<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Enochjeremiah on Medium]]></title>
        <description><![CDATA[Stories by Enochjeremiah on Medium]]></description>
        <link>https://medium.com/@enochjeremiah6?source=rss-fad9c0738e97------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*dNyyuSiyy-vWHmV9</url>
            <title>Stories by Enochjeremiah on Medium</title>
            <link>https://medium.com/@enochjeremiah6?source=rss-fad9c0738e97------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 11:18:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@enochjeremiah6/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Getting Data from Different Sources in Power BI: A Practical Guide]]></title>
            <link>https://medium.com/@enochjeremiah6/getting-data-from-different-sources-in-power-bi-a-practical-guide-ad8b6fa9ab90?source=rss-fad9c0738e97------2</link>
            <guid isPermaLink="false">https://medium.com/p/ad8b6fa9ab90</guid>
            <category><![CDATA[microsoft-power-bi]]></category>
            <category><![CDATA[excel]]></category>
            <category><![CDATA[microsoft]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[data-analytics]]></category>
            <dc:creator><![CDATA[Enochjeremiah]]></dc:creator>
            <pubDate>Fri, 03 Apr 2026 23:06:15 GMT</pubDate>
            <atom:updated>2026-04-03T23:06:15.759Z</atom:updated>
            <content:encoded><![CDATA[<h4>How to connect, import, and transform data from files, databases, APIs, and cloud platforms using Power BI Desktop</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R_dx71UPBfQeQB87eIzMIw.png" /><figcaption>Getting Data from Different Sources in Power BI: A Practical Guide</figcaption></figure><h3>Introduction</h3><p>One of Power BI’s most powerful capabilities is its ability to connect to virtually any data source, from a simple Excel file on your desktop to a live SQL database or a cloud-based API. Understanding how to pull data from different sources is the foundation of any meaningful data analysis project. Without clean, connected data, your dashboards are empty canvases.</p><p>In this guide, we will walk through the most commonly used data sources in Power BI, how to connect to each one, and best practices to keep your data pipeline clean and efficient. Whether you are just getting started or looking to expand your Power BI skills, this article covers everything you need to know.</p><h3>What Is Power BI’s Data Connectivity Model?</h3><p>Before diving into specific sources, it helps to understand how Power BI handles data connections. There are two primary modes:</p><p><strong>Import Mode</strong> : Data is loaded directly into Power BI’s in-memory engine (VertiPaq). This is the fastest mode for analysis but requires scheduled refreshes to stay current.</p><p><strong>DirectQuery Mode</strong> :Power BI sends queries live to the source database and returns results in real time. No data is stored inside Power BI, which is ideal for large datasets or data that changes frequently.</p><p><strong>Live Connection</strong> :Used specifically with Analysis Services models or Power BI datasets already published to the service.</p><p>For most beginner and intermediate users, Import Mode is the recommended starting point.</p><h3>Connecting to Data Sources: Step by Step</h3><p>To connect to any data source in Power BI Desktop, follow this standard starting path:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Rjk78Mj-kVk7QmoV6wkHsQ.png" /><figcaption>Home Page of Data Sources</figcaption></figure><blockquote><strong><em>Home tab → Get Data → Choose your source → Connect</em></strong></blockquote><p>Power BI organises its connectors into categories: File, Database, Power Platform, Azure, Online Services, and Other. Let’s go through the most important ones.</p><h3>1. Excel and CSV Files</h3><p>Excel and CSV are the most commonly used data sources for analysts, especially when working with business reports, survey results, or exported records.