The Western Cape province has the highest documented lifetime prevalence of common mental disorders in South Africa. To ensure the efficient, equitable and effective distribution of current resources, there is a need to determine the profile of patients requiring psychiatric admission.
To describe patients admitted to the acute adult admissions unit at Lentegeur Hospital.
Lentegeur Psychiatric Hospital is situated in Mitchells Plain, Cape Town, and serves about 1 million people from nearby urban and rural areas.
This retrospective study involved an audit of all patients (18–60 years of age) admitted between 01 January 2016 and 30 June 2016. The clinical records of 573 adult patients were examined.
The median age of the cohort was 29 years. Most patients (63%) were educated to the secondary level. Only 12% of the patients were employed, and 37% received disability grants. More than 90% of the patients presented with psychotic symptoms. Of these, 28% presented with a first-episode of psychosis. Of all patients, 20% were referred with manic symptoms and 7% with depressive symptoms. Many patients (62%) used substances concurrently in the period leading up to admission. Significantly more males (73%) used substances compared to females (38%). Cannabis was the most widely used substance (51%), followed by methamphetamine (36%). Recent violent behaviour contributed to 37% of the current admissions. A total of 70 patients (13%) tested positive for human immunodeficiency virus (HIV), and 49 (9%) tested positive for syphilis.
Substance use and a history of violence contributed to admissions in this population.
The Western Cape province has the highest documented lifetime prevalence of common mental disorders in South Africa.
Methamphetamines (35%) and cannabis (22%) are the most commonly abused substances in the Western Cape.
Crime, substance use and low socio-economic status are known to have an adverse impact on mental health.
Thus, for the improvement of mental health services and for the efficient, equitable and effective distribution of current resources, a better understanding of the profile of patients requiring psychiatric admission needs to be ascertained. Therefore, the aim of this study was to establish the profile of patients admitted to the acute adult admissions unit at LGH.
This study was a retrospective audit of all patients (18–60 years of age) admitted under the
Data were obtained by a retrospective review of the clinical records of all patients (
Data abstraction was undertaken by the principle investigator and included demographic information (age, gender, level of education, employment, drainage area [urban or rural], marital status, disability grant recipient), clinical information (history of substance use, violence or infectious disease) and psychiatric symptoms. Given the socio-political context, this study also focused on comparing data obtained from urban referral centres; that is, between Mitchells Plain (ethnic majority mixed race) and Khayelitsha (ethnic majority black African).
Substance use was defined as the ingestion, inhalation, or injection of illicit drugs; in particular, methamphetamine (Tik), cannabis, methaqualone (Mandrax), cocaine or heroin, as reported by either the patient, or caregiver, or if noted as observed or positively tested in the documentation from the referral hospital. Substance use also includes the use of alcohol at levels clinically deemed to amount to a use disorder. Only the use of such substances clinically deemed to be related to the onset and maintenance of the current episode was considered; that is, there had to be evidence of substance use, either at the time of current admission to LGH, or in the immediate few weeks preceding, and concurrent with the onset of the symptoms that resulted in the current admission.
The history of violence was defined as a physical assault to self or others (staff, other patients, family) and/or damage to property, from the time of onset of symptoms of the current episode, up to and including the current admission. Violence, however, excluded the ingestion of substances with the aim to self-injure. Violence predating the onset of the current episode was, therefore, excluded.
Demographic and clinical/behavioural data were summarised as means (± standard deviation) and median (25th – 75th percentiles) for continuous variables, as counts (percentages) for categorical variables, using Microsoft Excel. The distribution of variables across groups (gender, violence and drainage area) was compared using
The study was approved by the Health Research Ethics Committee of Stellenbosch University (S16/10/237), as well as the Western Cape Health Research Committee. A waiver of informed consent was granted for this retrospective study. All identifiable patient information was anonymised.
The demographic and clinical characteristics of 573 adult psychiatric patients who were admitted to Lentegeur Psychiatric Hospital between January 2016 and June 2016 are summarised in
Patient demographics and clinical characteristics with bivariate analysis across urban and rural drainage zones and in terms of gender.
