Positive Attitudes Concerning Infant Feeding- a Questionnaire for Women Living With HIV: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Positive Attitudes Concerning Infant Feeding- a Questionnaire for Women Living With HIV (NCT03294499), a registered phase not stated study. The primary outcome is Knowledge and Experiences of Breastfeeding Among Women Living With HIV.

2. Clinical Question

The clinical context described in the record is a difference between World Health Organisation guidance and British HIV Association guidelines. The former, as summarized in the record, advises women living with HIV who adhere to antiretroviral therapy to exclusively breastfeed for the first 6 months of an infant’s life and continue supplemental breastfeeding for up to 2 years. The latter advises exclusive formula feeding in the United Kingdom. The record describes the rationale for this difference as the relative safety of formula feeding in the United Kingdom alongside a low but persisting risk of HIV transmission through breastfeeding.

The study was designed to document women’s knowledge, views and reported practices in that setting. It was not designed as a comparison of feeding strategies or a test of whether one feeding practice is safer or more effective than another. The registry record does not report results that answer the questionnaire’s substantive questions.

3. Trial Design

PACIFY is recorded as a completed questionnaire study, with a planned enrollment of 94 participants. The intervention entry is a questionnaire, classified as an other intervention; no treatment allocation or control arm is described. The registry explicitly notes that this is not a randomized comparison and that allocation is not stated. Accordingly, the participant group should not be read as one arm of a controlled treatment trial.

The study summary describes antenatal participants in the third trimester and postnatal participants at three months after delivery. The primary-outcome description says participants were approached at a regular pre- or postnatal outpatient appointment, were given the patient information leaflet and consent form, and completed an anonymised questionnaire during the same visit if they agreed to participate. The recorded timeframe for the outcome is at the outpatient appointment, averaging one day.

The primary-outcome description reports recruitment by clinicians from 12 HIV clinics across England, in the South East, West Midlands and West Yorkshire, between June 2017 and June 2018. The brief summary also refers to clinics in London and Brighton. Questionnaires were collated at Imperial College Healthcare NHS Trust. The record lists the study start as 2017-05-31, primary completion as 2018-05-31, and status as completed.

The brief summary says the study aimed to analyse 100 questionnaires over a 3-6 month period, while the enrollment field records 94. These entries describe a target and an enrollment figure; they do not establish how many questionnaires were complete or analyzable. The record does not clarify whether any participant completed questionnaires at more than one stage, or how antenatal and postnatal responses were analyzed.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: 94 participants were recruited from 12 clinics in England

6. Primary Endpoint

The registry lists the primary outcome as Knowledge and Experiences of Breastfeeding Among Women Living With HIV, measured over At their pre-or postnatal outpatient appointment, an average of one day.

The primary outcome is titled “Knowledge and Experiences of Breastfeeding Among Women Living With HIV.” It is recorded in participants, with a count-of-participants parameter. Its description says that participants completed an anonymised questionnaire at a pre- or postnatal outpatient visit. In the results data, breastfeeding is abbreviated to BF.

This kind of outcome is used to describe what participants know, have experienced, report doing, or think about infant feeding. Those measures can help characterize the population’s understanding of guidance and the acceptability or feasibility of practices that might be relevant to future research or policy discussions. A questionnaire is suited to gathering reported views and experiences; it does not by itself establish a biological outcome or verify that reported practices occurred exactly as described.

The registry record does not report the questionnaire’s individual items, response options, scoring rules, or the way responses were combined into outcome categories. Nor does it provide participant counts for specific answers. Therefore, the title and description identify the broad subject of the endpoint, but do not show the distribution of knowledge, experiences or feeding practices. The record also does not report a separate primary endpoint result or an effect estimate.

The timing is described as a single outpatient appointment averaging one day, but the registry does not explain whether the same women contributed at both antenatal and postnatal time points. That distinction would matter for interpreting changes over time and for choosing an analysis that accounts for repeated responses. The record does not report such an analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins with the design. PACIFY is described as a questionnaire study, not a randomized trial of feeding approaches. The recorded activity was questionnaire completion during an outpatient visit. Since the registry does not describe assigned treatment groups or a comparator, treatment-effect methods for comparing randomized arms are not supported by the information in the record. The main statistical task would be to describe responses, but the registry does not post those response results.

What would a descriptive estimate represent?

For a questionnaire item, a count would represent the number of participants selecting a particular response, subject to the question’s wording and the completeness of the response data. A proportion would relate that count to an appropriate denominator, such as the number who answered that item. A denominator can vary across questions if answers are missing or not applicable. These are general principles; the registry does not report item counts, denominators, or proportions for PACIFY, so no such estimates can be interpreted.

A descriptive result would refer to the participants who were recruited and whose responses were included. It would not automatically represent all women living with HIV, all pregnant or post-partum women, or all clinics. Generalizability depends on who was invited, who agreed, who completed the questionnaire, and how the recruited participants compare with the wider population. The record gives the clinic recruitment description but does not report invitation counts, refusal counts, participant characteristics, or response completeness.

What does a confidence interval add?

