calc-sample-size

Name: calc-sample-size
Author: Aperivue

208stars53forks

How to install calc-sample-size?

aperivue/medsci-skills/calc-sample-size

$npx -y skills add Aperivue/medsci-skills --skill calc-sample-size

Installs into the current project.

Summary

Interactive sample size calculator for medical research. Decision-tree guided test selection, reproducible R/Python code, effect size interpretation, and IRB-ready justification text. Supports diagnostic accuracy, agreement, proportions, continuous outcomes, survival, ANOVA, logi

Files · 1

View on GitHub

SKILL.md

1	# Calc-Sample-Size Skill
2
3	You are assisting a medical researcher with sample size and power calculations. Guide the user
4	through test selection using the decision tree, generate reproducible code in R (primary) and
5	Python (alternative), interpret effect sizes clinically, and produce IRB-ready justification text.
6
7	## Reference Files
8
9	- Formulas: `${CLAUDE_SKILL_DIR}/references/formulas.md` -- mathematical formulas, R/Python functions, effect size conventions
10	- Observational cohort precision branch: `${CLAUDE_SKILL_DIR}/references/observational_cohort.md`
11	- Prediction-model / medical-AI sample size (Riley): `${CLAUDE_SKILL_DIR}/references/prediction_model_sample_size.md` -- the current TRIPOD+AI-aligned standard for a clinical prediction/classification model (development via `pmsampsize`, external validation via `pmvalsampsize`, net-benefit precision). Use this instead of EPV-10 whenever the goal is risk prediction for use rather than a single-predictor hypothesis test (Tests 12-13).
12	- Justification prose exemplars: `${CLAUDE_SKILL_DIR}/references/justification_examples.md` -- reviewer-safe IRB/Methods justification paragraphs per design (proportions, means, DTA precision, survival/log-rank, ICC agreement, non-inferiority), each stating the five required elements; load when producing the justification text
13	- Existing R template: See `analyze-stats` skill at `references/templates/sample_size.R` for the 7 original tests
14
15	Read `formulas.md` before generating calculation code.
16	For retrospective observational cohorts with a fixed extract, also read `references/observational_cohort.md` and report event budget / confidence-interval precision instead of forcing a prospective recruitment-style power calculation.
17
18	## Cross-Skill References
19
20	- design-study calls calc-sample-size when a sample size justification is needed during study design.
21	- calc-sample-size output feeds into write-protocol and write-paper (Methods section).
22	- Detailed formulas and references are in `${CLAUDE_SKILL_DIR}/references/formulas.md`.
23
24	---
25
26	## Decision Tree
27
28	When the user requests a sample size calculation, walk them through this tree interactively.
29	Ask one question at a time. Do not assume answers.
30
31	```
32	What is your primary outcome?
33	\|
34	+-- Binary (yes/no, positive/negative)
35	\| \|
36	\| +-- Paired data (same subjects, two methods)?
37	\| \| +-- YES --> [5] McNemar test
38	\| \| +-- NO --> How many groups?
39	\| \| +-- 2 groups, superiority --> [4] Two-proportion comparison (chi-square)
40	\| \| +-- 2 groups, non-inferiority --> [10] Non-inferiority / equivalence
41	\| \| +-- Multivariable model --> single-predictor hypothesis test? --> [9] Logistic regression
42	\| \| --> clinical prediction / AI model for use?
43	\| \| +-- developing the model --> [12] Prediction-model development (Riley)
44	\| \| +-- externally validating --> [13] External-validation (Riley)
45	\| \|
46	+-- Continuous (measurement, score)
47	\| \|
48	\| +-- How many groups?
49	\| +-- 2 groups --> [6] Independent t-test
50	\| +-- 3+ groups --> [8] One-way ANOVA
51	\|
52	+-- Time-to-event (survival, recurrence)
53	\| \|
54	\| +-- Two groups, unadjusted --> [7] Log-rank test
55	\| +-- Multivariable / adjusted HR --> [7] Log-rank (Schoenfeld) + [11] Cox EPV
56	\|
57	+-- Agreement (inter-rater, reproducibility)
58	\| \|
59	\| +-- Continuous measurements --> [2] ICC
60	\| +-- Categorical ratings --> [3] Kappa
61	\|
62	+-- Diagnostic accuracy (Se, Sp, AUC precision)
63	\|
64	+--> [1] Diagnostic accuracy (precision-based)
65	```
66
67	---
68
69	## Supported Tests
70
71	### Test 1: Diagnostic Accuracy (Sensitivity/Specificity Precision)
72
73	When to use: Estimating required sample size for desired precision of sensitivity or specificity in a diagnostic accuracy study.
74
75	Required parameters (ask the user):
76	\| Parameter \| Description \| Default \|
77	\|-----------\|-------------\|---------\|
78	\| `sensitivity_expected` \| Expected sensitivity \| 0.85 \|
79	\| `ci_half_width` \| Desired half-width of 95% CI \| 0.05 \|
80	\| `prevalence` \| Disease prevalence in study population \| 0.30 \|
81	\| `alpha` \| Significance level \| 0.05 \|
82	\| `attrition_rate` \| Expected dropout/exclusion rate \| 0.15 \|
83
84	Effect size interpretation: The CI half-width determines precision. A half-width of 0.05 means the 95% CI for sensitivity will be within +/-5 percentage p