Variables | Overall |
Referring centre |
Gender |
||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Urban |
Rural |
Males |
Females |
||||||||||||||
% | IQR | % | IQR | % | IQR | % | IQR | % | IQR | ||||||||
Population | 573 | - | - | 459 | 80.0 | - | 114 | 20 | - | < 0.001 |
383 | 67.0 | - | 190 | 33.0 | - | < 0.01 |
Median age | 29 | - | 24–40 | 29 | - | 25–39 | 29.5 | 24.0–43.5 | 0.868 | 28 | - | 24–35 | 33 | - | 26–46 | < 0.01 |
|
Male | 383 | 67 | - | 315 | 69.0 | - | 68 | 60 | - | 0.076 | - | - | - | - | - | - | - |
- | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - | 0.79 | |
Primary (Grades 1–7) | 147 | 26 | - | 114 | 25.0 | - | 33 | 29 | - | - | 98 | 26.0 | - | 49 | 26.0 | - | |
Secondary (Grades 8–12) | 362 | 63 | - | 295 | 64.0 | - | 67 | 59 | - | 0.547 | 245 | 64.0 | - | 118 | 62.0 | - | |
Tertiary | 34 | 6 | - | 25 | 5.0 | - | 9 | 8 | - | - | 20 | 5.0 | - | 14 | 7.0 | - | |
Unknown | 30 | 5 | - | 25 | 5.0 | - | 5 | 4 | - | - | 20 | 5.0 | - | 10 | 2.0 | - | |
Substance use | 353 | 62 | - | 287 | 62.5 | - | 66 | 58 | - | 0.390 | 281 | 73.0 | - | 72 | 38.0 | - | < 0.01 |
Married | 64 | 11 | - | 14 | 12.0 | - | 50 | 11 | - | 0.739 | 30 | 8.0 | - | 34 | 18.0 | - | < 0.01 |
Employed | 68 | 12 | - | 50 | 11.0 | - | 18 | 16 | - | 0.144 | 47 | 12.0 | - | 21 | 11.0 | - | 0.69 |
Disability grant | 211 | 37 | - | 173 | 38.0 | - | 37 | 32.5 | - | 0.279 | 128 | 34.0 | - | 83 | 44.0 | - | 0.02 |
Psychosis | 517 | 90 | - | 411 | 89.5 | - | 105 | 92 | - | 0.487 | 354 | 92.0 | - | 163 | 85.0 | - | < 0.01 |
First-episode psychosis | 159 | 28 | - | 122 | 27.0 | - | 37 | 33 | - | 0.198 | 107 | 28.0 | - | 52 | 27.0 | - | 0.92 |
Violence | 211 | 37 | - | 167 | 36.5 | - | 44 | 39 | - | 0.745 | 159 | 41.5 | - | 53 | 28.0 | - | < 0.01 |
Mania | 114 | 20 | - | 90 | 20.0 | - | 24 | 21 | - | 0.793 | 63 | 16.0 | - | 51 | 27.0 | - | < 0.01 |
Depression | 41 | 7 | - | 27 | 6.0 | - | 14 | 12 | - | 0.025 |
8 | 2.0 | - | 33 | 17.0 | - | < 0.01 |
Self-harm | 48 | 8 | - | 36 | 8.0 | - | 12 | 10.5 | - | 0.450 | 32 | 8.0 | - | 16 | 8.0 | - | 1.0 |
Full medical | 509 | 89 | - | 406 | 88.5 | - | 103 | 90 | - | 0.662 | 336 | 88.0 | - | 174 | 91.0 | - | 0.26 |
Syphilis | 49 | 9 | - | 37 | 8.0 | - | 12 | 11 | - | 0.456 | 26 | 7.0 | - | 23 | 12.5 | - | 0.04 |
HIV tested | 504 | 89 | - | 410 | 91.0 | - | 94 | 83 | - | 0.027 |
377 | 67.0 | - | 189 | 33.0 | - | 0.04 |
HIV-positive | 70 | 13 | - | 59 | 13.0 | - | 11 | 10 | - | 0.426 | 18 | 5.0 | - | 52 | 28.0 | - | < 0.01 |
, Indicates statistical significance at
The majority of patients (80%) were admitted from the urban referral centres (
Men represented the majority (67%) of the cohort and, compared to women, were significantly more likely to have used substances (73% vs. 38%), present with psychosis (92% vs. 85%), have violence contribute to their admission (42% vs. 28%), and be tested for HIV (67% vs. 33%) (
Of all patients, 234 (41%) were from Mitchells Plain, 213 (37%) were from Khayelitsha and 126 (22%) were from other rural referral sites (
Demographic and clinical characteristics of patients from Mitchells Plain and Khayelitsha.
