Married 4 (6,0) 3 (4,3) 7 (5,2) Widowed 0 (0,0) 1 (1.5) 1 (0.7) Divorced 0 (0,0) 1 (1,5) 1 (0,7) Employment Employed 20 (29,8) 8 (11,6) 28 (20.6) Unemployed 47 (70,2) 61 (88,4) 108 (79.4) Documented social habits of patient sample The majority of patients (78,7%) did not consume alcohol as per documented records. In terms of gender distribution this constituted 68,7% of men and 88,4% of women. (Table 2). 6 Table 2:Documented social habits of patient sample Description Men = n (%) Women =n (%) Smoker Yes 14 (21) 6 (9) Smoker No 53 (79) 63 (91) Total 67 (100) 69 (100) Cannabis 3 (37.5) 2 (33,3) Cannabis and Opioid 0 (0,0) 2 (33,3) Opioid 5 (62,5) 2 (33,4) Total 8 (100) 2 (100) Alcohol No 46 (68,7) 61(88,4) Alcohol Yes 21 (31,3) 8 (11,6) Total 67 (100) 69 (100) Recorded clinical classification of patients A majority of the patients diagnosed with attempted suicide (65.2%) were women, followed by men (55,2%). Figure 1 shows an overview of the diagnosis recorded on files according to gender. 7 Figure 1: Recorded clinical classification of patients History of attempts Most patients were documented as a first attempt (79,4%) of parasuicide/attempted suicide. Most of these were women (82.6%) who attempted parasuicide/suicide for the first time with the rest of the women (17,4%) having previous attempts. This was then followed by men (76.1%) who attempted parasuicide/suicide for the first time and 23.9% of men who had previous parasuicide/attempted suicide. (Figure 2) 0 10 20 30 40 50 60 70 80 90 Men Women Total Parasuicide Attempted suicide 0 20 40 60 80 100 120 Men Women Total First attempt Previous attempts 8 Figure 2: History of attempts Clinical profile With regard to the psychiatric clinical profile, (16 %) patients were known to have had major depression disorder (MDD): women (64,2%) and men (58,3%). This was followed by 30,8% of the patients with newly diagnosed MDD, a personality disorder (3,8%) and schizophrenia (3.8%). With regard to the medical clinical profile, patients had the following: HIV/AIDS (41,7%), other illnesses [such as diabetes mellitus, hypertension and thyroid disease (33,3%)], epilepsy (12,5%), chronic pain (4,2%), both epilepsy and Chronic pain (4,2%). Figure 3: Documented psychiatric clinical profile 0 2 4 6 8 10 12 14 16 18 Newly diagnosed MDD Personality d/o Known MDD Schizophrenia Total Women Men 9 Figure 4: Documented medical profile Methods used and documented reasons for attempts In respect of the methods used, 93,4% of the patients used self-poisoning, men and women comprising 88,1% and 98,6% respectively. This was followed by blunt or sharp objects (3,7%), hanging (0,7%), and other methods such as drinking crushed glass (1,4%) and self-poisoning with hanging (0,7%). (Table 3) The methods mentioned were accompanied by the following documented reasons as listed in Table 3 some patients: battled with childhood adversity (4,4%); personal loss of loved ones (2,9%); interpersonal conflict (74,3%) comprising 68,7% and 79,7% for men and women respectively; financial difficulties (7,4%), experience of crime or trauma or abuse (2,2%), interpersonal conflict and financial difficulty (2,2%); financial difficulty and experience of crime trauma or abuse (0,7%) while the rest had other reasons (5,9%). (Table 3) 0 5 10 15 20 25 30 epilepsy HIV/AIDS Chronic pain Others epilepsy and Chronic pain HIV/AIDS+ other Total Women Men Table 3: Methods used and documented reasons for attempts Men Women Total n=136(%) n=67(%) n=69(%) Methods Used n (%) Self-poisoning 59 (88,1) 68 (98,5) 127 (93,4) Hanging 1 (1,5) 0 (0,0) 1 (0,7) 10 Reasons for Attempt Childhood adversity 5 (7.5) 1 (1.4) 6 (4.4) Personal loss of loved ones 3 (4,4) 1 (1,4) 4 (2,9) Interpersonal conflict 46 (68,6) 55 (79,7) 101 (74,3) Financial difficulty 5 (7.5) 5 (7,2) 10 (7.4) Experience of crime trauma or abuse 1 (1,5) 2 (3,0) 3 (2,2) Other reasons 5 (7,5) 3 (4,3) 8 (5,9) Interpersonal conflict and financial difficulty 1 (1,5) 2 (3,0) 3 (2,2) Financial difficulty and experience of crime trauma or abuse 1 (1,5) 0 (0,0) 1 (0,7) Factors associated with self-harming behaviours Factors statistically and significantly related to self-harming behaviours were: employment status (p value 0,030), alcohol use or not (p value 0,002) as well as history of attempts (p value <0,0001). There were no other variables amongst demographics, clinical profile, methods used and documented reasons that had statistical significance. Table 4 Factors associated with self-harming behaviours Diagnosis Parameter Attempted (%) Parasuicide (%) P value n Mean (±) Median IQR Min/min Male 82 27,7 (±10,9) 24 (21-33) 14/74 37 (45,1) 54 28,1 (±9,76) 26 (21-33) 15/61 30 (55,6) 0,835 0,631 0,293 Blunt or sharp object 5 (7,5) 0 (0,0) 5 (3,7) Other methods (drank crushed glass) 2 (2.9) 0 (0,0) 2 (1.5) Self-poisoning and blunt or sharp object 0 (0,0) 1 (1,5) 1 (0.7) 11 Female 45 (54.9) 24 (44,4) Total 82 (100) 54 (100) black 64 (78.1) 38 (70.4) 0,275 coloured 4 (4,9) 1 (1.9) Indian 1 (1,2) 0 (0,0) white 13 (15.8) 15 (27.7) Total 82 (100) 54 (100) 0.797 Divorced 1 (1.2) 0 (0,0) Married 3 (3.7) 4 (7.4) Single 77 (93.9) 50 (92.6) Widowed 1 (1.2) 0 (0,0) Total 82 (100) 54 (100) Employed 22 (26.8) 6 (11.1) 0.030* Unemployed 60 (73,2) 48 (88.9) Total 82 (100) 54(100) Smoker 8 (100) 12 (100) 0,405 Total 8 (100) 12 (100) Cannabis 2 (50) 3 (30) Cannabis and opioid 1 (25) 1 (10) Opioid 1 (25) 6 (60) Total 4 10 Alcohol yes 72 (87.8) 35 (64,8) 0.002* Alcohol no 10 (12,2) 19 (35.2) Total 82 (100) 54 (100) New MDD 5 (41,7) 11 (78.6) 0,066 Known MDD 6 (50,0) 2 (14.3) Personality 1 (8,3) 0 (0,0) Schizophrenia 0 (0,0) 1 (7.1) Total 12 (100) 14 (100) Chronic pain 1 (5,9) 0 (0,0) 0,195 epilepsy 1 (5,9) 0 (0.0) Epilepsy and chronic pain 1 (5,9) 2 (28.6) HIV/AIDs 9 (52.9) 1 (14,3) HIV/AIDS and other 1 (5,9) 0 (0,0) Other 4 (23.5) 4 (57,1) Total 17 (100) 7 (100) First attempt 76 (92.7) 32 (60.4) <.0001* Previous attempts 6 (7.3) 21 (39.6) Total 82 (100) 53 (100) Childhood adversity 1 (1,2) 5 (9,3) 0,182 Crime/trauma 3 (3,7) 0 (0,0) 12 Financial difficulties 5 (6,1) 5 (9,3) Financial difficulties and crime/trauma 1 (1,2) 0 (0,0) Interpersonal conflict 64 (78,1) 37 (68,5) Interpersonal conflict and financial difficulties 1 (1,2) 2 (3,7) Personal loss of loved ones 3 (3,7) 1 (1,8) Other 4 (4,8) 4 (7,4) Total 82 (100) 54 (100) Blunt or sharp object 1 (1,2) 4 (7,4) 0,088 Hanging 1 (1,2) 0 (0,0) Other 2 (2,4) 0 (0,0) Poison 78 (95,1) 49 (90,7) Poison and blunt or sharp object 0 (0,0) 1 (1,9) Total 82 (100) 54 (100) *statistically significant Discussion Key findings The aim of this study was to explore the factors associated with intentional self-harming behaviour among patients admitted to an Ekurhuleni district hospital. Therefore, this study showed that the most common characteristics of patients