</p><p><strong>How to Connect:</strong></p><ol><li>Open Power BI Desktop</li><li>Click <strong>Home → Get Data → Excel Workbook</strong> (or <strong>Text/CSV</strong> for CSV files)</li><li>Browse to your file and click <strong>Open</strong></li><li>In the Navigator pane, select the table or sheet you want to load</li><li>Click <strong>Load</strong> to import directly, or <strong>Transform Data</strong> to open Power Query for cleaning</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XwxOYk55TyoTsOO4qz9CGQ.png" /><figcaption>Excel File Option</figcaption></figure><p><strong>What to Watch For:</strong></p><ul><li>Make sure your Excel file uses proper tabular formatting (no merged cells, consistent headers)</li><li>For CSV files, Power BI will detect the delimiter automatically, but you can change it in Power Query if needed</li><li>Avoid selecting entire sheets with blank rows, always select named tables where possible</li></ul><p><strong>Pro Tip:</strong> Convert your Excel ranges to proper Tables (Insert → Table in Excel) before connecting. Named tables import more reliably and show up clearly in the Power BI Navigator.</p><h3>2. SQL Server Database</h3><p>SQL Server is one of the most widely used enterprise databases, and Power BI has native, deep support for it.</p><p><strong>How to Connect:</strong></p><ol><li>Go to <strong>Get Data → SQL Server</strong></li><li>Enter the <strong>Server name</strong> and optionally the <strong>Database name</strong></li><li>Choose your data connectivity mode (Import or DirectQuery)</li><li>Authenticate using Windows or SQL Server credentials</li><li>Select your tables or write a custom SQL query in the advanced options</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HxDTV07PiA3D7qhOJm6h6w.png" /><figcaption>SQL Import and Direct Query</figcaption></figure><p><strong>Using a Custom SQL Query:</strong></p><p>Instead of loading a full table, you can filter at the source:</p><pre>SELECT CustomerID, CustomerName, Region, TotalPurchases<br>FROM Sales.Customers<br>WHERE TotalPurchases &gt; 5000<br>  AND Region = &#39;Lagos&#39;</pre><p>This reduces the data loaded into Power BI and improves performance.</p><p><strong>What to Watch For:</strong></p><ul><li>Always use DirectQuery if your database is updated in real time and you cannot wait for scheduled refreshes</li><li>Avoid loading unnecessarily wide tables, select only the columns your report actually needs</li><li>For large organisations, work with your database administrator to get read-only access credentials</li></ul><h3>3. Web / Online URLs</h3><p>Power BI can scrape data from web pages or pull structured content from public URLs, useful for fetching tables from government portals, Wikipedia, or publicly available datasets.</p><p><strong>How to Connect:</strong></p><ol><li>Go to <strong>Get Data → Web</strong></li><li>Paste the URL of the webpage containing your data</li><li>Power BI will detect any tables present on the page</li><li>Select the table you want and click <strong>Load</strong> or <strong>Transform Data</strong></li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zXRPPH0Vuai5lkdd70f7yQ.png" /><figcaption>Web Option</figcaption></figure><p><strong>Example Use Case:</strong></p><p>Imagine you want to pull a live table of currency exchange rates or a list of country population statistics from a government website. Rather than downloading a file manually each time, Power BI can fetch it automatically on refresh.</p><p><strong>What to Watch For:</strong></p><ul><li>Web scraping only works well on pages with clearly defined HTML tables</li><li>Dynamically loaded content (JavaScript-rendered tables) will not be visible to Power BI’s web connector</li><li>Always check the website’s terms of use before automating data pulls</li></ul><h3>4. REST APIs (Using Web Connector + JSON Transformation)</h3><p>Many modern data platforms expose data through REST APIs, think social media platforms, CRM systems, or custom business applications. Power BI can connect to these through the Web connector and then parse the JSON response.</p><p><strong>How to Connect:</strong></p><ol><li>Go to <strong>Get Data → Web</strong></li><li>Switch to the <strong>Advanced</strong> tab</li><li>Enter the API endpoint URL, and add any required headers (such as Authorization tokens)</li><li>Click <strong>OK, </strong>Power BI fetches the raw JSON</li><li>In Power Query, click <strong>Convert to Table</strong>, then expand the nested records</li></ol><p><strong>Example:</strong></p><p>Suppose you are connecting to a weather API that returns JSON like this:</p><pre>{<br>  &quot;city&quot;: &quot;Lagos&quot;,<br>  &quot;temperature&quot;: 32,<br>  &quot;humidity&quot;: 78,<br>  &quot;condition&quot;: &quot;Partly Cloudy&quot;<br>}</pre><p>Power Query will parse this into structured columns automatically after you expand the record.