Parental awareness, attitudes and decision-making of off-label use in children: a systematic review and meta-analysis

Abstract

Objective

To systematically evaluate parental awareness, acceptance, refusal rate and influencing factors of off-label use in children, aiming to provide evidence for optimising informed consent and physician-parent communication.

Methods

We searched PubMed, Web of Science, Elsevier ScienceDirect, CNKI, Wanfang and VIP from inception to March 2026 for observational, mixed-methods and qualitative studies on parental awareness, attitudes and decision-making about off-label use in children. A random-effects single-group rate meta-analysis pooled awareness and refusal rates; conditional acceptance rates were summarised narratively due to extreme heterogeneity in operational definitions; unconditional acceptance rates were reported as individual study estimates. Influencing factors and psychological experiences were synthesised thematically.

Results

Seven studies were included. The pooled parental awareness rate was 20.2% (95% CI: 10.6% to 31.9%), conditional acceptance ranged from 20.6% to 85.1% due to inconsistent operational definitions, unconditional acceptance was reported in two studies with individual estimates of 18.2% and 22.7%, and refusal was 36.5% (95% CI: 14.6% to 61.8%). Influencing factors fell into three themes: child clinical characteristics, parental cognition and emotion, and physician communication and qualification.

Conclusion

Parental awareness of off-label use in children is low. Part of parents accept conditionally; unconditional acceptance is limited, and refusal rates are relatively high. These three themes interact throughout the decision-making process. Postdecision caregiving experiences may lead to long-term scepticism and influence future attitudes. In clinical practice, informed consent should shift from procedural notification to trust-centred communication. Prescription and drug supply processes should be optimised to reduce parental care burden.

Introduction

Off-label use is pervasive in paediatric clinical practice. Although it may afford indispensable options for children who lack approved alternatives,¹ such use is also associated with substantial risks, including adverse drug reactions,² financial burdens on families³ and potential legal and ethical disputes.⁴ Against this backdrop, the quality of physician-parent communication and the informed consent process becomes a critical determinant of whether off-label treatment can be delivered ethically and effectively.

These realities place parents in a precarious position. When clinicians propose off-label treatment, parents must provide informed consent under conditions of clinical uncertainty, often without adequate information or support. Yet the evidence base for optimising this consent process remains fragmented. Existing studies have documented that off-label prescribing is common. Yackey et al⁵ documented a 28.1% off-label prescribing rate in the USA, and Lyu Chunlong et al⁶ reported 38.7% for antidepressants in Chinese children with mental disorders, but the supporting evidence remains weak: Van der Zanden et al⁷ found only 14% backed by high-quality data. Moreover, paediatric prescribing information is frequently inadequate. Tilen et al⁸ found that only 55.1% of Swiss drug inserts provided paediatric guidance. This evidence gap directly complicates physician–parent communication: clinicians lack robust data to convey risks and benefits clearly, while parents struggle to make genuinely informed decisions. Compounding this problem, parental awareness of off-label use is alarmingly low. Bang et al⁹ reported that only 30% of parents knew the term ‘off-label use’, which suggests that most consent conversations occur without parents understanding the fundamental nature of the treatment being proposed. In response to these challenges, Lenk et al¹⁰ stressed the importance of strengthening medication knowledge dissemination and risk communication. Polakova et al¹¹ concluded that parents of children with life-limiting conditions urgently require adequate informational support and emotional counselling, and called for standardised communication.

When clinicians propose an off-label medication regimen, parents as legal guardians face decisions under clinical uncertainty. Although existing studies have separately documented parents’ level of awareness, attitudes towards acceptance and selected influencing factors, most are limited by single-centre designs, small sample sizes and unidimensional assessments. Consequently, there is a lack of pooled quantitative estimates of key indicators and of an integrative framework of influencing factors, a gap that directly undermines the ability of clinicians and institutions to design evidence-based informed consent processes and communication strategies. The present study addresses this gap by providing the first systematic review and meta-analysis that quantifies parental awareness, acceptance and refusal rates, alongside a synthesis of the factors influencing their decision-making. We aim to generate actionable evidence that can inform clinical communication practices, institutional protocols and policy frameworks for paediatric off-label prescribing.

