"""Generate synthetic rehabilitation workforce & training dataset for SSA. Research-based parameterization: - WHO Rehabilitation 2030: SSA has 0.5-2 rehab professionals per 100K vs 60+ in high-income; massive shortage across all cadres. - WCPT: <5000 physiotherapists for all of SSA; most concentrated in urban areas; rural coverage near zero. - ISPO: ~0.5 prosthetists/orthotists per million in SSA. - SSA context: Few training institutions; brain drain; task-shifting to CHWs/nurses; limited CPD opportunities. """ from __future__ import annotations from pathlib import Path import numpy as np import pandas as pd SEED = 42 N_PER_SCENARIO = 10_000 YEAR_RANGE = np.arange(2010, 2025) YEAR_WEIGHTS = np.linspace(0.85, 1.3, len(YEAR_RANGE)) YEAR_WEIGHTS = YEAR_WEIGHTS / YEAR_WEIGHTS.sum() SCENARIOS = { "urban_training_institution": { "setting_probs": {"university": 0.30, "teaching_hospital": 0.25, "rehab_training_school": 0.20, "private_institution": 0.25}, "cadre_probs": {"physiotherapist": 0.25, "occupational_therapist": 0.12, "speech_therapist": 0.08, "prosthetist_orthotist": 0.10, "rehab_nurse": 0.15, "psychologist": 0.08, "social_worker": 0.08, "rehab_physician": 0.05, "audiologist": 0.04, "other": 0.05}, "staff_density_per_100k": 2.0, "training_places": 50, "retention_rate": 0.60, "cpd_access_pct": 0.30, "task_shifting_pct": 0.15, }, "district_service_delivery": { "setting_probs": {"district_hospital": 0.35, "health_centre": 0.25, "community_programme": 0.20, "outreach": 0.20}, "cadre_probs": {"physiotherapist": 0.20, "rehab_nurse": 0.20, "rehab_assistant": 0.15, "cbr_worker": 0.15, "occupational_therapist": 0.08, "prosthetist_orthotist": 0.05, "social_worker": 0.07, "other": 0.10}, "staff_density_per_100k": 0.5, "training_places": 15, "retention_rate": 0.40, "cpd_access_pct": 0.10, "task_shifting_pct": 0.35, }, "rural_task_shifted": { "setting_probs": {"health_post": 0.30, "community_home": 0.25, "cbr_programme": 0.25, "mobile_clinic": 0.20}, "cadre_probs": {"cbr_worker": 0.30, "chw_rehab_trained": 0.25, "rehab_assistant": 0.15, "nurse_task_shifted": 0.15, "physiotherapist": 0.05, "other": 0.10}, "staff_density_per_100k": 0.1, "training_places": 5, "retention_rate": 0.25, "cpd_access_pct": 0.03, "task_shifting_pct": 0.70, }, } SCENARIO_FILES = { "urban_training_institution": "rehab_wf_urban.csv", "district_service_delivery": "rehab_wf_district.csv", "rural_task_shifted": "rehab_wf_rural.csv", } def _choice(rng, prob_map): keys = list(prob_map.keys()) weights = np.array(list(prob_map.values()), dtype=float) weights = weights / weights.sum() return rng.choice(keys, p=weights) def _simulate_scenario(name, params, seed): rng = np.random.default_rng(seed) records = [] for idx in range(N_PER_SCENARIO): year = int(rng.choice(YEAR_RANGE, p=YEAR_WEIGHTS)) setting = _choice(rng, params["setting_probs"]) cadre = _choice(rng, params["cadre_probs"]) age = int(np.clip(rng.normal(35, 10), 22, 65)) sex = rng.choice(["male", "female"], p=[0.40, 0.60]) years_experience = int(np.clip(rng.exponential(6), 0, 35)) # Training qualification = rng.choice(["degree", "diploma", "certificate", "on_job_training"], p=[0.20, 0.30, 0.25, 0.25]) training_institution = rng.choice(["local_university", "foreign_trained", "regional_school", "in_service"], p=[0.30, 0.10, 0.25, 0.35]) cpd_received = int(rng.random() < params["cpd_access_pct"]) supervision_regular = int(rng.random() < 0.20) mentorship = int(rng.random() < 0.10) # Workforce position_filled = int(rng.random() < params["retention_rate"]) vacancy = int(not position_filled) task_shifted = int(rng.random() < params["task_shifting_pct"]) workload_patients_week = int(np.clip(rng.poisson(20), 2, 80)) burnout = int(workload_patients_week > 30 and rng.random() < 0.35) emigration_intent = int(rng.random() < 0.25) rural_willingness = int(rng.random() < 0.15) # Infrastructure equipment_available = int(rng.random() < 0.30) dedicated_space = int(rng.random() < 0.25) consumables_available = int(rng.random() < 0.35) # Competencies assessment_competent = int(rng.random() < 0.50) treatment_competent = int(rng.random() < 0.45) assistive_device_competent = int(rng.random() < 0.25) community_rehab_competent = int(rng.random() < 0.20) paediatric_competent = int(rng.random() < 0.15) # Outcomes patients_seen_month = int(np.clip(rng.poisson(workload_patients_week * 4), 5, 300)) patient_satisfaction = int(rng.random() < (0.55 if assessment_competent else 0.30)) functional_outcomes_measured = int(rng.random() < 0.15) referral_network = int(rng.random() < 0.20) # Policy rehab_in_health_policy = int(rng.random() < 0.25) budget_for_rehab = int(rng.random() < 0.10) regulation_registered = int(qualification in ("degree", "diploma") and rng.random() < 0.50) record = { "record_id": f"{name[:3].upper()}-{idx:05d}", "scenario": name, "year": year, "setting": setting, "cadre": cadre, "age": age, "sex": sex, "years_experience": years_experience, "qualification": qualification, "training_institution": training_institution, "cpd_received": cpd_received, "supervision_regular": supervision_regular, "position_filled": position_filled, "vacancy": vacancy, "task_shifted": task_shifted, "workload_patients_week": workload_patients_week, "burnout": burnout, "emigration_intent": emigration_intent, "rural_willingness": rural_willingness, "equipment_available": equipment_available, "dedicated_space": dedicated_space, "assessment_competent": assessment_competent, "treatment_competent": treatment_competent, "assistive_device_competent": assistive_device_competent, "patients_seen_month": patients_seen_month, "patient_satisfaction": patient_satisfaction, "functional_outcomes_measured": functional_outcomes_measured, "rehab_in_health_policy": rehab_in_health_policy, "budget_for_rehab": budget_for_rehab, "regulation_registered": regulation_registered, } records.append(record) return pd.DataFrame(records) def main(): output_dir = Path("data") output_dir.mkdir(parents=True, exist_ok=True) for idx, (name, params) in enumerate(SCENARIOS.items()): df = _simulate_scenario(name, params, SEED + idx * 211) df.to_csv(output_dir / SCENARIO_FILES[name], index=False) print(f"Saved {name} -> {SCENARIO_FILES[name]}") if __name__ == "__main__": main()