</p><p><strong>What to Watch For:</strong></p><ul><li>APIs that require OAuth2 authentication need additional setup, consider using a custom connector or Power Automate for complex auth flows</li><li>Paginated APIs require custom M code to loop through pages</li><li>Always test your API in a browser or tool like Postman before connecting in Power BI</li></ul><h3>5. SharePoint and OneDrive</h3><p>For teams working within Microsoft 365, SharePoint and OneDrive are natural homes for shared Excel files, CSV reports, and lists. Power BI connects to these natively.</p><p><strong>How to Connect (OneDrive/SharePoint Folder):</strong></p><ol><li>Go to <strong>Get Data → SharePoint Folder</strong> or <strong>OneDrive</strong></li><li>Enter your SharePoint site URL (e.g., <a href="https://yourcompany.sharepoint.com/sites/YourSite)">https://yourcompany.sharepoint.com/sites/YourSite)</a></li><li>Authenticate with your Microsoft 365 account</li><li>Browse and select the file or folder you want to connect</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u-2nLF2wiejm8pS1McpNxQ.png" /></figure><p><strong>SharePoint Lists:</strong></p><p>SharePoint lists work like database tables. You can connect directly:</p><ol><li><strong>Get Data → SharePoint Online List</strong></li><li>Enter the site URL</li><li>Select the list you want to import</li></ol><p><strong>What to Watch For:</strong></p><ul><li>Connect to the SharePoint folder, not individual files, when you are combining multiple Excel files stored in the same location</li><li>Permissions matter, your Power BI account must have at least read access to the SharePoint site</li></ul><h3>6. Azure Services (Azure SQL, Blob Storage, Data Lake)</h3><p>Power BI integrates deeply with Microsoft Azure, making it an ideal choice for organisations using cloud infrastructure.</p><p><strong>Azure SQL Database:</strong></p><p>The connection process is nearly identical to SQL Server, but you will use your Azure SQL Server address (format: yourserver.database.windows.net) and authenticate using your Azure credentials or SQL authentication.</p><p><strong>Azure Blob Storage:</strong></p><ol><li>Go to <strong>Get Data → Azure → Azure Blob Storage</strong></li><li>Enter your storage account name</li><li>Authenticate with your storage account key or connection string</li><li>Browse your containers and select your files (CSV, Parquet, JSON, etc.)</li></ol><p><strong>Azure Data Lake Storage Gen2:</strong></p><p>Ideal for organisations with large-scale data engineering pipelines. Connect via <strong>Get Data → Azure → Azure Data Lake Storage Gen2</strong>, enter your storage URL, and browse your file system.</p><h3>7. Google Analytics and Other Online Services</h3><p>Power BI includes built-in connectors for popular SaaS platforms, including Google Analytics, Salesforce, Dynamics 365, and more.</p><p><strong>Connecting to Google Analytics:</strong></p><ol><li>Go to <strong>Get Data → Online Services → Google Analytics</strong></li><li>Sign in with your Google account</li><li>Select the account, property, and view you want to import</li><li>Choose the dimensions and metrics you need</li></ol><p><strong>What to Watch For:</strong></p><ul><li>Google Analytics 4 (GA4) requires the GA4 connector, not the older Universal Analytics connector</li><li>Set sensible date ranges — pulling all historical data can be very slow</li><li>Be aware of data sampling in Google Analytics for large date ranges</li></ul><h3>Combining Multiple Data Sources</h3><p>One of Power BI’s most powerful features is the ability to combine data from different sources in a single report. Here is how to do it:</p><p><strong>Step 1:</strong> Connect to each source individually using the steps above.</p><p><strong>Step 2:</strong> Use <strong>Power Query</strong> to clean and shape each table.</p><p><strong>Step 3:</strong> In the <strong>Model view</strong>, create relationships between tables using common key columns (e.g., CustomerID, ProductCode, Date).</p><p><strong>Step 4:</strong> Build your report visuals using data from multiple connected tables.</p><p><strong>Example:</strong> You might combine a SQL database of transaction records with an Excel file of customer segments and a SharePoint list of product categories, all in a single Power BI model.