Variables | Overall ( |
Area |
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Mitchells Plain ( |
Khayelitsha ( |
|||||||||
% | IQR | % | IQR | % | IQR | |||||
Median age | 29 | - | 25–29 | 31 | - | 26–40 | 28 | - | 23–37.5 | 0.02 |
Male | 303 | 68 | - | 149 | 64.0 | - | 154 | 73.0 | - | 0.06 |
- | - | - | - | - | - | - | - | - | ||
Primary (Grades 1–7) | 112 | 25 | - | 68 | 29.0 | - | 44 | 21.0 | - | 0.13 |
Secondary (Grades 8–12) | 286 | 64 | - | 138 | 59.0 | - | 148 | 69.5 | - | |
Tertiary | 24 | 5 | - | 13 | 6.0 | - | 11 | 5.0 | - | |
Unknown | 25 | 6 | - | 15 | 6.0 | 10 | 5.0 | - | ||
Substance use | 276 | 62 | - | 141 | 60.0 | - | 135 | 63.0 | - | 0.56 |
Employed | 48 | 11 | - | 30 | 13.0 | - | 18 | 8.5 | - | 0.17 |
Violence | 166 | 37 | - | 94 | 40.0 | - | 72 | 34.0 | - | 0.20 |
Disability grant | 170 | 39 | - | 106 | 46.0 | - | 64 | 30.0 | - | < 0.01 |
Married | 50 | 11 | - | 39 | 17.0 | - | 11 | 5.0 | - | < 0.01 |
Psychosis | 403 | 90 | - | 205 | 88.0 | - | 198 | 93.0 | - | 0.08 |
FE psychosis | 118 | 26 | - | 50 | 21.0 | - | 68 | 32.0 | - | 0.01 |
Mania | 89 | 20 | - | 45 | 19.0 | - | 44 | 21.0 | - | 0.72 |
Depression | 27 | 6 | - | 18 | 8.0 | - | 9 | 4.0 | - | 0.16 |
Self-harm | 34 | 8 | - | 20 | 8.5 | - | 14 | 7.0 | - | 0.48 |
Full medical | 396 | 89 | - | 202 | 86.0 | - | 194 | 91.0 | - | 0.136 |
Syphilis | 35 | 8 | - | 20 | 9.0 | - | 15 | 7.0 | - | 0.60 |
Tested for HIV | 400 | 91 | - | 205 | 89.5 | - | 195 | 92.0 | - | 0.32 |
HIV-positive | 59 | 14 | - | 22 | 10.0 | - | 37 | 18.0 | - | 0.02 |
, Indicates statistical significance at
Violence was found to be a contributing factor to admission in 37% of the studied patients (
Demographic and clinical characteristics of patients where violence was reported.
Variables | Violence |
|||||||
---|---|---|---|---|---|---|---|---|
No ( |
Yes ( |
|||||||
% | IQR | % | IQR | |||||
30 | - | 25–42 | 28.0 | - | 24–34 | 0.09 | ||
Male | 224 | 58.5 | - | 159 | 41.5 | - | < 0.01 |
|
Female | 137 | 72.0 | - | 53 | 28.0 | - | ||
Primary (Grades 1–7) | 93 | 63.0 | - | 54 | 37.0 | - | 0.54 | |
Secondary (Grades 8–12) | 226 | 62.0 | - | 136 | 38.0 | - | ||
Tertiary | 25 | 73.5 | - | 9 | 26.5 | - | ||
Unknown | 17 | 57.0 | - | 13 | 43.0 | - | ||
Yes | 195 | 55.0 | - | 158 | 45.0 | - | < 0.01 |
|
No | 166 | 75.5 | - | 54 | 25.0 | - | ||
Yes | 42 | 66.0 | - | 22 | 34.0 | - | 0.68 | |
No | 319 | 63.0 | - | 190 | 37.0 | - | ||
Yes | 48 | 71.0 | - | 20 | 29.0 | - | 0.18 | |
No | 313 | 62.0 | - | 192 | 38.0 | - | ||
Yes | 128 | 61.0 | - | 82 | 39.0 | - | 0.47 | |
No | 233 | 64.0 | - | 130 | 36.0 | - | ||
Yes | 319 | 62.0 | - | 197 | 38.0 | - | 0.08 | |
No | 42 | 74.0 | - | 15 | 26.0 | - | ||
Yes | 101 | 63.5 | - | 58 | 36.5 | - | 0.92 | |
No | 260 | 63.0 | - | 154 | 37.0 | - | ||
Yes | 70 | 62.0 | - | 43 | 38.0 | - | 0.83 | |
No | 291 | 63.0 | - | 169 | 37.0 | - | ||
Yes | 32 | 72.0 | - | 9 | 22.0 | - | 0.04 |
|
No | 329 | 62.0 | - | 203 | 38.0 | - | ||
Yes | 26 | 54.0 | - | 22 | 46.0 | - | 0.21 | |
No | 335 | 64.0 | - | 190 | 36.0 | - |
, Indicates statistical significance at
Out of the 573 studied patients, the majority (62%) had used substances during the course of their current illness (
Demographic and clinical characteristics of patients who used substances prior to admission.
