medsci-skills 4.1.0

package/skills/analyze-stats/references/table-standards/table-standards.md

@@ -0,0 +1,287 @@
+# Publication Table Standards — Knowledge Base
+
+> Reference document for medical journal table formatting.
+> Source: YouTube tutorials, journal author guidelines, AMA Manual of Style, tool documentation.
+> Last updated: 2026-04-11
+
+---
+
+## 1. Universal Rules (All Medical Journals)
+
+1. **No vertical lines** — horizontal rules only (top, below header, bottom)
+2. **Editable Word tables** — use Word Insert > Table. Never submit images, Excel, or PDF
+3. **Sequential numbering** — Table 1, 2, 3... in order of first citation in text
+4. **Define all abbreviations** — in footnotes, independently for each table
+5. **Explanatory content in footnotes, not headings**
+6. **Display item limit** — most journals: tables + figures combined 4-5
+7. **Exact P values** — never just "significant/not significant"
+8. **Variability measures required** — always state mean (SD) or median (IQR)
+9. **Self-contained titles** — table title alone must convey content without reading text
+10. **No duplication** — tables and figures must not repeat the same data
+
+---
+
+## 2. Journal-Specific Differences
+
+### Footnote Markers
+
+| Journal Family | Marker System | Order |
+|---|---|---|
+| AMA (Radiology, JAMA, AJR, Rad:AI) | Superscript lowercase letters | a, b, c, d... |
+| NEJM | Symbols | *, †, ‡, §, ‖, ¶, **, ††... |
+| Lancet | Symbols | *, †, ‡, §, ¶, ‖, **, ††... |
+| European Radiology (Springer) | Superscript lowercase letters + asterisks for significance | a, b, c... and *, **, *** |
+
+### P Value Formatting
+
+| Journal | Case | Leading Zero | Examples |
+|---|---|---|---|
+| Radiology / JAMA / AJR / Rad:AI | Uppercase italic *P* | No | *P* = .03, *P* < .001 |
+| NEJM | Uppercase *P* | No | *P* = .04, *P* < .001 |
+| Lancet | Lowercase p | Yes | p = 0.03, p < 0.0001 |
+| European Radiology | Lowercase p (Springer) | Varies | p = 0.03 |
+
+### P Value Decimal Places
+
+| Value Range | JAMA/Radiology | NEJM | Lancet |
+|---|---|---|---|
+| > .01 | 2 decimals (.04) | 2 decimals (.04) | 2 sig figs (0.04) |
+| .01-.001 | 3 decimals (.003) | 3 decimals (.003) | Up to 4 decimals (0.0023) |
+| < .001 | *P* < .001 | *P* < .001 | p < 0.0001 |
+
+### Horizontal Lines
+
+| Journal | Rule |
+|---|---|
+| AMA journals | Top, below header, bottom. Additional sparingly |
+| NEJM | Minimize even internal horizontal lines |
+| Lancet | Exactly 3: top, below header, bottom |
+
+### 95% CI Format
+
+| Journal | Format |
+|---|---|
+| Radiology | (XX, XX) — comma separator |
+| JAMA | XX to XX — "to" separator |
+| Lancet | XX to XX or XX-XX |
+| NEJM | XX to XX |
+
+### Table Title Format
+
+| Journal | Format |
+|---|---|
+| AMA journals | **Table N.** Title in regular weight (bold "Table N." only) |
+| NEJM | Table N. Title |
+| Lancet | Table N: Title |
+
+---
+
+## 3. AMA Manual of Style (11th Ed) — Table Rules
+
+### Structure
+- **Column headers**: Bold
+- **Row headers (stub)**: Regular weight, sentence case
+- **Decking** (nested column headers): Maximum 3 levels
+- **Alignment**: Text = left-aligned, Numbers = center or decimal-aligned
+- **Units**: In column header parentheses, not repeated in cells
+- **Maximum size**: ~40 rows x 6-8 columns, fit within 1 page
+
+### Footnote System (AMA 11th)
+- **Table footnotes**: Superscript lowercase letters (a, b, c...)
+- **NOT symbols** — symbols (*, †, ‡) are for bottom-of-page text footnotes only
+- **Assignment order**: Top-to-bottom, left-to-right through the table
+- **Whole-table footnote**: Place superscript "a" at end of table title
+- **Reference numbers + footnotes**: Reference first, comma, then footnote letter (e.g., 5,b)
+
+### Footnote Placement Order
+```
+Below the table, in this sequence:
+
+1. General note (no marker, applies to entire table)
+   "Data are presented as median (IQR) unless otherwise noted."
+
+2. Abbreviation note
+   "ASA = American Society of Anesthesiologists, BMI = body mass index,
+    CI = confidence interval."
+   → Listed in order of appearance in table (left→right, top→bottom)
+   → Some journals (JAMA) require alphabetical order — check guide
+
+3. Specific notes (superscript letter markers)
+   "a Excludes patients with missing follow-up data."
+   "b Adjusted for age and sex."
+
+4. Probability notes (asterisk markers, if used)
+   "* P < .05; ** P < .01."
+```
+
+### P Value Rules
+- Uppercase italic: *P*
+- No leading zero: .05, not 0.05
+- Thin space around operators: *P* = .03
+- Only *P*, *α*, *β* omit leading zero; all other statistics keep it
+
+---
+
+## 4. Footnote System Specification
+
+### When to Use Symbols vs Letters vs Numbers
+
+| System | Markers | Use Case | Journals |
+|---|---|---|---|
+| Lowercase letters | a, b, c... | **Table footnotes (standard)** | AMA journals, EUR, most |
+| Symbols | *, †, ‡, §, ‖, ¶ | Table footnotes (traditional) | NEJM, Lancet |
+| Asterisks only | *, **, *** | **Probability notes only** | Springer journals |
+| Uppercase letters | A, B, C... | Table footnotes | JCI |
+| Numbers | 1, 2, 3... | Non-numeric tables only | Rarely used (confusion with references) |
+
+### Caption vs Footnote — What Goes Where
+
+| Content | Caption (above table) | Footnote (below table) |
+|---|---|---|
+| Table title | Yes | No |