</p><h3>Best Practices for Data Source Management</h3><p><strong>Use Power Query for Transformation, Not Excel</strong> Always apply data cleaning inside Power Query rather than in the source file. This makes your transformations reproducible and auditable.</p><p><strong>Minimise the Data You Load</strong> Only import the columns and rows your report actually needs. Use query folding (where supported) to push filtering back to the source database.</p><p><strong>Name Your Queries Clearly</strong> In Power Query, rename each query to reflect what it contains, for example, Sales_Transactions, Customer_Segments, or Product_Master. Avoid leaving defaults like Sheet1 or Table1.</p><p><strong>Document Your Data Sources</strong> Add descriptions to your queries and models so that other users (or future you) know where each table comes from and when it was last updated.</p><p><strong>Schedule Refreshes Appropriately</strong> Once published to Power BI Service, configure your dataset refresh schedule to match how frequently your source data changes, hourly, daily, or weekly.</p><h3>Conclusion</h3><p>Power BI’s broad range of data connectors is what makes it one of the most versatile business intelligence tools available today. Whether your data lives in an Excel file, a corporate SQL database, a cloud storage bucket, or a third-party API, Power BI can bring it all together in a single, unified model.</p><p>The key is to start with the right connector, apply transformations in Power Query, and build a clean data model before designing your visuals. Once you master data connectivity, the rest of Power BI opens up naturally.</p><p>As you continue building your Power BI skills, challenge yourself to connect to at least three different data sources in your next project, and explore how combining them can produce insights that no single source could deliver alone.</p><p><em>If you found this article helpful, consider sharing it with your network and leaving a comment below with the data sources you work with most. Happy analysing!</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ad8b6fa9ab90" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Data to Decisions: Inside a Healthcare Analytics Dashboard]]></title>
            <link>https://medium.com/@enochjeremiah6/from-data-to-decisions-inside-a-healthcare-analytics-dashboard-aa795d4e96e0?source=rss-fad9c0738e97------2</link>
            <guid isPermaLink="false">https://medium.com/p/aa795d4e96e0</guid>
            <category><![CDATA[exploratory-data-analysis]]></category>
            <category><![CDATA[microsoft-power-bi]]></category>
            <category><![CDATA[power-bi]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Enochjeremiah]]></dc:creator>
            <pubDate>Mon, 23 Feb 2026 10:28:13 GMT</pubDate>
            <atom:updated>2026-02-23T10:28:13.015Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*okx8N8KRyiTLFF5DAI3zDQ.png" /></figure><h3>A plain-English walkthrough of a Power BI hospital discharge report — what it shows, how it works, and why it matters</h3><h3>The Big Picture</h3><p>Imagine you’re a hospital administrator or a healthcare analyst. Every day, patients are admitted, treated, and discharged. Each one generates data — how long they stayed, what it cost, what illness brought them in, and where they went after. Multiply that by hundreds of hospitals across a state, and you’re sitting on an enormous amount of information.</p><p>The challenge isn’t collecting that data. It’s making sense of it.</p><p>This Power BI dashboard — built around a dataset called hospital_discharges — does exactly that. It answers three core questions for any healthcare organization:</p><ol><li><strong>Are our patients staying longer than they should be?</strong></li><li><strong>Are we spending more than comparable hospitals?</strong></li><li><strong>Who are our patients, and what risks do they carry?</strong></li></ol><p>Four pages. Three analytical lenses. One coherent story.</p><h3>Page 1: The Home Page</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WZJ5JFB3HJ40pvfHXBEXpQ.png" /><figcaption>Home Page</figcaption></figure><p>The home page is the front door — a visual welcome screen with navigation buttons that direct users to one of three analytical sections. It keeps things clean and accessible, especially for non-technical users who might be intimidated by a screen full of charts.