Materials and methods

Literature search

This systematic review was prospectively registered in PROSPERO (registration number: CRD420261376173) and conducted in accordance with the PRISMA 2020 guidelines. We systematically searched PubMed, Web of Science, Elsevier ScienceDirect, CNKI, Wanfang Data and VIP from database inception to March 2026, and manually screened the reference lists of included studies. Search terms included ‘off-label’, ‘unlicensed’, ‘child’, ‘pediatric’, ‘parent’, ‘caregiver’ and ‘guardian’. The complete PubMed search strategy is presented in table 1, and the strategies for the other databases were adapted accordingly.

Table 1. Search strategy for PubMed.

Step	Search terms
#1	off-label[tiab] OR “off label”[tiab] OR unlicensed[tiab] OR unapproved[tiab] OR “out of label”[tiab] OR “outside the label”[tiab] OR “Off-Label Use”[Majr]
#2	“Child”[Mesh] OR “child”[tiab] OR “children”[tiab] OR “pediatric”[tiab] OR “paediatric”[tiab] OR “infant”[tiab] OR “infants”[tiab] OR “newborn”[tiab] OR “neonate”[tiab] OR “adolescent”[tiab] OR “adolescents”[tiab]
#3	“Parents”[Mesh] OR “parent”[tiab] OR “parents”[tiab] OR “parental”[tiab] OR “caregiver”[tiab] OR “caregivers”[tiab] OR “guardian”[tiab] OR “guardians”[tiab] OR “family”[tiab] OR “families”[tiab]
#4	#1 AND #2 AND #3

Inclusion and exclusion criteria

Inclusion

(1) Participants were parents, legal guardians or primary caregivers of children (0–18 years); if studies included both parents and healthcare professionals, parental data had to be extractable. (2) Studies addressed parental awareness, attitudes, willingness to accept, decision-making processes or psychological experiences regarding paediatric off-label use. (3) Outcomes included awareness rates, acceptance rates, refusal rates, influencing factors or qualitative descriptions of parental perspectives. (4) Study designs included observational studies (cross-sectional, cohort, case–control), mixed-methods studies and qualitative studies.

Exclusion

(1) Studies involving only healthcare professionals (physicians, pharmacists, nurses). (2) Studies on adult patients’ self-decision-making. (3) Articles reporting only prevalence or prescribing patterns of off-label use without parental data. (4) Studies on informed consent in clinical trials rather than real-world clinical decision-making. (5) Reviews, systematic reviews, case reports, conference abstracts, commentaries. (6) Full-text unavailable.

Study selection and data extraction

The initial search identified 1521 records, of which 1178 remained after the removal of duplicates. Following title, abstract and full-text screening, seven studies ultimately met the inclusion criteria and were included in the review. The limited number of eligible studies reflects the global paucity of parent-centred empirical research in this domain. The study selection process is illustrated in figure 1.

Figure 1. Flowchart of study selection.

Data extraction was performed independently by two reviewers and cross-checked. Extracted information included: (1) study characteristics (first author, publication year, country); (2) study design (type, sample size, data collection method); (3) participant characteristics (gender, age, child’s condition, education, socioeconomic status); (4) outcomes (awareness rates, acceptance/refusal rates, influencing factors, qualitative findings). Data of outcomes were extracted from study tables, figures and narrative descriptions.

Four primary outcome measures were defined and extracted: (1) awareness rate: the proportion of parents who reported knowing or having heard of the term ‘off-label use’ or its equivalent; (2) conditional acceptance rate: the proportion of parents willing to accept off-label use under specified conditions (eg, only for life-threatening illness, or only when no approved alternatives exist); (3) unconditional acceptance rate: the proportion of parents willing to accept off-label use without specifying any conditions; (4) refusal rate: the proportion of parents who would refuse or who had refused off-label prescribing for their child.

Regarding outcome comparability and analytical approach: the included studies used fundamentally different operational definitions for ‘conditional acceptance’ (eg, ‘acceptance with careful monitoring for side effects’ vs ‘acceptance only for life-threatening conditions or when no alternative exists’), so these rates were not pooled and were instead reported narratively as original study estimates (20.6% to 85.1%). Unconditional acceptance was reported in only two studies, so individual study estimates (18.2% and 22.7%) were presented. Awareness and refusal rates had consistent operational definitions across studies and were, therefore, pooled using a random-effects model.