+| Study period/setting | Yes (brief) | No |
+| Abbreviation definitions | No | **Yes** (first footnote) |
+| Statistical test descriptions | No | **Yes** |
+| P value thresholds | No | **Yes** (probability note) |
+| Data source | No | **Yes** (source note) |
+| Missing data explanation | No | **Yes** (specific note) |
+| Sample size (N) | Column header or caption | Footnote (supplementary) |
+
+### Statistical Test Footnote Patterns
+
+**Pattern A — Individual markers (preferred when tests vary by row):**
+```
+Age, y                    65.3 (12.1)    62.1 (11.8)    .04ᵃ
+Sex, male                 53 (52)        48 (47)        .41ᵇ
+
+ᵃ Wilcoxon rank-sum test.
+ᵇ Fisher exact test.
+```
+
+**Pattern B — General note (preferred when same test for all rows of a type):**
+```
+P values were calculated using the Wilcoxon rank-sum test for continuous
+variables and the Fisher exact test for categorical variables.
+```
+
+---
+
+## 5. Common Mistakes Checklist
+
+### Structure
+- [ ] Binary variables: show only one level (Male 53%, not Male 53% / Female 47%)
+- [ ] Remove derivable columns (Total column when groups are shown)
+- [ ] Keep tables under 50 rows (excess → supplementary)
+- [ ] No sub-part numbering (Table 3A/3B → Table 3, Table 4)
+- [ ] Do not repeat all table data in Results text
+
+### Formatting
+- [ ] Consistent decimal places within each column
+- [ ] Units in column headers, not in cells
+- [ ] No vertical lines, minimal horizontal lines
+- [ ] Numbers right-aligned or center-aligned (not left)
+- [ ] Specify variability: "Mean (SD)" or "Median (IQR)" in header or footnote
+
+### Statistics
+- [ ] RCTs: P values in Table 1 are usually unnecessary (randomization)
+- [ ] Always name the statistical test (in footnote or general note)
+- [ ] Report effect sizes per clinically meaningful unit (OR per 10-year, not per 1-year)
+- [ ] CI notation consistent throughout (parentheses vs "to" — match journal style)
+- [ ] Never use "NS" — report exact P values
+
+### Submission
+- [ ] Editable Word table (not image/screenshot)
+- [ ] No color as sole information carrier (use bold/italic/symbols instead)
+- [ ] All abbreviations defined in footnotes
+- [ ] Each table's footnotes are self-contained (no "see Table 1 footnote")
+
+---
+
+## 6. Tool Recommendations
+
+### Primary Pipeline: R {gtsummary}
+
+**Why gtsummary:**
+- Built-in journal themes: JAMA, Lancet, NEJM
+- Auto-selects statistical tests (Wilcoxon, Fisher, Chi-square)
+- Auto-generates footnotes (test names, summary statistics)
+- Outputs to Word (via flextable), LaTeX (via huxtable), HTML (via gt)
+
+**Core API:**
+```r
+library(gtsummary)
+
+# Set journal theme FIRST
+theme_gtsummary_journal("jama")  # or "lancet", "nejm"
+theme_gtsummary_compact()
+
+# Table 1
+tbl <- df %>%
+  tbl_summary(
+    by = group,
+    type = list(age ~ "continuous2"),
+    statistic = list(
+      all_continuous() ~ c("{mean} ({sd})", "{median} ({p25}, {p75})"),
+      all_categorical() ~ "{n} ({p}%)"
+    ),
+    digits = list(all_continuous() ~ 1),
+    missing = "ifany"
+  ) %>%
+  add_p() %>%
+  add_overall() %>%
+  add_stat_label() %>%
+  bold_labels() %>%
+  modify_footnote(all_stat_cols() ~ "Mean (SD); Median (Q1, Q3); n (%)")
+
+# Export
+tbl %>% as_flex_table() %>% flextable::save_as_docx(path = "table1.docx")
+tbl %>% as_hux_table() %>% huxtable::to_latex() %>% writeLines("table1.tex")
+```
+
+**Regression table:**
+```r
+model <- glm(outcome ~ age + sex + bmi, data = df, family = binomial)
+
+tbl_regression(model, exponentiate = TRUE) %>%
+  add_global_p() %>%
+  bold_p(t = 0.05) %>%
+  bold_labels()
+```
+
+### Supporting Tools
+
+| Tool | Role | When to Use |
+|---|---|---|
+| gt | Fine-grained styling | Cell-level colors, custom footnote marks, heatmaps |
+| flextable | Word output engine | Final DOCX formatting, autofit, borders |
+| huxtable | LaTeX output engine | LaTeX code extraction |
+| tableone (R/Python) | Quick exploratory Table 1 | Early drafts, data exploration |
+| python-docx | Python Word tables | When pipeline is Python-only |
+| great_tables | Python gt port | Growing but immature for medical use |
+
+### Python Limitation
+Python has no equivalent to gtsummary's journal themes or automatic statistical test selection. For publication tables, R is strongly recommended. Python-only pipelines should use tableone for generation + python-docx for Word formatting, accepting manual style adjustments.
+
+---
+
+## 7. Format-Specific Notes
+
+### Word (DOCX) Submission
+- Use Word's Insert > Table function (never tab-separated)
+- Font: Times New Roman 10-12pt (table body can be 9-10pt)
+- Single spacing within cells
+- Remove all vertical borders; keep 3 horizontal lines
+- Bold headers, regular body text
+- Each table on a separate page, after References
+
+### LaTeX Submission
+- Use `booktabs` package (`\toprule`, `\midrule`, `\bottomrule`)
+- Never use `\hline` or `|` vertical separators
+- `\siunitx` for decimal alignment
+- `\multirow` / `\multicolumn` for merged cells
+- Generate via gtsummary → huxtable → to_latex() for consistency
+
+### HTML (Review/Proofing)
+- gt produces the highest quality HTML tables
+- Useful for co-author review before final Word/LaTeX export
+- Can include interactive features (sorting, filtering) for supplementary materials