</p><p><strong>Why this matters:</strong> Good dashboard design respects its audience. Not everyone who needs insights is a data analyst, and a clear starting point ensures that a hospital CEO and a data scientist can both find what they need.</p><h3>Page 2: LOS Comparison — How Long Are Patients Staying?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hxl9m-KVYWQrlBxCvu2lCw.png" /><figcaption>LOS Comparison Page</figcaption></figure><p><strong>LOS</strong> stands for <em>Length of Stay</em> — the number of days a patient spends in the hospital from admission to discharge. It’s one of the most closely watched metrics in healthcare because it directly impacts both cost and patient outcomes.</p><h3>What you see on this page:</h3><ul><li><strong>Three KPI cards</strong> showing total hospitals, total discharges, and average LOS days across all facilities</li><li><strong>A combo chart</strong> displaying average LOS days (line) and total discharges (bars) for the top 15 hospitals</li><li><strong>Top 3 and Bottom 3 bar charts</strong> highlighting which hospitals have the highest and lowest average LOS</li><li><strong>A Key Influencers visual</strong> that automatically identifies which patient characteristics (diagnosis, severity, risk of mortality, age, gender, disposition) are driving longer or shorter stays</li></ul><h3>The DAX behind the numbers</h3><p>DAX (Data Analysis Expressions) is the formula language that Power BI uses to calculate metrics. Think of it like Excel formulas, but far more powerful. Here are the core measures powering this page:</p><p><strong>Total Discharges</strong></p><pre>Total Discharges = COUNT(hospital_discharges[discharge_id])</pre><p><em>Plain English:</em> Count every row in the discharge table. Each row is one patient discharge. Simple, but foundational — almost every other calculation builds on this.</p><p><strong>Average LOS Days</strong></p><pre>Average LOS Days = AVERAGE(hospital_discharges[length_of_stay])</pre><p><em>Plain English:</em> Add up all the days patients stayed, then divide by the number of patients. This gives us the average experience across whatever filter is currently applied — whether that’s a single hospital or the entire state.</p><p><strong>% Var Average LOS Days</strong></p><pre>% Var Average LOS Days = <br>DIVIDE(<br>    [Average LOS Days] - [Average LOS Days ALL],<br>    [Average LOS Days ALL]<br>)</pre><p><em>Plain English:</em> How does this hospital’s average LOS compare to the state average? A positive number means patients are staying longer than average. Negative means they’re being discharged faster. This is the metric shown in tooltips on the top/bottom 3 bar charts.</p><p><strong>Average LOS Days ALL</strong> (the state benchmark)</p><pre>Average LOS Days ALL = <br>CALCULATE(<br>    [Average LOS Days],<br>    ALL(hospital_discharges)<br>)</pre><p><em>Plain English:</em> CALCULATE is DAX&#39;s most powerful function — it lets you recalculate a measure under a different filter context. ALL() removes all current filters, giving us the statewide average regardless of what hospital or slicer is selected. This becomes the benchmark that all hospitals are measured against.</p><p><strong>Total Surgeons</strong></p><pre>Total Surgeons = DISTINCTCOUNT(hospital_discharges[surgeon_id])</pre><p><em>Plain English:</em> Count the unique surgeons involved in discharges. This appears as a tooltip to give context — a hospital with high LOS but very few surgeons might just be handling more complex cases.</p><h3>The Key Influencers visual — no formula needed</h3><p>One of the most impressive visuals on this page requires zero DAX. The Key Influencers visual is a built-in Power BI AI feature. You point it at a target metric (Average LOS Days) and a set of explanatory fields, and it automatically runs statistical analysis to surface what’s most correlated with increases or decreases in that metric.</p><p>For this report, the influencers being analyzed include:</p><ul><li>APR Risk of Mortality (Minor, Moderate, Major, Extreme)</li><li>APR Severity of Illness Description</li><li>CCS Diagnosis Description</li><li>Patient Disposition (discharged to home, transferred, etc.)</li><li>Gender and Age Bins</li><li>Health Service Area</li><li>Surgical Program Size</li></ul><p><em>Why this matters for non-technical users:</em> Instead of needing to run a regression analysis or pivot table, a clinical director can open this page and instantly see “patients classified as Extreme severity are 3.2x more likely to have an LOS above 7 days.” That’s an actionable insight delivered without any formula knowledge.