Prescribing pattern of psychotropic medications in child psychiatric practice in a mental referral hospital in Botswana

Introduction There is a growing preference for psycho-pharmacological therapy over non-pharmacological care. The prescription pattern and the choice of psychotropic medications vary in different settings. Whilst newer agents and rational prescribing are favored in the more specialized settings, the pattern remains unclear in less specialized units, largely due to lack of data. The aims were to conduct a treatment audit in the only mental referral hospital in Botswana, which is a non-specialized child and adolescent care setting and see how it conforms to best practice. Methods A retrospective audit which involved the extraction of socio-demographic and clinical information from the records of patients who were ≤ 17 years and seen from January 1, 2012-July 31, 2016. Results A total of 238 files were used for this report. Mean age (SD) was 12.41 (4.1) years. Of the 120 (50.4%) patients who had pharmacological intervention, only 85(70.8%) had monotherapy. The most commonly prescribed psychotropic agents were antipsychotics (40%). Off-label use of antipsychotics and polypharmacy were 31.2% and 29.2% respectively. Conclusion The level of conformity to standard practice in terms of psychotropic prescribing in our setting is consistent with the reports from developed countries where more specialized care ostensibly exists. Further studies will be necessary to determine the scope of psychotropic use.


Introduction
Psychotropic medications use other than stimulants have not been adequately studied in children in terms of benefit and safety. Their use in children is mostly based on extrapolation of information from adult studies [1]. There is a growing preference for the use of psychotropic medications over other non-pharmacological interventionsin child and adolescent mental care [1][2][3]. Factors majorly responsible for this preference include severity or type of psychiatric disorder, the need for quick symptom relief and lack of multidisciplinary care, particularly in the low resourced countries [1,2]. Patterns of prescribing and the choice of psychotropic medications vary in different settings [4]. Whilst newer agents and rational prescribing (justifiable use) are favored in specialized settings [1][2][3]5], the pattern remains largely unclear in settings without specialized care units for children, mainly due to lack of data [6]. In Europe and America for instance, where specialized child and adolescent health care services are available, psychosocial management is more preferred, especially for mild to moderate conditions [1][2][3]5]. Newer agents such as Risperidone, olanzapine, fluoxetine are preferred to older ones like haloperidol and amitriptyline [1,2,7]. Nonetheless, they have also reported increase in rate of off-label use of psychotropic medications (psychotropic prescription outside its approved indications) [2,7,8] and polypharmacy among children and adolescent in these centers [2,[9][10][11]. Studies conducted in Africa among adult population, though scanty, have likewise shown a high rate of off-label use of psychotropics low preference for newer agents due to cost and polypharmacy [12,13]. It is however not clear the extent to which these apply to children and adolescents in centers without specialized care. Botswana spends about 5.8% of its Gross Domestic Product (GDP) on health and provides free medical for all the citizens, but only one percent of its total health budget goes to mental health care [14]. It operates three tiers of health care system (primary, secondary and tertiary), with mental health being part of all levels but has only one mental referral (tertiary) hospital which majorly provides services for adult and some children with mental conditions. Child and adolescent services are thus being provided by general adult mental specialists who are currently less than 10 in a population of 2.1 million people [15] and no evaluation of the effectiveness of psychopharmacological therapies currently on offer has been done. We therefore decided to conduct a treatment audit in the mental tertiary hospital in Botswana, which is a non-specialized child and adolescent care setting, and see how it conforms to rational drug prescription pattern in the developed countries. It was also hoped that the findings of this study would lay a foundation for biomedical research in pediatric mental health prescribing in this setting.

Methods
We conducted a retrospective audit of 238 children and adolescents who attended Sbrana psychiatric hospital (SPH) from the first of January, 2012 when the hospital record could easily be accessed to the thirty-first of July, 2016 when we decided to conduct this audit.
Location of study: SPH is located in Lobatse, South-Western district of Botswana and is staffed by six (6) general adult psychiatrists, ten multi-drug treatment when more than two medications were given at the same time. This pro-forma was designed by the researchers based on previous studies [2,16]. Data extraction was done by two psychiatrists from the hospital who agreed on every information needed before they were recorded in the pro-forma. However, files which information could not be agreed on or with incomplete documentation on variables of interest were excluded from the analysis to minimize coding bias. for sleep related problems, while promethazine as pro re nata was administered to 3(2.5%) patients during in-patient care ( Table 2).
The rate of 'off-label' prescribing among 48 who had antipsychotic medication (measured by use for unlicensed purposes) was 54.2%.