records reviewed was that of black single unemployed females aged between 14-24 years, who did not consume alcohol. This is in agreement with studies done in Limpopo in 2013 where more females attempted suicide compared to males. The three factors found to be associated with self-harming behaviours were: employment status, history of attempts and alcohol use this is in correlation with the study done in Johannesburg in 201624. Discussion of key findings A study done in Spain revealed similar findings with regards to the gender of patients, where females are more prone to self-harming behaviours. They found that 58.4% of suicide attempts were committed by women. The age group of patients in this study were younger, in the age bracket of 14-24, which is in alignment with many other studies that show a majority of suicide attempters are at a younger age (< 40 years) 6,9,10,12. With regards to social habits the majority 13 of patients in this study did not consume alcohol. This differs from the study in Durban in 2013 where a substantial number of patients smoked cigarettes and consumed alcohol.9 Most patients in this study sample did not have chronic illnesses or any psychiatric conditions. The clinical profile of patients in terms of psychiatric illness showed insignificant findings in this study with regard to self-harming behaviours. This does not correlate with the study done in Johannesburg in 201624 which suggested that experiencing a chronic illness not only increases the risk of self-harm in the affected patient but could also raises the risk in other household members. This difference could be attributed to the mostly younger sample in this study, as younger patients are unlikely to suffer from chronic illnesses The most prevalent method used in this study sample was self-poisoning, though not poison- type specific, as it was not indicated by the records. The finding correlates with the Africa- wide study6 that suggested that pesticide poisoning is a prominent method in Africa. This could be due to easy accessibility as compared to other methods. These findings differ from Naidoo and Schlebusch’s9 study where hanging surfaced as the leading method in the majority of suicide sufferers, all racial groups, both sexes and all ages, excluding the over-65-year age group. This study showed that the commonest documented reason for self-harm was interpersonal conflict. These findings differ from Obida, Clark and Govenders5 qualitative study which found that the commonest reason for suicide attempts was poverty and financial difficulties. However, the Durban study in 2013 found that interpersonal tribulations, poor family functioning and occurrence of previous attempts respectively may be a contributor in precipitating the impulsive suicidal behavioural action. The similarities between this study and the Durban study could be attributed to the urban setting of where the studies were done as opposed to the rural setting of the Limpopo study. Implications The majority of patients in this study had their first attempt of self-harming behaviour, this being the opposite of South African studies which show that majority of patients being both genders have had previous attempts of self-harm,9-10,13. This difference may be credited to the much younger sample of patients in this study starting at the tender age of 14 years old compared to the South African studies which show the majority of of youth committing self 14 harm are between 25-44 9,10,13. Strengths and Limitations Whilst this study is the first in Ekurhuleni that looks at documented factors relating to patients of self-harming behaviour the presented data is of one district hospital and so cannot be extrapolated to the rest of the district. The distinction between parasuicide and attempted suicide may have been confused as this is reliant on the clinician. The Failure to document some information in an emergency setting can also be seen as a limitation. Conclusions As human beings are made up of many complex aspects so are their behaviours. This study was able to point out important documented sociodemographic and clinical aspects of patients who are prone to self-harming behaviours. Furthermore, it was able to distinguish documented factors between attempted suicide and parasuicide which differs from previous studies on self- harm. As more studies show a shift in change towards the younger population committing self harm, including this one, should this not be an alarm on a human level about the challenges and outcry for help amongst this age group, who is yet to experience more challenges in their later life. This could be an indication that young black women who are unemployed feel like there is no reason to live if they loose or argue with a partner/family member. Besides being a medical problem, this should be a societal problem. In this regard in-depth qualitative studies looking specifically at the younger black female patients, can help find the specific challenges they face and how society can support them through those challenges. Acknowledgements Prof H.S Schoeman: Statistical analysis. Prof S Moosa: Supervision, support and guidance for the completion of the research report Dr S Agbo: Supervision during the protocol development and data collection The authors also wish to extend their gratitude to the hospital management at Bertha Gxowa Hospital for permission to perform the study. 15 Competing interests The authors declare that they have no financial or personal relationship(s) that may have inappropriately influenced them in writing this article. Author contributions This article was submitted by E.M as a requirement for the partial fulfilment of the degree of MMed (Family Medicine). E.M was the main author and S.M was the primary supervisor. All mentioned authors reviewed the drafts and approved the final version to be published. Funding contributions This study did not receive a grant from any funding agency in the public, commercial or not- for-profit agency. Data availability statement Data sharing is not applicable to this article, as no new data were created or analysed in this study Disclaimer The opinions and views of this article are those of the authors and do not reflect the official policy or position of any affiliated agency of the authors. References 1. Crosby AE, Ortega L, Melanson, C. Self-directed violence surveillance: uniform definitions and recommended data elements, Version 1.0. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2011. 2. Wilkinson P, Kelvin R, Roberts C, Dubicka B, Goodyer I. Clinical and psychosocial predictors of suicide attempts and nonsuicidal self-injury in the adolescent depression antidepressants and psychotherapy trial (ADAPT). Am J Psychiatry [serial online]. 2011 [cited 2018 Nov 06];168(5):495–501. Available from: https://doi.org/10.1176/appi.ajp.2010.10050718 16 3. Guan K, Fox KR, Prinstein MJ. Nonsuicidal self-injury as a time-invariant predictor of adolescent suicide ideation and attempts in a diverse community sample. J Consult Clin Psychol [serial online]. 2012 [cited 2018 Nov 06];80(5):842-849. Available from: http://dx.doi.org/10.1037/a0029429 4. Alberdi-Sudupe J, Pita-Fernández S, Gómez-Pardiñas SM, et al. Suicide attempts and related factors in patients admitted to a general hospital: a ten-year cross-sectional study (1997-2007). BMC Psychiatry [serial online]. 2011 [cited 2018 June 23];11:a51. Available from: https://doi.org/10.1186/1471-244X-11-51 5. International Society for the Study of Self-injury. What is self-injury? [homepage on the Internet]. May 2018. [cited 2018 Nov 06]. Available form: https://itriples.org/about-self-injury/what-is-self-injury 6. Mars B, Burrows S, Hjelmeland H, Gunnell D. Suicidal behaviour across the African continent: a review of the literature. BMC Public Health [serial online]. 2014 [cited 2018 Jun 01];14:606. Available from: https://dx.doi.org/10.1186/1471-2458-14-606 7. Madu SN, Matla MP. Family environmental factors as correlates for adolescent suicidal behaviours in the Limpopo Province of South Africa. Social Behaviour & Personality: an International Journal [serial online]. 2004 [cited 2018 Jun 01];32(4):341–353. Available from: https://dx.doi.org/10.2224/sbp.2004.32.4.341 8. Strydom MA, Pretorius PJ, Joubert, G. Depression and anxiety among Grade 11 and 12 learners attending schools in central Bloemfontein. S Afr J Psych [serial online]. 2012 [cited 2018 Jun 23];18(3):84-88. Available from: https://sajp.org.za/index.php/sajp/article/view/356/326 9. Naidoo SS, Schlebusch L. Sociodemographic and clinical profiles of suicidal patients requiring admission to hospitals south of Durban. S Afri Fam Pract [serial online]. 17 2013 [cited 2018 Jun 25];55(4):373–79. Available from: http://dx.doi.org/10.1080/20786204.2013.10874379 10. Naidoo SS, Schlebusch L. Sociodemographic characteristics of persons committing suicide in Durban, South Africa: 2006–2007. Afr J Prim Health Care Fam Med [serial online]. 2014 [cited 2018 Jun 23];6(1):e1–7. Available from: https://dx.doi.org/10.4102/phcfm.v6i1.568 11. Hawton K, Saunders KE, O'Connor R. Self-harm and suicide in adolescents. Lancet [serial online]. 2012 [cited 2018 Sept 02];379(9834):2373-2382. Available from: https://dx.doi.org/10.1016/S0140-6736(12)60322-5 12. Hoosen I, Smith P. Attempted suicide and deliberate self-harm. Ch.13. In: Primary care psychiatry: a practical guide for southern Africa, ed. by S.E. Baumann. Kenwyn: Juta; 2014, p.168-178. 13. Obida M, Clark C, Govender I. Reasons for parasuicide among patients admitted to Tshilidzini Hospital, Limpopo Province: a qualitative study. S Afr J Psych [serial online]. 2013 [cited 2018 Jun 24];19(4):a452. Available from: https://dx.doi.org/10.7196/SAJP.452 14. Muehlenkamp JJ, Claes, L, Havertape L, Plener PL. International prevalence of adolescent non-suicidal self-injury and deliberate self-harm. Child Adolesc Psychiatry Ment Health [serial online]. 2012 [cited 2018 Nov 06];6:a10. Available from: https://doi.org/10.1186/1753-2000-6-10 15. Glenn CR, Klonsky, ED. Nonsuicidal self-injury disorder: an empirical investigation in adolescent psychiatric patients. J Clin Child Adolesc Psychol [serial online]. 2013 [cited 2018 Nov 06];42(4):496–507. Available from: https://doi.org/10.1080/15374416.2013.794699 16. Nock, MK, Kessler, RC. Prevalence of and risk factors for suicide attempts versus suicide gestures: analysis of the national comorbidity survey. J Abnorm Psychol 18 [serial online]. 2006 [cited 2018 Nov 06];115(3):616–623. Available from: https://doi.org/10.1037/0021-843X.115.3.616 17. Klonsky ED, May AM, Glenn CR. The relationship between nonsuicidal self-injury and attempted suicide: converging evidence from four samples. J Abnorm Psychol [serial online]. 2013 [cited 2018 Nov 06];122(1):231–237. Available from: https://doi.org/10.1037/a0030278 18. Klonsky ED, Muehlenkamp, JJ. Self-injury: a research review for the practitioner. J Clin Psychol [serial online]. 2007 [cited 2018 Nov 06];63(11):1045–1056. Available from: https://doi.org/10.1002/jclp.20412 19. Asarnow JR, Porta G, Spirito A, et al. Suicide attempts and nonsuicidal self-injury in the treatment of resistant depression in adolescents: findings from the TORDIA study. J Am Acad Child Adolesc Psychiatry [serial online]. 2011 [cited 2018 Nov 06];50(8):772–781. Available from: https://doi.org/10.1016/j.jaac.2011.04.003 20. Zeppegno P, Gramaglia C, Castello LM, et al. Suicide attempts and emergency room psychiatric consultation. BMC Psychiatry [serial online]. 2015 [cited 2018 Jun 24];15:13. Available from: https://dx.doi.org/10.1186/s12888-015-0392-2Psychiatry [serial online]. 2011 [cited 2018 Nov 06];168(5):495–501. Available from: https://doi.org/10.1176/appi.ajp.2010.10050718 21. Paul R, Panchal S, Zimba K. Description of cases of parasuicides reported at University Teaching Hospital, Lusaka. Zambia: preliminary findings. Health Press Zambia Bull [serial online]. 2017 [cited 2018 Jun 24];1(4):14-18. Available from: http://znphi.co.zm/thehealthpress/2017/03/ 22. WHO. Practice Manual for Establishing and Maintaining Surveillance Systems for Suicide Attempts and Self-Harm [homepage on the Internet]. World Health Organisation [updated 2016; cited 2018 June 24]. Available from: http://apps.who.int/iris/bitstream/10665/208895/1/9789241549578_eng.pdf 19 23. Kar, N. Profile of risk factors associated with suicide attempts: a study from Orissa, India. Indian J Psychiatry [serial online]. 2010 [cited 2018 June 18];52(1):48–56. Available from: https://doi.org/10.4103/0019-5545.58895 24. Naicker N, de Jager P, Naidoo S, Mathee A. Household factors associated with self- harm in Johannesburg, South African Urban-Poor Households. PLoS One [serial online]. 