Quality assessment

Two reviewers independently assessed quality; disagreements were resolved by a third reviewer. Given the heterogeneity of the included study designs, we selected quality appraisal tools tailored to each design. For cross-sectional studies, we applied the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Analytical Cross-Sectional Studies, a standardised instrument for evaluating the risk of bias in observational research, particularly prevalence and incidence studies. Its items specifically address key methodological domains, including sample representativeness and measurement validity, that align with the effect sizes of interest in this review, such as awareness rates and acceptance willingness rates. For the qualitative study, we used the Critical Appraisal Skills Programme (CASP) Qualitative Checklist, an internationally recognised tool that facilitates systematic evaluation of qualitative research by assessing rigour, credibility and ethical considerations, thereby ensuring a comprehensive appraisal of the study’s trustworthiness.

Statistical analysis

Single-group rate meta-analysis was performed using R software (V.4.5.3). Raw rates were transformed using the Freeman-Tukey double arcsine method. Heterogeneity was assessed using the I² statistic; a random-effects model for I²> 50%. Pooled rates and 95% CIs were calculated for the awareness and refusal rates. For the conditional acceptance rate, individual study estimates are presented descriptively due to inconsistent operational definitions across studies. For the unconditional acceptance rate, which was reported in only two studies, individual estimates are presented rather than a pooled rate. Sensitivity analysis was conducted by omitting one study at a time. Given that fewer than 10 studies were available for each outcome, formal assessment of publication bias (eg, funnel plots, Egger’s test) was not conducted, as these methods lack statistical power with small numbers of studies. Influencing factors were summarised thematically.

Results

Basic characteristics of included studies and quality assessment results

Among the seven included studies, five were quantitative cross-sectional surveys and two were qualitative. They were conducted in India (n=2), Germany (n=1), Northern Ireland (n=1), Japan (n=1) and the United Kingdom (n=2). Sample sizes of the quantitative studies ranged from 94 to 1000, yielding a combined total of 2348 participants. S1, S2, S3 and S4 directly surveyed parents of affected children, parents of healthy children or adults considering hypothetical scenarios. S5 surveyed child psychiatrists who reported their experiences with parental refusal of off-label prescriptions; parental refusal data in S5 were thus physician-reported rather than parent-self-reported (table 2).

Table 2. Basic characteristics of included studies.

Study ID	First author	Year	Country	Study design	Data collection	Sample size
S1	Bang et al9	2014	India	Cross-sectional survey	Questionnaire	400
S2	Lenk et al10	2009	Germany	Cross-sectional survey	Questionnaire	94
S3	Saiyed et al21	2015	India	Cross-sectional survey	Questionnaire	407
S4	Mukattash et al22	2008	Northern Ireland	Cross-sectional survey	Interview	1000
S5	Tsujii et al23	2016	Japan	Cross-sectional survey	Questionnaire	447
S6	Husain et al24	2017	United Kingdom	Qualitative study	Semistructured interview	15
S7	Duncan et al25	2026	United Kingdom	Qualitative study	Semistructured interview	13

For quantitative studies (JBI), S1 (9/10), S3 (8/10), S4 (8/10) and S5 (7/10) were of high quality; S2 (6/10) was of moderate quality (table 3). For qualitative studies (CASP), S6 scored 7/10, S7 scored 9/10 and were rated high quality (table 4).

Table 3. JBI quality assessment results of quantitative studies.

Item	Assessment content	S1	S2	S3	S4	S5
1	Was the sample frame appropriate to address the target population?	Y	Y	Y	Y	Y
2	Were study participants sampled in an appropriate way?	Y	Y	Y	Y	Y
3	Was the sample size adequate?	Y	N	Y	Y	Y
4	Were the study subjects and the setting described in detail?	Y	Y	Y	Y	Y
5	Was the data analysis conducted with sufficient coverage of the identified sample?	Y	Y	Y	Y	Y
6	Were valid methods used for the identification of the condition?	Y	Y	Y	Y	Y
7	Was the condition measured in a standard, reliable way for all participants?	Y	Y	Y	Y	Y
8	Was there appropriate statistical analysis?	Y	N	N	N	Y
9	Was the response rate adequate, and if not, was the low response rate managed appropriately?	Y	N	Y	Y	N
10	Were the statistical analysis methods appropriate for the study design and research questions?	Y	Y	Y	Y	Y
Total score		9/10	6/10	8/10	8/10	7/10
Level		High	Moderate	High	High	High

JBI, Joanna Briggs Institute.