Discussion
The  [18]. Antipsychotic prescription outside its approved indications (off-label) was remarkable among others, accounting for 54.2% of its use and is comparable to a report from more specialized child care setting [2], thus suggesting that the practice is not restricted to settings without specialized care. The practice of using medications outside their licensed indications was previously known in adult; it is currently becoming a common practice in pediatric population [7,24]. Disorders such CD and hyperkinetic disorder, which are externalizing in nature, are often misdiagnosed as psychotic disorders thus informing the use of antipsychotics [2].
For example in the current study, 13 (29.5%) of those who had hyperkinetic disorder were given antipsychotics in the course of management while half of the 44 patients with CD equally had antipsychotics. Nevertheless, only one 1 patient had co-occurring psychosis and CD, while none had hyperkinetic disorder and psychosis. It is therefore possible that these disorders were previously misdiagnosed, hence the reason for being given antipsychotic trial.
Depot antipsychotic (flupenthixol) was used in three (4.9%) cases as in a previous study [2], but was restricted to older teens who had compliance issues, though it has not been licensed for use below the age of 18 years [2]. Benzodiazepines (which was used for sleep problems) prescribing was exclusively off-label, accounting for 11.6% of psychotropic use in the current study similar to a report from the United Kingdom [2]. Psychotropic medications e.g., antipsychotics and benzodiazepines have been used for various purposes for which they are not approved in psychiatry with good results. These include agitation, aggression,obsessive compulsive disorder, posttraumatic stress disorder and sleep problems [2,7,8]. There is therefore a need for more research into other potential benefits of psychotropic medications especially as it affect pediatric mental health.
Polypharmacy, defined as the prescription of two or more medications concurrently to a patient [9], is another notable feature in our study, as 29.2% rate was found. This is not peculiar to our center as authors from the developed [2,9,10] and developing countries [13,25] have reported an increasing trend of multidrug treatment among children with mental conditions. In this audit, the most common pharmacological pair were antipsychotics plus stimulants and anticonvulsants plus stimulants, while the most common three combinations were antipsychotic-anticonvulsant-stimulant. Albeit the rationale behind the clinical decisions concerning psychotropic coprescribing was not routinely documented, these combinations suggest that disorders which often presented with aggressiveness and externalizing behavioral symptoms are mostly responsible for polypharmacy in our setting as reported by an earlier author [10].
Polypharmacy is sometimes associated with risks which include adverse effects of drug interactions and increased toxicity, especially when liver enzyme inhibitors are part of the combinations [9]. This practice may not necessarily be an improper prescribing pattern as it is useful in some clinical situations, such as, the treatment of adverse effect of another agent, co-existing condition e.g., seizure and psychosis, immediate relief of symptoms while waiting for the main medication to act and so on. A consensus statement issued by some Psychiatrists in the United Kingdom rationalizes some cases of transitory polypharmacy, comprising making a gradual change from one psychotropic to another (Royal College of Psychiatrists, 1993) [9]. Moreover, one author is of the opinion that the "demerits of polypharmacy are not contained in 'where' the drugs are being used but in 'how' these drugs are being used" [9], further buttressing the fact that rational (justifiable use) polypharmacy could be beneficial. A new generation of research is required to guide optimal medication management and to identify the clinical situations where psychotropic co-prescribing is superior to monotherapy in terms of clinical outcomes.

Competing interests
Authors declare no competing interests.

Authors' contributions
AA conceived the idea; AA was part of data collection; AA, JA and PO drafted the manuscript. All authors read and approved the manuscript.

Acknowledgments
Special thanks to Dr. Frank-Hatitchki Bechedza, a consultant psychiatrist at SPH for her assistance in data extraction, the record staffs and the management of SPH for allowing this work to be a reality and the reviewers of this manuscript.