Modelling policy interventions to improve patient access to rural dermatology care

Abstract

Timely access to dermatology care is poor across the US, especially in underserved geographical areas. Rural regions with fewer practising dermatologists and constrained resources often experience insufficient care access and health outcomes, underscoring the importance of addressing these disparities. However, potential interventions are difficult to compare due to their disruptiveness, time and resource requirements, and institutional resistance, given their uncertain impacts. Queueing and computer simulation models were used to analyse several potential interventions to reduce dermatology appointment delays and gain insights into dynamics and structural inter-relationships. Model logic, candidate interventions, and cost–benefit considerations were developed from mixed-methods analyses of rural access processes and barriers. Sensitivity analyses were conducted. The best of ten investigated interventions reduced internal dermatology access delays from roughly 150 weeks with 95% provider utilisation currently to 0.49 and 72%, respectively. Two other interventions reduced travel for external dermatology care by an estimated 68.1%. Model logic and inputs were developed from the literature and a six-facility rural health system, which may differ in other geographic regions. Model simplifications may not capture all access dynamics, and resources required for some interventions may not be available. Model-based analysis of rural care access disparities can help evaluate and screen potential interventions, develop useful insights, and identify policies worth further evaluating or testing in actual practise. In our rural setting, two interventions appear cost-effective in reducing patient access delays and provider over-utilisation; two others performed too poorly to warrant investing resources to implement or test in practise.

Appendices

Appendix

Table 4 Thematically grouped primary care stakeholder suggestions for improving access to dermatology care (without hiring new dermatologists or dermatology advanced practise clinicians). PCP, primary care provider

Interventions

Table 4 summarises key themes that were observed from our mixed-methods analysis of patient surveys and staff interviews. Themes were grouped via thematic analysis into five primary categories (clinician education, scheduling, resource identification, patient education, and communication and technology) and informed the specific interventions evaluated in this study (Table 1). Numeric inputs for each intervention are summarised in Table 2.

Simulation Model

The pseudocode used in all simulation models is summarised in Figure 4.

Fig. 4

Simulation model logic pseudocode summarising the general calculations, random number generation, and calculation of performance measures

Cost considerations

Table 5 summarises key patient and health system cost categories to consider for each intervention, although cost considerations will vary in any given setting (e.g., reimbursement rates, geographical location, overhead).

Table 5 Summary of anticipated cost–benefit variables by intervention scenario. All costs are considered ongoing unless otherwise noted (*, i.e., one-time costs). Support staff is defined as medical or nursing assistants and scheduling support. NPC, non-physician clinician (i.e., nurse practitioner or physician assistant); PCP, primary care provider; QALY, quality-adjusted life-year; QOL, quality of life; RVU, relative value unit