package/skills/analyze-stats/references/table-standards/table-types/diagnostic_accuracy.md

@@ -0,0 +1,36 @@
+# Table: Diagnostic Accuracy Results
+
+## Reporting Guidelines
+- **STARD**: Sensitivity, Specificity, PPV, NPV with 95% CIs
+- **TRIPOD-AI / CLAIM**: For AI model performance
+
+## Standard Structure
+
+```
+Table 2. Diagnostic Performance of [Model/Test] for [Condition]
+
+                     Sensitivity       Specificity       PPV              NPV              AUC
+                     (95% CI)          (95% CI)          (95% CI)         (95% CI)         (95% CI)
+
+Model A              0.92 (0.87-0.96)  0.85 (0.79-0.90)  0.78 (0.71-0.84) 0.95 (0.91-0.97) 0.94 (0.91-0.97)
+Model B              0.88 (0.82-0.93)  0.90 (0.85-0.94)  0.83 (0.76-0.88) 0.93 (0.89-0.96) 0.93 (0.89-0.96)
+P value              .12               .08               .21              .34              .45
+
+95% CIs were calculated using the Wilson score method. AUC comparison
+by DeLong test.
+```
+
+## Rules
+- **Always include 95% CIs** for all metrics (STARD requirement)
+- **CI method**: Specify in footnote (Wilson score, Clopper-Pearson, DeLong)
+- **Threshold**: State the decision threshold used (e.g., "at Youden optimal threshold")
+- **Per-class results**: For multi-class, show per-class + macro/micro average
+- **Comparison P values**: DeLong test for AUC, McNemar for sensitivity/specificity
+- **Decimal places**: 2-3 for proportions (0.92), 2-3 for AUC (0.94)
+- **Reader studies**: Include per-reader AND pooled results
+
+## Key Footnote Content
+- CI calculation method
+- Comparison test used (DeLong, McNemar, bootstrap)
+- Threshold selection method
+- Whether results are per-patient or per-lesion