</p><h3>Page 3: Cost Comparison — What Are We Spending?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DAbwrIjzub6gY7uFGUXXxg.png" /><figcaption>Cost-Comparison Page</figcaption></figure><p>Length of stay and cost are closely related — the longer a patient stays, the more it costs. But the relationship isn’t always linear. This page digs into cost independently.</p><h3>What you see on this page:</h3><ul><li><strong>Four KPI cards</strong> showing total hospitals, total discharges, average LOS days, and average cost per discharge</li><li><strong>A scatter plot</strong> with average LOS on the X-axis, average cost per discharge on the Y-axis, bubble size representing total discharges, and color representing health service area — all by hospital</li><li><strong>Top 3 and Bottom 3 bar charts</strong> for cost per discharge</li><li><strong>A Key Influencers visual</strong> targeting average cost per discharge</li></ul><h3>The DAX behind cost analysis</h3><p><strong>Average Cost per Discharge</strong></p><pre>Average Cost per Discharge = AVERAGE(hospital_discharges[total_charges])</pre><p><em>Plain English:</em> The mean cost across all discharges in the current filter context.</p><p><strong>% Var Average Cost per Discharge</strong></p><pre>% Var Average Cost per Discharge = <br>DIVIDE(<br>    [Average Cost per Discharge] - [Average Cost per Discharge ALL],<br>    [Average Cost per Discharge ALL]<br>)</pre><p><em>Plain English:</em> Same pattern as the LOS variance — how does a hospital’s average cost compare to the statewide benchmark? This appears in tooltips on the bar charts.</p><p><strong>Average Cost per Discharge ALL</strong></p><pre>Average Cost per Discharge ALL = <br>CALCULATE(<br>    [Average Cost per Discharge],<br>    ALL(hospital_discharges)<br>)</pre><p><em>Plain English:</em> The statewide average cost, calculated with all hospital filters removed — the benchmark.</p><h3>Reading the scatter plot</h3><p>The scatter chart is arguably the most insight-dense visual in the entire report. Each bubble is a hospital. Its position tells you two things simultaneously: where it sits on LOS (left = short stays, right = long stays) and where it sits on cost (bottom = low cost, top = high cost). The size of the bubble tells you how many discharges they handle — volume context.</p><p>The ideal hospital sits in the <strong>bottom-left quadrant</strong>: short stays, low cost. Hospitals in the <strong>top-right</strong> are outliers worth investigating.</p><p>Color-coding by health service area adds a regional dimension — you might notice that hospitals in one region consistently cluster in a higher-cost zone, which could indicate regional pricing differences or patient population differences.</p><h3>Page 4: Hospital Profile — Who Are Our Patients?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3zaNSG-LbVnMzDOUsmFGMw.png" /><figcaption>Hospital’s Profile Page</figcaption></figure><p>The first two pages compare hospitals against each other. This page zooms into a <em>single</em> hospital selected via a slicer and answers: what is the profile of the patient population at this facility?</p><h3>What you see on this page:</h3><ul><li><strong>A hospital name card</strong> showing the selected facility (driven by a measure called Title Selected Facility)</li><li><strong>Two gauge charts</strong> — one for LOS vs. state benchmark, one for cost vs. state benchmark — showing at a glance whether this hospital is above or below average</li><li><strong>Two column charts</strong> breaking down total discharges by illness severity and by risk of mortality</li><li><strong>Two donut charts</strong> showing discharges by diagnosis description and by patient disposition (where patients go after discharge)</li></ul><h3>The DAX powering the gauges</h3><p><strong>Average LOS Days ALL</strong> and <strong>Average Cost per Discharge ALL</strong> (as defined above) serve as the <em>target</em> values on the gauge charts. The needle shows where the selected hospital sits; the target line shows the state average. If the needle is past the target, that hospital is underperforming on that metric.