Staloff Jonathan, Gunnink Eric, Rojas Jorge, Wong Edwin S, Hagan Scott, Nelson Karin, Reddy Ashok
Center of Innovation for Veteran-Centered and Value-Driven Care, Veterans Affairs Puget Sound Health Care System, Seattle, Washington
Department of Family Medicine, University of Washington, Seattle, Washington.
Ann Fam Med. 2025 Jul 28;23(4):285-293. doi: 10.1370/afm.240360.
Secure messaging has increased in primary care, but we lack information on who uses it at a high level. Characterizing this population is important to inform policy about patients most likely to be affected by secure messaging reimbursement and coverage decisions; to contextualize the secure messaging work burden for primary care; and to understand how secure messaging complements use of other primary care access points. Our objectives were to describe characteristics of high users, assess their overall primary care and emergency department use, and determine the proportion of total Veterans Health Administration primary care messages exchanged by this group.
We conducted a 1-year retrospective cross-sectional analysis and multivariate logistic regression analysis of all veterans assigned to primary care on October 1, 2022 who exchanged at least 1 secure message with primary care by September 30, 2023. We calculated descriptive statistics and assessed whether certain veteran characteristics were associated with high user status (use in the 95th or higher percentile of all messages exchanged annually).
Patients were more likely to be high users if they were older (adjusted odds ratio [aOR] for age ≥75 years = 2.49); had highest frailty (fourth quartile aOR = 2.65) or high physical comorbidity (Gagne score of 2 aOR = 1.17 and score of 3 aOR = 1.14); lived with schizophrenia or bipolar disorder (aOR = 1.23); or had posttraumatic stress disorder, anxiety, or depression (aOR = 1.16) ( < .05 for each). Patients were less likely to be high users if they identified as Black (aOR = 0.74), Hispanic (aOR = 0.80), or Asian/Pacific Islander/Native Hawaiian (aOR = 0.85), or were male (aOR = 0.78) ( < .05 for each). High users exchanged 30.5% of all messages. Few high users of secure messaging were also high users of other primary care encounters (only 14.0% were high users of in-person visits, 11.3% of video visits, and 11.8% of telephone visits) or the emergency department (8.0% of visits).
A small percentage of veterans exchange a disproportionate share of total secure messages. Most high users of secure messaging do not use other primary care or emergency department services at the same high frequency as secure messaging. High users commonly have high physical comorbidity, are frail, and live with mental health conditions, and racial disparities exist.
安全消息传递在初级保健中有所增加，但我们缺乏关于谁大量使用它的信息。了解这一人群的特征对于制定政策很重要，这些政策涉及最有可能受到安全消息传递报销和覆盖范围决策影响的患者；将初级保健中安全消息传递的工作负担置于背景中；以及了解安全消息传递如何补充其他初级保健接入点的使用。我们的目标是描述高用户的特征，评估他们对初级保健和急诊科的总体使用情况，并确定该群体在退伍军人健康管理局初级保健消息交换总量中所占的比例。
我们对2022年10月1日分配到初级保健的所有退伍军人进行了为期1年的回顾性横断面分析和多变量逻辑回归分析，这些退伍军人在2023年9月30日前至少与初级保健交换了1条安全消息。我们计算了描述性统计数据，并评估了某些退伍军人特征是否与高用户状态相关（每年交换的所有消息中处于第95百分位或更高百分位的使用情况）。
如果患者年龄较大（年龄≥75岁的调整优势比[aOR]=2.49）；虚弱程度最高（第四四分位数aOR=2.65）或身体合并症较多（加涅评分为2时aOR=1.17，评分为3时aOR=1.14）；患有精神分裂症或双相情感障碍（aOR=1.23）；或患有创伤后应激障碍、焦虑或抑郁症（aOR=1.16）（每项P<0.05），则他们更有可能成为高用户。如果患者自认为是黑人（aOR=0.74）、西班牙裔（aOR=0.80）或亚裔/太平洋岛民/夏威夷原住民（aOR=0.85），或者是男性（aOR=0.78）（每项P<0.05），则他们成为高用户的可能性较小。高用户交换了所有消息的30.5%。很少有安全消息传递的高用户也是其他初级保健就诊的高用户（只有14.0%是面对面就诊的高用户，11.3%是视频就诊的高用户，11.8%是电话就诊的高用户）或急诊科就诊的高用户（8.0%的就诊）。
一小部分退伍军人交换的安全消息占总数的比例过高。大多数安全消息传递的高用户使用其他初级保健或急诊科服务的频率与安全消息传递的频率不同。高用户通常身体合并症较多、身体虚弱且患有精神健康疾病，并且存在种族差异。