2016 [cited 2020 Dec 26];11(1):e0146239. Available from: https://doi.org/10.1371/journal.pone.0146239. Appendices Appendix I: Data collection tool A modified Data collection tool from Practice Manual for Establishing and Maintaining Surveillance Systems for Suicide Attempts and Self-Harm 4: *Please see MS Excel attachment (Appendix I) 20 Appendix II: Ethical clearance 21 22 Appendix III: Permissions 23 PERMISSION LETTER TO CONDUCT MMED RESEARCH AT BERTHA GXOWA HOSPITAL 02 July 2018 The Chief executive officer Bertha Gxowa hospital Germiston Dear sir/Madam Request for permssion to conduct MMed research at Bertha Gxowa hospital My name is Dr Elsie Mugisha, a second year registrar in the department of family medicine at the university of the Witwatersrand. As part of the MMED studies,the depatment of Family medicine requires that I undertake a research at a health care center. The title of the research is “Factors associated with parasuicide among patients in Bertha Gxowa hospital”. The aim of this study is to determine the factors associated with parasuicide among patients admitted to the hospital. The study will be a cross sectional study using the medical records of patients admitted between June 2016 and June 2018. Data wil be collected by means of a data collection tool which includes sociodemographic characteristics,clinical profile, method used and reason for parasuicide. There will be no contact with patients during this study and confidentiality will be maintained at all times. I would like request permission to conduct my Mmed research at Bertha Gxowa Hospital and access files from the records department once I have obtained approval from the Ekurhuleni research committee and WITS Health research ethics committee. I look forward to your positive response. Kind Regards, 24 ______________________ Dr ENV Mugisha Registrar Ekurhuleni district 25 Appendix IV: Turnitin report 26 27 Appendix V: Proofreading certificate From the desk of F. E. Meyer P. o. box 885 2041 HOUGTON e-mail: meyer.fe576@gmail.com https://meyerfe576.wixsite.com/editormysite (Note: two full stops: one after 576 and one after wixsite January 27th, 2021 ENGLISH PROOF-READING/EDITING To Whom It May Concern The journal article entitled, ‘Factors associated with intentional self-harming behaviours among patients admitted to Bertha Gxowa Hospital’ to be submitted by Dr E.N.V. MUGISHA et al has been proof-read and edited for proper English language, syntax, grammar, punctuation, British English spelling and overall style by F.E. MEYER. As requested, document formatting was applied to the journal article according to the Journal of Primary Health Care & Family Medicine (PHCFM) guidelines for authors. The 24 references were checked according to the Vancouver referencing style as per the journal’s guidelines. The research content and/or the authors’ intentions were not altered during the process. Blue and yellow highlighting will require the corresponding author’s attention. The date on the footnote on this page corresponds with that of the attached proof-read and edited sections of the journal article. Any further proof-reading or editing required by the author(s), due to changes or amendments to this journal article will be accompanied by an updated covering page and will replace this covering page. F.E. Meyer ____________ F.E. MEYER B.A. (UNISA) Footnote: Proof reader’s/Editor’s covering page Edited journal article 2020/01/27 MUGI/02 28 Appendix VII: Research proposal FACTORS ASSOCIATED WITH INTENTIONAL SELF-HARMING BEHAVIOURS AMONG PATIENTS ADMITTED IN BERTHA GXOWA HOSPITAL. Dr ENV Mugisha Student no: 300023 Department of Family Medicine 29 1 Table of Contents Declaration ............................................................................................................................... ii Dedication ............................................................................................................................... iii Acknowledgements ................................................................................................................ iii Abstract .................................................................................................................................... iv Background ......................................................................................................................... iv Aim: ...................................................................................................................................... iv Methods: .............................................................................................................................. iv Results: ................................................................................................................................. iv Conclusion: .......................................................................................................................... iv Keywords ............................................................................................................................. iv Table of Contents ..................................................................................................................... v Nomenclature .......................................................................................................................... vii List of figures ............................................................................................................................. i List of tables............................................................................................................................. ii Introduction .............................................................................................................................. 1 Research methods .................................................................................................................... 2 Study design .......................................................................................................................... 2 Setting.................................................................................................................................... 3 Study population .................................................................................................................. 3 Data collection ...................................................................................................................... 3 Data analysis ......................................................................................................................... 4 Ethical considerations .......................................................................................................... 4 Results ....................................................................................................................................... 4 Discussion ............................................................................................................................... 12 Conclusions ............................................................................................................................. 14 Acknowledgements ................................................................................................................ 14 Competing interests ........................................................................................................... 15 Author contributions ......................................................................................................... 15 Funding contributions ....................................................................................................... 