When we use DISTICT in JAVA SDK, we find that the index usage is 0. How should we optimize it?

[kkxuwork]

When we use DISTINCT in the JAVA SDK, we find that the index usage is 0. In fact, we have created a consistent index, as shown below, but we are still prompted about a potential composite index. How should we optimize it? Thanks

SQL: SELECT DISTINCT U.NAME, U.AGE, U.GENDER, U.SYSTEM,U.AMOUNT,U.INFOR FROM USER U WHERE U.REGION = 'CG' AND ARRAY_CONTAINS(['DTO','PAG'], U.BUSINESSUNIT) AND HBASE.IMPORTTIME >= '20240410' OFFSET 0 LIMIT 100;

compositeIndexes:
[
{
"path": "/REGION",
"order": "ascending"
},
{
"path": "/BUSINESSUNIT",
"order": "ascending"
},
{
"path": "/IMPORTTIME ",
"order": "ascending"
}
]

Metrics:
Index Utilization : 0.00 %
Potential Composite Indexes Metrics :
┌──────────────────────────┬────────────────────────┬──────────────────┐
│Index Document Expressions │Index Plan Full Fidelity │Index Impact Score│
├──────────────────────────┼────────────────────────┼──────────────────┤
│[/REGION ASC, /BUSINESSUNIT ASC, /IMPORTTIME ASC]│ false│ High│
└──────────────────────────┴────────────────────────┴──────────────────┘