package/skills/analyze-stats/references/table-standards/table-types/meta_analysis.md

@@ -0,0 +1,58 @@
+# Tables: Meta-Analysis
+
+## Reporting Guidelines
+- **PRISMA 2020**: Study characteristics + pooled results
+- **PRISMA-DTA**: For diagnostic test accuracy meta-analyses
+
+---
+
+## Table A: Characteristics of Included Studies
+
+```
+Table 1. Characteristics of Included Studies
+
+Author, Year   Country   Design   N      Population       Index Test        Reference Standard   Quality
+Kim 2023       Korea     Retro    450    Suspected PE     CTPA AI (v2.1)    Expert consensus     Low risk
+Smith 2024     USA       Prosp    1200   ED patients      CTPA AI (v3.0)    Pulmonary DSA        Some concerns
+...
+
+Retro = retrospective, Prosp = prospective, PE = pulmonary embolism,
+CTPA = CT pulmonary angiography, ED = emergency department,
+DSA = digital subtraction angiography.
+```
+
+### Rules
+- **Column order**: Author/Year, Country, Design, N, Population, Index test, Reference standard, Quality/RoB
+- **Author format**: First author surname + year
+- **Study design**: Use standard abbreviations (Retro, Prosp, RCT)
+- **Quality assessment tool**: QUADAS-2 (DTA), RoB 2 (RCT), NOS (observational)
+- **Quality rating**: "Low risk" / "Some concerns" / "High risk" (QUADAS-2 terms)
+
+---
+
+## Table B: Pooled Results / Summary Estimates
+
+```
+Table 2. Pooled Diagnostic Accuracy Estimates
+
+                    k    N       Pooled Estimate (95% CI)    I²     P_het
+
+Sensitivity         12   3400    0.91 (0.87-0.94)           78%    <.001
+Specificity         12   3400    0.88 (0.83-0.92)           65%    .002
+Positive LR         12   3400    7.58 (5.12-11.2)           72%    <.001
+Negative LR         12   3400    0.10 (0.07-0.15)           69%    .001
+DOR                 12   3400    75.8 (42.1-136.5)          58%    .008
+
+k = number of studies, N = total participants, CI = confidence interval,
+LR = likelihood ratio, DOR = diagnostic odds ratio.
+Pooled estimates from bivariate random-effects model.
+I² = Higgins inconsistency statistic; P_het from Cochran Q test.
+```
+
+### Rules
+- **k and N**: Always report number of studies and total participants
+- **Heterogeneity**: I² + P from Cochran Q test (mandatory)
+- **Model**: State random-effects vs fixed-effects in footnote
+- **Subgroup analyses**: Separate rows or separate table
+- **Prediction interval**: Include for random-effects if k >= 3
+- **Forest plot complement**: Table complements but does not replace forest plot

package/skills/analyze-stats/references/table-standards/table-types/model_comparison.md

