Use of flow-charts by nurses dealing with mental patients: an evaluation in Lesotho

Bulletin of the World Health Organization, 66 (4): 507-514 (1988) © World Health Organization 1988 Use of flow-charts by nurses dealing with mental patients: an evaluation in Lesotho K. MEURSING1 & V. WANKIIR2 General nurses without previous training in psychiatry were instructed, over a period of thirteen hours, in the use of eightflow-charts for the identification and management of mental health conditions. They then prepared management plans, with the aid oftheflow- charts, for 105 patients with suspected mental health problems who were seen in three outpatient clinics. The same patients were also seen by trained mental health workers who made a diagnosis and wrote up a management plan, which subsequently was compared with the management plans devised by the nurses. Seventy-eight of the 105 patients (74%) were identified and treated correctly by the nurses. A total of 32 mistakes were made, 17 of which were due to the nurses and 15 to defects in the flow-charts. On the basis of these findings, suggestions are made for improvements in the flow-charts and in the way the nurses are instructed in their use. INTRODUCTION In recent years, attention has been drawn to the considerable number of mental health patients who have no access to any kind of psychiatric service in the developing countries. It is estimated that at least 1% of any population at any time is suffering from a seriously incapacitating mental disorder such as schizophrenia and affective disorders, while at least 10% of the population would have been affected by such a condition at some time in their life (5). The extent of other forms of mental disorder (psycho- neuroses, emotional disorders, and personality prob- lems) is more difficult to define, but there is no evidence to support the view that such disorders are less common in developing countries than elsewhere. Studies show that between 13 % and 20% of the out- patients attending general medical clinics in devel- oping countries are in fact suffering from a mental health problem (1, 2). Mental health care services and specialized mental health workers are rare in developing countries, or only available in the urban areas. In these countries, it is not uncommon to find one psychiatrist serving a population of one or two million people. In the last ten years WHO, as well as other agencies and individuals, has been urging governments, health planners and administrators to decentralize the mental X Clinical Psychologist, formerly at Scott Hospital, Morija, Lesotho; now at Mpilo Hospital, Bulawayo, Zimbabwe. Requests for reprints should be sent to K. Meursing, 18 Clark Road, Suburbs, Bulawayo, Zimbabwe. 2 Psychiatric Nurse Tutor, Maseru, Lesotho. health care services and to integrate them as much as possible with the general health services (2, 5)." In practical terms this means that a large part of mental health care must be rendered by front-line health workers, i.e., health workers without -specialized training in or knowledge of mental health. The role of the trained mental health workers such as psychi- atrists, clinical psychologists, psychiatric nurses, and psychiatric social workers would then be to deal with referred cases, and to instruct, supervise, and support the front-line workers. In response to this new approach, efforts have in recent years been directed towards developing simple and practical methods to help front-line health workers to identify and manage a range of priority mental health conditions effectively.b One such effort, by Essex & Gosling, was the development of flow-charts for the identification of mental health problems (3, 4). These charts are problem-oriented and do not aim to provide a diagnosis, but with their aid the front-line health workers should be able to recognize the presenting problems and provide im- mediate appropriate treatment and follow-up in the community. The charts also enable the health worker to select patients who need to be referred to a higher level and to indicate, with reasonable accuracy, whether the presenting problem is caused by a physi- cal or mental illness. a Mental health. Report of the WHO Regional Expert Panel on Mental Health. Brazzaville, 1979 (AFRO Technical Report Series No. 7, unpublished document). b Training in mental health for primary health care workers. Report of a Workshop, March-April 1981. Brazzaville, WHO Regional Office for Africa (unpublished document AFR/MH/12). 