sql优化，表优化

执行计划输出：SELECT STATEMENT, GOAL = CHOOSE                  12 1 189
 SORT ORDER BY                                  12 1 189
  NESTED LOOPS                                   9 1 189
   NESTED LOOPS                                   9 1 185
    HASH JOIN SEMI                           8 1 32
     TABLE ACCESS FULL NEWDOC T_INHOSPITAL_ACTION           2 218 4142
     VIEW SYS VW_NSO_1                           5 274 3562
      SORT GROUP BY                           5 274 4658
       TABLE ACCESS FULL NEWDOC T_INHOSPITAL_ACTION   2 274 4658
    TABLE ACCESS BY INDEX ROWID NEWDOC T_IN_PATIENT           1 1 153
     INDEX UNIQUE SCAN NEWDOC PK_T_IN_PATIENT                 58
   INDEX UNIQUE SCAN NEWDOC PK_T_SICKBEDINFO         750 3000
以下是sql部分
select 
a.id,a.patient_name,a.gender_code,a.birthday,a.pid,a.age,a.sick_doctor_id, a.sick_doctor_name,a.sick_area_id,a.sick_area_name,a.section_id, a.section_name,a.in_time,a.state,a.sick_bed_id,a.sick_bed_no,a.age_unit, a.Sick_Degree,
a.NURSE_LEVEL,a.insection_id,a.insection_name,a.in_area_id, a.in_area_name,a.Die_flag,a.Sick_Group_Id,Marriage_State,Medicare_NO,Home_Address, Homepostal_Code,Home_Phone,Office,Office_Address,Office_Phone,Career,Relation_Name,Relation,
 Relation_Address, Relation_Phone,RelationPOS_Code,OfficePOS_Code,InHospital_Count,Birth_Place,Folk_Code, CERT_ID,Pay_Manner,IN_Circs,native_place,a.card_id,a.patient_Id 
from t_in_patient a  
inner join t_inhospital_action b 
on a.id=b.patient_Id
inner join t_sickbedinfo c
on a.sick_bed_id=c.bed_id 
where a.section_id=5251 and  b.action_state=4 and 
b.id in (select max(id) from t_inhospital_action group by patient_Id) 
order by cast(sick_bed_id as number)
2个关联的表有索引了。查询数据只有13行，速度比较慢。