The Association of Established Primary Care with Postoperative Outcomes Among Medicare Patients with Digestive Tract Cancer

Primary care plays a crucial role in the management of all aspects of patient care, including cancer prevention, detection, and intervention. Access to these essential services may not be possible for all patients because PC shortages of 50,000 physicians disadvantage over 74 million Americans.^1,2 Previous studies have suggested that established and consistent use of PC services before surgery may improve the odds for favorable postoperative outcomes.^9,10 The current study was important as we specifically examined the impact that established PC had on individuals undergoing surgical interventions for hepatopancreatobiliary and colorectal malignancies. Notably, in the current study, roughly four in five Medicare beneficiaries who underwent surgery for digestive tract cancer had a least one visit with a generalist outpatient PC provider in the year before the index surgical procedure. Patients who lacked established PC were more likely to be a racial/ethnic minority, live in a rural area with high social vulnerability, and receive treatment at hospitals with a lower procedural volume. Established PC was associated with an increased likelihood to achieve optimal postoperative results and be associated with reduced expenses. Nevertheless, individuals with numerous PC appointments (>10/year) did not experience a benefit from more PC encounters, likely due to the fact that the extreme number of PC visits was a surrogate for higher baseline patient comorbidities.

Social determinants of health may shape access to care, a factor known to play a role in health care disparities. Moreover, social vulnerability has been linked to worse surgical outcomes across a range of medical specialties.^13,15,18,19 In the current study, patients who lacked established PC were more likely to be a racial/ethnic minority (no PC [22.1%] vs PC [18.4%]), socially vulnerable (no PC [35.6%] vs PC [32.6%]), or live in a rural area (no PC [26.5%] vs PC [14.1%]) (Table 1). These findings are consistent with previous research highlighting disparities in access to health care among vulnerable populations. For example, rural areas face significant access disparities due to cultural, educational, financial, and infrastructure limitations, resulting in poorer health outcomes, a higher comorbidity burden, and challenges recruiting health care professionals and maintaining related services.^20,21 Similarly, minority patients are more likely to lack access to PC providers than White patients; in addition, PC providers caring for minority patients are less inclined to adopt comprehensive preventive measures or possess board certification.^4,5 As a result, rural, minority, and socially vulnerable patients receive fewer preventive services (i.e., vaccination or comorbidity screening) despite equivalent insurance enrollment, and thus may be at a higher risk for adverse events.^22‐24 To this point, our own group previously noted that patients with high social vulnerability are more likely to experience fragmentation of surgical care.²⁵

Established PC has been associated with improved perioperative outcomes. For example, a previous study by Roberts et al.⁹ demonstrated that patients with PC exposure had lower mortality after emergency general surgery in both the in-hospital and the postoperative settings. Likewise, Hyer et al.¹⁰ demonstrated that preoperative continuity of care (defined as at least 4 outpatient appointments) in the year before hepatopancreatobiliary surgery was associated with lower risk for readmission and lower surgical-related health care expenditures. In the current study, patients with established PC were 14% more likely to achieve a TO (OR, 1.14) even after adjusting for confounding variables; in addition, patients with established PC had a corresponding lower risk of complications (OR, 0.89), extended hospital stay (OR, 0.81), readmission (OR, 0.94), and mortality (OR, 0.87) (Table 2). The benefits of established PC may occur through various multifactorial mechanisms. In terms of primary prevention, PC providers can influence patients to adopt lifestyle changes that mitigate risky behaviors, thereby decreasing the chance of cancer development.²⁶ Among patients with cancer, PC is crucial to earlier detection and timely referral for therapy, as well as management of non-cancer morbidities to limit treatment related comorbidity.^27,28 In the current study, individuals with established PC had a slightly higher incidence of pancreatic and hepatobiliary cancer diagnoses, as well as a tendency toward a lower disease stage upon presentation and intervention. There are inherent challenges to early detection, and many patients have nonspecific subtle symptoms especially with the initial stages of disease. In turn, lack of PC evaluation on a regular basis may lead to delayed presentation.^29,30

Moreover, beyond the impact on in-hospital outcomes, established PC also was associated with discharge disposition and health care expenses. Notably, patients lacking PC were less inclined to be discharged to home based on their own independent self-care and were more prone to require advanced care such as skilled nursing (Table S1). The higher resource utilization related to management of adverse postoperative events, as well as the increased need for advanced care resulted in increased health care expenditures among patients who lacked PC in the year prior to surgery. In particular, PC was associated with a 4% to 5% reduction in expenditures during the perioperative period (Fig. 1). Beyond cancer, continuity of care and consistent management of chronic conditions such as congestive heart failure, chronic obstructive pulmonary disease, and diabetes mellitus has been linked to reduced resource utilization, as evidenced by decreased rates of hospitalization, emergency department visits, and complications.³¹ Hyer et al.¹⁰ and Hussain et al.³² conducted two studies illustrating that maintaining continuity in established primary and surgical care led to a 12% and 20% decrease in annual total costs linked to the operative treatment of hepatopancreatic and colorectal cancers, respectively. These trends remain consistent across various surgical subspecialties, such as orthopaedic management of shoulder pathology.³³ Without effective patient discharge instructions and counseling, the responsibility of long-term management can fall on PC services. For instance, patients experiencing common wound-related symptoms after a laparoscopic cholecystectomy frequently sought care at PC clinics rather than attending surgical outpatient follow-up visits, thereby imposing a considerable burden on these community services.³⁴

The current study should be interpreted in the context of several limitations. As with any a retrospective study, data maintained in the registries may contain inherent misclassification and reporting biases.³⁵ The use of the Medicare dataset inherently limited the analysis to individuals 65 years old or older enrolled in Medicare, whereas the SEER database includes only data from specific counties and states. For example, trends by hospital volume in SEER-participating hospitals may not be representative of other facilities nationally. As a result, the findings might not be applicable to a nationwide younger demographic with private insurance, thus restricting the generalizability of the results. Moreover, our findings could have been affected by self-selection bias. Patients who have established PC may be more inclined to engage with practices that facilitate earlier detection and adopt beneficial preventive lifestyles, such as exercise, good diet, and avoidance of alcohol or smoking.³⁶ Conversely, patients who avoid or cannot use PC may delay seeking necessary cancer care, potentially predisposing them to disease progression and poorer postoperative outcomes. Furthermore, patients without established PC likely had more undiagnosed or undocumented comorbidities that may not have been fully captured by the claims data yet still complicated the postoperative course. Similarly, patients who had a single visit to PC solely for preoperative medical clearance might have been classified under “established care” despite not receiving comprehensive care. However, this differentiation could not be made using retrospective coding data. Although incorporating the SEER data enhances the fundamental oncologic clinical understanding, encompassing disease stage and severity, reliance on claims data might not comprehensively capture patient-specific intricacies in care, such as surgical factors and detailed comorbidity status.

In conclusion, one in five patients with digestive tract cancer lacked established PC within the year before surgical resection. Patients without established PC had later-stage disease diagnosed, often resided in rural areas, and were treated at hospitals with lower surgical volume. Established PC was associated with a higher probability of achieving ideal postoperative outcomes and lower costs. In contrast, patients with a very high number PC appointments (>10/year) experienced no improvement or even worse outcomes, likely due to these patients having more comorbidites that necessitated more frequent PC visits. Initiatives to improve access to care may improve prevention measures, multidisciplinary coordination of oncologic services, as well as postoperative surgical outcomes.