PostgreSQL我有这样的表：

create table if not exists some_table
               (
        some_id         varchar(40)              not null,      
        
        created_at      timestamp with time zone not null,
        
        constraint some_pkey
            primary key (some_id)

下一步我在created_at字段上创建索引

create index concurrently if not exists some_table__created_at
        on statement (created_at);

我无法使用条件在created_at上创建部分索引，因为此条件包含可变值
如果我像这样使用查询：

EXPLAIN ANALYSE
select t1.created_at
FROM some_table t1
where t1.created_at < '2023-06-19 10:17:20.830627+00:00';

索引扫描正在工作

Index Only Scan using statement__created_at on statement t1  (cost=0.42..5991.71 rows=210245 width=8) (actual time=0.016..27.510 rows=210277 loops=1)
  Index Cond: (created_at < '2023-06-19 10:17:20.830627+00'::timestamp with time zone)
  Heap Fetches: 0
Planning Time: 0.098 ms
Execution Time: 37.884 ms

但我想得到字段some_id与使用索引扫描

EXPLAIN ANALYSE
select t1.some_id
FROM statement t1
where t1.created_at < '2023-06-19 10:17:20.830627+00:00';

现在我得到了

Seq Scan on statement t1  (cost=0.00..6535.50 rows=210245 width=37) (actual time=0.007..33.904 rows=210277 loops=1)
  Filter: (created_at < '2023-06-19 10:17:20.830627+00'::timestamp with time zone)
  Rows Removed by Filter: 2723
Planning Time: 0.099 ms
Execution Time: 44.463 ms

我尝试在2列上使用index（some_id和created）。对我没什么帮助。
也许我应该使用not btree索引或其他东西？