**Background:** Achieving glycemic control in Type 1 diabetes mellitus (T1DM) is critical to reduce complications, and carbohydrate counting (CC) is a key strategy for matching prandial insulin to carbohydrate intake. However, CC requires high mathematical and literacy skills, and many patients struggle with accurate estimation of carbohydrate content, leading to post-prandial hypo- or hyperglycemia. New technologies like bolus calculators and apps are not accessible to all populations, including older adults, those with lower socioeconomic status, and those with limited technological skills. To address these barriers, the Diabetes Clinic of Soroka University Medical Center developed a Simple Carb Counting (SCC) tool—a low-tech, individualized approach using personalized tables.
**Methods:** This open-label randomized controlled trial was conducted at Soroka University Medical Center, a tertiary 1200-bed hospital in Israel serving a diverse population including Bedouin, Arab, Jewish, ultra-Orthodox, and Ethiopian communities. Eligible participants were adults over 18 with T1DM, A1C ≥ 8.5%, treated with insulin pump or multiple daily injections. Exclusion criteria included pregnancy, lactation, severe renal failure, heart failure, or active cancer treatment. Of 107 recruited, 85 were randomized: 41 to regular carbohydrate counting (RCC) and 44 to SCC. All participants received up to six free 60-minute instructional sessions with a registered dietitian over six months. The RCC group received commercial carbohydrate-counting booklets and instruction on using websites/apps, personal insulin-to-carb ratios (I:C), insulin sensitivity (IS), correction factors, and glucose targets. The SCC group received two personalized tables in their native language: Table 1 listed insulin units needed to correct pre-meal blood glucose to target; Table 2 listed the participant's commonly consumed foods with carbohydrate content and corresponding insulin units per usual portion. For pump users, carbohydrate amounts were listed in grams for bolus calculator entry. The primary endpoint was A1C at 6 months. Secondary outcomes included PAID5 questionnaire scores (Problem Areas in Diabetes Scale), weight, lipid levels, and compliance (number of sessions attended). Statistical analysis used paired t-tests for within-group changes and Student's t-test for between-group differences, with stratification by age, sex, education level, and diabetes duration.
**Key Results:** Baseline characteristics were similar between groups: mean age 43.1 years (range 18-74), 47% women, mean diabetes duration 15.8 years, 40% on insulin pumps, mean A1C 9.9%. All participants improved A1C from baseline to 6 months (9.9% [13.2 mmol/L] vs 8.6% [11.1 mmol/L], p=0.001). Among participants aged 40 and older (mean age 55.2), only the SCC group showed significant A1C improvement (9.6% to 8.6%, p=0.002), while the RCC group did not (9.7% to 9.2%, p=0.09). Among younger participants (mean age 29.6), both groups improved significantly (RCC: 10.2% to 8.3%, p=0.001; SCC: 10.3% to 8.3%, p=0.002). No differences were found by education level: SCC participants with above and below 12 school years showed similar improvements (−1.4% vs −1.3%, p=0.7). Newly diagnosed participants (<5 years, n=13) showed greater A1C improvement than those with longer duration (−6.1 vs −0.7, p=0.005; −3.4 vs 0.9, p=0.032). Compliance was good (mean 4.8 of 6 sessions), with women more compliant than men (5.3 vs 4.4 visits, p=0.01). PAID5 scores improved across all participants (10.6 to 9.5, p=0.023), with women showing significant improvement (11.5 to 9.9, p=0.04) while men did not (9.9 to 9.18, p=0.27). Two RCC patients were hospitalized for DKA; no hypoglycemia-related hospitalizations occurred in either group.
**Clinical Implications:** The SCC tool offers a feasible, low-tech alternative to standard carbohydrate counting that may be particularly beneficial for older adults with T1DM who struggle with complex CC methods. Its personalized, culturally adaptable approach—using patients' own food preferences and native language—may improve glycemic control in populations with limited access to technology or lower health literacy. The non-inferiority of SCC across education levels suggests it can help reduce disparities in diabetes outcomes. The greater benefit observed in newly diagnosed patients underscores the importance of early, accessible diabetes education. The improvement in emotional well-being (PAID5 scores) across both groups highlights the value of structured diabetes education itself. Limitations include single-center design, one dietitian delivering the intervention, exclusion of pediatric patients, and need for larger studies in lower socioeconomic and culturally diverse populations to establish generalizability.