D458 Task 1 Iceberg Model Analysis Example

More Clients, More Missed Visits: An Iceberg Model Analysis of a Composite Home Care Agency That Grew Faster Than Its Caregivers

Student Name

Leavitt School of Health, Western Governors University

D458: Introduction to Systems Thinking for Health Professionals, Task 1

Course Instructor

Month Day, Year

More Clients, More Missed Visits: An Iceberg Model Analysis of a Composite Home Care Agency That Grew Faster Than Its Caregivers

The Scenario

Hearthside Home Care, a composite agency, was started six years ago by two sisters, Renee and Dana Castillo, both former certified nursing assistants. They began with eight clients, older adults in their own town who needed help bathing, dressing, cooking and taking medications on time, and they did many of the visits themselves. Word spread through churches and physicians' offices, and within three years the agency served 60 clients. Then a regional Medicaid managed care plan offered a contract to serve its members in three neighboring counties, and Hearthside accepted. To cover the costs of expansion, a second office, a scheduling software license and a leased van, the sisters drew on a business line of credit.

Today Hearthside serves 210 clients with 95 caregivers. Missed visits, shifts in which no caregiver arrives, have risen from about 1% of scheduled visits to 7%. Families call the office angry, the managed care plan has issued a warning, and Renee and Dana spend their evenings covering shifts themselves. Their response has been to recruit harder: they have run hiring advertisements every month and offered a $300 sign-on bonus. Yet the agency hires about as many caregivers each quarter as it loses.

Why the Iceberg Model

Systems thinking looks at how the parts of a situation are connected and how those connections produce behavior over time. A central idea is that a system's structure largely determines its behavior, so the same problems keep appearing, no matter who is involved, until the structure changes (Meadows, 2008). Systems methods are especially useful for problems that are complex, involve many actors and do not respond to simple fixes (Peters, 2014). The iceberg model organizes that thinking into four levels: events, which everyone can see; patterns, which appear over time; structures, which produce the patterns; and mental models, the beliefs that created and maintain the structures. The lower the level, the more lasting the change an intervention there can produce.

Events

The events are what the sisters see every day. A caregiver calls out at 6:00 a.m. and a client with dementia goes without breakfast and medication until noon. A family member finds their father alone after a scheduled visit never happened and calls the managed care plan. A new caregiver quits after her second week. Renee cancels her own plans to cover a Sunday shift. Each event is treated as a separate emergency and resolved by whoever is available.

Patterns

When the events are viewed over time, patterns appear. Missed visits cluster on weekends, early mornings and holidays, and in the two most rural counties, where clients live farthest apart. They rose sharply each time the agency added a batch of new clients from the managed care contract, then eased slightly before rising again. About half of new caregivers leave within 90 days, and those who leave most often worked split shifts in the rural counties. The sisters themselves are working more hours each month, and the scheduling coordinator has changed three times in two years. Recruiting advertisements and bonuses produce a burst of hires, followed by a burst of departures.

Structures

Several structures produce these patterns. First, pay: caregivers are paid by the visit, not by the hour, and they are not paid for travel time between clients. A rural caregiver may drive 25 minutes between one-hour visits, so her effective hourly pay falls below that of a retail job in town. The Medicaid rate the agency receives is fixed, and the line of credit's interest payments consume the margin that could have raised wages. Second, scheduling: one coordinator builds every schedule by hand in the new software, with no float pool of caregivers to cover call-outs and no clustering of clients by geography. Third, growth: the managed care contract has no cap on referrals, and the agency accepts every one because each new client brings revenue that helps pay the debt. Fourth, support: new caregivers receive two days of orientation, then work alone in clients' homes, with no mentor and no one to call after hours except the owners.

These structures form a reinforcing loop. Missed visits create pressure to hire quickly; rushed hiring and thin orientation lead to early turnover; turnover leaves more shifts uncovered, producing more missed visits. A second loop runs through the owners: as they cover more shifts, they have less time to fix scheduling or pay, so the structures that produce missed visits stay in place. The research on direct care workers points to how much structure matters. In a national study of more than 18,000 direct care workers, the estimated probability of turnover was lower for unionized than for nonunionized workers, 37.4% versus 45.0%, and differed by employer ownership (Gusoff et al., 2026), suggesting that pay, voice and working conditions shape who stays.

Mental Models

Beneath the structures are beliefs the sisters have never questioned. The first is that growth equals success: every referral accepted is a sign the agency is thriving, so saying no feels like failure. The second is that caregivers are replaceable: when one leaves, the answer is another advertisement, not a change in the job. The third is that the owners must personally solve every problem, a belief that made sense with eight clients and now keeps them too busy to change anything. A fourth belief, shared by the managed care plan, is that a fixed visit rate is fair regardless of distance, which leaves rural visits underfunded. These mental models explain why the agency keeps recruiting harder instead of redesigning the work.

Where a Systems View Would Intervene

The iceberg points to the places below the waterline where change lasts. At the level of mental models, the sisters could redefine success as visits completed and caregivers retained rather than clients enrolled, and pause new referrals until missed visits fall below 2%. At the level of structures, they could pay for travel time, build routes that cluster clients geographically, create a small float pool paid a standby rate, pair each new caregiver with an experienced mentor for the first month, and negotiate a rural add-on with the managed care plan using their own data on travel time. Recruiting would still matter, but it would fill a job people want to keep. Responding only to events, by covering more shifts themselves, will keep the cycle turning.

Conclusion

Hearthside's missed visits are not a string of bad luck or unreliable workers. They are the predictable output of a system that pays by the visit, grows without limits and relies on its owners as the only backup. The iceberg model moves attention from each crisis to the patterns, structures and beliefs that keep producing them, and it shows that the most effective changes, paying for travel, limiting growth and redefining success, are ones the sisters had not considered while they were busy covering shifts.

References

Gusoff, G. M., Jang, H., Spertus, D., Abraham-Aggarwal, K., Avgar, A., & Sterling, M. (2026). Unionization, ownership status, and direct care worker turnover. JAMA Network Open, 9(4), e264636. https://doi.org/10.1001/jamanetworkopen.2026.4636

Meadows, D. H. (2008). Thinking in systems: A primer (D. Wright, Ed.). Chelsea Green Publishing.

Peters, D. H. (2014). The application of systems thinking in health: Why use systems thinking? Health Research Policy and Systems, 12, 51. https://doi.org/10.1186/1478-4505-12-51