Actionable medications in Australian Vietnam War veterans: Implications for pre-emptive pharmacogenetic testing

Abstract

Purpose of research

Describe prescribing patterns in Australian Vietnam veterans, identify CYP2C19-metabolised medications using established pharmacogenetic (PGx) resources, characterise CYP2C19 profiles of veterans and assess the potential clinical impact of these medications.

Major findings

Among 283 veterans with CYP2C19 profiles, 256 reported current medications use, with a mean prescribed medication of 5.4. Of these, 89 veterans (34.7%) were prescribed at least one medication with CYP2C19 PGx recommendation. Notably, 52 veterans (58.4%) had CYP2C19 profiles that may be at risk of therapeutic failure or adverse effects. Prescribed medications also included six CYP2C19 inhibitors and one inducer, with potential to induce phenoconversion and impact drug metabolism.

Conclusion

Veterans experienced high levels of polypharmacy and frequently carried CYP2C19 phenotypes associated with increased risk of therapeutic failure or adverse effects. The presence of CYP2C19 inhibitors and inducers raises the potential for phenoconversion, where CYP2C19 profiles may be placed at risk. Given their similarities to the broader older population, these findings suggest that both groups may benefit from pre-emptive PGx testing. Furthermore, ongoing monitoring of drug–gene and drug–drug interactions remains essential to optimise medication safety and efficacy in these high-risk individuals.

Keywords

pharmacogenetics CYP2C19 phenoconversion polypharmacy Vietnam veterans

Introduction

Australian Vietnam veterans are now an ageing population with a high burden of physical and mental health comorbidities.¹ These complexities make medication management challenging, increasing the risk of drug interactions, contraindications and adverse effects.² Multiple comorbidities often lead to polypharmacy (five or more medications) and psychotropic polypharmacy (two or more psychotropic drugs).³ A systematic review and meta-analysis examining general and psychotropic polypharmacy in military and veteran populations found a high pooled prevalence of both – general polypharmacy was reported at 49% across five studies, and psychotropic polypharmacy at 36% across 15 studies.⁴ For individuals with post-traumatic stress disorder (PTSD) selecting effective psychotropics is challenging, often involving prolonged trial and error and the use of multiple agents with limited efficacy.^5,6 This approach can lead to poor health outcomes, underscoring the need for more personalised and efficient prescribing.⁷

Pharmacogenetics (PGx) is gaining traction as a tool to guide drug selection and dosing.⁸ The Royal College of Pathologists of Australasia (RCPA) highlighted in 2017 that many Australians were prescribed gene–drug-associated medications, with a significant proportion likely carrying variants warranting therapy adjustments.⁹ In 2024, the RCPA released PGx testing guidelines for 35 common medications to improve efficacy and reduce adverse reactions in Australia (https://www.rcpa.edu.au/Library/Practising-Pathology/Pharmacogenomic-Indications-in-Australia). However, Medicare subsidised PGx tests remain limited.¹⁰ Whilst PGx should not be used in isolation to guide treatment decisions, the general recommendation is to use this approach in conjunction with existing clinical data, including Therapeutic Drug Monitoring and knowledge of known drug–drug interactions.¹¹

CYP2C19 is a key enzyme in the metabolism of various medications.¹² While CYP2C19 activity is multifactorial, single nucleotide polymorphisms (SNPs) determine metabolic activity and thus medication response.¹² Individuals can be classified into ultra-rapid (UM, two gain-of-function), rapid (RM, one gain-of-function), normal (NM, two wildtype), intermediate (IM, one non-functional) and poor metabolisers (PM, two non-functional).¹³

This study aims to (1) examine medication use in Australian Vietnam veterans, (2) identify CYP2C19-metabolised drugs using available PGx resources (former PharmGKB site now ClinPGx, Clinical Pharmacogenetics Implementation Consortium (CPIC), Dutch Pharmacogenetics Working Group (DPWG) and RCPA), (3) characterise PGx profiles of the veterans and determine the prevalence of CYP2C19-metabolised medication use and (4) assess the potential clinical impact for veterans prescribed these medications.

