gms | German Medical Science

Background: Each year the regional representatives of the Statutory Health Insurances (SHI) and the Association of SHI Physicians in Germany negotiate the implementation of outpatient drug costs controlling on the federal state level. Due to the Supply Support Act (“Versorgungsstärkungsgesetz”) passed by the federal legislator in 2015 far-reaching reforms were possible. From 2017 on Morbidity Related Groups (MRG) are used for drug prescription management in Schleswig-Holstein. Prescription data structured by the international Anatomical Therapeutic Chemical Classification System (ATC) with German specifications due to admission law is the only basis of data necessary. The physicians are informed based on MRG considerations and drug advice is given to improve economic results as well as supply quality. The MRG system has to be as congruent as possible with the benefit assessment of innovative drugs by the Federal Joint Committee (G-BA), which is the highest decision-making body of the joint self-government of physicians, dentists, hospitals and health insurance funds in Germany.

Materials and methods: Prospective and retrospective calculations done for a period of one year are divided into quarters because of billing and other important analyses done quarterly. Approx. 12.5m patient-related prescriptions of >2m patients are combined with master data (pharmaceutical database of >200.000 products). Quarterly each patient of every physician is assigned to a unique basic MRG defined by the third level of the ATC (four characters= therapeutic/pharmacological subgroup) with the highest cost. In analogy to the DRG system in the hospital setting this basic MRG will be extended by a severity level depending on age group (five year steps), polypharmacy and prescription intensity separated by medical specialist groups. If the mean costs of all patients are above a certain threshold value, costs are used instead of prescription intensity due to large drug cost heterogeneity in a relatively small amount of basic MRG. If the drug expenditures of a patient are a fraction or multiple of 5 or 10 compared with the MRG mean value in the underlying group corrections for under- and overvaluations due to statistical outliers are applied for approx. 3% of the patients.

Results: Each physician received a risk adjusted guaranteed prospective MRG budget value for the drug prescriptions in 2017 as well as in 2018. If the patient structure changes (larger number/increased morbidity) the practice will get an increased retrospective drug budget value. If a physician has drug expenditures exceeding more than 107% of the risk adjusted value in the respective quarter and the guaranteed prospective drug budget value an audit by a consulting physician and a consulting pharmacist is offered. During the audit the risk adjusted groups of patients with higher expenses are discussed. This may change the prescription behavior of the physician or can give hints for an improvement of the MRG system. In outpatient treatment no main diagnoses are stated, there is only a differentiation between acute and continuous diagnosis as well as the degree of diagnoses confidence. Using the basic MRG one can select main diagnoses out of all documented diagnoses using increased and reduced risk-adjusted probabilities. Additionally one can get a hint of poor diagnostic quality if there is no match. This connection of MRG defined by the third level of ATC and diagnoses justifies the label “morbidity related”. The basic MRG patient types can be compared with the patient types of all medical specialist groups. The group most similar might be different to the group defined by admission law. In this case the MRG risk adjustment might lead to poor results and should be corrected by the drug economic examination institution (“Prüfungsstelle”). Alternatively one can consider the one to three most similar physicians for every physician and can use graph cluster methods to define medical specialist groups well adapted to prescription behavior.

Conclusion: The application of the MRG system leads to a much better risk adjustment compared to the predecessor based only on four age or three insurance status groups. Thereby activities for the improvement of drug economic results as well as quality of supply are more targeted. Additionally a best practice approach compared with an average group approach might improve the outcome, but there are several assessment criteria to determine “best” (from economic to guideline orientation, complicated by multimorbidity aspects). The number of patients assigned to an ATC groups is not equally distributed, in descending order it follows a gamma distribution. Depending on the affected patients the different resolution levels of the ATC system might be of use. Further improvements could be made by promotion or restriction of coefficients for MRG groups derived from economic evaluation or evaluation of the medical quality of supply. Innovations positively evaluated by the benefit assessment of the Federal Joint Committee (G-BA) are adequately rated in the MRG system, the same holds true for orphan drugs. A special strength of the MRG system is the patient orientation with respect to polypharmacy and multimorbidity in contrast to most guidelines. The MRG classification with respect to patients and physicians can be used in modern methods of statistics, differential equations and graph theory. The focus of the MRG system is the improvement of patient care not at least through better conditions for physicians.