C. Davey, A. B. Desai, P. M. Shajahan
e-mail
polash.shajahan@lanarkshire.scot.nhs.uk
SMJ 2006 51(4): 49
We
aimed to assess the quality of out-patient clinic letters in relation to GPs
expectations. A questionnaire was devised and sent to 65 GPs to ascertain what
they felt to be important in a psychiatric out-patient clinic letter.
Introduction
Out-patient
letters between secondary and primary care are an important method of
communication. Previous studies into this area have shown that GPs prefer short
letters conveying rapid information on the diagnosis or problems, management and
specific requests to them1 Structured
letters with headings are preferred by GPs 2,3 and improve comprehension, additionally they are found to take
no longer to read than unstructured letters. One very early study on the use of
structured letters found that the chief advantages of structured letters over
conventional letters were that the writer could state concisely the patient’s
problems and management, the reader could see at a glance what the writer’s
views were and that these letters were shorter 4.
Lanarkshire is a county in Scotland with a population of approximately 550,000. Psychiatric services are divided into three districts. Our local district has a population of approximately 200,000. This population generates secondary psychiatric case records for about 6000 people. The quantity of correspondence generated by this population in terms of letters from psychiatrists to GPs is considerable. In this study we asked GPs how they rated key points contained in an out-patient clinic letter including how they rated a formatted letter pattern. We compared these findings with the analysis of the most recent out-patient clinic letters of 250 psychiatry patients in order to find out how successful we were at providing the information GPs wanted in these letters.
Methods
Survey
of General Practitioners
A questionnaire was sent out to 65 GPs in our district asking if the following were necessary, beneficial or not necessary within a review letter: Formatted letter pattern, clinic date, diagnosis/problems, update of symptoms, mental state examination, opinion/summary, medication, dose of medication, non-medical recommendations and follow-up.
Below the questionnaire a space was allocated for any comments, which the responding GPs may have had regarding this issue. In addition the responding GP was invited to provide his or her details, for example, name, address, although this was optional.
Examination
of Psychiatric Records
Using an Electronic Document Management System where all medical and nursing correspondence in secondary psychiatric services are archived, the most recent out-patient clinic letter of 250 patients was examined. Whether or not a formatted letter pattern was followed was noted. A formatted letter was considered to be one that covered specifically the headings mentioned in the questionnaire and was laid out in a pre-determined template. Non-formatted letters were examined to establish whether they mentioned, at any time, information related to the above-mentioned headings. Non-medical recommendations included for example, involvement or referral to Community Psychiatric Nurses, Social Work Departments or Psychology.
The word count for each letter was also recorded starting from the name of the addressee and ending with the signatory’s name. This was to find whether there was a relationship between letter format, content and length.
Results
Table 1: GP Questionnaire
|
|
NECESSARY |
BENEFICIAL |
NOT
NECESSARY |
|
|
|
|
|
|
FORMAT |
25% |
50% |
25% |
|
DATE |
92% |
8% |
0% |
|
DIAGNOSIS/PROBLEMS |
97% |
3% |
0% |
|
UPDATE
OF SYMPTOMS |
65% |
35% |
0% |
|
MENTAL
STATE EXAM |
49% |
43% |
8% |
|
OPINION/SUMMARY |
92% |
8% |
0% |
|
MEDICATION |
100% |
0% |
0% |
|
MEDICATION
DOSE |
100% |
0% |
0% |
|
NON
MEDICAL RECOMMENDATIONS |
16% |
78% |
6% |
|
FOLLOW UP |
89% |
8% |
3% |
37 GPs (57%) responded to the questionnaire.
Table One shows that almost all GPs considered date, diagnosis, psychiatrist opinion, medication, medication dose and follow up necessary.
Figure 1: Content of correspondence
68% (170/250) of out patient clinic letters examined in this study were found to use a formatted letter pattern versus 32% (80/250), which were unformatted. All formatted letters included the date, patient’s diagnosis/problems, update of symptoms, medication list and follow up arrangements. The remainder of the headings were mentioned in almost all of the formatted letters except for non-medical recommendations which featured less frequently in both formatted and unformatted letters. The clinic date, update of symptoms, medication and follow up were mentioned in most unformatted letters. Mental State Examination and the consultant’s opinion of the patient featured in just over half of these. The patient’s diagnosis was documented in only one fifth of unformatted letters.
Few comments were returned with the questionnaire. These included positive feedback for our current letter format, a recommendation that some letters were too long and general advice regarding the length and the need for a concise letter.
Discussion
One of the main findings of this study was that formatted letters consistently out-performed unformatted letters in all areas.Not all GPs, however, considered a formatted letter to be necessary. A preferred out-patient clinic letter, according to all GPs, should contain the date, diagnosis/problems, update of symptoms, opinion, medication, medication dose, and follow up. On average this information was conveyed in 98.9% of the formatted letters compared to only 75.9% of the unformatted ones.
Surprisingly, only
1 in 5 unformatted letters included the diagnosis/problems.
The inclusion of the diagnosis/problems in each letter is not only a
reminder to the GP, but is informative to the clinician who has never met the
patient before 5, for example, out of hours doctors or doctors
covering a new clinic. Likewise, letters specifying mental state examination,
which was not considered to be important by all GPs, can also be a valuable
source of reference in these circumstances 6.
Non-medical recommendations were not considered to be necessary by GPs and this particular area was only referenced in an average of 66% of all letters. The psychiatric out-patient review letters which were sent to GPs are often copied to our nursing and other non-medical colleagues, for example community psychiatric nurses and psychologists. The non-medical advice from the psychiatrist may be beneficial for these non-medical and paramedical professionals.
Formatted letters have previously been described as being shorter than unformatted letters 4. However, our findings were that formatted letters were 30% longer than unformatted letters after correction for headings. We can explain this by the fact that the formatted letter, without fail, provides more information, therefore requiring more words to communicate this. One major advantage however, of the formatted letter is that the provision of headings allows the reader to quickly find the parts of the letter in which he or she is most interested and pass over parts of the letter which are less immediately important to him or her.
Our study found that two thirds of our out-patient letters were currently produced using the formatted template. Our findings show these are consistently more informative, providing information specific to the needs of the GP. This is worth considering in an attempt to give our letters clarity and to facilitate both our GP and hospital colleagues.
References
1. Melville C, Hands S. & Jones P. Randomised trial of the effects of structuring clinic correspondence. Archives of Disease in Childhood 2002, 86, 374-375.
2. Salathia K. & Mcilwaine WJ. General practitioner and hospital letters. Ulster Medical Journal1995, 64, 46-50.
3. Tattersall MHN, Butow PN, Brown JE. et al Improving doctors’ letters. Medical Journal of Australia 2002, 177, 516-520.
4. Rawal J, Barnett P & Lloyd BW. Use of structured letters to improve communication between hospital doctors and general practitioners. British Medical Journal 1993, 307, 1044.
5. Lloyd BW & Barnett P. Use of Problem Lists in Letters Between Hospital Doctors and General Practitioners. British Medical Journal, 1993, 306, 247.
6. Campbell B, Vanslembroek K, Whitehead E., et al Views of doctors on clinical correspondence: questionnaire survey and audit of content of letters. British Medical Journal 2004, 328, 1060-1061.