diff --git a/src/stepcount/utils.py b/src/stepcount/utils.py
index 10f2d69..eb031f9 100644
--- a/src/stepcount/utils.py
+++ b/src/stepcount/utils.py
@@ -41,9 +41,9 @@ def read(
     - resample_hz (str, optional): The resampling frequency for the data. If 'uniform', it will use `sample_rate`
       and resample to ensure it is evenly spaced. Default is 'uniform'.
     - sample_rate (float, optional): The sample rate of the data. If None, it will be inferred. Default is None.
-    - csv_start_row (int, optional): The row number to start reading from (0-indexed, excluding header).
-      Only applies to CSV files. Default is None (read from the beginning).
-    - csv_end_row (int, optional): The row number to stop reading at, inclusive (0-indexed, excluding header).
+    - csv_start_row (int, optional): The file row number where the header is located (0-indexed).
+      Rows before this are skipped. Only applies to CSV files. Default is None (header at row 0).
+    - csv_end_row (int, optional): The file row number to stop reading at, inclusive (0-indexed).
       Only applies to CSV files. Default is None (read to the end).
     - csv_time_format (str, optional): Format string for parsing the time column (e.g., '%Y-%m-%d %H:%M:%S.%f').
       Only applies to CSV files. Default is None (auto-detect).
@@ -82,7 +82,8 @@ def read(
                 if any(i < 0 for i in [tidx, xidx, yidx, zidx]):
                     raise ValueError(f"csv_txyz_idxs must be non-negative integers, got: '{csv_txyz_idxs}'")
                 # Read header to get column names at those indices
-                header = pd.read_csv(filepath, nrows=0).columns.tolist()
+                # Skip csv_start_row rows to reach the actual header row
+                header = pd.read_csv(filepath, nrows=0, skiprows=csv_start_row).columns.tolist()
                 max_idx = max(tidx, xidx, yidx, zidx)
                 if max_idx >= len(header):
                     raise ValueError(f"Column index {max_idx} out of range. CSV has {len(header)} columns.")
@@ -95,18 +96,20 @@ def read(
                 if csv_end_row < csv_start_row:
                     raise ValueError(f"csv_end_row ({csv_end_row}) must be >= csv_start_row ({csv_start_row})")
 
-            # skiprows: skip rows after header if csv_start_row is specified
-            # csv_start_row is 0-indexed (excluding header), so skiprows skips file rows 1 to csv_start_row
-            skiprows = None if csv_start_row is None else range(1, csv_start_row + 1)
+            # skiprows: skip rows before the header if csv_start_row is specified
+            # csv_start_row is 0-indexed file row where the header is located
+            # skiprows=N skips rows 0 to N-1, making row N the header
+            skiprows = csv_start_row
 
             # nrows: number of data rows to read
-            # csv_end_row is inclusive and 0-indexed (excluding header)
+            # csv_end_row is file row to stop at (inclusive, 0-indexed)
+            # Data rows are from (csv_start_row + 1) to csv_end_row
             if csv_end_row is None:
                 nrows = None
             elif csv_start_row is None:
-                nrows = csv_end_row + 1
+                nrows = csv_end_row  # rows 1 to csv_end_row = csv_end_row rows
             else:
-                nrows = csv_end_row - csv_start_row + 1
+                nrows = csv_end_row - csv_start_row  # rows (csv_start_row+1) to csv_end_row
 
             # Common read_csv kwargs
             read_kwargs = dict(