The Quality of Referral Letters in Two Health Centers in Riyadh

Introduction

Referral is the process of one physician requesting another physician to examine a patient to obtain advice or management ^[1]. It is an important aspect of patient care, but if the quality of patient care is to be guaranteed, it is essential that referral be accomplished successfully. Important steps in this process include: communication of the need and purpose of the referral to the consultant, communication of the consultant's findings and recommendations to the referring physician, and mutual decision on continuing care ^{[2, 3]}. Failure in communication can occur at any of these steps.

The referring physician may fail to phrase a clear- cut question ^[4], inadequate information may be provided ^[5], the consultant may not address the referring physician's question ^[4,6], he may fail to communicate his findings to the referring physician ^[1,3], or he may make recommendations that are not clear or may even be inappropriate ^[6,7]. In 1984, the Ministry of Health in the Kingdom of Saudi Arabia adopted primary health care as an important objective of the fourth five-year plan (1985-1990) ^[8]. Successful primary health care services require complete coordination between the primary, secondary (general hospitals), and tertiary (specialist hospitals or specialist units in general hospitals) levels of care through a structured referral system, aimed at offering the best possible and most cost-effective services to all citizens. In Saudi Arabia, the obligatory referral system was implemented in all regions to improve the organization of services and hence provide better patient care.

There are predesigned standard forms for referrals which obtain important relevant clinical and social information, along with brief notes on the investigations done and the resulting findings and recommendations ^[9]. No attempt has been made to evaluate the referral system even though evaluation is an integral part of the managerial process. However, improvement in the quality of patient care depends on a referral system that is correctly used by both the primary care physician and the consultant. Therefore, the current study examines some of the problems related to this process, focusing on the quality of referral letters and feedback reports.

METHODS

Two health centers were selected for the study. Diriyah Health Center (DHC) is located on the outskirts of Riyadh, north of the city. It serves a catchment area with a total population of 12,940. The other center, Rawada Health Centre (RHC), is located in the eastern part of Riyadh. It serves a catchment area with a total population of 26,965. DHC was chosen because.it is located outside Riyadh and has a small catchment population, while RHC has a large catchment population and is located inside Riyadh. Both centers had been using the referral system for one year and both are located about the same distance from their referral hospitals. Both health centers also follow the same guidelines for referral.

A review of the records of referrals during the month of May 1989, corresponding to one year from establishment of the obligatory referral system, was conducted for the two centers. Of 221 referrals from DHC, 57 referral letters were chosen randomly, compared with 43 letters from among 225 referrals at RHC, all during the month of study. After a random start, a systematic sampling method was used to select one in every four letters from DHC and one in every five from RHC.

Data on a number of items, including the demographic data of the patient referral, the consultant to whom he was referred, and the interval between referral and the actual appointment with the consultant, were recorded. In addition, seven components that were considered an essential part of an ideal referral letter were documented; these included: present history, past medical history, results of physical examination, previous diagnostic tests, primary health care physician's diagnosis, previous therapy, and reason(s) for referral. The quality of each referral letter was graded by granting one point for each correctly provided component, with a total score ranging from 0 to 7.

Similarly, the presence or absence of six components in the feedback reports was also noted; these included: history of the problem, result of physical examination, diagnosis, decision on future patient's care, statement about follow up of the patient, and whether the recommendation was clear. Each feedback report was also graded on the basis of one point for each correctly done entry, with a total score ranging from 0 to 6.

The data were processed on a microcomputer using a statistical package (SYSTAT) to test statistical significance of results. The letters were later scored as poor or good, based on the scores. The chi-square test was used to test for significant differences between the two centers.

RESULTS

A total of 100 referral letters and consultation reports from the two health centers were studied: 57 from Diriyah and 43 from Rawada. The referral rate was 4.3 per 100 patient visits for DHC and 4.8 per 100 patient visits for RHC during the month of study. Regarding the important reasons for referral, the most frequent are for general therapeutic advice (43.2%), general diagnostic evaluation (24.8%), routine specialty examination (11.7%), and sophisticated tests not available in the centers (11.7%); the reasons were vague and nonspecific in 17.6% of the referrals from DHC. A patient could also have more than one reason for referral. Patients were referred to 12 different specialists. The most frequent were obstetric/gynecology (26%), dermatology (15%), ophthalmology (12%), otolaryngology (9%), and general medicine (9%). There was no significant difference between the two health centers in this regard. Of the referrals from DHC, 24.6% were to pediatricians, while the corresponding rate from RHC was only 2.3%. However, this difference may be due to the fact that 21% of the patients seen at RHC were children as compared with 40% at DHC. Most patients (82.1%) were seen by the consultant within less than one week of referral; 36.9% were seen on the same day and 45.2% within one week. These results are presented in Table 1.