Table 4. CASP quality assessment results of qualitative studies.

Item	Assessment content	S6	S7
1	Was there a clear statement of the aims of the research?	Y	Y
2	Is a qualitative methodology appropriate?	Y	Y
3	Was the research design appropriate to address the aims of the research?	Y	Y
4	Was the recruitment strategy appropriate to the aims of the research?	Y	Y
5	Was the data collected in a way that addressed the research issue?	Y	Y
6	Has the relationship between researcher and participants been adequately considered?	N	N
7	Have ethical issues been taken into consideration?	N	Y
8	Was the data analysis sufficiently rigorous?	Y	Y
9	Is there a clear statement of findings?	N	Y
10	How valuable is the research?	Y	Y
Total score		7/10	9/10
Level		High	High

CASP, Critical Appraisal Skills Programme.

Meta-analysis results

Table 5 summarises the meta-analysis results.

Table 5. Summary of meta-analysis results.

Outcome	Number of studies	Total sample	Number of events	Pooled rate (95% CI)	I²	Model	Sensitivity analysis
Awareness rate	4	1901	331	20.2% (10.6% to 31.9%)	95.90%	Random-effects	Robust
Conditional acceptance rate	4	1889	831	20.6%–85.1% (not pooled)	/	Not pooled	Not performed
Unconditional acceptance rate	2	1388	270	22.7%; 18.2% (not pooled)	/	Not pooled	Not performed
Refusal rate	3	1482	633	36.5% (14.6% to 61.8%)	97.20%	Random- effects	Affected by S2

The pooled awareness rate was 20.2% (95% CI: 10.6% to 31.9%, I²=95.9%). Leave-one-out sensitivity confirmed robustness (figure 2a). Conditional acceptance rates were reported in four studies but were not pooled due to fundamental differences in operational definitions across studies: Bang et al (India) reported 20.6%, Mukattash et al (Northern Ireland) 41.9%, Saiyed et al (India) 61.9% and Lenk et al (Germany) 85.1%. Individual study estimates are presented descriptively. The pooled refusal rate was 36.5% (95% CI: 14.6% to 61.8%, I²=97.2%) (figure 2b). For unconditional acceptance rates, we present individual estimates rather than a pooled rate: Mukattash et al reported 18.2% and Bang et al reported 22.7%. S2 was the main heterogeneity source. After excluding it, refusal increased the pooled estimate from 36.5% to 48.2% (S2 alone: 14.9% refusal).

Figure 2. Forest plots of parental awareness and refusal rates of off-label use in children. (a) awareness rate; (b) refusal rate.

Given the small number of studies contributing to each outcome, formal publication bias assessment was not performed.

Multidimensional factors influencing parental decision-making

Thematic synthesis identified three core dimensions influencing parental decisions on paediatric off-label use: child clinical characteristics, parental cognition and emotion, and physician communication and qualifications.

Child clinical characteristics

Disease severity and the availability of alternative treatments were pivotal. Saiyed et al reported that 62% of parents would consent to off-label use for a life-threatening illness despite known risks. Lenk et al found that acceptance was largely confined to severe or life-threatening scenarios, with higher acceptance among parents of chronically ill children than among those of healthy children (72% vs 51%, p=0.11).

Parental cognition and emotion

Cognitive factors included awareness of the term ‘off-label’ perceived legality and educational level. Bang et al found that only 30% of parents were aware of off-label prescribing, while 73% mistakenly believed it was illegal. Mukattash et al noted that university-educated parents were significantly less likely to request a switch to an approved alternative (p<0.05). Emotional responses, captured qualitatively by Husain et al, included fear of medication supply disruption, distrust of primary care physicians and frustration due to poor communication. Parents also reported substantial caregiving burdens, such as managing prescription cycles and pharmacy coordination-leading to exhaustion. Duncan et al extended these findings by revealing that most UK parents administering age-inappropriate formulations at home had never actively considered whether their child’s medicines were off-label or unlicensed; instead, they followed prescriber instructions with passive trust. When the off-label status was not disclosed, parents experienced underlying anxiety about the safety of manipulation practices. Duncan et al also highlighted that the emotional burden was compounded by practical challenges: parents described having to travel long distances to hospital pharmacies, order unlicensed liquid specials weeks in advance and manage frequent brand switches that disrupted established administration routines, all of which contributed to a sustained sense of uncertainty and vigilance.