In general, a confidence interval communicates statistical uncertainty around an estimate. A narrow interval indicates greater precision under the assumptions used to calculate it; a wide interval indicates less precision. The interval does not measure all sources of uncertainty. In particular, it does not remove selection bias, nonresponse bias, measurement error, or limitations in how the participants were recruited. No confidence intervals are reported for the PACIFY primary outcome, and the available registry information is insufficient to assess the precision of any specific questionnaire result.

A p-value, in general, evaluates how compatible data are with a specified statistical model or null hypothesis. It is not a measure of the size or importance of an effect, and it cannot compensate for a weak design or incomplete outcome reporting. The registry lists no analyses. It therefore provides no p-values, model estimates or tests to interpret. There is also no reported treatment contrast for which a comparative p-value would be meaningful.

Would repeated responses require special handling?

The study summary identifies antenatal and postnatal participants, but the registry does not state whether the same women completed the questionnaire at both stages. If participants contributed repeated responses, observations from one person would generally be related rather than independent. General methods for repeated measures account for that within-person relationship. If different participants contributed at each stage, the analysis would instead need to describe those groups and their comparability. Because the record does not clarify the response structure or report an analysis, neither interpretation can be confirmed.

Questionnaire outcomes may also involve multiple items or several response categories. In general, analyzing many outcomes raises a multiplicity issue: the more separate statistical tests undertaken, the greater the chance of at least one apparently unusual result arising by chance if no adjustment or prespecified hierarchy is used. A study may address this through prespecified primary outcomes, adjustment, or emphasis on estimation rather than isolated significance tests. The registry record does not report the number of questionnaire items analyzed, a multiplicity strategy, or an analysis plan. No approach should be inferred from the endpoint title alone.

What are planning calculations?

Sample-size planning asks how much information a future study might need to detect a specified effect under stated assumptions. The registry includes a planning-illustration description based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that this questionnaire study randomized participants between two arms or used that design. The record does not report planning output values, a sample-size justification, or the effect estimate needed to assess the illustration. Planning calculations are not observed power and do not establish that the study had adequate precision for its descriptive aims.

For event counts in general, over-dispersion means that variation is greater than a simple model assumes. Ignoring it can make uncertainty estimates too small. The registry does not report a count model for questionnaire responses or an analysis of over-dispersion; this concept does not imply that such an issue arose in PACIFY. The appropriate analysis depends on the actual outcome format, response structure and missingness, none of which are fully described in the registry record.

The primary outcome is recorded as a participant count, but no analysis populations or derivation rules are reported. Without those details and the underlying outcome summaries, the methods cannot be independently assessed or reproduced from the registry entry. The appropriate reading is therefore descriptive: the record documents a questionnaire study and its stated aims, not a reported inferential analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical analysis plan, a hierarchy of primary and secondary questionnaire outcomes, or a method for handling multiple comparisons. It lists two outcome measures and no analyses. The broad primary-outcome title does not establish how many individual questions were evaluated or whether any responses were combined into scales. Consequently, the risk of selective emphasis across questions cannot be assessed.

In general, multiple testing matters when many hypotheses are examined: even if each test uses a conventional threshold, the chance of at least one nominally significant finding can increase as the number of tests grows. Prespecifying a primary question, adjusting for multiplicity, or emphasizing estimates and uncertainty are possible approaches, but the registry does not report which, if any, were used. No significance findings are posted, so there is no basis for making a multiplicity-adjusted interpretation.

Missing data handling is also not reported. The registry does not say how many questionnaires were incomplete, how item nonresponse was treated, whether any responses were excluded, or whether missing values were imputed. A count for a particular response would need its question-specific denominator; the overall enrollment figure cannot be assumed to be that denominator.

The analysis populations are not defined. There is no reported distinction between enrolled participants, questionnaire completers, or participants with data available for a particular item. The record also does not explain whether antenatal and postnatal responses came from separate participants or from repeat participation. These omissions limit assessment of both the validity and precision of any outcome summary, and prevent reconstruction of an analysis population from the registry entry.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The PACIFY Study safety record lists 94 participants at risk in the serious adverse event field and 0 serious adverse events. It also lists 94 at risk in the other adverse event field and 0 other adverse events, with 0 deaths. These figures are safety entries for the study group identified as PACIFY Study.

They should be read as registry-reported counts, not as a comparative safety analysis. The study description is a questionnaire study, and no feeding-strategy comparison or randomized allocation is described. The counts therefore do not establish whether breastfeeding or formula feeding is safer, and they do not compare outcomes between feeding practices.

The registry does not provide event definitions, the method of adverse-event collection, the period over which events were assessed, or details of any monitoring procedures. It also does not report whether events outside the recorded categories were assessed. A zero count in the record means that no events are entered in that category; it is not proof that an event was impossible or that every potential harm was measured. The safety entries do not replace the missing primary questionnaire results.

12. What the Result Does, and Does Not, Mean

The study addresses a practical and sensitive question: what women living with HIV know and experience about infant feeding, and how they view guidance that differs across health authorities. Its stated aims include understanding current or recent practices and whether mothers would be willing and able to follow special monitoring and guidance if recommendations changed. This makes the questionnaire’s subject relevant to understanding information needs and the context in which feeding decisions are discussed.

