diff --git a/.env.sample b/.env.sample
index 418957b..fafb8f9 100644
--- a/.env.sample
+++ b/.env.sample
@@ -4,3 +4,5 @@ FLYWHEEL_GROUP=""
 ORTHANC_CREDENTIALS=""
 ORTHANC_HOST=""
 ORTHANC_PORT=80
+PHI_BUCKET_NAME="phi-data-bucket"
+SUBJECT_ID_MAPPING_PATH="s3://phi-data-bucket/subject_id_mapping.csv"
diff --git a/README.md b/README.md
index b44488b..f41bd7c 100644
--- a/README.md
+++ b/README.md
@@ -31,6 +31,8 @@ Then, customize its contents with a text editor:
 - For `FLYWHEEL_GROUP`, specify either `d3b` or an alternative group created for testing (e.g., your name).
 - For `ORTHANC_CREDENTIALS`, use your Orthanc username and password specified like `username:password`.
 - For `ORTHANC_HOST`, specify the hostname (minus `http(s)://`) that you use to access Orthanc.
+- For `PHI_DATA_BUCKET_NAME`, specify the bucket name where the ETL should backup NIfTI files.
+- For `SUBJECT_ID_MAPPING_PATH`, specify the [path](https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html) to the CSV file containing subject ID mappings.
 
 Next, run `update` to build the container image and initialize the database:
 
diff --git a/deployment/terraform/batch.tf b/deployment/terraform/batch.tf
index 468fb18..6760208 100644
--- a/deployment/terraform/batch.tf
+++ b/deployment/terraform/batch.tf
@@ -104,6 +104,9 @@ resource "aws_batch_job_definition" "default" {
     orthanc_host        = var.orthanc_host
     orthanc_port        = var.orthanc_port
 
+    phi_data_bucket_name    = var.d3b_phi_data_bucket_name
+    subject_id_mapping_path = var.subject_id_mapping_path
+
     image_deid_etl_log_level = var.image_deid_etl_log_level
   })
 }
diff --git a/deployment/terraform/job-definitions/image-deid-etl.json.tmpl b/deployment/terraform/job-definitions/image-deid-etl.json.tmpl
index e1ee71a..185f297 100644
--- a/deployment/terraform/job-definitions/image-deid-etl.json.tmpl
+++ b/deployment/terraform/job-definitions/image-deid-etl.json.tmpl
@@ -27,6 +27,14 @@
             "name": "ORTHANC_PORT",
             "value": "${orthanc_port}"
         },
+        {
+            "name": "PHI_DATA_BUCKET_NAME",
+            "value": "${phi_data_bucket_name}"
+        },
+        {
+            "name": "SUBJECT_ID_MAPPING_PATH",
+            "value": "${subject_id_mapping_path}"
+        },
         {
             "name": "IMAGE_DEID_ETL_LOG_LEVEL",
             "value": "${image_deid_etl_log_level}"
diff --git a/deployment/terraform/variables.tf b/deployment/terraform/variables.tf
index 6b156e0..67dfbf7 100644
--- a/deployment/terraform/variables.tf
+++ b/deployment/terraform/variables.tf
@@ -274,6 +274,10 @@ variable "orthanc_port" {
   type = number
 }
 
+variable "subject_id_mapping_path" {
+  type = string
+}
+
 variable "image_deid_etl_log_level" {
   type    = string
   default = "INFO"
diff --git a/src/image_deid_etl/image_deid_etl/__main__.py b/src/image_deid_etl/image_deid_etl/__main__.py
index 0d86d0d..6a6808e 100644
--- a/src/image_deid_etl/image_deid_etl/__main__.py
+++ b/src/image_deid_etl/image_deid_etl/__main__.py
@@ -4,7 +4,6 @@
 import os
 import sys
 import tempfile
-import time
 
 import boto3
 import flywheel
@@ -31,6 +30,11 @@
         "You must supply a valid Flywheel group in FLYWHEEL_GROUP."
     )
 
+PHI_DATA_BUCKET_NAME = os.getenv("PHI_DATA_BUCKET_NAME")
+if PHI_DATA_BUCKET_NAME is None:
+    raise ImproperlyConfigured(
+        "You must supply a valid S3 bucket in PHI_DATA_BUCKET_NAME."
+    )
 