4909 507- 508 K. MEURSING & V. WANKIIRI The object of the present study was to evaluate the usefulness of these flow-charts in actual practice. Specifically, the aims were to determine: whether general nurses with no training in psy- chiatry could be taught to use these charts; -whether a management plan, devised by a nurse with the help of the flow-charts, differed significantly from a treatment plan developed by a trained mental health worker using a psychiatric approach. MATERIALS AND METHODS The use of the flow-charts was taught to ten general nurses who had no previous training in psychiatry or psychiatric nursing, after which each of them devised a management plan, with the help of the charts, for a total of 105 patients with a suspected mental health problem who were seen in three outpatient clinics. Each of these plans was compared to the corres- ponding plan devised by a trained mental health worker (a psychiatrist, a psychiatric nurse, or a general doctor with knowledge and experience in psy- chiatry), who had seen the same patient on the same day. Discrepancies between the two plans were rated according to their importance; in cases of serious mistakes the cause was sought, i.e., whether it was due to misapplication by the nurse or to an inherent defect in the flow-chart. Flow-charts and evaluation forms By using the flow-charts developed by Essex & Gosling (3), every patient with a disturbed mental state or behaviour can be classified into one of the following eight categories: (1) "Violence to others", (2) "Violence to self', (3) "Delusions (including hallucinations)", (4) "Withdrawal", (5) "Abnormal speech", (6) "Abnormal behaviour", (7) "Anxious- ness", and (8) "Depression". A basic rule governing the use of these charts is that the health worker must fit each case or problem into the first appropriate category indicated in the charts, as recommended in the original flow-charts. The flow-charts used in our study were specially adapted to local practices and resources. Our first chart, on attempted violence to others (Fig. 1), com- pared with the one by Essex & Gosling, shows an additional box and an extra item (1.3) for violence associated with hallucinations or delusions. Fig. 2 shows our flow-chart concerned with delusions (in- cluding hallucinations).c c Because of space limitations, the remaining six flow-charts (on "Violence to self", "Withdrawal", "Abnormal speech", "Over- activity or unusual behaviour", "Anxiousness" and "Depres- sion") are not included in this article. The original models, on which they are based, will be found in the publication by Essex & Gosling (3). To compare the two management plans, prepared by the nurses using the flow-charts and by the trained mental health workers (MHW), evaluation forms were constructed on which each of them indicated how the patient they saw should be managed and treated. The main points for comparison were: -the flow-chart box selected by the nurse and the diagnosis made by the MHW; -type of drug treatment recommended; -any need for treatment of a physical disease; -whether social management (e.g., counselling for alcohol and drug problems, or for other serious problems) was needed; -need for referral. The researcher also noted down on each evaluation form the following information: -a short history of the illness, the presenting problems, and why the patient decided to go to the clinic; -a description of the manner in which the nurse selected the flow-chart box. Scoring the differences Differences between the assessments by the nurse and the MHW were scored according to the serious- ness of the risk to the patients or to those around them, as follows: -complete agreement between the two was scored as "1"; -a score of "2" was given if the difference posed no risk to the patient or those around them, or where there was no essential difference in the treatment (e.g., when different drugs were chosen but both belonged to the same group, such as major tranquil- lizers); -a score of "3" was given when the difference posed a serious risk to the patient and those around them, such as when the nurse proposed no education on alcohol use for a patient who was judged by the MHW to be suffering from alcohol-related problems, or if the drugs prescribed belonged to entirely dif- ferent groups. Selection ofpatients Patients in this study were first-time attenders or had not attended the clinic for at least one year, and none of them was already under psychiatric treat- ment. These patients presented with a wide variety of problems (Table 1). Criteria for exclusion from the research were: -age younger than 10 years; USE OF MENTAL HEALTH FLOW-CHARTS BY NURSES Fig. 1. Violence to others: flow-chart on presenting problems and action to be taken Presenting problems Immediate action Follow-up Attempted violence to others 1.1 * Record symptoms and other necessary details * Treat injuries if present * Observe and restrain patient until effects of alcohol have gone * Police help may be needed 1.2 * Record symptoms and other necessary details * Restrain and make continuous observation until effects of drug have gone * Treat injuries if present 1.3 * Record symptoms and other necessary details * Give thioridazine or chlor- promazine 100 mg twice or three times daily * Refer. If not possible: -admit for observation OR ask relatives to observe patient at home -give above-mentioned drugs to relatives with instructions on use by the patient 1.4 * Record symptoms and other necessary details * Restrain patient if still violent * Give minor tranquillizers (e.g., diazepam or chlordiazepoxide) * Make continuous observation, restrain and support * Treat injuries if present 1.5 * Treat injuries if present * Give minor tranquillizer (e.g., chlordiazepoxide or diazepam) * Inform doctor * Police help may be needed * Re-examine patient after effects of alcohol have gone and use flow-charts for any other symptom or abnormal behaviour * Educate patient and family about dangers of excess alcohol * Re-examine patient and use