National Health Insurance Scheme Awareness and Benefit-statement Endorsement among Urban- and Rural-classified Respondents in a Ghanaian Household Survey

Abstract

Aims: To describe self-reported awareness of Ghana's National Health insurance scheme (NHIS), affirmative responses to broad benefit statements and their distribution across recorded urban and rural groups. Study Design: A descriptive cross-sectional household survey with exploratory comparisons by recorded settlement classification. Place and Duration of Study: Four districts in the present Northern Region of Ghana constituted the intended sampling frame; interviews took place from March 2025 to February 2026. Methodology: A four-stage household selection procedure was specified. Fourteen positively phrased benefit statements were administered; one for which 80.0% of responses were missing was excluded, leaving a 13-item endorsement count (0–13). The three descriptive score bands were 11–13, 8–10 and 0–7. Statistical intervals and tests did not account for clusters or sampling weights. Results: Among 595 analysed records (194 urban; 401 rural), 591 (99.3%) reported NHIS awareness. The endorsement bands contained 49 (8.2%), 290 (48.7%) and 256 (43.0%) respondents, respectively. Affirmative responses were least frequent for physiotherapy (27.9%), specialist consultation (39.8%), general surgery (40.2%) and basic dental care (40.5%). The highest-band proportion was 10.8% in the urban group and 7.0% in the rural group (unadjusted difference 3.8 percentage points; 95% CI −0.8 to 9.4; nominal P = .11); the three-band comparison yielded nominal P = .26. Internal consistency of the 13-item index was low (KR-20 = 0.35). Geographic fields recorded districts outside the intended four for 46 records and other regions for 15 records. Conclusion: General NHIS awareness was frequently reported, whereas endorsements of individual benefit statements varied considerably. The unvalidated, low-reliability score, conditional benefit entitlements, discrepancies in geographic fields and lack of design-based analysis prevent conclusions about regional prevalence, validated literacy or urban–rural equivalence.