How to Forecast Annual Consumable and Maintenance Spend When Running a Multi-Site Sleep and Respiratory Service in 2026

Feb 20, 2026

Forecasting annual consumable and maintenance spend across multiple sleep and respiratory sites is not simply a budgeting exercise. It is a clinical operations discipline. Done well, it protects service continuity, prevents stockouts of critical consumables, and gives finance teams the confidence to approve your budget. Done poorly, it leads to reactive purchasing, cost blowouts, and avoidable equipment downtime. This guide gives respiratory and sleep service managers a practical, structured approach to spend forecasting that works across sites of different sizes, activity levels, and equipment mixes.

TL;DR

  • Multi-site respiratory and sleep services face unique forecasting challenges due to variable patient volumes, diverse equipment fleets, and site-specific consumable usage patterns.

  • Effective spend forecasting combines historical data analysis, consumption rate modelling, and predictive maintenance planning.

  • Centralising your data and standardising categories across sites is the single highest-leverage action you can take before building any forecast.

  • Software that eliminates data silos and integrates clinical activity with procurement signals dramatically improves forecast accuracy.

  • Starting simple and refining iteratively beats waiting for a perfect model that never gets built.

Why Is Spend Forecasting Harder for Multi-Site Respiratory and Sleep Services?

Multi-site forecasting is harder than single-site forecasting because variability compounds. A single site has one patient throughput pattern, one equipment fleet, and one team's consumption habits. Multiply that across five, ten, or twenty sites and the complexity grows non-linearly.

Specific challenges in respiratory and sleep services include:

  • Diverse equipment fleets: Sites may run different spirometry, polysomnography, or CPAP titration devices from different manufacturers, each with unique consumable requirements and maintenance schedules.

  • Variable patient volumes: Referral patterns, seasonal respiratory illness trends, and waitlist lengths differ significantly by geography and site type.

  • Consumable variability: Mouthpieces, filters, nasal cannulas, electrode sets, and calibration gases all have different consumption rates, shelf lives, and supplier lead times.

  • Decentralised purchasing: Without central visibility, individual sites over-order to feel safe, creating waste and inflating total spend.

  • Maintenance unpredictability: Equipment age, usage intensity, and service contract terms vary across sites, making maintenance cost forecasting feel like guesswork.

According to Verdantis, fragmented MRO (maintenance, repair, and operations) spend data is one of the primary drivers of cost inefficiency in multi-site operations. The principle applies directly to clinical services managing distributed equipment fleets.

What Data Do You Need Before You Can Build a Forecast?

A reliable forecast is built on clean, categorised, historical data. Before modelling anything, gather the following across all sites:

Data Category

What to Capture

Patient activity

Tests performed by type (spirometry, sleep study, CPAP titration, etc.) per site per month

Consumable usage

Units consumed per test type, per site, per period

Consumable costs

Unit costs by supplier, including any contract pricing

Equipment inventory

Device type, age, manufacturer, and current service contract status

Maintenance history

Dates, nature, and costs of past maintenance events per device

Supplier lead times

Average days from order to delivery for key consumable categories

HUB Industrial Supply notes that accurate consumable forecasting for large-scale operations depends on understanding both consumption rates and project scope before any modelling begins. For respiratory services, "project scope" translates directly to planned patient activity and test volumes for the coming year.

Once data is gathered, clean it. Remove anomalies caused by COVID-related shutdowns, equipment failures, or one-off bulk purchases that do not reflect normal operations. According to Zycus, cleaning and normalising historical spend data is a non-negotiable step before any forecasting model can produce reliable outputs.

How Do You Model Consumable Spend Across Sites?

The most practical approach for respiratory and sleep services is a consumption rate model, built bottom-up from test activity.

Step-by-step approach:

  1. Define your consumable categories by test type (e.g., spirometry consumables, sleep study consumables, calibration gases).

  2. Calculate a consumption rate for each category: units consumed per test, averaged across the last 12 to 24 months of clean data.

  3. Project test volumes for the coming year by site, using referral trends, waitlist data, and any planned capacity changes.

  4. Multiply consumption rate by projected volume to get a unit forecast per category per site.

  5. Apply current unit costs (adjusted for any known supplier price changes) to convert unit forecasts into dollar forecasts.

  6. Aggregate across sites and add a contingency buffer, typically 10 to 15%, to account for demand variability.

EZO recommends conducting frequent consumption audits to study usage patterns over time, noting that this practice is one of the most reliable ways to improve inventory and spend forecast accuracy. For multi-site services, this means building site-level consumption reviews into your quarterly operations rhythm, not just your annual budget cycle.

How Do You Forecast Maintenance Spend Without Guessing?

Maintenance spend forecasting sits at the intersection of equipment age, usage intensity, and service contract coverage. The goal is to move from reactive (fixing things when they break) to predictive (anticipating when things will need attention).