flow- charts for any other symptom or abnormal behaviour when effects of drug have gone * Educate patient and family about dangers of drug abuse * See patient in one week * Provide continuous support during the period of grief * Get village health worker, religious leader and/or others to give support at home * Discuss with village elders how to control this behaviour when it arises -patients with grand mal epilepsy but without any other psychiatric problems; -patients receiving psychiatric treatment. but one gave up because she found the course "too difficult". The course, which made use of the modi- fied charts from the originals by Essex& Gosling (3), lasted an average of 13 hours and comprised the fol- lowing four stages: VJ n&&I3Ova (1) for explaining the method of classifying a pre- The nurses were taught the use of the flow-charts in senting problem into the first appropriate category small groups of 2-5 persons. Originally ten started, (3 hours); 509 [netrwrtinn nf nlxreve K. MEURSING & V. WANKIIRI Fig. 2. Delusions (including hallucinations): flow-chart on presenting problems and action to be taken Presenting problems Immediate action Follow-up Holds incredible beliefs OR sees or hears things others cannot (not related to culture) 3.1 * Diagnose and treat cause of fever 3.2 * Admit to clinic No * If symptoms still present after one day, refer. If not possible: - manage as for 3.4 but Suspected of taking drugs known to Yes without referral cause these symptoms 0 No| Onset after tragic event OR emotional shock OR severe stress Yes 3.3 * Record symptoms and other within past four weeks necessary details * Interview members of family about cause of emotional upset * If patient can be treated at home, give chlorpromazine or thioridazine to relatives and instruct them how to give it to the patient (100 mg twice No or thrice daily) * If treatment at home is difficult or risky, refer to doctor, hospital or M.O.U.' * Help patient and family to overcome shock * Find others in the community to rsupport the patient and family Work or family life affected Yes 3.4 * Record symptoms and other necessary details * Give a major tranquillizer (i.e., chlorpromazine or thioridazine) if agitated * Continuous observation * Refer to doctor, hospital or M.O.U. If not possible, instruct relatives to observe patient at home No * Give relatives thioridazine or chlorpromazine and instruct how to give it to the patient (100 mg twice or thrice daily) * Consider contacting other community members who might be of help 3.5 * Reassure patient and family that the problem is not serious * Educate about dangers of drug and alcohol abuse * See regularly at home or in clinic * If patient is better, consider if drugs can be stopped * Refer as soon as possible * See patient at least once weekly * If patient does not reattend, visit at home * See patient in three weeks * Re-use chart and if still at 3.5, then discharge e Mental observation unit. 510 USE OF MENTAL HEALTH FLOW-CHARTS BY NURSES Table 1. Distribution of diagnostic categories, by sex, among the 105 patients in the study, as determined by the trained mental health workers No. of No. of males females Total Psychoses: Alcoholic psychosis 10 4 14 Alcohol/marihuana psychosis 6 0 6 Marihuana psychosis 6 0 6 Paranoid psychosis 4 2 6 Dementia 3 3 6 Reactive psychosis 1 5 6 Schizophrenia 1 2 3 Suicide attempt, psychotic state 0 2 2 Febrile hallucinations 2 0 2 Others (pellagra, puerperal, 3 4 7 epileptic psychosis) Total 36 22 58 Non-psychotic complaints: Alcohol, other complaints 3 0 3 Reactive depression 5 15 20 Suicide attempt, reactive 1 2 3 Anxiety neurosis 2 6 8 Hysteria 0 3 3 Epilepsies 3 0 3 Impotence 2 0 2 Others 3 2 5 Total 19 28 47 (2) for working on "cases" from the manual and preparing the management plans (5 hours); (3) for using the flow-charts and evaluation forms in role play (2 hours); (4) using the flow-charts and evaluation forms on real patients and in contact with their families, under supervision (3 hours). Procedure during research A newly arrived patient with a suspected mental health problem would be presented to the nurse and the MHW on the same day, but separately. Both of them were given access to the same sources of infor- mation, such as family members or existing files. After interviewing the patient and, if considered necessary, performing a physical examination, the MHW and the nurse wrote up their management plans. The researcher then completed the evaluation form by briefly noting the history of the patient, and how the nurse arrived at her chosen management. RESULTS AND DISCUSSION The diagnoses on the 105 patients, as judged by the MHW are set out in Table 1. Of the total number of cases treated by the nurses, 78 were treated correctly; 32 mistakes (i.e., scores of "3") were made, of which 17 were judged to be due to an error by the nurses and 15 were caused by a defect in the charts (Table 2). Mistakes made by the nurses The errors were most often related to the decision on whether the patient was deluded and hallucinating, or not. In seven cases the nurses decided that the patients were hallucinating, while the MHW thought they were not; and in three cases the nurses over- looked hallucinations, which according to the MHW were present (Table 2). It