Variables | Overall ( |
Area |
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Mitchells Plain ( |
Khayelitsha ( |
|||||||||
% | IQR | % | IQR | % | IQR | |||||
29 | - | 24–40 | 37 | - | 28–37 | 27 | - | 23–33 | < 0.01 |
|
Male | 383 | 100.0 | - | 102 | 27.0 | - | 281 | 73.0 | - | < 0.01 |
Female | 190 | 100.0 | - | 118 | 62.0 | - | 72 | 38.0 | - | |
Primary (Grades 1–7) | 147 | 100.0 | - | 60 | 41.0 | - | 87 | 59.0 | - | < 0.01 |
Secondary (Grades 8–12) | 363 | 100.0 | - | 127 | 35.0 | - | 236 | 65.0 | - | |
Tertiary | 34 | 100.0 | - | 22 | 65.0 | - | 12 | 35.0 | - | |
Unknown | 30 | 100.0 | - | 12 | 40.0 | - | 18 | 60.0 | - | |
Urban | 459 | 100.0 | - | 172 | 37.5 | - | 287 | 62.5 | - | 0.39 |
Rural | 114 | 100.0 | - | 48 | 42.0 | - | 66 | 58.0 | - | |
Yes | 64 | 100.0 | - | 38 | 59.0 | - | 26 | 41.0 | - | < 0.01 |
No | 509 | 100.0 | - | 182 | 36.0 | - | 327 | 64.0 | - | |
Yes | 68 | 100.0 | - | 27 | 40.0 | - | 41 | 60.0 | - | 0.90 |
No | 505 | 100.0 | - | 193 | 38.0 | - | 312 | 62.0 | - | |
Yes | 211 | 100.0 | - | 105 | 50.0 | - | 106 | - | - | < 0.01 |
No | 362 | 100.0 | - | 115 | 32.0 | - | 247 | - | - | |
Yes | 517 | 100.0 | - | 193 | 37.0 | - | 324 | 63.0 | - | 0.09 |
No | 56 | 100.0 | - | 27 | 48.0 | - | 29 | 52.0 | - | |
Yes | 159 | 100.0 | - | 44 | 28.0 | - | 115 | 72.0 | - | < 0.01 |
No | 414 | 100.0 | - | 176 | 42.5 | - | 238 | 57.5 | - | |
Yes | 212 | 100.0 | - | 54 | 25.5 | - | 158 | 74.5 | - | < 0.01 |
No | 361 | 100.0 | - | 166 | 46.0 | - | 195 | 54.0 | - | |
Yes | 114 | 100.0 | - | 57 | 50.0 | - | 57 | 50.0 | - | < 0.01 |
No | 459 | 100.0 | - | 163 | 35.5 | - | 296 | 64.5 | - | |
Yes | 41 | 100.0 | - | 26 | 63.0 | - | 15 | 37.0 | - | < 0.01 |
No | 532 | 100.0 | - | 194 | 36.5 | - | 338 | 63.5 | - | |
Yes | 48 | 100.0 | - | 17 | 35.0 | - | 31 | 65.0 | - | < 0.01 |
No | 525 | 100.0 | - | 203 | 39.0 | - | 322 | 61.0 | - | |
Yes | 510 | 100.0 | - | 188 | 37.0 | - | 322 | 63.0 | - | 0.03 |
No | 63 | 100.0 | - | 32 | 51.0 | - | 31 | 49.0 | - | |
Yes | 49 | 100.0 | - | 12 | 24.5 | - | 37 | 75.5 | - | 0.05 |
No | 524 | 100.0 | - | 208 | 40.0 | - | 316 | 60.0 | - | |
Yes | 505 | 100.0 | - | 192 | 38.0 | - | 313 | 62.0 | - | 0.69 |
No | 68 | 100.0 | - | 28 | 41.0 | - | 40 | 59.0 | - |
, Indicates statistical significance at
Cannabis was the most widely used substance (51%), followed by methamphetamine (36%) (
Relationship between specific substances, referral centre and gender.