15 Data availability statement ................................................................................................ 15 Disclaimer ........................................................................................................................... 15 References ............................................................................................................................... 15 Appendices .............................................................................................................................. 19 Appendix I: Data collection tool ....................................................................................... 19 Appendix II: Ethical clearance ......................................................................................... 20 30 Appendix III: Permissions ................................................................................................ 22 Appendix IV: Turnitin report .......................................................................................... 25 Appendix V: Proofreading certificate .............................................................................. 27 Appendix VII: Research proposal .................................................................................... 28 References: ............................................................................................................................... 42 Introduction Background 31 Intentional self-harm may be defined as self-injurious/harmful behaviour that is self- directed and deliberately results in injury or the potential for injury to oneself, often as a response to feelings of extreme psychological distress or emotional pain (1). It includes cutting, burning or hitting oneself, bingeing or starvation, abuse of drugs or alcohol, and overdosing on prescription medications (20-22). This risky behaviour can have a fatal outcome (13). Intentional self-harming behaviour encompasses parasuicide and attempted suicidal behaviours that are often used interchangeably in literature (1). Parasuicide is referred to as nonsuicidal self-injury (NSSI) and attempted suicide as suicidal self-injury (SSI) (13). The relationship between parasuicide and attempted suicide is a complex one and there is an overlap consensus between the two (14,17). Research findings indicate that parasuicide behaviour predicts suicide attempts and high co-occurrence are encountered (15,19). Thus, differentiation between the two relies more on the lethal nature, method, prevalence, frequency and function (16,18). In 1985, the WHO European regional multicentre study was launched to monitor and examine non-fatal suicidal behaviour. Researchers from 15 countries agreed that one of the main problems for epidemiological research in this area was the lack of a common terminology (4). Moreover, they agreed that there was a need for reasonable terms and definitions of non-fatal suicidal acts from a theoretical point of view that could at the same time be used and applied to surveillance of the aforementioned. The WHO European multicentre study group used ‘parasuicide’ as an umbrella term that covered terms such as ‘attempted suicide’, ‘deliberate self- harm’ and ‘self-poisoning’. As the study expanded to more countries, the term ‘parasuicide’ was found to be challenging in practice as it was difficult to translate in some cultures (4). During 1994, the term ‘parasuicide’ was replaced with ‘attempted suicide’ as the term adopted at the WHO European multicentre study. However, even in relation to hospital data records, the use of the term ‘attempted suicide’ was fluid because it meant very different things to different people (4). Intentional self-harming behaviours have increased globally over the past decade and have been identified as one of the major contributors to the high disease and 32 healthcare burden in many low- to middle-income countries (10). Patients who engage in intentional self-harming behaviours are 38 times more likely commit suicide (2). Motivation/ justification The justification of this study is to document the patterns of intentional self-harm cases which will contribute to the understanding of self-harming behaviours more fully and thereby assist health care workers with the best ways of identifying, reaching and treating those who struggle with self harming behaviours. This research may assist on a provincial and national level in regard to relevant documentation and proper follow-up of patients with self-harming behaviours. Literature Review Research into the epidemiology and cause of intentional self-harming behaviours is hampered by the lack of agreement on terminology and definitions; that is further complicated by the varying levels of suicidal intent and diverse motives reported by people engaging in self-harming behaviours (4). Moreover, there are also major procedural disparities between countries for recording these suicides. During 2014, the WHO’s global report on suicide prevention identified a need for guidance on the surveillance of self-harming behaviour cases presenting to hospitals. (6) A committed hospital-based surveillance system of self-harm incidents range from: countries’ registries, such as in Ireland to subnational registries at the regional level, such as in the United Kingdom of Great Britain and Northern Ireland; whereas in Derby, Manchester and Oxford, data are collected on intentional self-harming patients presenting to hospital emergency departments. In most low- and middle- income countries there are no hospital-based surveillance systems for suicide attempts. Intentional self-harm is usually the last step of a sequence of processes starting with a death wish, suicidal ideation, suicidal contemplation and suicidal behaviours (2). It is the second leading cause of death in the 15-29 years’ age group (3). Important contributors to self-harm and suicide in this young age group include a family history 33 of intentional self-harming behaviours and psychiatric illnesses, psychological, familial, social, and cultural factors. The effects of media are also important, with the internet presently playing an important role. (11) Annually, at least one million people worldwide commit suicide. Intentional self- harming behaviours outnumber actual suicides by an equally shocking figure ranging from 10-20 times more per year (4). A patient who has engaged in one or more acts of intentional self harm makes it the single most important predictor of death by suicide (4). In a South African study, it was estimated that the average European rate for attempted suicide in persons aged 15 years and above was 140:100 000 for males and 193:100 000 for females (5). There is scarcity of information on self-harming behaviours in Africa. A systematic review