Amoroso P J, Smith G S, Bell N S
U.S. Army Research Institute for Environmental Medicine (Amoroso), Natick, Massachusetts, USA.
Am J Prev Med. 2000 Apr;18(3 Suppl):174-87. doi: 10.1016/s0749-3797(00)00110-0.
Accurate injury cause data are essential for injury prevention research. U.S. military hospitals, unlike civilian hospitals, use the NATO STANAG system for cause-of-injury coding. Reported deficiencies in civilian injury cause data suggested a need to specifically evaluate the STANAG.
The Total Army Injury and Health Outcomes Database (TAIHOD) was used to evaluate worldwide Army injury hospitalizations, especially STANAG Trauma, Injury, and Place of Occurrence coding. We conducted a review of hospital procedures at Tripler Army Medical Center (TAMC) including injury cause and intent coding, potential crossover between acute injuries and musculoskeletal conditions, and data for certain hospital patients who are not true admissions. We also evaluated the use of free-text injury comment fields in three hospitals.
Army-wide review of injury records coding revealed full compliance with cause coding, although nonspecific codes appeared to be overused. A small but intensive single hospital records review revealed relatively poor intent coding but good activity and cause coding. Data on specific injury history were present on most acute injury records and 75% of musculoskeletal conditions. Place of Occurrence coding, although inherently nonspecific, was over 80% accurate. Review of text fields produced additional details of the injuries in over 80% of cases.
STANAG intent coding specificity was poor, while coding of cause of injury was at least comparable to civilian systems. The strengths of military hospital data systems are an exceptionally high compliance with injury cause coding, the availability of free text, and capture of all population hospital records without regard to work-relatedness. Simple changes in procedures could greatly improve data quality.
准确的伤害原因数据对于伤害预防研究至关重要。与民用医院不同，美国军事医院使用北约标准化协议（STANAG）系统进行伤害原因编码。据报道，民用伤害原因数据存在缺陷，这表明有必要对STANAG进行专门评估。
使用陆军总伤害与健康结果数据库（TAIHOD）来评估全球范围内陆军医院收治的伤害情况，特别是STANAG创伤、伤害及发生地点编码。我们对特里普勒陆军医疗中心（TAMC）的医院程序进行了审查，包括伤害原因和意图编码、急性损伤与肌肉骨骼疾病之间的潜在交叉情况，以及某些并非真正入院的医院患者的数据。我们还评估了三家医院中自由文本伤害注释字段的使用情况。
对全军伤害记录编码的审查显示，虽然非特定编码似乎被过度使用，但在原因编码方面完全符合要求。对一家医院的少量但深入的记录审查显示，意图编码相对较差，但活动和原因编码良好。大多数急性损伤记录和75%的肌肉骨骼疾病记录中都有特定伤害史的数据。发生地点编码虽然本质上不具体，但准确率超过80%。对文本字段的审查在80%以上的病例中提供了伤害的更多细节。
STANAG意图编码的特异性较差，而伤害原因编码至少与民用系统相当。军事医院数据系统的优势在于对伤害原因编码的极高合规性、自由文本的可用性，以及涵盖所有人群的医院记录，而不考虑与工作的相关性。程序上的简单改变可以大大提高数据质量。
