Objective. To explore the application of the stratified nursing mode of the prediction model constructed based on case system data in the nursing of patients with acute renal failure (ARF). Methods. A total of 84 patients with ARF confirmed in the hospital were enrolled between February 2020 and February 2022. According to the simple random grouping method, they were divided into an observation group and a control group, 42 cases in each group. The control group was given routine nursing while the observation group was given stratified nursing of the prediction model constructed based on case system data. All were nursed for 2 months. Results. There was no significant difference in general data such as gender, age, body mass index (BMI), serum creatinine (Scr), hemoglobin (Hb), and albumin between the two groups (
P
>
0.05
). Age >60 years, weight fluctuation >2 kg during dialysis, vascular blockage or infection, coronary heart disease, diabetes mellitus, chronic hepatopathy and stroke, bleeding tendency, and neuromuscular abnormalities were high-risk factors for ARF patients, hypertension, thyroid abnormalities, hyperlipidemia, persistent or repeated blood volume overload, and usage of antihypertensive drugs were moderate-risk factors for ARF patients, and nonpermeability dehydration was a low-risk factor of ARF patients. The scores of nursing satisfaction and treatment compliance in the observation group were significantly higher than those in the control group (
P
<
0.05
). After 2 months of nursing, scores of SAS, SDS, and SPBS in both the groups were significantly decreased (
P
<
0.05
), which were significantly lower in the observation group than those in the control group (
P
<
0.05
). Conclusion. The stratified nursing mode of the prediction model constructed based on case system data is conducive to timely and targeted nursing, with high patient satisfaction and cooperation, and a better psychological state.