Variables | Overall ( |
Referral centre |
Gender |
|||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Urban ( |
Rural ( |
Males ( |
Females ( |
|||||||||
% | % | % | % | % | ||||||||
Substance use | 353 | 62.0 | 287 | 62.5 | 66 | 58 | 0.39 | 281 | 73 | 72 | 38 | < 0.01 |
Methamphetamine | 204 | 36.0 | 168 | 37.0 | 36 | 32 | 0.33 | 163 | 43 | 41 | 22 | < 0.01 |
Cannabis | 288 | 50.0 | 234 | 51.0 | 54 | 47 | 0.53 | 239 | 62 | 49 | 26 | < 0.01 |
Methaqualone | 81 | 14.0 | 60 | 13.0 | 21 | 18 | 0.18 | 71 | 19 | 10 | 5 | < 0.01 |
Heroin | 16 | 3.0 | 14 | 3.0 | 2 | 2 | 0.55 | 13 | 3 | 3 | 2 | 0.29 |
Alcohol | 100 | 17.5 | 81 | 18.0 | 19 | 17 | 0.89 | 73 | 19 | 27 | 14 | 0.16 |
, Indicates statistical significance at
The aim of our study was to establish the profile of adult admissions to the acute wards at LGH. Most patients were men (67%), had a secondary level of education (63%), were unemployed (88%), had a history of concurrent substance abuse (62%) and/or concurrent violence (37%), and presented with psychotic symptoms (90%).
A previous study found that patients admitted to LGH between 01 August 2012 and 31 January 2013 were also mostly men (65.6%) and aged younger than 35 years (58%).
The SACENDU data indicate high levels of substance use, especially illicit drug use, among the general population within the LGH drainage regions.
Psychosis was the predominant finding in more than 90% of the cases, with a slightly higher prevalence among men. First-episode psychosis constituted only 28% of all admissions. This might suggest that the current community mental health care system is failing to maintain remission. This could be due to ill-equipped primary health care services and a lack of psychosocial rehabilitation facilities.
Notably, 37% of the overall admissions in the current study were found to have displayed violent behaviour during the period leading up to, or upon, admission. In contrast, Chaput et al. reported that only 7.4% of visits to participating psychiatric emergency services in Quebec, Canada were marked by current aggression.
Rothärmel et al. found substance use and the severity of psychotic symptoms to be significant factors associated with violent behaviour in psychotic patients.
In terms of the 72-h observation requirements of the
Combined unemployment statistics for Khayelitsha and Mitchells Plain showed an unemployment rate of 32% in 2014.
The LGH catchment area is characterised by severe socio-economic challenges. Violence and other contact crimes abound within its communities.
Comorbid medical illness also requires further consideration in our psychiatric population. Although uniform testing for HIV is not included in standard operating procedures at LGH, this is advised where there is a high index of suspicion based on clinical grounds. HIV infection brings with it other concerns, such as associated medical complications and related cost of care.
A limitation of this retrospective study is that our data depended on the accuracy of the original source, as reliance was almost solely placed on self-reported data and collateral information. This is particularly relevant to our substance use data, as very few patients had toxicology done at referral centres. The sample period of 6 months may also limit the accuracy of data, especially regarding seasonal variance, but we believe a large cohort allows for sufficient generalisation of data. Lastly, the retrospective nature of our study limits the inference of causality.
Our study highlighted the profile of patients referred to LGH and identified salient factors impacting care. In describing the profile of acute adult patient admissions to LGH, we identified substance use and a propensity for violence as significant factors influencing the likelihood of admission. These factors place strain on available resources, and complicate treatment. This study thus argues for the expansion and capacitation of mental health services particularly at the tertiary level in the Western Cape and potentially across the broader mental health platform in South Africa. As this study may be suggestive of particular medico-legal risk at both corporate and clinical governance levels at medical institutions, we invite planners within the Department of Health, as well as other stakeholders in the government, to take heed of this burgeoning crisis and implement specific strategies for addressing these problems, before the effectiveness of mental health services as a whole is further undermined.
We acknowledge Dr Alfred Musekiwa (Centre for Evidence-based Health Care, Stellenbosch University) for his help with data analysis. We also thank Dr Karis Moxley (Department of Psychiatry, Stellenbosch University) for critical feedback, writing assistance and technical editing.
The authors have declared that no competing interests exist.
H.F. collected all data and wrote the manuscript. R.W. conceived and supervised the project and cowrote the manuscript. R.A. and J.P. provided critical feedback and contributed to the final version of the manuscript.
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Data sharing is not applicable to this article as no new data were created or analysed in this study.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.