reported that data on self-harming behaviours was only available from 11 countries on the African Continent. The self- harming behaviours incidence rates for seven countries varied from 0.1 per 100000 in Ghana to 100 per 100000 in Namibia, the outstanding four countries reported lifetime prevalence estimates for suicide attempts collected mainly from surveys (6). In South Africa, age, race and city play independent roles in sex-specific suicide rates (5). Younger age groups of both genders are at a higher risk, with females being at higher risk than males in many studies (8). In South Africa studies have shown that suicide rates have been increasing steadily in all population groups over the past decade, although the prevalence varies considerably across age and race groups. Among the Indian, Black and Coloured race groups/population, the third major cause of death after homicide and natural death is suicide, whereas among the White race group/population suicide is the second major cause of death (9). Research has shown that the risk factors include child adversity, interpersonal loss, interpersonal conflict, recent migration and financial issues (12). There is a strong association between suicide and mental illness (12). A metanalysis of the rates of suicide for psychiatric illnesses revealed an increased risk of suicide among patients with eating disorders. The risk of suicide is seven times higher for those with personality disorder. The lifetime risk of suicide for individuals who suffer 34 from schizophrenia is 10 % (13). For post-partum psychiatric disorders, the risk of suicide is increased especially during the first year (13). Patients with cancer have twice the suicide rate compared to the general population and it is five times more common in those with epilepsy. Studies have also shown increased rates of suicidal ideation in patients with chronic pain and HIV/AIDS (13). A wide range of methods are used by the people committing suicide and are dependent on different factors (10). In South Africa, the statistics for the past decade have shown that hanging and shooting are preferred methods followed by self- poisoning with agents such as pesticides and poisons, drug overdoses, self-gassing and self-immolation (10). Aim and objectives Aim The aim of this study is to explore the factors associated with intentional self-harming behaviour among patients admitted to the Bertha Gxowa Hospital. Specific objectives • Determine the sociodemographics of patients admitted with intentional self- harming behaviour. • Explore the clinical profile of patients admitted with intentional self-harming behaviour. • Determine the methods used by patients admitted to the hospital with intentional self-harming behaviour. • Determine the documented reasons for the intentional self-harming behaviour by patients admitted to the hospital. • Determine the association between the demographics, clinical profile, methods used, documented reasons and intentional self-harming behaviour. Methods Study design A descriptive cross-sectional study of patients’ medical records will be used. This will involve retrospective review of medical records of patients admitted for intentional self-harm between June 2016 and June 2018. 35 Site of study: The Bertha Gxowa Hospital is the only district hospital in the Ekurhuleni Health District. The hospital has 230 approved beds. It provides level one health care services to Germiston - a town with a population of 3942533 and surrounding communities. These services include: primary health care facilities; 24-hour accident emergency and trauma; 72-hour observations for psychiatric patients; medico-legal services; surgery excluding orthopaedics; obstetrics and gynaecology; allied services and pharmacy. Study population The study population will include all patients that were admitted with intentional self- harming behaviour to the Bertha Gxowa Hospital during June 2016 and June 2018. The sample size and sampling This will be purposive sampling of all medical records of patients admitted during the period June 2016 and June 2018 by the researcher based on the inclusion and exclusion criteria. Inclusion criteria • All files/folders with a diagnostic label of parasuicide or attempted suicide on patients admitted to the adult medical wards for intentional self-harm cases/incidents during the period from June 2016 to June 2018. Exclusion criteria • Patient folders/files with inadequate data and completely illegible records will be excluded from the study. Measurement tool/instrument A modified data collection tool adapted from the World Health Organization Practice Manual for establishing and maintaining surveillance systems for suicide attempts and self-harm will be used (6). This tool was designed by WHO to be used by 36 countries to establish public health surveillance systems for suicide attempts and self-harm cases presenting to general hospitals, based on medical records (6). This tool will be used in this research study to collect the following objective variables: • Diagnosis (attempted suicide or parasuicide); • Sociodemographic aspects (sex, date of birth, age, nationality, ethnicity, religion, marital status, employment status, alcohol or drug use and educational status); • Clinical profile (eating disorders, newly diagnosed major depressive disorder, personality disorders, patients known with major depressive disorder, schizophrenia, post-partum psychiatric disorder, epilepsy, HIV/AIDS, cancer, chronic pain and, others); • Intentional self-harming methods used (self-poisoning, hanging, etc…); and • Documented reasons for intentional self-harm (child adversity, interpersonal loss, interpersonal conflict, recent migration and financial issues, experience of crime, trauma or abuse and others). Validity will be secured with inclusion of all files except the stipulated exclusion criteria. This will help to eliminate selection bias. All records will be measured in the same way with the same tool which has been modified according to current literature and research findings on intentional self-harming behaviours. Data collection: All files of admitted patients with the label of intentional self-harming behaviours between June 2016 and June 2018 will be identified and retrieved from the hospital record archives with the assistance of the record clerks. Inclusion and exclusion criteria will be applied by the researcher. All patient files that meet the inclusion criteria will have a label sticker adhered to them for identification purposes and to prevent duplication. Confidentiality will be maintained where a study number will be allocated with a separate coding sheet for the patients file number. 