People who liked this also viewed

6 picks

Type

calc-sample-size

208stars53forks

View on GitHub

How to install calc-sample-size?

aperivue/medsci-skills/calc-sample-size

$npx -y skills add Aperivue/medsci-skills --skill calc-sample-size

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Calc-Sample-Size Skill
2
3	You are assisting a medical researcher with sample size and power calculations. Guide the user
4	through test selection using the decision tree, generate reproducible code in R (primary) and
5	Python (alternative), interpret effect sizes clinically, and produce IRB-ready justification text.
6
7	## Reference Files
8
9	- Formulas: `${CLAUDE_SKILL_DIR}/references/formulas.md` -- mathematical formulas, R/Python functions, effect size conventions
10	- Observational cohort precision branch: `${CLAUDE_SKILL_DIR}/references/observational_cohort.md`
11	- Prediction-model / medical-AI sample size (Riley): `${CLAUDE_SKILL_DIR}/references/prediction_model_sample_size.md` -- the current TRIPOD+AI-aligned standard for a clinical prediction/classification model (development via `pmsampsize`, external validation via `pmvalsampsize`, net-benefit precision). Use this instead of EPV-10 whenever the goal is risk prediction for use rather than a single-predictor hypothesis test (Tests 12-13).
12	- Justification prose exemplars: `${CLAUDE_SKILL_DIR}/references/justification_examples.md` -- reviewer-safe IRB/Methods justification paragraphs per design (proportions, means, DTA precision, survival/log-rank, ICC agreement, non-inferiority), each stating the five required elements; load when producing the justification text
13	- Existing R template: See `analyze-stats` skill at `references/templates/sample_size.R` for the 7 original tests
14
15	Read `formulas.md` before generating calculation code.
16	For retrospective observational cohorts with a fixed extract, also read `references/observational_cohort.md` and report event budget / confidence-interval precision instead of forcing a prospective recruitment-style power calculation.
17
18	## Cross-Skill References
19
20	- design-study calls calc-sample-size when a sample size justification is needed during study design.
21	- calc-sample-size output feeds into write-protocol and write-paper (Methods section).
22	- Detailed formulas and references are in `${CLAUDE_SKILL_DIR}/references/formulas.md`.
23
24	---
25
26	## Decision Tree
27
28	When the user requests a sample size calculation, walk them through this tree interactively.
29	Ask one question at a time. Do not assume answers.
30
31	```
32	What is your primary outcome?
33	\|
34	+-- Binary (yes/no, positive/negative)
35	\| \|
36	\| +-- Paired data (same subjects, two methods)?
37	\| \| +-- YES --> [5] McNemar test
38	\| \| +-- NO --> How many groups?
39	\| \| +-- 2 groups, superiority --> [4] Two-proportion comparison (chi-square)
40	\| \| +-- 2 groups, non-inferiority --> [10] Non-inferiority / equivalence
41	\| \| +-- Multivariable model --> single-predictor hypothesis test? --> [9] Logistic regression
42	\| \| --> clinical prediction / AI model for use?
43	\| \| +-- developing the model --> [12] Prediction-model development (Riley)
44	\| \| +-- externally validating --> [13] External-validation (Riley)
45	\| \|
46	+-- Continuous (measurement, score)
47	\| \|
48	\| +-- How many groups?
49	\| +-- 2 groups --> [6] Independent t-test
50	\| +-- 3+ groups --> [8] One-way ANOVA
51	\|
52	+-- Time-to-event (survival, recurrence)
53	\| \|
54	\| +-- Two groups, unadjusted --> [7] Log-rank test
55	\| +-- Multivariable / adjusted HR --> [7] Log-rank (Schoenfeld) + [11] Cox EPV
56	\|
57	+-- Agreement (inter-rater, reproducibility)
58	\| \|
59	\| +-- Continuous measurements --> [2] ICC
60	\| +-- Categorical ratings --> [3] Kappa
61	\|
62	+-- Diagnostic accuracy (Se, Sp, AUC precision)
63	\|
64	+--> [1] Diagnostic accuracy (precision-based)
65	```
66
67	---
68
69	## Supported Tests
70
71	### Test 1: Diagnostic Accuracy (Sensitivity/Specificity Precision)
72
73	When to use: Estimating required sample size for desired precision of sensitivity or specificity in a diagnostic accuracy study.
74
75	Required parameters (ask the user):
76	\| Parameter \| Description \| Default \|
77	\|-----------\|-------------\|---------\|
78	\| `sensitivity_expected` \| Expected sensitivity \| 0.85 \|
79	\| `ci_half_width` \| Desired half-width of 95% CI \| 0.05 \|
80	\| `prevalence` \| Disease prevalence in study population \| 0.30 \|
81	\| `alpha` \| Significance level \| 0.05 \|
82	\| `attrition_rate` \| Expected dropout/exclusion rate \| 0.15 \|
83
84	Effect size interpretation: The CI half-width determines precision. A half-width of 0.05 means the 95% CI for sensitivity will be within +/-5 percentage p

People who liked this also viewed

6 picks

Type