A birth can be recorded as clinically uncomplicated while the person giving birth leaves feeling unheard, pressured, or excluded from decisions. During pregnancy loss, a medical record may document the diagnosis and treatment without capturing whether the news was delivered privately, a support person was present, or there was time to absorb what had happened.
These experiences belong in our assessment of care quality. Mortality, morbidity, and other clinical outcomes are essential indicators, but they cannot tell us everything about how care was delivered. To understand that, we need to ask the people receiving it.
Patient-reported experience measures, or PREMs, make aspects of care such as communication, respect, consent, and decision-making visible in data. In my research on maternity care and health equity, these measures have helped us examine differences that clinical records alone would miss.
Measuring Experience Rather Than Satisfaction
“Were you satisfied with your care?”
“Were your options explained before you made a decision?”
The above two questions ask different things. Satisfaction reflects an overall judgment, shaped by expectations and what someone believes was available. Experience questions focus more directly on what happened.
That distinction matters for improvement. A service can respond to evidence that patients didn’t receive enough information, felt uncomfortable asking questions, or underwent procedures without permission. A general satisfaction score offers less guidance about what needs to change.
PREMs also differ from patient-reported outcome measures, or PROMs. PROMs assess symptoms, functioning, health status, or quality of life. PREMs assess the experience of receiving care. Both can contribute to evaluation, alongside clinical outcomes and access indicators.
Choosing a measure requires care. A systematic review by Nilvér and colleagues identified 36 instruments addressing childbirth experiences, including satisfaction, fear, support, and control. This breadth is useful, but the instruments are not interchangeable. A questionnaire on fear of childbirth cannot replace a measure of informed decision-making. pmc.ncbi.nlm.nih.gov
The starting point should therefore be the question we want answered: are we assessing communication, autonomy, respectful treatment, mistreatment, or the experience of a particular care pathway?
What Respect and Autonomy Look Like in Data
Several measures help turn these concepts into specific questions. The seven-item Mothers Autonomy in Decision Making scale, or MADM, assesses how care providers facilitate participation in decisions. The Mothers on Respect Index (MORi) examines respectful interactions, including whether people feel comfortable asking questions or accepting or declining care. Mistreatment indices capture experiences such as threats, ignored requests for help, privacy violations, and abuse; researchers need to specify which version they use. Springer Nature Link
These measures let us examine care beyond a single overall positive or negative rating. Someone may receive clear information but feel unable to decline a recommendation. Another person may feel respected during routine appointments yet encounter poor communication during an emergency. A single global score can obscure those distinctions.
In a U.S. study I co-authored with Niles and colleagues, participants receiving midwifery care in community birth settings reported greater autonomy and respect, and less mistreatment, than those receiving physician care in hospitals. These were associations in a cross-sectional survey, rather than proof that a particular provider or setting caused the differences. They nevertheless identify questions worth investigating about care relationships and organizational conditions. link.springer.com
Our research on coercion and non-consent similarly shows why consent belongs in assessments of maternity-care quality. Completing a procedure and obtaining meaningful permission for it are separate dimensions of care. Evaluation must account for both. Wiley Online Library
Averages Can Hide Inequities
A service-wide average can look reassuring while particular groups report substantially poorer treatment. For equity research, who the data represent—and whose experiences are combined into an average—matters as much as the overall score.
In our Canadian study of weight-based disparities, we analyzed autonomy, respect, and mistreatment among 3,280 participants. Higher pre-pregnancy body mass index was associated with poorer experiences across these domains, with several associations remaining after adjustment for demographic and socioeconomic characteristics. Indigenous and other racialized participants also reported inequities in care. Springer Nature Link
These findings should prompt scrutiny of how care is delivered. They should not be interpreted as evidence that a person’s body size or identity causes disrespect. Nor does statistical adjustment alone establish how different forms of disadvantage interact.
For health-system monitoring, I would pair overall results with carefully chosen comparisons by population and care setting. Those comparisons should be developed with relevant communities, protect confidentiality, and acknowledge when sample sizes are too small to support reliable conclusions. Analysis categories should reveal inequity without reducing people to a label.
Pregnancy Loss Requires Specific Questions
A measure designed around birth may miss the most important aspects of care during miscarriage, stillbirth, or neonatal death. Even a reliable scale can leave a significant part of the experience unmeasured.
In our Canadian pregnancy-loss study, led by Wendy Hall, we introduced the six-item Compassionate Disclosure of Loss Index, or CDL. It addresses privacy, the presence of a support person, time to process the loss, kind and caring communication, perceived provider skills, and information about options for examination of the baby.
The study found that participants experiencing miscarriage were less likely than those experiencing later loss to report highly compassionate disclosure. Participants grouped as Indigenous, Black, and people of color had lower odds of reporting high autonomy and respectful care. The CDL showed promising internal consistency and a single-factor structure in this sample; further testing can examine how it performs in other populations and settings. onlinelibrary.wiley.com
For me, this illustrates a central measurement principle: we need to include early loss in research while distinguishing experiences across types and timing of loss. A questionnaire should let us identify different care needs rather than assume one pathway fits everyone.
It also gives services concrete questions to act on. Was there privacy? Could someone supportive be present? Was information communicated with care? Was time provided to understand it?
Turning Measurement Into Better Care
Collecting responses is only the beginning. I would judge a PREM program by whether it informs decisions and leads to changes patients can recognize.
Three commitments would make measurement more useful:
- Choose measures with patients and communities. Check that the questions cover what matters to the people receiving care, including those whose experiences are often overlooked. Test whether translated or adapted versions retain their meaning.
- Interpret scores alongside context. Examine variation between groups, care pathways, and settings. Use qualitative accounts to understand what a score alone cannot explain, and consider recall periods, missing responses, and ceiling effects.
- Assign responsibility for action. Return findings to patients, clinical teams, and managers. Identify a specific change, name who will carry it forward, and assess whether experiences improve afterward.
Not every improvement requires a new questionnaire. Evidence of poor communication might prompt changes to consultation practices; reports of non-consented care should trigger a review of consent processes; gaps in compassionate disclosure can inform bereavement-care protocols and staff support.
As an economist and quantitative researcher, I see PREMs as a way to connect measurement with accountability. What we choose to count influences what receives attention. If respect, autonomy, and communication remain outside routine evaluation, services can more easily overlook them even when clinical targets are met.
The practical test is whether patient-reported evidence changes how care is delivered. A health system that measures experience should be able to show patients what it learned, what it changed, and whether the change helped.
This article draws on selected research on maternity-care measurement and studies I co-authored. Links to the original publications appear throughout.