seems that differentiating correctly between "normal" experiences and halluci- nations is a major problem for the nurses. For example, treatment for severe mental disorder was sometimes started unnecessarily because the patient reported hearing sounds of aeroplanes, running water, cars, and telephones ringing, or complained of "feeling something moving in the head", or said he felt bewitched. These expressions are very common in the culture of Lesotho in people who are anxious or Table 2. Analysis of mistakes made by nurses and those due to the flow-charts Total number of patients 105 No. correctly treated 78 No. of mistakes made 32 Mistakes by nurses in using the flow-charts: Identified hallucinations in error 7 Failed to identify hallucinations 3 Overprescribed drugs 3 Chose wrong box 2 Wrote wrong dose of drugs 2 Total 17 Mistakes due to the flow-charts: Hallucinations after alcohol withdrawal not recognized Pellagra not recognized Psychosis in suicide attempt not recognized Problems in dealing with relatively rare syndromesa Dosages of drugs too high 5 3 2 4 Total 15 a Temporal lobe epilepsy, epileptic psychosis, Parkinson's disease. 511 K. MEURSING & V. WANKHRI feeling unwell, and are not necessarily indications of severe mental disorder. It is no simple matter to distinguish correctly between culturally appropriate and inappropriate sensations. Since almost 60% of the mistakes made by the nurses were related to this difficulty, it seems worth while to devote extra time on this subject during instruction. Mistakes due to theflow-charts Fifteen mistakes were due to the flow-charts them- selves (Table 2). The problem most often encountered was in the third category concerning delusions (including hal- lucinations) that were due to recent alcohol use. The wording in the box ("smells of alcohol or known to have been drinking heavily today"), thus formulated, covers hallucinations caused by an acute overdose of alcohol, but not those due to alcohol withdrawal which can persist for weeks or months after the drinking has stopped (6). Fifteen hallucinating patients were judged by the MHWs to be suffering from such withdrawal symptoms. In ten of these cases, a discrepancy score of "3" was prevented because the nurses had selected a more or less appro- priate box (e.g., where hallucinations affected work and family life) and added "education on alcohol use" in their management plan on their own initiat- ive. In the remaining five patients the alcohol prob- lem was overlooked. A second defect in the flow-charts is that they do not provide a check for patients with pellagra, which in Lesotho is a frequent cause of psychiatric symp- toms. This disease was present in six patients, and had been overlooked in three (Table 3). Thirdly, in the flow-chart on "Violence to self' the box inquiring whether the violence was a reaction to a recent life problem precedes the box where the presence of hallucinations, delusions or bizarre behaviour is checked. Since the nurse must select the first appropriate box she meets, the psychotic state in two cases, in which both a reactive element and psychotic symptoms were present, was overlooked. Reversal of the sequence of these two boxes would solve this problem. An important issue is whether the nurses could correctly recognize a physical illness when this was the cause of the presenting problems. In 42 of our 105 patients such an underlying physical condition was present and 32 of them (76%) were correctly recog- nized (Table 3). The physical conditions missed were: -the above-mentioned patients with hallucinations due to alcohol withdrawal; -the above-mentioned pellagra cases; -two cases of temporal lobe epilepsy that resulted in outbursts of hallucinations. This investigation shows that physical conditions were mostly overlooked because of limitations in the flow-charts. It should be noted that our modification of the first flow-chart "Violence to others" before this investi- gation was started, i.e., by introducing the box con- cerning hallucinations and delusions (see Fig. 1), was a significant improvement. This box was used cor- rectly by the nurses in seven cases. Attitudes of the nurses towards theflow-charts In general, the nurses appreciated the charts and did not think the system was especially difficult. It was noted, however, that when they had not used the charts for some time, they soon tended to forget the basic rules. Whenever the nurses felt the charts were in some way deficient, confusion and demotivation were soon observed. CONCLUSIONS Our first objective was to decide whether general nurses, as a category of intermediate-level health worker, could be trained in the use of flow-charts. This proved to be the case in nine out of ten nurses (one gave up during the training); the relatively small number of their mistakes was evenly divided so that all nine of them performed reasonably well. The second objective was to determine whether the treatment proposed by a nurse using the flow-chart would differ significantly from treatment by a trained Table 3. Extent to which various physical conditions were correctly recognized Physical Total Correctly conditions present recognized Fever 2 2 Epilepsy 4 2 Concussion 1 1 Drug overdose 1 1 Pellagra 6 3 Alcohol and/or