Physician communication and qualifications

Tsujii et al found that physician disclosure of off-label status to parents (OR=2.73, 95% CI 1.09 to 6.82) and to the child (OR=1.70, 95% CI 1.12 to 2.58) was associated with increased parental refusal. Additionally, each additional year of physician experience reduced the odds of parental refusal by 2% (OR=0.98, 95% CI 0.96 to 1.00, p=0.03). Qualitative evidence from Husain et al indicated that perfunctory communication (eg, handing over medication without explanation) left parents feeling dismissed and eroded trust. Complementing this, Duncan et al found that UK parents were frequently excluded from shared decision-making processes surrounding off-label prescribing: several parents reported receiving no discussion about formulation choices (‘I had no conversation with the doctors’), while others experienced inconsistent information across primary and secondary care interfaces, such as General Practitioners (GPs) refusing to prescribe medicines initiated by specialist paediatricians without explanation.

These communication failures left parents navigating complex medication supply logistics without adequate guidance, further underscoring that trust in off-label use depends not merely on disclosure of status but on meaningful engagement that includes families in treatment planning.

Discussion

Reasons and heterogeneity of low parental awareness rate

Our study revealed a pooled parental awareness rate of paediatric off-label prescribing of only 20.2%, with substantial heterogeneity across included studies (I²=95.9%). These findings derive from studies conducted in five countries, suggesting that low parental awareness is a consistent finding across diverse healthcare contexts. A systematic review by Balan et al¹² similarly reported parental awareness rates ranging from 14% to 35%, in striking contrast to 69%–93% in physicians and 73% in pharmacists. This awareness gap has direct clinical implications, in that when parents do not understand the concept of off-label use, informed consent becomes a procedural formality rather than meaningful shared decision-making. The gap arises from three interacting factors: physician non-disclosure, ambiguous legal environments and parental cognitive characteristics.

The predominant cause is non-disclosure by physicians. Studies report disclosure rates as low as 69% in Scotland¹³ and 4.8% in Jordan.¹⁴ In Spain, Piñeiro Pérez et al¹⁵ reported that fewer than half of paediatricians informed parents, and only 22% documented it in the medical record. The reasons for non-disclosure include fear of provoking unnecessary parental anxiety, the belief that disclosure may reduce treatment adherence, and in some cases, clinicians’ own inadequate recognition of off-label prescribing, Chalumeau et al¹⁶ found that 92% of French paediatricians failed to identify the off-label status of their own prescriptions.

Legal ambiguity compounds this barrier. In India, off-label prescribing is legally prohibited, yet the dearth of paediatric medicines renders it unavoidable in clinical practice. Kannan et al¹⁷ found that nearly half believed the practice to be illegal, 90.2% held the erroneous belief that a drug approved for adults could not be used in children, and only 55.7% informed families. When physicians themselves harbour misconceptions about the legal framework, the likelihood that parents receive accurate information becomes low. In Saudi Arabia, Almintakh et al¹⁸ reported that 47% of physicians were unaware of whether national regulations existed, and 64.2% considered their own knowledge insufficient, underscoring that the absence of institutional guidance engenders inconsistent practices. In contrast, while the United Kingdom legally permits off-label prescribing, informed consent practices have not improved commensurately. Husain et al documented parents being repeatedly questioned by pharmacists during medication dispensing, revealing failures in system coordination that further undermine information delivery.

