Referral letters to the psychiatrist in Nigeria: is communication adequate?

Objective: Over half of patients with psychiatric disorders are first seen by primary care physicians before referral to the psychi -atrist. An efficient referral system is, therefore, important to offering quality care to such patients. Communication between phy sicians is often sub-optimal and referral letters to specialists sometimes provide inadequate information. The current study’s aim was to observe the prevalence of deficiencies in the referral letters to a department of psychiatry in a Nigerian teaching hospital. Methods: All referral letters to the Department of Psychiatry, University College Hospital Ibadan, Nigeria over a three- year period from January 1, 2010, to December 31, 2012, were retrieved and assessed for quality using the Consultation and Referral Request Letter Assessment Tool. Result: A majority (>80%) of the referral letters had no information on the current medication list, relevant psychosocial history, outline of management to date, results of investigations to date, and known allergies. Conclusion: Deficits in communication or information transfer through referral letters to the psychiatrist are common. Inter ventions such as the use of standardized formats for such letters may facilitate more efficient communication.

Background: Psychiatric disorders contribute significantly to the total global burden of diseases 1 . Over half of patients with psychiatric disorders are first seen by primary care physicians before being referred to the psychiatrist. The reasons for referral include a doubtful diagnosis, poor results from treatment and emergency situations 2 . An efficient referral system is therefore not only important to offering quality care to such patients and their relatives, but also to the society by saving unnecessary waste of time and avoidable expenditure. Quality communication is crucial to a safe and efficient referral process. Furthermore, the referral process is a critical component of quality clinical care. An efficient referral system is a two-way process that involves the transfer of relevant clinical information from the referring physician (e.g. primary care physician) to another physician (e.g. specialist) and vice versa.
Vehicles for referrals take many forms (letters, telephone, e-mail, on-line link etc.) depending on the culture and geographic location. For example in many African and European countries, referral letters are the primary means of communication between primary care physicians and specialists whereas in the USA a phone referral may be more prevalent. It is crucial for a referring physician working in sub-Saharan Africa to know what information the referral letter should contain, since the literacy rate in the region is very low and patients may not be educated enough to inform doctors to whom they are referred about pertinent information such as medical illnesses they are being treated for or the medications they have been placed on.
Apart from information transfer, referral letters also serve as a means of educating physicians. The psychiatrist, for example, can serve as a teacher and consultant to the primary care physician since such a psychiatrist deals with serious mental and emotional problems that are beyond the skills of the primary care physician 3 . Collapse in communication through a poor referral system can lead to poor continuity of care, repetition of investigations, increasing poly-pharmacy, delayed diagnoses, increased health care costs and decreased quality of care 4,5 . Communication between the referring physicians and other physicians is often sub-optimal and referral letters to specialists sometimes provide inadequate amount of information [6][7][8] . It is not uncommon to get referrals with missing reports of previous investigations, insufficient detail in the referral letter, absence of an explanation for the referral or statements such as ''the above named is a known psychiatric patient He/She is being referred to you or ''Mr. X has been behaving abnormally He/she is being referred to you ''.
Even though proficiency in written communication is essential to the practice of physicians, most physicians do not receive any training in the art of writing referral letters9. Communication among physicians can always be improved upon and an excellent way to start is by measuring its deficiencies. Few studies have investigated the information content of doctors' letters, especially in sub-Saharan Africa. The current study's aim was to investigate the quality of referral letters to a department of psychiatry in a Nigerian teaching hospital.

Methods
This study was conducted at the University College Hospital (UCH), Ibadan Nigeria. UCH is an academic tertiary care teaching hospital and referral centre located in Ibadan, Southwest Nigeria. The estimated population of South West Nigeria is about 28 million 10. The hospital receives referrals specifically from health facilities in the South West geopolitical zone of Nigeria and frequently from all geopolitical zones of the country.
All referral letters to the Department of Psychiatry, University College Hospital Ibadan, Nigeria over a three-year period from January 1, 2010, to December 31, 2012, were retrieved from their case notes and assessed for quality using the Consultation and Referral Request Letter Assessment Tool 11 .The tool has 18 items that assesses content , style and overall appreciation of referral letters. The 18 items are divided into 15 content and 3 style items. It also includes a 5-point Likert scale to provide an overall rating for each referral letter. The current study only assessed the content of the letters. Data extracted included patient demographics, initial statement identifying the reason for the referral, past medical history, past surgical history, relevant psychosocial history, current medication list, allergies, relevant clinical findings, results of investigations to date, outline of management to date, provisional diagnosis or clinical impression, and a statement of what was expected from the referral.