Table 1. Interval between referral and appointment with consultants in the two health centers.

The legibility of the letters and reports was also evaluated. The letters were poor in 31%, fair in 37%, and good in 32%. Legibility of the consultation (feedback) reports was good in only 9%, fair in 51%, and poor in 40%. None of the letters or reports were typewritten. The frequency of specific components of a standard referral letter are shown in Table 2. Information on the present history was available in 96.2%, reason for referral in 85.7%, physical examination findings in 76.4%, and diagnosis in 60.2%. Past history was available in only 25.4% and a statement about current or previous therapy in only 13.9%.

Table 2. Frequency distribution of referral letters according to relevant items in primary health centers.

There are significant differences between the two centers with regard to information given on past history, examination findings, diagnosis, therapy, and reason for referral (P <0.05).

The frequency of important components contained in consultation reports for the two health centers is compared in Table 3. The most frequent information recorded was final diagnosis (75.8%), while the least recorded was the present history (19.9%). There was a significant difference between the two health centers only in relation to information given on physical examination (P < 0.01). Seventy-four percent of the referral letters were rated good, compared with 53% of the feedback reports. Findings are presented in Table 4 and significant differences are seen in the quality of letters for the two centers (P < 0.01).

Table 3. Frequency of components of referral feedback reports.

Table 4. Quality of referral letters unit reports in two health centers.

DISCUSSION

In this study, the rate of feedback from consultants was only 40% for DHC and 34% for RHC. Published studies have reported consultation feedback rates ranging from 50 to 88% in different settings ^[2,3,4,6]. There are standard forms for referral letters for use in the two health centers. Despite this, certain important clinical information was lacking. For example, past medical history and some statement about current or previous therapy were mentioned in only a few letters. Referring physicians are expected to supply the consultant with all relevant clinical information, and the consultant is required to provide feedback for the referring physician if the quality of a patient's care is to be maintained ^[2].

Although the reason for referral was available in most letters from DHC, it was vague and nonspecific in about 20%. This deficiency may influence the consultant's opinion and lead to misinterpretation of the referral information. The referring physician should furnish the consultant with clear and specific questions. Also·, even though the majority of patients were referred for management, final diagnosis was not specified by the consultant in about a quarter of the letters and the findings from physical examination were mentioned in less than half the letters. The consultant may deem this less important, but the primary care physician needs information relevant to the patient's condition. Without such information, the referring physician is hampered in dealing with the problem, and the educational purpose of the referral process is hindered. Decision on who should care for the patient was not specified in 56% of all the referrals in our study, and instructions for follow up with the consultant were given in only half of the reports. There is evidence that consultants’ recommendations are considered if they are provided ^[10]. The primary care physician, consultant, and patient all share responsibility for continuing care. In the current study, consultants failed to make their recommendations clear in more than half of the reports. Consultants’ recommendations are more likely to be followed if they are limited in number, clear, specific, and focus on issues central to current patient care ^[5–7]·

The quality of referral letters and reports was further evaluated by scoring each component. The overall frequency of poor referral letters was 26% while the overall frequency of poor reports was 47%, though this was less so for the DHC than the RHC. On the other hand, the frequency of poor consultant reports was higher for the RHC. This finding is similar to that reported elsewhere, that poor referrals yield poor consultant reports ^[6]. Several factors may have contributed to the poor quality of the letters and reports in our study. The system is new and the physicians were not sufficiently oriented to its objectives and the process itself. The lack of appropriate means of communication between the centers and hospitals adds to the problem. The consultants in the hospitals did not seem to recognize the importance of providing the primary care physician with sufficient and appropriate information. Both parties did not seem to appreciate the importance of complete referral forms and the negative effect on patients’ care. Regular meetings between the primary care physician and consultant should be held to discuss feedback on the referral system and its problems. This will improve the interaction between the primary care center and the referral hospital. Reorientation of physicians to the system is also mandatory. Consultants should also hold certain clinical sessions in the primary health care centers. However, work on a large sample is urgently needed to investigate the influence of physicians’ personal characteristics and other factors on the quality of the referral. There were few referrals for each referring physician, which ruled out study of the effect of their individual demographic characteristics on the quality of referral letters.

If the quality of referral, and hence patient care, is to improve, the primary care physicians must provide the consultants with enough relevant information. Consultants must in turn answer the questions raised and provide the referring physicians with clear-cut, specific, and appropriate recommendations. Further studies are needed to establish to what extent communication between the primary and secondary care providers is a problem, and appropriate solutions to such problems need to be specified.