Parental cognitive characteristics also drive heterogeneity. The lowest awareness rate was reported by Mukattash et al, 14% in a Northern Irish community sample containing many childless adults, who are less engaged with paediatric issues. Conversely, Lenk et al observed the highest rate (35%) among German parents of children with renal disease, who were actively caregiving. The disease context activates information-seeking motivation, making parents more likely to acquire and process relevant information. Part of the parents in the Lenk et al study held a university degree or higher, whereas Bang et al found no significant association between awareness and educational level in India. This discrepancy suggests that, in settings where legal frameworks are inadequate and physician disclosure is absent, the informational advantage conferred by higher education may be overridden by systemic information blocking. Mukattash et al¹⁹ also reported no significant differences in parental awareness across socioeconomic strata. Thus, the observed heterogeneity in awareness fundamentally reflects the interaction of institutional, situational and individual factors: studies reporting higher awareness rates correspond to contexts with relatively clear legal frameworks, active disease situations and higher educational attainment, whereas lower awareness rates emerge in settings characterised by ambiguous regulations, general-population sampling and low rates of physician disclosure. From a clinical practice perspective, these findings suggest that improving parental awareness cannot rely solely on parent education materials. Clinicians should proactively disclose off-label status during consultations, explain the rationale in accessible language and tailor information delivery to the parent’s educational background and emotional state. Institutions should complement this by providing standardised, legally informed disclosure templates and ensuring pharmacy staff are equipped to reinforce rather than contradict the information provided during prescribing.

Beyond the factors discussed above, the four studies also differed in disease contexts and sampling frames (parents of children with chronic renal disease actively engaged in care vs community samples, including individuals without direct paediatric caregiving responsibilities) and in data collection modalities (face-to-face interviews vs telephone surveys vs self-administered questionnaires). These structural differences mean that the pooled estimate of 20.15% should be interpreted as a weighted average across highly diverse contexts rather than a single global figure.

Hierarchical structure of parental acceptance willingness

We identified a consistent pattern across studies: conditional acceptance was the most frequently reported response category (ranging from 20.6% to 85.1%), unconditional acceptance remained limited, and refusal was non-negligible. This pattern signals a critical challenge for clinical practice, in that most parents do not fully embrace off-label treatment but instead adopt a cautious, contingent stance that places ongoing monitoring and communication burdens on both families and clinicians.

Conditional acceptance reflects a passive risk-benefit trade-off rather than genuine autonomy, often occurring when alternatives are absent. Bang et al found that 73% of parents mistakenly regarded such practice as illegal, and 57% requested a switch to approved medications. Similarly, Mukattash et al reported that 62.4% perceived off-label drugs as unsafe, with 39.9% demanding replacements. Balan et al¹² also indicated that parental perceived drug safety declined markedly once they were informed of off-label use, with most parents tending to either request switching to approved medicines or closely monitor adverse reactions during medication use.

Oyeku et al²⁰ found that even when their children were already on the drug, 89% of parents expressed severe concerns about potential carcinogenic risks, and 81% worried about the lack of Food and Drug Administration (FDA) paediatric approval. Treatment acceptance and resolution of internal doubts often coexist. Parental ‘conditional acceptance’ is essentially a defensive authorisation. Under information asymmetry and decision-making pressure, parents delegate final decision-making authority to physicians, while retaining control by closely monitoring side effects, thereby alleviating anxiety caused by cognitive dissonance.

High heterogeneity was observed in the conditional acceptance rate across the four included studies. Banga et al (S1, India) reported 20.6% (80/388), Mukattash et al (S4, Northern Ireland) reported 41.9% (419/1000), Saiyed et al (S3, India) reported 61.9% (252/407) and Lenk et al (S2, Germany) reported 85.1% (80/94). This fourfold range reflects a fundamental construct validity problem, as none of the studies used the term conditional acceptance or provided a uniform operational definition. We extracted the original definitions used in each study and present these in table 6. What each study actually measured was, therefore, qualitatively different: S1 and S4 captured cautious behavioural acceptance with active monitoring, S3 captured risk-tolerance-driven acceptance under extreme clinical duress, and S2 captured conditional agreement under constrained therapeutic alternatives. These are not different operationalisations of a single construct; they capture meaningfully different parental response patterns.

Table 6. The original definitions of conditional acceptance.

Study ID	The original definition
S1	Although they would allow use of the drug in the child, they would check carefully for side effects.
S2	When there is no other possibility; a life-threatening condition or only in extreme situations.
S3	Their child was having a life-threatening condition like cancer or heart failure and off-label medicine was necessary for survival, but that it also caused an adverse drug reaction like hair loss or anaemia.
S4	They would use the medicine in the child but check the child carefully for side effects.

