Patient Portal Use Among Older Adults With Dementia Diagnosis

This cohort study assesses the level of engagement with an electronic health management system among patients with recently diagnosed dementia and their caregivers.

Discussion | Primary care spending varied significantly by insurance type and area. The estimated 5.3% spending on primary care for the Medicare population in 2019 was similar to the 4.9% estimate in 2015. 1 The estimated $461 spending for those with group private insurance in 2019 was in between the estimated amounts for a commercially insured sample using narrow ($291) and broad ($538) definitions of primary care. 2 The study is limited by MEPS being subject to underreporting of medical visits, although MEPS produces nationally representative estimates of primary care spending for all insurance sources. The degree of underreporting varies little by population characteristics: age, sex, race and ethnicity, and insurance source, 4 and MEPS data are used to derive official estimates of national health care spending. 5 Imputation of some price data produces unbiased expenditure estimates, but MEPS variance estimates do not account for imputation. Several states have established targets for insurers' share of primary care spending under the assumption that higher shares are better. 6 Because this share varies by characteristic, adjusting targets for some characteristics, such as age, may be needed, although having different targets for other characteristics, such as race and ethnicity, is problematic. Consensus on the ideal share of medical spending on primary care may be difficult to reach, but the large fraction of the population with no primary care spending, especially among uninsured individuals, seems likely to be suboptimal.

Patient Portal Use Among Older Adults With Dementia Diagnosis
The patient portal has a growing role in health system navigation. Persons with dementia have especially complex health needs and less ability to perform electronic health management tasks than those without dementia. 1 Persons with dementia and their care partners have a range of information needs that could be addressed through the patient portal, 2 but little is known about patient portal practices in this population. We conducted a cohort study of older adults' patient portal interactions at a large academic health system by receipt and timing of dementia diagnosis.
Methods | We included patients aged 65 years or older with 2 or more evaluation and management visits during any 24 months within a 5-year period (October 3, 2017-October 2, 2022). The Johns Hopkins School of Medicine Institutional Review Board approved this study and waived the informed consent requirement because it used secondary data. We followed the STROBE reporting guideline. Patient portal activity was computed from date-or timestamped interactions, which contain login and session information. 3 Portal user and used portal accounts were categorized as registered accounts with 1 or more logins. To measure portal activity, we created the portal activity metric, the ratio of number of portal sessions to number of clinical encounters. We examined the portal activity metric for all portal users and their proxy (with shared-access credentials) by dementia diagnosis. We then compared several measures of portal activity by older adults with recently diagnosed dementia each month for the 12 months before vs after diagnosis.
Statistical significance before vs after diagnosis was determined using a paired t test, with 2-sided P = .05 indicating significance. Statistical analysis was performed with Stata 17 (StataCorp LLC).  (Table). Only 6.4% of patients had diagnosed dementia. Those with a diagnosis were similarly likely to be registered for the patient portal (71.2% vs 71.5%; P = .69) but were more likely to have a registered care partner with shared ac-cess to their portal account (10.4% vs 3.3%; P < .001) than those without a diagnosis.

Results
Persons with vs without dementia diagnosis had lower portal activity metric (3.88 vs 5.35; P < .001) but were similarly likely to be a portal user (65.5% vs 65.8%; P = .66) and sender of a similar number of messages from their portal account (28.77 vs 29.14; P = .77). More portal messages originated from registered care partners of patients with vs without a diagnosis (19.50 vs 13.85; P = .03) (Table).
Portal activity metric was significantly higher in the 12 months after than before dementia diagnosis (3.34 vs 2.02). Monthly number of messages sent (4.31 vs 3.32; P < .001) and number of sessions (6.54 vs 4.17; P < .001) were also higher in the 12 months after than before diagnosis (Figure).
Discussion | This analysis found that older adults with and without diagnosed dementia were similarly likely to be registered for the patient portal. Although minimal in both groups, persons with dementia were 3 times more likely to have a registered care partner, and these care partners were actively engaged in messaging. Additionally, portal activity was significantly greater after a diagnosis. A study limitation was use of electronic medical records, which are subject to missing and inaccurate data, including dementia, which is often underdiagnosed. We found that older adults with dementia and their care partners relied on the information and functionality afforded by the patient portal. These results, in conjunction with gaps in dementia care quality and the importance of care partner engagement and support, 5 have implications for modalities of systems-level dementia care initiatives that leverage the patient portal, including efforts to remedy the low uptake of shared-access or proxy portal registration. 6 Additionally, the results highlight the need to better support all patients, including those who desire or rely on care partners, through consumer-oriented health information technologies.

Evaluating the PEN-FAST Clinical Decision-making Tool to Enhance Penicillin Allergy Delabeling
Patient-reported penicillin allergy is associated with inappropriate prescribing, antibiotic resistance, and adverse outcomes. 1 Various clinical decision-making tools for penicillin allergy have been developed to guide antibiotic selection and delabeling strategies, 1-3 yet further validation is required to promote their use. PEN-FAST is a clinical decision-making tool with a high negative predictive value (NPV) that can identify patients with low-risk penicillin allergy who do not require skin testing prior to oral penicillin