drug abuse 28 23 Total 42 32 512 USE OF MENTAL HEALTH FLOW-CHARTS BY NURSES mental health worker. Our nurses, who were instruc- ted in the use of the flow-charts for an average of thirteen hours, made 17 mistakes in the 105 cases. About 60% of these mistakes were related to the decision on whether the patient was hallucinating or not, which appeared to be complicated by cultural factors. It is worth while therefore to spend extra time during training on the distinction between culturally appropriate and inappropriate feelings and expressions. Fifteen mistakes were related to problems inherent in the flow-charts. For example, alcohol withdrawal hallucinations were not correctly identified in cases where alcohol had not been consumed for more than 24 hours, and pellagra as a cause of psychosis and abnormal behaviour could not be identified by using the charts. However, physical diseases underlying the mental health symptoms were correctly recognized by the nurses in 76% of the cases where they were present. Modifications in the charts to include alcohol withdrawal hallucinations and abnormal behaviour associated with pellagra would further improve the performance of the nurses. The use of these flow-charts therefore offers a quick and handy means for nurses and other health workers to deal with mental health patients. After their training, a prolonged period of supervision seems essential so that problems encountered in the actual use of the charts can be discussed. A steady flow of cases is necessary to keep the nurses in prac- tice; seeing only occasionally a patient with a mental problem may result in difficulties in using the charts and subsequent demotivation. Subjects for future research could be a follow-up of the nurses, after a long interval, to see whether the charts are still being used and how correctly, and a comparison of the performance of nurses after a short course of practical psychiatry with that of another group after an equal period of training in using the flow-charts. ACKNOWLEDGEMENTS We thank everybody who cooperated in this work, particularly Dr J. Orley, Dr L. Mohapeloa (Director of Mental Health Services, Lesotho), Mr L. D. Africa, Dr E. T. Vos, and all the nurses and others who gave their time and energy. This research was supported by a grant from WHO. RESUME UTILISATION DE DIAGRAMMES DE SANTE MENTALE PAR DES INFIRMIERES: EVALUATION AU LESOTHO Dans le cadre de cette etude, neuf infirmieres sur dix ont suivi un cours de breve duree pour apprendre a utiliser une serie de huit diagrammes de sante mentale s'inspirant de ceux qui avaient ete concus par Essex & Gosling en 1982. II s'agit d'une methode qui a pour objet de donner aux agents de sante de niveau intermediaire les connaissances et les competences necessaires pour reperer et prendre en charge correctement toute une serie d'affections mentales cou- rantes apres une breve periode de formation. Notre but etait de savoir: - s'il etait possible d'apprendre a des infirmieres non specialisees a utiliser correctement ces diagrammes; - si un plan de prise en charge fonde sur l'utilisation de ces diagrammes differait beaucoup d'un plan de prise en charge du meme patient propose par un agent ayant requ une for- mation specialisee en sante mentale. Le cours a dure 13 heures, a la suite de quoi neuf infir- mieres ont examine 105 patients qui avaient des problemes de sante mentale. Au total, 32 erreurs ont ete faites, dont 17 dues a une meprise des infirmieres et 15 a des lacunes dans les diagrammes. L'erreur la plus frequente de la part des infirmieres a consiste a interpreter comme signes de psychose des sensations et des experiences anodines et compatibles avec le contexte culturel. Des lacunes dans les diagrammes ont empeche les infirmieres d'identifier 1) les hallucinations provoquees par le sevrage alcoolique et 2) la pellagre comme cause de psychose et de comportement anormal. n faudrait modifier les diagrammes pour combler ces lacunes et insister particulierement pendant la formation sur la distinction entre les sensations et manifestations compatibles avec le contexte culturel et celles qui ne le sont pas. Une assez longue periode d'appui et d'encadrement des agents de sante qui utilisent les diagrammes est necessaire. 513 514 K. MEURSING & V. WANKIIRI REFERENCES 1. HARDING, T. W. ET AL. Mental disorders in primary health care-a study oftheir frequency and diagnosis in four developing countries. Psychological medicine, 10: 231-241 (1980). 2. WHO Technical Report Series, No. 698, 1984. (Mental health care in developing countries: a critical appraisal of research findings: report of a WHO Study Group). 3. ESSEX, B. & GOSLING, H. Programme for identifi- cation and management of mental health problems. Harlow, Churchill Livingstone, 1982 (Tropical Health Series). 4. ESSEX, B & GOSLING, H. An algorithmic method for management of mental health problems in developing countries. British journal ofpsychiatry, 143: 451-459 (1983). 5. WHO Technical Report Series, No. 564, 1975. (Organization of mental health services in developing countries: sixteenth report of the WHO Expert Com- mittee on Mental Health). 6. VICTOR, M. & ADAMS, R. D. Alcohol. In: Harrison's Principles of internal medicine, 8th edition. McGraw- Hill, 1977, pp. 707-716.