Current studies examining the effects of Afghanistan's conflict transition on the performance of health systems, health service delivery, and health outcomes are outdated and small in scale and do not span all essential reproductive, maternal, newborn, and child health interventions. To evaluate associations of conflict severity with improvement of health system performance, use of health services, and child nutrition outcomes in Afghanistan during the 2003 to 2018 reconstruction period. This population-based survey study included a sequential cross-sectional analysis of individual-level panel data across 2 periods (2003-2010 and 2010-2018) and a difference-in-differences design. Surveys included the 2003 to 2004 and 2010 to 2011 Multiple Indicator Cluster Surveys and the 2018 Afghanistan Health Survey. Afghanistan's 2013 National Nutrition Survey was used to assess nutritional outcomes, and the annual Balanced Scorecard data sets were used to evaluate health system performance. Participants included girls and women aged 12 to 49 years and children younger than 5 years who completed nationally representative household surveys. All analyses were conducted from January 1 through April 30, 2019. Provinces were categorized as experiencing minimal-, moderate-, and severe-intensity conflict using battle-related death data from the Uppsala Conflict Data Program. Health intervention coverage was examined using 10 standard indicators: contraceptive method (any or modern); antenatal care by a skilled health care professional; facility delivery; skilled birth attendance (SBA); bacille Calmette-Guérin vaccination (BCG); diphtheria, pertussis, and tetanus vaccination (DPT3) or DPT3 plus hepatitis B and poliomyelitis (penta); measles vaccination; care-seeking for acute respiratory infection; oral rehydration therapy for diarrhea; and the Composite Coverage Index. The health system performance was analyzed using the following standard Balanced Scorecard composite domains: client and community, human resources, physical capacity, quality of service provision, management systems, and overall mission. Child stunting, wasting, underweight, and co-occurrence of stunting and wasting were estimated using World Health Organization growth reference cutoffs. Responses from 64 815 women (mean [SD] age, 31.0 [8.5] years) were analyzed. Provinces with minimal-intensity conflict had greater gains in contraceptive use (mean annual percentage point change [MAPC], 1.3% vs 0.5%; P < .001), SBA (MAPC, 2.7% vs 1.5%; P = .005), BCG vaccination (MAPC, 3.3% vs -0.5%; P = .002), measles vaccination (MAPC, 1.9% vs -1.0%; P = .01), and DPT3/penta vaccination (MAPC, 1.0% vs -2.0%; P < .001) compared with provinces with moderate- to severe-intensity conflict after controlling for confounders. Provinces with severe-intensity conflict fared significantly worse than those with minimal-intensity conflict in functioning infrastructure (MAPC, -1.6% [95% CI, -2.4% to -0.8%]) and the client background and physical assessment index (MAPC, -1.0% [95% CI, -0.8% to 2.7%]) after adjusting for confounders. Child wasting was significantly worse in districts with greater conflict severity (full adjusted β for association between logarithm of battle-related deaths and wasting, 0.33 [95% CI, 0.01-0.66]; P = .04). Associations between conflict and maternal and child health in Afghanistan differed by health care intervention and delivery domain, with several key indicators lagging behind in areas with higher-intensity conflict. These findings may be helpful for planning and prioritizing efforts to reach the United Nations' Sustainable Development Goals in Afghanistan.