37 The variables will be recorded in the data collection sheets by the researcher on a daily basis until all eligible files have been reviewed. The data will then be entered into a Microsoft Excel software file for analysis by the researcher. During the data collection period, all files will be kept in the hospital record locker and secured. After the completion of the data collection, the files will be handed back to the hospital record staff. Data analysis: Microsoft Excel software will be used to capture data. The descriptive and inferential statistics will be done using statistica data analysis software. Data analysis will be mainly descriptive statistics in the form of frequencies and percentages for the categorical data such as age, sex and race. In order to evaluate the differences in proportions between categorical variables (sociodemographics, clinical profile, documented reasons and methods) and intentional self-harming behaviour, the chi- square test will be used. Continuous data will be summarised using means with standard deviations if normally distributed. Skewed continuous data will be summarized using median with interquartile range. A multivariable logistic regression will be used to determine the factors from the demographics, clinical profile, methods used and documented reasons for intentional self-harming behaviour. Intentional self harming behaviours will be subdivided into parasuicide and attempted suicide. Statistical significance will be considered at 2-sided alpha level of 0.05 The data will be stored safely on the researcher’s personal hard-disk drive with back- up support and password protected. Table 1: analysis guide table: Specific objectives Variables Type of data Data analysis 38 Sociodemographics 1. Age 2. Sex 3. Race 4. Education 5. Employment 6. Religion 7. Drug use 8. Alcohol use 9. Marital status 1. Continuous 2. Categorical 3. Categorical 4. Categorical 5. Categorical 6. Categorical 7. Categorical 8. Categorical 9. Categorical • Continuous data will be summarised using mean with standard deviations if it is normally distributed. If the continuous data is skewed it will be summarized using median with interquartile range • Categorical data will be summarized with frequencies and percentages and presented in tables or graphs. Clinical profile 1. Eating disorder newly diagnosed MDD 2. Personality disorder 3. Known MDD 4. Schizophrenia 5. Post-partum psychiatric disorder 6. Epilepsy HIV/AIDS 7. Cancer 8. Chronic pain 9. Others 1. Categorical 2. Categorical 3. Categorical 4. Categorical 5. Categorical 6. Categorical 7. Categorical 8. Categorical 9. Categorical • Categorical data will be summarized with frequencies and percentages and presented in tables or graphs. Methods used 1. Self poisoning 2. Hanging 3. Blunt or sharp object 4. Others 1. Categorical 2. Categorical 3. Categorical 4. Categorical • Categorical data will be summarized with frequencies and percentages and presented in tables or graphs. Documented reasons 1. Childhood adversity 2. Interpersonal loss 1. Categorical 2. Categorical • Categorical data will be summarized with frequencies 39 3. Interpersonal conflict 4. Recent migration 5. Financial difficulties 6. Experience of crime, trauma or abuse 7. other 3. Categorical 4. Categorical 5. Categorical 6. Categorical 7. Categorical and percentages and presented in tables or graphs Associations between sociodemographics, clinical profile, documented reasons, methods used and intentional self-harming behaviours. 1. Sociodemographics 2. Clinical profile 3. Documented reasons 4. Methods used 1. Categorical 2. Categorical 3. Categorical 4. Categorical • Univariate logistic regression will be used. Significant variables from the univariate logistic regression will be used in the multivariable logistic regression to determine the association. Pilot study: A pilot study will be conducted in a hospital in the district to test if the protocol is feasible. A minimum of ten patient folders will be reviewed The aim of the pilot study is to test the research process (e.g. duration), acceptability, suitability, understanding and the validity of the elements in the data collection tool of the study. This will be done after approval has been obtained from the Human Research Ethics Committee, University of the Witwatersrand, Johannesburg. Limitations: • Inadequate record keeping, by health care workers, of required information. • Loss of patient records/missing continuation notes/sheets from previous years. 40 • Loss of patient files from the medical wards and emergency department regarding patients admitted for intentional self-harm. • Patients wrongly diagnosed with intentional self-harm behaviours. In such a case concerning missing data, loss of records or inadequate record keeping this information will be documented on a daily tracking sheet by the researcher. Once documented on the tracking sheet, all missing data will be reported Bias Bias may occur while reviewing patient records through over-interpretation or under- interpretation of data by the researcher. In such a case those records will be reviewed by the supervisor. Ethical considerations: Prior to the commencement of the study, permission from the management at the Bertha Gxowa Hospital and the Ekurhuleni District will be obtained. Ethical approval will be obtained from the Health Research and Ethics Committee, University of the Witwatersrand, Johannesburg. All data will be collected in a private room and all information will be locked in a cupboard where access will only be allowed to the researcher and supervisor. Timing: As abovementioned, once the researcher has received permission from the management at the Bertha Gxowa Hospital, University of the Witwatersrand Human Research Ethics Committee and the Ekurhuleni District, the study will commence. The study schedule is as follows: Table 2: study schedule J u n J u l A u g S e p O c t N o v D e c J a n F e b M a r A p r M a y J u n J u l A u g S e p O c t N o v D e c J a n Literature review Preparing protocol 41 Protocol assessment Ethics application Collecting data Data analysis Writing-up thesis Writing-up paper Funding: This research proposed budget will be self- funded. Table 3: budget Items Number needed Unit cost Transport Monthly basis R500 Stationary Reams of paper, pens, photocopying R300 Statistician 1 At no cost Proof reading +- 60 pages for research report with R50 per page R3000 Plan for utilization / dissemination of results The study will form part of the research report, in partial fulfilment of the Master of Medicine, Family Medicine degree. The results will be shared with the Gauteng Department of Health, the health management at the Ekurhuleni District and the managers of the Bertha Gxowa Hospital. Once completed, the study findings will also be published in peer-reviewed journals. 