Key inputs for maintenance forecasting:

  • Equipment age and expected service life for each device across all sites

  • Manufacturer-recommended service intervals and associated costs

  • Historical maintenance event frequency and cost per device type

  • Current service contract coverage and what is excluded

  • Planned equipment replacements or additions

Research published in Frontiers in Manufacturing Technology explored how predictive maintenance models using real-time data analytics can reduce machine downtime and improve operational insights. While the study focused on industrial manufacturing, the finding that shifting from reactive to predictive maintenance meaningfully reduces unplanned costs is directly relevant to clinical equipment management.

Infodeck recommends building maintenance budgets around three cost categories: planned preventive maintenance, unplanned corrective maintenance (budgeted as a percentage of asset replacement value), and capital maintenance reserves for end-of-life equipment. Applying this structure across your equipment fleet gives finance teams a framework they recognise and trust.

How Do You Turn a Forecast Into a Budget That Gets Approved?

A forecast is a number. A budget is a story. Finance teams approve stories that are credible, defensible, and tied to operational outcomes.

  • Segment your forecast into consumables, planned maintenance, and unplanned maintenance reserves. Do not present a single lump sum.

  • Tie spend to activity. Show that your consumable forecast is directly proportional to projected test volumes. This makes the budget self-justifying if volumes increase.

  • Benchmark where possible. Reference industry norms for maintenance spend as a percentage of asset value to validate your estimates.

  • Quantify the cost of under-budgeting. Deferred maintenance on spirometry or polysomnography equipment creates clinical risk and compliance exposure, not just operational inconvenience.

  • Present a range, not a single figure. A base case, upside, and downside scenario demonstrates analytical rigour and gives decision-makers options.

According to Harvard Business Review, credible forecasting requires understanding the drivers of demand, not just projecting past trends forward. For respiratory services, this means understanding what is driving referral growth or decline at each site, and reflecting that in your activity projections.

How Can Rezibase Help With Spend Forecasting?

Accurate spend forecasting depends on having accurate, centralised activity data. When clinical activity data lives in disconnected systems across sites, building a reliable consumption rate model becomes a manual, error-prone exercise.

Rezibase, Australia's most advanced cloud-based respiratory and sleep reporting platform, gives multi-site services a single source of truth for clinical activity across all locations. Because Rezibase is vendor-neutral and integrates with existing hospital systems including Patient Administration Systems and Electronic Medical Records, it captures test volume data regardless of which equipment manufacturer a site uses. This makes it significantly easier to extract the site-level activity data that underpins a reliable consumable spend forecast.

For services currently using Respiro and considering a move to Rezibase, the transition is designed to be straightforward. Data migration is handled as part of the onboarding process, so historical activity data, which is essential for building your consumption rate baselines, is preserved and accessible from day one.

Frequently Asked Questions

How far back should I look when analysing historical consumable spend?
Twelve to twenty-four months of clean data is generally sufficient. Go further back only if you need to identify long-term trends or if recent data was distorted by unusual events.

How do I handle sites with very different patient volumes in a single forecast?
Build the forecast bottom-up at the site level, then aggregate. Applying a single average across sites with very different activity levels will produce figures that are wrong for every site.

What contingency buffer should I apply to a consumable forecast?
A 10 to 15% buffer is standard for services with reasonably stable volumes. Higher variability or new sites warrant a larger buffer of up to 20%.

How often should I update the forecast during the year?
Review against actuals quarterly. Update projections if patient volumes deviate significantly from plan or if supplier pricing changes.

What is the biggest mistake services make when forecasting maintenance spend?
Budgeting only for planned maintenance and ignoring unplanned corrective maintenance. A reserve of 1 to 3% of asset replacement value for unplanned events is a widely used benchmark.

How do I standardise consumable categories across sites that use different equipment?
Map consumables to test type rather than device model. A spirometry mouthpiece is a spirometry mouthpiece regardless of which device it is used with. This allows cross-site comparison even with different equipment fleets.

Is it worth investing in forecasting software for a small number of sites?
Even for two or three sites, the time saved in manual data aggregation and the improvement in forecast accuracy typically justifies the investment within the first budget cycle.

About Rezibase

Rezibase is a cloud-based respiratory and sleep reporting platform built by respiratory scientists for respiratory scientists. Trusted by over 35 sites across Australia and the UK, including NHS and NSW Health, Rezibase offers a vendor-neutral, all-inclusive solution covering reporting, accreditation, administration, and integrations with hospital systems. It is designed to reduce clinical risk, eliminate data silos, and make life easier for the scientists and clinicians who run respiratory and sleep services every day.

To learn more or to start a free 30-day trial, visit rezibase.com.

References

SaaS vs. Perpetual Licensing in Clinical Diagnostics: A 5-Year Total Cost of Ownership Model for Respiratory Lab Managers ›

SaaS vs. Perpetual Licensing in Clinical Diagnostics: A 5-Year Total Cost of Ownership Model for Respiratory Lab Managers ›