Induced prescriptions requested by specialists at reference hospitals have been a well known phenomenon for years in most European Community countries. In 1991 induced prescription was defined as "any prescription generated by another physician or requested by the patient, which the general practitioner agrees to write although he or she does not agree with the diagnosis or therapeutic indication; in the case of [prescriptions requested by] specialist physicians, [it is understood that] the general practitioner did not schedule the appointment with the patient."1 Several Spanish studies have tried to quantify and evaluate prescriptions induced by hospital physicians that are written by primary care physicians. Induced prescriptions have been estimated to account for 11.6% to 35.8%2-5 of all prescriptions written by family doctors. However, the considerable variability in the methods used in published studies makes their results difficult to compare, although all studies reveal a situation that has arisen because of the central role of the primary care physician as the agent who writes the patient's prescriptions in order to ensure continuity of care.
Evidently, prescriptions generated at a different level of care can involve high costs for the family physician both in terms of resources (time needed to see the patient and write the prescription) and in terms of emotional wear, especially when the primary care physician does not agree with the prescribed treatment, with the need for treatment, or when he or she feels that the information provided by the specialist is insufficient to ensure adequate follow-up of the patient.
In this connection the study by Fernández Liz and colleagues centers on induced prescriptions requested by reference hospital physicians, and provides results in three areas that deserve attention:
1. A relatively low level of disagreement between the family physician and the hospital physician over the treatment prescribed (total disagreement =5%, partial disagreement =14%), and a high percentage of cases (63%) in which the family physician agreed completely with the prescribed medication. It should be noted that the authors provided no information on the degree of agreement for the remaining 18% of the prescriptions.
2. Perception by participating family physicians of the quality of information provided by the specialist for follow-up of the patient. The diagnosis was not indicated for 11.2% of the patients, and the duration of treatment was not specified for 45% (seen mainly in outpatient clinics).
3. Identification of differences between the active principles prescribed for hospital inpatients (only those included in the hospital formulary) and those prescribed for outpatients.
It should be born in mind that as in earlier studies, the results are not easy to extrapolate or generalize because of the methodological characteristics of the study and because of the characteristics of the participating health centers (proximity to a third-level reference hospital, urban health centers, lack of continued care services at some centers). Moreover, 30% of the family physicians at the health centers included in the study declined to participate. However, no details are given on their reasons for refusing, their main characteristics, or the possible differences between this subgroup and the population of primary care physicians at participating centers as a whole.
Nevertheless, the results of the study, although they cannot be generalized, make it possible to quantify a feature that should be of concern not only to family physicians but also to health administrators.
In view of these data, it is evident that not all induced prescriptions are undesirable, although a small percentage (5%-14%) give rise to considerable dissatisfaction on the part of family physicians and their patients. However, these figures are not high enough to dispute the importance of centralized prescribing by family physicians as a measure aimed at maintaining continuity of care, although they do reveal the need for mechanisms that would make it possible for specialists at reference hospitals to share responsibility for the prescriptions they generate. This evidently requires the joint preparation of hospital guidelines for the use of medications for outpatients in collaboration with family physicians at centers in the hospital's catchment area, and common policies to ensure the quality of prescribing practices. Also needed are measures to foment the use of new information and communication technologies at the primary care-hospital interface, the use of a unique, machine-readable medical record for each patient, and the establishment of electronic prescriptions on a national level.
Key Points
* Spanish studies have estimated induced prescriptions to account for 11.6% to 35.8% of all prescriptions written by family doctors.
* Although a small percentage of induced prescriptions (5%-14%) give rise to dissatisfaction on the part of family physicians, these figures are not high enough to dispute the importance of centralized prescribing by family physicians.
* The joint preparation by hospital and family physicians of guidelines for the use of medications, and good communication between primary care and hospital physicians, can improve the situation.