The pooled refusal rate was 36.5%, while a stronger refusal tendency was reported in the Indian study by Bang et al, in which 57% of parents requested switching to approved drugs and 73% viewed off-label use as illegal. This phenomenon is directly associated with India’s legal context, where off-label use is explicitly prohibited by law; once informed, parents tend to equate such practice with illegality and insecurity. Tsujii et al’s study in Japan found that physician disclosure itself was independently associated with an increased risk of parental refusal (OR=2.73), and the association remained significant after adjusting for confounding factors. This counterintuitive finding suggests that in countries where drug package inserts carry strong legal evidentiary weight and off-label prescribing may entail litigation risks, parental interpretation of physician disclosure tends to be a risk warning rather than informed authorisation, thereby triggering defensive refusal. Accordingly, refusal depends not only on individual parental attitudes but is profoundly constrained by national legal traditions, medical litigation culture and the quality of physician communication strategies.

For clinical practice, the predominance of conditional acceptance signals that parents need ongoing reassurance and structured follow-up rather than a one-time consent signature. Physicians should frame off-label discussions as collaborative decision-making processes, explicitly invite questions and provide clear channels for reporting concerns. When refusal occurs, clinicians should explore whether it stems from clinical misunderstanding, legal anxiety or past negative experiences, and respond with tailored education or alternative treatment discussions rather than viewing refusal as non-compliance. Reducing the care burden on families, through streamlined prescription renewal, coordinated pharmacy services and accessible medication counselling, may also shift conditional acceptance towards unconditional acceptance by addressing the logistical anxieties that underlie parental hesitation.

Three-dimensional factors influencing parental decision-making

We identified three core dimensions: children’s clinical characteristics, parental cognitive and emotional status, and physicians’ communication behaviours and professional qualifications. These three dimensions are not isolated; instead, they function sequentially throughout the decision-making process, forming a progressive logical chain ranging from external situational pressure and internal psychological evaluation to external regulatory factors.

First, child clinical characteristics establish the situational context and decisional pressure. When a child faces a life-threatening illness with few or no approved alternatives, the perceived benefit of off-label use outweighs potential risks for most parents. This finding aligns with previous work on parental willingness to enrol children in clinical trials, where acceptance increased proportionally with disease severity. In such high-stakes scenarios, parents become more receptive to risk and more dependent on physician guidance, which amplifies the influence of the other two dimensions.

Second, parental cognitive and emotional status acts as the internal driving force of decision-making. The widespread lack of awareness (pooled rate 20.2%) and common misconception that off-label use is illegal (73% in Bang et al) mean that when a physician does disclose off-label status, parents may perceive it as a warning of illegality or danger rather than a routine clinical explanation. This cognitive distortion can trigger defensive refusal, especially when amplified by fear and distrust. The qualitative finding that prior negative caregiving experiences accumulate into long-term scepticism suggests that each encounter with off-label prescribing can reshape future decision-making—a recursive effect rarely captured in cross-sectional surveys. Duncan et al found that most UK parents administering age-inappropriate formulations at home had ‘not really thought about’ the off-label or unlicensed status of their child’s medicines, simply following prescriber instructions without questioning, a passive trust that facilitates short-term adherence but precludes genuine informed consent.

Third, physician communication and qualifications act as the critical moderating variables. The counterintuitive finding that disclosure is associated with higher refusal rates does not negate the ethical necessity of informed consent. Rather, it underscores that how disclosure is delivered matters more than whether it occurs. When disclosure is procedural, unidirectional and devoid of reassurance or risk-benefit contextualisation, parents interpret it as a red flag. Conversely, experienced clinicians (who showed a 2% reduction in refusal odds per year of practice) likely embed disclosure within trust-centred dialogue that acknowledges parental anxiety, explains therapeutic necessity and addresses caregiving burdens proactively. This is consistent with Mukattash et al,¹⁹ who found that although most healthcare staff acknowledged the need for parental notification, actual disclosure was often perfunctory. Duncan et al provide complementary UK evidence: parents described being excluded from formulation decision-making entirely and receiving inconsistent information across primary and secondary care, with prescription rejections from GPs causing delays and distress.

