【背景・目的】当研究班が在宅医を対象に実施した質問紙調査（以下「在宅医調査」）では、複数疾患併存状態高齢者に関し在宅療養後方支援病院が提供する診療情報提供書の記載内容や情報共有方法について、病院医に対する多種多様な要望が寄せられた。しかしその中には病院医には提供困難な情報（例：増悪時の再入院確約）や、在宅医個人の医療観・力量に依存する要望（例：在宅診療への詳細な助言）が散見されたことから、病院医側の視点での精査を目的とするインタビュー調査を実施した。
【方法】機縁法により協力を得た病院医に対し40-70分の半構造化面接を実施した（慶應義塾大学看護医療学部研究倫理委員会承認307）。在宅医調査結果（診療情報提供書記載事項・情報共有手段、病院医への要望全般）を提示し意見を尋ねた。
【結果】複数疾患併存状態の在宅移行期高齢者に関する診療情報提供書を作成した経験のある病院医7名（男性6名・女性1名）に協力を得た。医学部卒後5-40年、首都圏在勤、専門は総合診療科、総合内科、家庭医療科、リハビリテーション科／神経内科、血液内科、救急科、心臓血管外科であった。
在宅医調査結果に対する研究協力者（病院医）の意見は、①病院医も在宅医同様記載すべきと考える情報（例：インスリン量）、②必要性は理解するものの現実的には取得が難しい情報や他職種が提供すべき情報（例：住居環境）、③診療科により捉え方の異なる情報（例：輸血を拒否する宗教的背景について一般内科系医師と心臓血管外科・血液内科の医師では対応が異なる）、④病院医と在宅医で重要性の捉え方が異なる情報（例：患者家族関係）の４通りに大別された。
【考察・結論】
３年間にわたる本研究においては、当初の目的を病院医向けテンプレートの策定としていたが、今回の調査により病院医・在宅医間のみならず、各診療科間に見解の相違が見られたことを踏まえ、汎用型ではなく、診療科毎にある程度特化したチェックリストの方が有用と思われる。また病院医・在宅医の意見の相違の背景には、専門性や医師としての能力・経験に加え、病態や患者心情・家族関係を静的（＝退院時から不変）・動的（＝刻一刻と変化する）いずれの要因として理解しているかの違いが反映されているように思われる。インタビュー対象を拡大し上記２点の検証を行うことが本研究の次の課題である。
Background and Purpose: In a previous questionnaire survey of home care physicians, we received a wide variety of requests toward hospital physicians regarding the types of information provided to them of elderly multimorbidity patients. However, some of the requests included information difficult for hospital physicians to provide (e.g., assurance of readmission) while others appeared to depend on the medical views and clinical competence of individual home care physicians (e.g., detailed advice on home medical care). To examine the legitimacy of these requests from hospital physicians' points of view, an interview study was conducted with hospital physicians with experience in providing medical information of multimorbidity elderly patients to home care physicians.
Methods: Semi-structured interviews lasting 40-70 minutes were conducted with hospital physicians recruited through a snowball sampling method (Keio University Faculty of Nursing and Medical Care Research Ethics Committee Approval, #307). The results of the questionnaire survey were presented to the participants and their opinions were asked.
Results: Seven hospital physicians (six male and one female) participated in the study: they have 5-40 years of clinical experience, work in the Tokyo metropolitan area, and specialize in general internal medicine, family medicine, rehabilitation medicine/neurology, hematology, emergency medicine or cardiovascular surgery.
The study participants' comments on the questionnaire responses can be roughly divided into the following categories: a) hospital physicians agree with home care physicians in that particular types of information should always be shared (e.g., insulin level); b) hospital physicians understand the need for but feel that certain types of information are hard to obtain / provide (e.g., home environment); c) perceived importance of certain types of information varies across medical departments (e.g., religious reasons to refuse blood transfusion); and d) hospital and home care physicians disagree on the importance of certain types of information (e.g., patient-family relationships).
Discussion and Conclusion: The current research project had initially set out to propose a template for hospital physicians to use when providing medical information of multimorbidity elderly patients to home care physicians. However, based on the fact that the survey revealed differences in opinions not only between hospital physicians and home care physicians but also across medical departments, a checklist tailored for the needs of individual departments could be more practical and useful. In addition, the differences in opinions between hospital and home care physicians seem to reflect different views of pathological conditions, patient sentiments, and family relationships. Home care physicians seem to regard them as relatively static (i.e., unchanging from the time of discharge) while hospital physicians see them as relatively dynamic (i.e., changing from moment to moment). Given that, the next step in this research project will be to expand the interview and verify these issues.