@@ -0,0 +1,36 @@
+# Table: Model Performance Comparison
+
+## Reporting Guidelines
+- **TRIPOD-AI**: AI model development/validation
+- **CLAIM**: AI in medical imaging
+
+## Standard Structure
+
+```
+Table 3. Comparison of Model Performance on Test Set
+
+Model          AUROC (95% CI)       Sensitivity (95% CI)  Specificity (95% CI)  P vs Baseline
+
+Baseline CNN   0.87 (0.83-0.91)    0.82 (0.75-0.88)     0.85 (0.79-0.90)     —
+Proposed Model 0.93 (0.90-0.96)    0.90 (0.84-0.94)     0.89 (0.84-0.93)     .003
+Reader A       0.88 (0.84-0.92)    0.85 (0.78-0.90)     0.86 (0.80-0.91)     .52
+Reader B       0.86 (0.81-0.90)    0.83 (0.76-0.89)     0.84 (0.78-0.89)     .71
+
+AUROC compared by DeLong test. Sensitivity/specificity at Youden optimal
+threshold. 95% CIs by Wilson score interval.
+```
+
+## Rules
+- **Baseline/reference model**: Always include (first row, P = — or "Ref")
+- **Reader comparison**: Include human readers when available
+- **Same test set**: All models evaluated on identical held-out data
+- **Threshold**: Specify how threshold was chosen (Youden, fixed sensitivity, etc.)
+- **Bootstrap CIs**: State number of iterations if used (e.g., "1000 bootstrap replicates")
+- **Multiple test sets**: Separate sections or sub-tables for internal vs external validation
+- **Calibration**: Include Brier score or calibration slope if relevant
+
+## AI-Specific Additions (CLAIM)
+- Software version / model architecture in footnote or methods
+- Training set size in footnote
+- Hardware used (if inference time reported)
+- Whether results are per-image or per-patient

package/skills/analyze-stats/references/table-standards/table-types/regression_results.md

@@ -0,0 +1,50 @@
+# Table: Regression Results
+
+## Reporting Guidelines
+- **STROBE**: Observational studies — OR, HR, RR with 95% CI
+- **TRIPOD**: Prediction models — coefficients or OR with CI
+
+## Standard Structure
+
+```
+Table 3. Multivariable Logistic Regression for [Outcome]
+
+Variable              Univariable              Multivariable
+                      OR (95% CI)    P Value   OR (95% CI)      P Value
+
+Age, per 10 y         1.22 (1.08-1.38) .002   1.18 (1.03-1.35)  .02
+Sex, male vs female   1.45 (0.98-2.14) .06    1.38 (0.91-2.09)  .13
+BMI, per 5 kg/m²      1.31 (1.12-1.53) <.001  1.25 (1.06-1.47)  .008
+Stage III vs I-II     2.87 (1.92-4.29) <.001  2.54 (1.65-3.91)  <.001
+
+OR = odds ratio, CI = confidence interval, BMI = body mass index.
+```
+
+## Rules
+- **Reference category**: Always state (e.g., "male vs female", "Stage III vs I-II")
+- **Clinically meaningful units**: "per 10 years" not "per 1 year" for continuous
+- **Effect measure**: Match study design — OR (logistic), HR (Cox), RR (log-binomial), β (linear)
+- **Always show both univariable AND multivariable** (or justify omission)
+- **Model fit statistics**: AIC, C-statistic, or Hosmer-Lemeshow in footnote or separate row
+- **Variable selection method**: State in footnote (e.g., "Variables with P < .10 in univariable entered multivariable")
+- **Collinearity**: Note if VIF checked, in footnote
+
+## gtsummary Code
+```r
+# Univariable
+uv <- tbl_uvregression(
+  data, y = outcome, method = glm,
+  method.args = list(family = binomial),
+  exponentiate = TRUE
+)
+
+# Multivariable
+mv <- glm(outcome ~ age + sex + bmi + stage, data = data, family = binomial)
+mv_tbl <- tbl_regression(mv, exponentiate = TRUE) %>%
+  add_global_p() %>%
+  bold_p(t = 0.05)
+
+# Merge
+tbl_merge(list(uv, mv_tbl),
+          tab_spanner = c("Univariable", "Multivariable"))
+```

package/skills/analyze-stats/references/table-standards/table-types/table1_demographics.md

@@ -0,0 +1,51 @@
+# Table 1: Baseline Demographics / Characteristics
+
+## Reporting Guidelines
+- **RCT**: CONSORT (Table 1 should NOT include P values — randomization makes them irrelevant)
+- **Cohort/Cross-sectional**: STROBE (P values optional, SMD preferred for propensity-matched)
+- **Diagnostic**: STARD (patient demographics + index test characteristics)
+
+## Standard Structure
+
+```
+Table 1. Baseline Characteristics of Patients
+
+                          Group A (n=XXX)    Group B (n=XXX)    P Value
+Age, y                    65.3 (12.1)        62.1 (11.8)        .04
+Sex
+  Male                    53 (52)            48 (47)            .41
+  Female                  49 (48)            54 (53)
+BMI, kg/m²               24.8 (3.2)         25.1 (3.5)         .52
+...
+
+Data are presented as mean (SD) for continuous variables and n (%)
+for categorical variables.
+```
+
+## Rules
+- **Binary variables**: Show only one level (e.g., Male only; Female is implied)
+- **Continuous variables**: Mean (SD) if normal; Median (IQR) if skewed. State which in footnote
+- **Categorical variables**: n (%)
+- **Column headers**: Include group size — "Group A (n=XXX)"
+- **Units**: In row label — "Age, y" or "Age, years"
+- **Missing data**: Report as "Missing" row or footnote stating N with complete data
+- **P values in RCTs**: Omit (CONSORT recommendation) or include for observational studies
+- **SMD**: Preferred over P values for propensity-matched studies
+
+## gtsummary Code
+```r
+tbl_summary(
+  data, by = group,
+  type = list(age ~ "continuous2"),
+  statistic = list(
+    all_continuous() ~ c("{mean} ({sd})"),
+    all_categorical() ~ "{n} ({p}%)"
+  ),
+  digits = list(all_continuous() ~ 1, all_categorical() ~ c(0, 1)),
+  missing = "ifany",
+  missing_text = "Missing"
+) %>%
+  add_p() %>%         # omit for RCTs
+  add_overall() %>%
+  bold_labels()
+```