Finally, a previously underexplored dimension, post-decision caregiving burden, emerged from the qualitative data. Although not a direct antecedent of the initial decision, the exhaustion from managing prescription cycles, pharmacy coordination and drug shortages feeds back into parents’ overall attitudes toward off-label use. This means that refusal or conditional acceptance in a future scenario may be driven as much by past logistical struggles as by clinical risk perception. Healthcare institutions should, therefore, optimise prescription workflows and drug supply processes to reduce this hidden burden. In this regard, clinical pharmacists can play a pivotal role: through medication counselling, adverse drug reaction monitoring, dose individualisation and coordination with prescribing physicians, pharmacists can alleviate parental anxiety, reduce caregiving burden and serve as an accessible source of ongoing information and reassurance.

In summary, parental decision-making about paediatric off-label use is a dynamic process shaped by the interplay of situational pressure (illness severity), internal psychological state (cognition and emotion) and external communication quality (physician factors). These findings carry actionable implications for clinical practice. For clinicians, disclosure of off-label status should be embedded within trust-centred dialogue that acknowledges parental anxiety, explains therapeutic necessity in accessible terms and proactively addresses caregiving burdens, rather than delivered as a unilateral procedural notification. For institutions, standardised communication protocols for off-label use should be established, prescription workflows streamlined and clinical pharmacists integrated into multidisciplinary care teams to provide ongoing medication support. For health systems, the persistent information asymmetry between physicians and parents indicates that improving informed consent requires structural investments in communication infrastructure, not merely individual clinician training. Only by addressing all three domains, namely clarifying risk-benefit rationale, correcting misconceptions and building trust through sustained, multiprofessional engagement, can the transition from procedural notification to genuine shared decision-making be achieved.

Limitations

The study has several limitations. The total number of included studies was small, reflecting the scarcity of parent-centred research in this field. This may restrict the comprehensiveness of our analysis of parental psychological experiences and decision-making processes. Variations across studies in measurement tools, outcome definitions and research settings contributed substantially to the high heterogeneity and may reduce the stability of the pooled estimates. For the unconditional acceptance rate, only two studies were available; we, therefore, present individual estimates rather than a pooled rate, and this outcome should be interpreted with very low certainty. In addition, evidence from S5 was derived from physician reports of parental refusal rather than direct parent self-report; such data may be subjected to information bias and measurement bias, as physician perceptions of refusal may not accurately reflect parental reasoning or experiences.

Conclusion

This systematic review and meta-analysis provides the first integrated synthesis of evidence on parental awareness, acceptance willingness and influencing factors concerning paediatric off-label use. Among the five countries represented in this review, parental awareness of off-label use in children was consistently low. Across the included studies, conditional acceptance was the most frequently reported response category, with individual study estimates ranging from 20.6% to 85.1%; unconditional acceptance remained limited; and pooled refusal rates were relatively high. Influencing factors span three themes: child clinical characteristics, parental cognition and emotion, and physician communication and qualifications. Illness severity and the availability of alternatives determine the decisional pressure parents face; parental cognitive level and trust in the healthcare system shape how medical recommendations are interpreted; and physician communication style and experience further moderate the final decision. Moreover, postdecision caregiving burdens, while not directly affecting the immediate decision, may accumulate into generalised scepticism towards off-label use and influence attitudinal orientations in future similar contexts.

The value of this study lies in providing an evidence base to inform clinical management and the refinement of informed consent practices. For clinicians, informed consent should shift from procedural notification to trust-centred, substantive communication that addresses parental concerns while articulating therapeutic necessity. This requires not merely disclosing off-label status but engaging in dialogue that corrects misconceptions, acknowledges anxiety and invites ongoing questions. For healthcare institutions, standardised communication protocols for off-label use should be established, and prescription workflows and medication supply processes should be optimised to reduce the burdens imposed on families, and clinical pharmacists should be integrated into multidisciplinary teams to provide sustained medication counselling and adverse effect monitoring. For policymakers, legal frameworks and informed consent standards should be further clarified to provide unambiguous guidance for clinical practice, while professional training programmes should equip clinicians with the communication skills necessary for navigating off-label consent conversations.

Given the limited number of included studies and substantial heterogeneity, future research should prioritise large-scale empirical investigations across diverse healthcare settings to further validate and enrich the multidimensional framework of influencing factors proposed herein.

Data availability statement

The data supporting the findings of this study are available within the respective articles cited in this review.