</p><p><strong>Title Selected Facility</strong></p><pre>Title Selected Facility = <br>SELECTEDVALUE(hospital_discharges[facility_name], &quot;All Facilities&quot;)</pre><p><em>Plain English:</em> SELECTEDVALUE returns the currently selected value from a slicer — in this case, the hospital name. If no hospital is selected (or more than one is), it defaults to &quot;All Facilities.&quot; This measure drives the dynamic title card at the top of the page, making the report feel personalized.</p><h3>Reading the column and donut charts</h3><p>These charts don’t need complex DAX — they use Total Discharges broken down by categorical fields from the data:</p><ul><li><strong>Illness Severity</strong> (Minor, Moderate, Major, Extreme) — a high proportion of “Major” or “Extreme” cases explains why a hospital might have higher LOS and costs</li><li><strong>Risk of Mortality</strong> — similar logic; a hospital serving sicker patients isn’t necessarily inefficient</li><li><strong>Diagnosis Description</strong> — which diagnoses are most common at this facility?</li><li><strong>Patient Disposition</strong> — did most patients go home? Get transferred? This affects LOS calculations significantly</li></ul><p>Together, these visuals provide <em>context</em> for the numbers on the other pages. A hospital with a high average LOS might look alarming in the LOS Comparison page — but if the Hospital Profile shows 60% of their patients are classified as “Extreme” severity, the story changes entirely.</p><h3>The Data Model: What’s Under the Hood</h3><p>The report draws from two tables:</p><p><strong>hospital_discharges</strong> — the main fact table. Each row is one patient discharge and includes fields like:</p><ul><li>facility_name — the hospital</li><li>length_of_stay — days in hospital</li><li>total_charges — cost</li><li>apr_severity_of_illness_description — patient severity</li><li>apr_risk_of_mortality — mortality risk classification</li><li>ccs_diagnosis_description — what they were admitted for</li><li>patient_disposition — where they went after discharge</li><li>health_service_area — regional grouping</li><li>gender, Age Bins — demographic fields</li></ul><p><strong>surgical_program_size_summary</strong> — a supporting lookup table linking hospitals to their surgical program size category, used as an influencer in the Key Drivers visuals.</p><h3>Five Insights Any User Can Draw Immediately</h3><p>You don’t need to be a data analyst to use this dashboard. Here’s what a non-technical stakeholder can walk away with on first use:</p><ol><li><strong>“Which hospitals should I investigate first?”</strong> — Go to LOS or Cost Comparison. Sort by the top 3 bar chart. The hospitals with the highest positive % variance relative to state average are your starting point.</li><li><strong>“Is our hospital’s performance explained by patient complexity?”</strong> — Open the Hospital Profile for your facility. If your severity and mortality columns are skewed toward Major/Extreme, high LOS and cost are expected, not alarming.</li><li><strong>“What’s driving long stays across the state?”</strong> — The Key Influencers visual on the LOS Comparison page will tell you directly, in plain language.</li><li><strong>“Are expensive hospitals also keeping patients longer?”</strong> — The scatter plot on Cost Comparison answers this visually in seconds.</li><li><strong>“How many unique surgeons are contributing to our discharge volume?”</strong> — The Total Surgeons card and tooltip provide this instantly.</li></ol><h3>Conclusion</h3><p>This dashboard is a masterclass in the principle that good analytics isn’t about showing everything — it’s about surfacing the right things for the right audience. The LOS Comparison page speaks to operational efficiency teams. The Cost Comparison page speaks to finance. The Hospital Profile speaks to clinical leadership. And the home page makes sure everyone can find their page without a tutorial.</p><p>The DAX formulas follow a consistent pattern: calculate the metric, then calculate the same metric without filters for benchmarking, then calculate the variance between them. Once you understand that three-measure pattern, you can read — and build — dashboards like this one.</p><p>Data tells a story. This dashboard makes sure it’s one people can actually read.</p><p><em>Built with Microsoft Power BI · Data: New York State hospital discharge records · Dashboard theme: HealthStat</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aa795d4e96e0" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>