package/skills/analyze-stats/references/table-standards/tool-comparison.md

@@ -0,0 +1,79 @@
+# Publication Table Tools — Comparison
+
+## R Packages
+
+| Feature | gtsummary | gt | flextable | tableone | kableExtra | huxtable |
+|---|---|---|---|---|---|---|
+| **Primary Use** | Table 1 + regression | Fine styling | Word output | Quick Table 1 | LaTeX/PDF | Multi-format |
+| **Auto Statistics** | Yes (test selection) | No | No | Yes (p, SMD) | No | huxreg() only |
+| **Journal Themes** | JAMA, Lancet, NEJM | No | Custom | No | No | No |
+| **Footnotes** | Via gt engine | Native (excellent) | Native | Manual | Native | Native |
+| **Word Output** | Via flextable | Limited | **Best** | Via knitr | Limited | Yes |
+| **LaTeX Output** | Via huxtable | Native | Limited | Via knitr | **Best** | Native |
+| **HTML Output** | Via gt | **Best** | Yes | Via knitr | Yes | Yes |
+| **Maintenance** | Active (MSKCC) | Active (Posit) | Active | Maintenance | Maintenance | Maintenance |
+
+## Python Libraries
+
+| Feature | great_tables | tableone (py) | python-docx | pandas Styler |
+|---|---|---|---|---|
+| **Primary Use** | gt port | Quick Table 1 | Word tables | HTML/LaTeX |
+| **Auto Statistics** | No | Yes (p, SMD) | No | No |
+| **Journal Themes** | No | No | No | No |
+| **Footnotes** | tab_footnote() | No | Manual | No |
+| **Word Output** | No | No | **Native** | No |
+| **LaTeX Output** | No | Via export | No | to_latex() |
+| **Maturity** | Growing | Stable | Stable | Built-in |
+
+## Recommended Pipelines
+
+### Best: R gtsummary → Word
+```r
+theme_gtsummary_journal("jama")
+theme_gtsummary_compact()
+tbl %>% as_flex_table() %>% flextable::save_as_docx(path = "table.docx")
+```
+
+### Best: R gtsummary → LaTeX
+```r
+tbl %>% as_hux_table() %>% huxtable::to_latex()
+# OR
+tbl %>% as_gt() %>% gt::gtsave("table.tex")
+```
+
+### Python-only (acceptable, not ideal)
+```python
+# Generation
+from tableone import TableOne
+table1 = TableOne(df, columns=cols, categorical=cats, groupby='group', pval=True)
+# Export to CSV, then format with python-docx
+```
+
+## Key Insight
+> **Python lacks gtsummary-equivalent journal themes and auto-statistics.**
+> For publication tables in medical journals, R is strongly preferred.
+> Python pipelines require manual style adjustments that R handles automatically.
+
+## Tool Selection Decision Tree
+
+```
+Need Table 1 or regression table?
+  → gtsummary (+ journal theme)
+
+Need pixel-perfect custom styling?
+  → gt (HTML) or flextable (Word)
+
+Submitting to Word-only journal?
+  → gtsummary → as_flex_table() → save_as_docx()
+
+Submitting LaTeX?
+  → gtsummary → as_hux_table() → to_latex()
+  → OR kableExtra for direct LaTeX control
+
+Python-only constraint?
+  → tableone (generation) + python-docx (formatting)
+  → Accept manual style work
+
+Quick exploratory table?
+  → tableone (R or Python) — fastest setup
+```