Results
There were 284 referrals to the Department during this period. Over half,(58.1%) of the referrals came from the general out-patient department (i.e. referrals from the primary care physicians) 23.6 % of the referrals came from other hospitals Table 1  However they were less likely to contain items such as a description of relevant collateral history, the past medical history, relevant clinical findings, the current medication list, relevant psychosocial history, outline of management to date, results of investigations to date, allergies, and past surgical history. The results in the current study are similar to findings in some studies from other climes. For example in the study by Newton et al (1994), 95% of the referral letters indicated the reason for referral while we found 94.4%. Similarly, relevant psychosocial history was only mentioned in 11.9% of the referrals compared with 18% by Newton et al 12 . Also, our study showed that 20% of the referrals indicated the relevant medical history. This is lower than the 29-62% that has been previously reported [12][13][14] . Unlike the study by Newton et al where only 5% of the referrals indicated a statement of what was expected from the referral, about 70% of the referrals in the current study indicated a statement of what was expected from the referral 12 .
It is noteworthy that in 2.5% of the referrals, the source of referral was not indicated. This suggests that such patients were probably not attended to, since the source of referral is usually necessary to locate a patient in need of such psychiatric evaluation or treatment. If it is accepted that improvement in communication between the referring physician and the psychiatrist is desirable, then it is necessary to consider how to rectify the deficiencies indicated in the current study. An effective way to enhance information content and communication in referral letters to specialists is the use of form letters. [15][16][17] Form letters are letters written from templates, rather than being specially composed for specific recipients. They contain more information than non-form letters 17 and have been shown to improve the quality of referral letters 18 . In addition, physicians prefer to receive such letters as referrals 15 . In addition, because such letters use laid out titles, it allows the reader to easily identify the information desired, thus saving time for both reader and writer.
Another way in which communication using referral letters can be enhanced is for physicians to get trained in the art of referral letter writing. Attending letter-writing training courses can significantly improve the content of referral letters of doctors. Conducting a letter-writing course may, however, be an expensive intervention 7 . It may also be beneficial for such letter-writing courses to be introduced to the medical curriculum both at the undergraduate and postgraduate levels.
Even though our study was conducted in one specialty (psychiatry), it is unlikely that the referral letters to other specialties would be different from what has been observed in the present study. One limitation of this study is that it was done at a single large tertiary care teaching and referral centre. The problems faced by physicians at this site may be different from the challenges at smaller institutions. Furthermore, we did not look at the designation of the writers of the referrals. It would have been beneficial to segregate the letters into the different cadres of doctors in the teaching  hospital (i.e. specialist/ consultants, residents, medical officers, house officers).
However, often times in clinical practice consultants seldom write referrals. The younger colleagues in the team are usually saddled with that responsibility. It is however of strategic importance to mandate senior residents to cross check outgoing referrals on behalf of their consultants. Again, we described the deficiencies using a tool that was developed for use within a family medicine practice. This may not reflect informational elements that may be of more relevance to psychiatrists when receiving referrals. It would be beneficial to have a group of psychiatrists establish a checklist of items that they find particularly important for their practice as a tool for assessing the adequacy of the information contained in referral letters. For example, some of the missing elements that were discovered through this study, such as psychosocial history, may not be elements that non-psychiatrist physicians routinely collect during their examinations and, therefore, one would not expect such information to be available to be included in referral letters.

Conclusion
Deficits in communication or information transfer through referral letters to the psychiatrist are common. Interventions such as the use of standardized formats for such letters may facilitate a more efficient communication.