 # Configure Python's logging module. The Django project does a fantastic job explaining how logging works:
 # https://docs.djangoproject.com/en/4.0/topics/logging/
@@ -229,7 +233,9 @@ def add_fw_metadata(args) -> int:
 
 def s3_backup_niftis(args) -> int:
     local_path = f"{args.program}/{args.site}/"
-    s3_path = f"s3://d3b-phi-data-prd/imaging/radiology/{args.program}/{args.site}/"
+    s3_path = (
+        f"s3://{PHI_DATA_BUCKET_NAME}/imaging/radiology/{args.program}/{args.site}/"
+    )
 
     return os.system("aws s3 sync " + local_path + "NIfTIs/ " + s3_path + "NIfTIs/")
 
diff --git a/src/image_deid_etl/image_deid_etl/main_pipeline.py b/src/image_deid_etl/image_deid_etl/main_pipeline.py
index 78f6b13..473a471 100644
--- a/src/image_deid_etl/image_deid_etl/main_pipeline.py
+++ b/src/image_deid_etl/image_deid_etl/main_pipeline.py
@@ -1,9 +1,8 @@
-# NOTE: cbtn-all is HARD-CODED in cbtn_subject_info, will need to change this when ADAPT sets up routine CSV updating
 import logging
-import boto3
-import pandas as pd
-import io
 
+import pandas
+
+from image_deid_etl.exceptions import ImproperlyConfigured
 from image_deid_etl.external_data_handling import *
 from image_deid_etl.images_no_save import *
 
@@ -11,12 +10,12 @@
 
 todays_date = datetime.now().strftime('%Y-%m-%d')
 
-# point to cbtn-all CSV file from s3 using boto3 & default AWS profile
-table_fn = 'cbtn-all_identified_2022-03-17.csv'
-bucket_name = 'd3b-phi-data-prd'
-obj_path = f'imaging/{table_fn}'
-s3_client = boto3.client('s3')
-obj = s3_client.get_object(Bucket=bucket_name, Key=obj_path)
+SUBJECT_ID_MAPPING_PATH = os.getenv("SUBJECT_ID_MAPPING_PATH")
+if SUBJECT_ID_MAPPING_PATH is None:
+    raise ImproperlyConfigured(
+        "You must supply a valid string path in SUBJECT_ID_MAPPING_PATH."
+    )
+
 
 def subject_info(local_path, program, file_dir, validate=0):
     # site_name = local_path.split('/')[1]
@@ -29,7 +28,7 @@ def subject_info(local_path, program, file_dir, validate=0):
     if program == 'cbtn':
         # get CBTN Subject IDs
         try:
-            cbtn_all_df = pd.read_csv(io.BytesIO(obj['Body'].read()))
+            cbtn_all_df = pandas.read_csv(SUBJECT_ID_MAPPING_PATH)
         except IndexError as error:
             logger.error("Missing CBTN subject ID .csv file from internal EIG database: %r", error)
         try:
diff --git a/src/image_deid_etl/requirements.txt b/src/image_deid_etl/requirements.txt
index b326088..2a04b16 100644
--- a/src/image_deid_etl/requirements.txt
+++ b/src/image_deid_etl/requirements.txt
@@ -1,3 +1,4 @@
+boto3==1.21.34
 flywheel-sdk==15.8.0
 nibabel==3.2.2
 numpy==1.22.3
@@ -8,6 +9,6 @@ psycopg2==2.9.3
 pydicom==2.2.2
 python-magic==0.4.25
 requests==2.27.1
+s3fs>=0.4.0 # Relaxing the version constraint to avoid botocore incompatibilities.
 sqlalchemy==1.4.32
 urllib3==1.26.8
-boto3==1.21.34
diff --git a/src/image_deid_etl/setup.cfg b/src/image_deid_etl/setup.cfg
index d1d8467..585ce0e 100644
--- a/src/image_deid_etl/setup.cfg
+++ b/src/image_deid_etl/setup.cfg
@@ -9,6 +9,7 @@ url = https://github.com/d3b-center/image-deid-etl
 packages = find:
 python_requires = >=3.9,<3.10
 install_requires =
+    boto3
     flywheel-sdk
     nibabel
     numpy
@@ -19,9 +20,9 @@ install_requires =
     pydicom
     python-magic
     requests
+    s3fs
     SQLAlchemy
     urllib3
-    boto3
 
 [options.entry_points]
 console_scripts =