Question

假设我们有2个表：

Patients（身份证，姓名）
Patients_Treatments（PatientID，treatment_code）

我希望查询能够检索至少接受id='999999999'

我尝试了很多组合，没有任何效果，我得到的是至少有一种'9999999999'治疗方法的患者。

Answer 1

一种方法在having子句中使用自连接和比较：

select pt2.patientId
from patient_treatments pt join
     patient_treatments pt2
     on pt.treatment_code = pt2.treatment_code and pt.patientid <> pt2.patientid
where pt.id = '999999999'
group by pt2.patientId
having count(pt2.treatment_code) = (select count(*) from patient_treatments pt where pt.id = '999999999');

注意：此版本假定Patient_Treatments中没有重复项。

如果您在数据中有重复项，则可以使用count(distinct)：

having count(distinct pt2.treatment_code) = (select count(distinct pt.treatment_code) from patient_treatments pt where pt.id = '999999999');

Answer 2

这个想法是

（1）选择属于患者的治疗代码“999999999”

（2）仅选择治疗代码与属于患者的治疗代码之一匹配的治疗记录“999999999”

（3）通过patient_id对这些进行分组

（4）使用HAVING语句和COUNT（DISTINCT）仅选择与患者“999999999”具有相同数量的不同治疗代码的患者ID。

select pt.Patient_ID,count(distinct pt.treatment_code)
from
Patient_treatments pt
inner join
(select distinct treatment_code
 from 
 Patient_treatments pt
 where pt.Patient_ID="999999999"
 )t1
 on t1.treatment_code=pt.treatment_code
 where pt.Patient_ID<>"999999999"
 group by pt.Patient_ID
 having count(distinct pt.treatment_code)=
 (select count(distinct treatment_code) 
  from 
  Patient_treatments pt
  where pt.Patient_ID="999999999"
  );

使用以下模式，仅选择患者1： -

  Create table Patient_Treatments
  (
    Patient_ID varchar(10),
    Treatment_code varchar(10)
    );


   Insert into Patient_Treatments
    values
    ("1","abc"),
    ("1","def"),
    ("1","def"),
    ("1","ghi"),
    ("2","abc"),
    ("2","def"),
    ("2","def"),
    ("3","ghi"),
    ("999999999","abc"),
    ("999999999","def"),
    ("999999999","ghi"),
    ("999999999","ghi")

http://sqlfiddle.com/#!9/03a84b

协助SQL查询

2 个答案: