Pulmonary function tests are performed by trained technologists or registered respiratory therapists and interpreted by physicians, most often pulmonologists. That split between doing and reading is important because each role carries different responsibilities and different failure modes. A poorly coached test produces data no physician can salvage, and a perfectly executed test can still be misread. Understanding who does what, and where things go wrong at each stage, helps you know what to expect and what questions to ask when you walk into the lab.
The Person Running the Test
In most hospital-based pulmonary function laboratories, the person strapping the nose clip on you and coaching you through the breathing maneuvers is a registered respiratory therapist (RRT) or a certified pulmonary function technologist (CPFT). These are credentialed healthcare professionals who have undergone specific training in how to operate spirometers, body plethysmographs, and diffusion-capacity equipment. Their job goes well beyond pressing buttons: they calibrate the equipment each day, verify that the device is reading accurately, position you correctly, demonstrate the maneuver, coach you through it, and judge in real time whether each effort meets technical quality standards before moving on.
Coaching is a bigger deal than most people realize. Spirometry, the most common pulmonary function test, requires you to inhale as deeply as possible and then blast the air out as hard and as long as you can. That is surprisingly difficult to do correctly, and many patients cut the exhale short, hesitate at the start, or fail to seal their lips around the mouthpiece. The technologist’s verbal cues and encouragement during the maneuver directly affect how usable the data turn out to be.1PubMed Central. Coaching patients during pulmonary function testing: A practical guide A good technologist can coax three reproducible efforts out of a nervous patient in minutes; a poor one may produce a chart full of technically unacceptable blows that the interpreting physician has to work around or throw out entirely.
In primary care offices, the picture changes. The person performing your spirometry might be a medical assistant or practice nurse rather than a dedicated respiratory therapist. That is not inherently a problem, but the training gap shows up in the data. One study of spirometry quality in general practice found that only about 19% of tests performed by practitioners who had received workshop training met full acceptability criteria, and among untrained practitioners that figure dropped to roughly 5%.2PubMed. Spirometry in primary care practice: the importance of quality assurance and the impact of spirometry workshops Those are sobering numbers, and they explain why specialists sometimes repeat spirometry that was already done at a primary care visit.
Why the Technologist’s Skill Matters More Than You’d Think
Pulmonary function testing is one of the few areas in medicine where the quality of the data depends almost entirely on how well the patient cooperates, and how well the technologist extracts that cooperation. A blood draw either gets blood or it doesn’t; an X-ray either captures the image or the tech repositions you. But spirometry is a maximal-effort test that asks you to perform at your physical limit while a stranger watches. The technologist’s ability to motivate, demonstrate, and troubleshoot in real time is the single largest variable in data quality.
Research backs this up. In the large Lung Health Study, which tracked thousands of participants, institution of site visits and technician performance monitoring was associated with improved test quality grades and maintenance of high quality grades over time. During only about 2% of test sessions were participants unable to produce three acceptable maneuvers when the technologists were well supervised.3American Review of Respiratory Disease. Spirometry in the Lung Health Study: 1. Methods and Quality Control That is an impressive figure and speaks to what dedicated, well-trained technologists can achieve under a structured quality program.
One lab that implemented a focused training intervention saw adherence to acceptability and repeatability criteria jump from 61% to 92% over four years, while a comparison lab without the same intervention stayed flat at around 60–65%.4PubMed. Adherence to acceptability and repeatability criteria for spirometry in complex lung function laboratories Similarly, in an industrial health setting in Iran, the proportion of spirometry tests meeting all acceptability criteria rose from about 15% before an intensive training program to over 70% afterward.5PubMed Central. Quality of spirometry tests and pulmonary function changes among industrial company workers in Iran: a two-year before-and-after study following an intensive training intervention The takeaway is clear: the technologist’s competence is not just a nice-to-have, it is the foundation that everything else rests on.
Who Interprets the Results
Once the breathing maneuvers are recorded and the software generates the numbers, a physician reviews them. In a hospital pulmonary function lab, that physician is usually a pulmonologist. They look at the flow-volume loops and the numerical values, compare them with reference ranges for your age, sex, height, and ethnicity, and issue a written interpretation that goes into your medical record. The interpretation typically states whether the pattern is normal, obstructive (airflow is limited, as in asthma or COPD), restrictive (lung volumes are reduced, as in pulmonary fibrosis), or mixed.
The American Thoracic Society and European Respiratory Society jointly publish technical standards for how those interpretations should be made, including which reference equations to use and where to draw the line between normal and abnormal. These guidelines were updated in 2021, replacing some of the fixed cutoff values that had been in use for decades with statistically derived lower limits of normal.6PubMed. ERS/ATS technical standard on interpretive strategies for routine lung function tests That shift matters because it means the “abnormal” threshold now varies by age and sex rather than being a single number applied to everyone. In practice, though, adoption of these updated standards is uneven, and not every lab or clinician uses the same criteria.
In primary care, interpretation often falls to the family physician or internist who ordered the test. That is perfectly appropriate for straightforward spirometry used to screen for or monitor asthma and COPD. Spirometry is within the scope of any well-trained primary care provider, and several clinical resources emphasize that the key is understanding quality criteria, using proper reference equations, and knowing when to refer complex cases to a specialist.7PubMed Central. Basic spirometry testing and interpretation for the primary care provider Where primary care interpretation gets into trouble is when the test quality is poor or when the clinical picture calls for more advanced testing like lung volumes or diffusion capacity, which require more specialized knowledge to read.
How Often Doctors Disagree on What the Results Mean
One of the less reassuring aspects of pulmonary function testing is how much physician interpretations can vary. This has been documented for decades. A classic study asked 26 pulmonary physicians to interpret results from ten consecutive PFTs from a single laboratory and found frequent disagreement in their assessment of respiratory impairment.8PubMed. Variability in interpretation of pulmonary function tests That was published in 1979, and the problem has not gone away.
A more recent multicenter European study put 120 pulmonologists to the test on 50 PFT cases, yielding 6,000 independent interpretations. Pattern recognition matched ATS/ERS guidelines in about 74% of cases on average, with individual physicians ranging from 56% to 88%. When asked to make a specific diagnosis based on the PFT plus clinical information, the average accuracy dropped to roughly 45%, with individual scores ranging from 24% to 62%.9European Respiratory Journal. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests A quarter-correct rate at the low end is not reassuring, and the wide spread shows that the expertise of the individual reader matters enormously.
Part of the disagreement traces to the criteria used. A review of published research found at least eleven different definitions of airflow obstruction in the literature, with the most common being an FEV1/FVC ratio below 70%, used in about a third of studies.10Respiration. Analysis of Variability in Interpretation of Spirometric Tests When doctors are not even using the same threshold to decide what counts as obstruction, it is no surprise their interpretations diverge. The 2021 ATS/ERS standards aim to reduce this kind of variability, but old habits and old criteria persist in many practices.
Training’s Effect on Interpretation Accuracy
If you are wondering whether all this variability is fixable, the answer is partly yes. The same Iranian industrial-health study that documented dramatic improvements in test quality also tracked physician interpretation accuracy. Before the training intervention, doctors interpreted spirometry correctly about 53% of the time. Afterward, that figure jumped to roughly 91%.5PubMed Central. Quality of spirometry tests and pulmonary function changes among industrial company workers in Iran: a two-year before-and-after study following an intensive training intervention That is a striking improvement, and it suggests that much of the error in PFT interpretation comes not from inherent difficulty but from insufficient training and practice. Physicians who do not read PFTs regularly tend to lose the skill, much like any other pattern-recognition task.
This is one reason referral labs staffed by pulmonologists who read dozens of PFTs a week tend to produce more reliable interpretations than settings where a generalist reads one or two a month. Volume matters for expertise, and expertise matters for accuracy.
Pediatric Pulmonary Function Testing
Testing children adds a layer of complexity to both the performance and interpretation sides. Young children often cannot perform the sustained, maximal-effort maneuvers that standard spirometry demands. Infants and toddlers require specialized infant PFT equipment and sedation or tidal-breathing techniques that are generally available only at specialized centers.11PubMed Central. Lung Function Tests in Infants and Children The technologists running these tests need training specific to pediatric techniques, and the interpreting physician is typically a pediatric pulmonologist.
Even for school-age children who can cooperate with standard spirometry, interpretation carries pitfalls. Reference values for children are more sensitive to small errors in height measurement, and lung growth is not linear, which means a child’s results can shift between “normal” and “abnormal” depending on which reference equation is used. Pediatric PFT specialists have cautioned that results need to be interpreted carefully and correlated with the child’s clinical condition rather than taken at face value.12PubMed. How to avoid misinterpreting lung function tests in children: a few practical tips If your child has had PFTs and the results seem borderline, it is reasonable to ask whether the reference equations used were appropriate for their age and size.
Beyond Spirometry: Who Handles Advanced Tests
Spirometry is the most common pulmonary function test, but the full PFT suite can include lung volume measurement (by body plethysmography or gas dilution), diffusing capacity for carbon monoxide (DLCO), and sometimes bronchoprovocation testing. These tests require more sophisticated equipment and more nuanced interpretation.
Diffusing capacity, for example, measures how efficiently gas crosses from the air spaces of the lungs into the blood. A low DLCO can point to conditions like interstitial lung disease, emphysema, or pulmonary vascular disease. Interpreting it requires understanding how lung volume affects the measurement: at half the expected lung volume, the DLCO will be about 80% of predicted, which might look mildly low even in a healthy person who simply has small lungs.13Chest. Interpretation of Diffusing Capacity Sorting out whether a low DLCO is real or an artifact of low lung volume is a judgment call that typically falls to a pulmonologist rather than a generalist.
Body plethysmography and bronchoprovocation testing are almost exclusively performed in dedicated pulmonary function laboratories by experienced technologists and interpreted by pulmonologists. These tests are rarely done in primary care. If your doctor orders one, you will likely be referred to a hospital or academic center lab.
Preoperative Pulmonary Assessment
One common reason people encounter PFTs outside of a respiratory clinic is before surgery, especially lung resection. Surgeons and anesthesiologists use PFT results to estimate whether a patient can tolerate having part of a lung removed. In this context, the test is still performed by a technologist and interpreted by a pulmonologist, but the results feed into risk-stratification models used by the surgical team. Preoperative PFTs are considered essential for lung resection surgery, though for other types of surgery their routine use is more debatable and should be tailored to the individual patient’s risk profile.14PubMed Central. Comprehensive Strategies for Preoperative Pulmonary Risk Evaluation and Management The interpreting physician in this scenario may be a pulmonologist providing a consultation, or an anesthesiologist experienced in perioperative lung function assessment.
Coding, Billing, and Supervision Requirements
Behind the scenes, there are regulatory rules about who can supervise and bill for PFTs. In the United States, Medicare has specific participation rules, and laboratories performing PFTs must have appropriate testing supervision in place, meet ATS testing standards, and ensure that the administrative structure supports compliance with all regulations.15PubMed. Pulmonary function testing: coding and billing issues In practical terms, this means a qualified physician must be available to oversee the testing, and the interpretation must be documented by a physician or other authorized provider. The technologist performs the test, but a physician’s involvement is required for the interpretation to be billable as a professional service. This regulatory structure reinforces the two-person model: one person does, another reads.
The Growing Role of Artificial Intelligence
Given the documented variability in human interpretation, there has been growing interest in using AI to assist with or even replace parts of the interpretation process. The results so far are promising. In the European multicenter study mentioned earlier, AI software trained on over 1,500 historical cases outperformed the 120 pulmonologists in both pattern recognition and diagnosis.9European Respiratory Journal. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests
A recent review aggregating AI performance across PFT interpretation studies reported that AI achieved roughly 87% accuracy against a gold standard, compared with about 66% for pulmonologists evaluated in the same studies.16PubMed Central. Application of Artificial Intelligence in the Interpretation of Pulmonary Function Tests Another study found AI achieving 92% overall diagnostic accuracy with strong agreement with pulmonologist interpretations.17PubMed Central. AI-Assisted Interpretation of Pulmonary Function Tests: Enhancing Diagnostic Precision in Respiratory Medicine These tools are not yet standard in clinical practice, but they are moving in that direction. The likely near-term model is AI-assisted interpretation, where the software flags patterns and suggests diagnoses for a physician to confirm or override, rather than fully autonomous reading.
AI does not help with the test-performance side, though. No algorithm can coach a patient through a spirometry maneuver or recognize that a cough artifact invalidated a blow. The human technologist remains essential for data acquisition. Where AI has the most to offer is in reducing the kind of interpretive variability documented in those multicenter studies, especially in settings where the interpreting physician may not have extensive PFT experience.
Remote and Smartphone-Based Testing
An emerging frontier is home-based pulmonary function monitoring using smartphones or portable spirometers. A pilot study tested whether a smartphone’s built-in microphone could measure PFT parameters via a cloud platform, allowing patients to test freely without time or location constraints while a physician monitored results remotely.18PubMed Central. Remote Pulmonary Function Test Monitoring in Cloud Platform via Smartphone Built-in Microphone The concept is appealing, particularly for patients with chronic lung conditions who need frequent monitoring. However, the technology is still early. The accuracy of smartphone microphones is lower than laboratory-grade equipment, and no trained technologist is present to coach the patient or verify maneuver quality. For now, remote PFT monitoring is better suited to tracking trends over time in a known patient than to making a new diagnosis.
Portable handheld spirometers that connect to tablets or phones are more developed and are already used in some primary care and occupational health settings. The same quality concerns apply: without a trained person watching the maneuver and checking the flow-volume curve in real time, the data quality can be unreliable. The device does the measuring, but a human still needs to ensure the patient’s effort was adequate. Some newer devices incorporate automated quality-grading algorithms that flag poor efforts, which partially fills the gap left by the absent technologist, but partially is the key word.