42 References: 1. Crosby, A., Ortega, L. & Melanson, C. 2011. Self-directed violence surveillance: uniform definitions and recommended data elements, Version 1.0. Atlanta (GA): Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. 2 Zeppegno, P., Gramaglia, C. & Castello, L.M. et al. 2015. Suicide attempts and emergency room psychiatric consultation. BMC Psychiatry 15:13. doi: https://dx.doi.org/10.1186/s12888-015-0392-2 [Accessed 2018-06-24] 3 Paul, R., Panchal, S. & Zimba, K. 2017. Description of cases of parasuicides reported at University Teaching Hospital, Lusaka. Zambia: preliminary findings. Health Press Zambia Bull 1(4):14-18. Available: http://znphi.co.zm/thehealthpress/2017/03/ [Accessed 2018-06-24] 4 World Health Organization. 2016. Practice Manual for Establishing and Maintaining Surveillance Systems for Suicide Attempts and Self-Harm. Available: http://apps.who.int/iris/bitstream/10665/208895/1/9789241549578_eng.pdf. [Accessed 2018-06-24] 5 Obida, M., Clark, C. & Govender, I. 2013.Reasons for parasuicide among patients admitted to Tshilidzini Hospital, Limpopo Province: a qualitative study. S Afr J Psych 19(4):222–225.doi: https://dx.doi.org/10.7196/SAJP.452 [Accessed 2018-06-24] 43 6 Mars, B., Burrows, S. & Hjelmeland, H. et al. 2014. Suicidal behaviour across the African Continent: a review of the Literature. BMC Public Health 14:606. doi: https://dx.doi.org/10.1186/1471-2458-14-606 [Accessed 2018-06-01] 7 Madu, S.N. & Matla, M.P. 2004. Family environmental factors as correlates for adolescent suicidal behaviours in the Limpopo Province of South Africa. Social Behaviour & Personality: An International Journal 32(4):341–354. doi: https://dx.doi.org/10.2224/sbp.2004.32.4.341 [Accessed 2018-06-01] 8 Strydom, M.A., Pretorius, P.J. & Joubert, G. 2012. Depression and anxiety among Grade 11 and 12 learners attending schools in central Bloemfontein. S Afr J Psych 18(3): 84-88. Available: https://sajp.org.za/index.php/sajp/article/view/356/326 [Accessed 2018-06-23] 9 Naidoo, S.S. & Schlebusch, L. 2013. Sociodemographic and clinical profiles of suicidal patients requiring admission to hospitals South of Durban. South African Family Practice 55(4): 373–79. doi: http://dx.doi.org/10.1080/20786204.2013.10874379 [Accessed 2018-06- 25] 10 Naidoo, S.S. & Schlebusch, L. 2014. Sociodemographic characteristics of persons committing suicide in Durban, South Africa: 2006–2007. Afr J Prim Health Care Fam Med 6(1):e1-7. doi: https://dx.doi.org/10.4102/phcfm.v6i1.568 [Accessed 2018-06-23] 11 Hawton, K., Saunders, K.E. & O'Connor, R. 2012.Self-harm and suicide in adolescents. Lancet 379(9834):2373-2382. 44 doi: https://dx.doi.org/10.1016/S0140-6736(12)60322-5 [Accessed 2018-09-02] 12. Hoosen, I. & Smith, P. Attempted suicide and deliberate self-harm. Ch.13. In: Primary care psychiatry: a practical guide for southern Africa, ed. by S.E. Baumann. Kenwyn: Juta, 2014, pp.168-178. 13. International Society for the Study of Self-injury. May 2018. What is self-injury? Available https://itriples.org/about-self-injury/what-is- self-injury [Accessed 2018-11-06] 14. Muehlenkamp, J.J., Claes, L. & Havertape, L. et al. 2012. International prevalence of adolescent non-suicidal self-injury and deliberate self-harm. Child Adolesc Psychiatry Ment Health 6:10. doi:10.1186/1753-2000-6-10. [Accessed 2018-11-06] 15. Glenn, C.R. & Klonsky, E.D. 2013. Nonsuicidal self-injury disorder: an empirical investigation in adolescent psychiatric patients. J Clin Child Adolesc Psychol 42(4):496–507. doi: 10.1080/15374416.2013.794699. [Accessed 2018-11-06] 16. Nock, M.K. & Kessler, R.C. 2006. Prevalence of and risk factors for suicide attempts versus suicide gestures: analysis of the national comorbidity survey. J Abnorm Psychol 115(3):616–623. doi: 10.1037/0021-843X.115.3.616. [Accessed 2018-11-06] 17. Klonsky, E.D., May, A.M. & Glenn, C.R. 2013. The relationship between nonsuicidal self-injury and attempted suicide: converging evidence from four samples. J Abnorm Psychol 122(1):231–237. doi: 10.1037/a0030278 [Accessed 2018-11-06] 45 18. Klonsky, E.D. & Muehlenkamp, J.J. 2007. Self-injury: a research review for the practitioner. J Clin Psychol 63(11):1045–1056. doi: 10.1002/jclp.20412. [Accessed 2018-11-06] 19. Asarnow, J.R., Porta, G. & Spirito, A. et al. 2011. Suicide attempts and nonsuicidal self-injury in the treatment of resistant depression in adolescents: findings from the TORDIA study. J Am Acad Child Adolesc Psychiatry 50(8):772–781. doi: 10.1016/j.jaac.2011.04.003. [Accessed 2018-11-06] 20. Wilkinson, P., Kelvin, R. & Roberts, C. et al. 2011. Clinical and psychosocial predictors of suicide attempts and nonsuicidal self-injury in the adolescent depression antidepressants and psychotherapy trial (ADAPT). Am J Psychiatry 168(5):495–501. doi: 10.1176/appi.ajp.2010.10050718. [Accessed 2018-11-06] 21. Guan, K., Fox, K. R. & Prinstein, M.J. 2012. Nonsuicidal self-injury as a time- invariant predictor of adolescent suicide ideation and attempts in a diverse community sample. Journal of Consulting and Clinical Psychology 80(5):842- 849. http://dx.doi.org/10.1037/a0029429 [Accessed 2018-11-06] 46 Appendices Modified Data collection tool from Practice Manual for Establishing and Maintaining Surveillance Systems for Suicide Attempts and Self-Harm (4): *Please see MS Excel attachment (Appendix II) 47 Appendix I 48 Appendix III PERMISSION LETTER TO CONDUCT MMED RESEARCH AT THE BERTHA GXOWA HOSPITAL 02 July 2018 The Chief Executive Officer Bertha Gxowa Hospital Germiston Dear Sir/Madam Request for permssion to conduct MMed research at the Bertha Gxowa Hospital My name is Dr Elsie Mugisha, a second year registrar in the Department of Family Medicine, University of the Witwatersrand, Johannesburg. As part of the MMED studies,the Depatment of Family Medicine requires that I undertake a research study at a health care center. The title of the research is, ‘Factors associated with parasuicide among patients in Bertha Gxowa hospital.’ The aim of this study is to determine the factors associated with parasuicide among patients admitted to the hospital. The study will be a cross-sectional study using the medical records of patients admitted between June 2016 and June 2018. Data wil be collected by means of a data collection tool which includes sociodemographic characteristics,clinical profile, method used and reason for parasuicide. There will be no contact with patients during this study and confidentiality will be maintained at all times. I would like request permission to conduct my MMed research at the Bertha Gxowa Hospital and access files from the records department once I have obtained approval 49 from the Ekurhuleni Research Committee and the WITS Health Research Ethics Committee. I look forward to your positive response. Kind regards, ______________________ Dr ENV Mugisha Registrar Ekurhuleni District