However, the registry entry does not report the responses. It cannot show how many participants understood particular recommendations, favored a feeding approach, reported breastfeeding or formula feeding, or felt able to meet monitoring requirements. It also does not report whether views differed between antenatal and postnatal participants. The enrollment figure and stated questionnaire-analysis aim do not substitute for those outcome data.

The study is descriptive rather than a randomized comparison. Even with questionnaire findings, reported attitudes and practices would not establish that a feeding approach caused a health outcome. The record contains no comparison of infant or maternal outcomes by feeding practice, and no basis for drawing conclusions about the relative safety or effectiveness of breastfeeding and formula feeding.

More broadly, a questionnaire captures responses as participants understand and report them. Interpretation depends on the questions, response choices, completion rates and population recruited. The registry does not report item wording, scoring, missingness or participant characteristics, so readers cannot assess these measurement and representativeness issues in detail.

The appropriate conclusion from the registry record is limited: it documents a completed study intended to explore knowledge, experiences, attitudes and reported infant-feeding practices among women living with HIV. It does not provide the results needed to characterize those responses or to make clinical comparisons. The study’s stated aims should not be mistaken for findings.

13. Why This Trial Matters Statistically

Statistically, PACIFY illustrates the difference between a study question and a reported result. A clearly relevant question about knowledge and experience can motivate data collection, but interpretation requires response counts, denominators, outcome definitions and information about missing data. Without those elements, a reader cannot estimate how common a view or practice was or how precise that estimate might be.

It also demonstrates why design and purpose must be kept aligned. A questionnaire can describe self-reported knowledge, attitudes and practices in a recruited population. It does not become a comparative clinical trial merely because the topic involves alternative feeding practices. Since the registry does not describe randomized groups, there is no treatment contrast to interpret and no basis for attributing differences in responses or health outcomes to a feeding approach.

The clinical context includes differing guidance described in the study summary. For research in such a context, clear reporting of how questions were framed, how participants were recruited, and how responses were summarized would help readers understand what the findings represent. The registry reports recruitment through clinics and the broad questionnaire purpose, but it does not report the questionnaire’s item-level results or its analysis plan.

Finally, planning and observed evidence are distinct. A planning illustration is a hypothetical calculation under assumptions; it is not a description of the study’s actual precision or observed power. The record provides no output values or reported outcome estimates with which to assess precision. This is a useful reminder that enrollment alone does not communicate what the data show: the endpoint results and their denominators are essential.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

A central limitation of the registry entry is the absence of primary outcome results. The outcome title and study aims are available, but no response distributions, counts for questionnaire items, estimates or uncertainty measures are reported. This prevents assessment of what participants knew, experienced or practiced and how precisely those findings were measured.

The questionnaire itself is not described in sufficient detail for measurement appraisal. Although the primary-outcome description refers to an anonymised questionnaire and notes that breastfeeding is abbreviated to BF in the data table, the registry does not report item wording, response options, scoring, validation, or rules for combining responses. Those details would be needed to judge whether the measure captures the stated concepts consistently and interprets them as intended.

Recruitment is described through HIV clinics, but the record does not report how many people were approached, how many declined, or the characteristics of participants and nonparticipants. Selection and nonresponse could affect how well the sample reflects the wider population. The brief summary refers to clinics in London and Brighton, while the outcome description identifies recruitment across England and names the South East, West Midlands and West Yorkshire. The registry does not explain the relationship between these descriptions.

The record does not clarify whether antenatal and postnatal participants were different people or whether some participants contributed responses at both stages. Nor does it explain missing questionnaire responses, analysis populations, or any adjustment for repeated measurements. The stated aim to analyse 100 questionnaires and recorded enrollment of 94 do not indicate the number of complete responses or the analytic denominator.

Finally, no analysis plan or statistical results are reported. The registry does not explain multiplicity handling, provide confidence intervals, or give enough information to evaluate planning assumptions. The safety counts lack reported event definitions and collection details. These limitations restrict interpretation to the documented aims and conduct; they do not establish that the study procedures were inadequate.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Positive Attitudes Concerning Infant Feeding- a Questionnaire for Women Living With HIV — NCT03294499.
• Registry record last updated 2023-02-02; status last verified 2022-05.
• Results first posted 2023-02-02.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT03294499 is the completed PACIFY questionnaire study, sponsored by Imperial College London. Its stated focus is knowledge and experiences of breastfeeding among women living with HIV, including attitudes toward infant feeding, understanding of guidance, reported practice, and willingness or ability to follow special monitoring if guidance changed.

The record lists enrollment of 94 and an intervention of questionnaire, with no randomized comparison described. It reports recruitment at pre- or postnatal outpatient appointments and an anonymised questionnaire completed during the visit. Recruitment is described between June 2017 and June 2018. The registry lists a study start of 2017-05-31 and primary completion of 2018-05-31.

No primary outcome results, item-level response data, statistical analyses, missing-data methods or analysis populations are reported. Safety entries list 0 serious adverse events, 0 other adverse events and 0 deaths. These entries do not provide a comparative assessment of feeding practices.