Materials and methods

Ethics

The study was approved by the relevant Human Research Ethics Committees.

Participants and genetic data

Data collection and evaluations undergone by veterans have been reported in a prior publication.¹⁴ Briefly, the data was obtained from the PTSD Initiative study conducted at Gallipoli Medical Research, Brisbane, Australia, which occurred between 2014 and 2015. The study included male veterans who previously served in the Australian Defence forces during the Vietnam War. A total of 311 veterans consented, and their data was considered for subsequent evaluations.

Psychological and physical outcome measures

Veterans underwent a series of interviews and evaluations of various psychological and physical conditions of interest as summarised in Table 1. They were considered to have depressive symptoms if depression anxiety and stress scales (DASS-21) depression scores were greater than five, indicating mild to extremely severe depression.¹⁵

Table 1.

Summary of evaluations and conditions assessed.

Evaluation or test	Conditions assessed
Psychiatric evaluation (DSM-5 criteria)	PTSD
Mini-International Neuropsychiatric Interview (MINI)	Major depressive disorder (MDD)
	Generalised anxiety disorder (GAD)
	Alcohol and substance abuse
	Suicide risk
Alcohol use disorders identification test (AUDIT) and general medical questionnaire	Excessive alcohol consumption
	Alcohol-use disorders
Structured Sleep History questionnaire	Sleep disturbance
Epworth Sleepiness Scale (ESS)	Daytime sleepiness
Montreal Cognitive Assessment (MoCA)	Cognitive functioning and impairment
Medical screening questionnaires	Comorbid medical conditions
Clinician-Administered PTSD Scale for Diagnostic and Statistical Manual of Mental Disorders-5 (CAPS-5)	PTSD
Depression, Anxiety, Stress Scale (DASS-21)	Depression, anxiety and stress

Following quality control of genotype data obtained from blood samples, 28 veterans were removed due to missing genotypes, relativeness or genotyping error, leaving 283 individuals. As not all veterans reported current medications, results relating to medications were reported only on 256 individuals.

Medications

Detailed methodologies on the classification of current medications taken by participants have been described in a previous publication.¹⁶ Briefly, current medications were recorded and coded using the Anatomical Therapeutic Chemical (ATC) classification system. Emphasis was placed on central nervous system (N class medications) drugs, excluding anaesthetics (N01). Lithium was considered an independent class (N05AN), and all benzodiazepines in hypnotic and sedative class (N05CD) were placed with anxiolytics (N05B) to capture within-class polypharmacy. Medications typically used for cardiovascular conditions but considered off-label for anxiety disorders were also considered. These medications included beta-blockers (C07A) and prazosin (C02CA01).

Medications lacking an ATC code were excluded. Where ATC codes differed but the active ingredient was the same (e.g. short-acting and long-acting insulin), these were treated as a single medication.

PGx recommendations on CYP2C19 were sourced from CPIC and DPWG in the former PharmGKB site now ClinPGx (https://www.clinpgx.org/), and the PGx indications in Australia developed by the RCPA (https://www.rcpa.edu.au/Library/Practising-Pathology/Pharmacogenomic-Indications-in-Australia).

Medications affected by CYP2C19 classified as either substrates (metabolised by CYP2C19), inhibitors (reduce CYP2C19 activity) or inducers (enhance CYP2C19 activity), along with their corresponding levels of evidence according to the Flockhart Table.¹⁷

Genotyping

Handling of blood samples for genotyping was described in a previous publication.¹⁸ Blood samples were collected from veterans and sent to the Australian Genome Research Facility (AGRF) which carried out DNA extraction and genotyping. Samples were stored in a −20°C environment. The MACHERY-NAGEL NucleoSpin L (MACHEREY-NAGEL GmbH & Co. KG, Dueren, NRW, Germany) was used to extract DNA from 2 mL of blood. DNA quality was evaluated using resolution on a 0.8% agarose gel at 130 V for a period of 60 min and normalised to 200 ng of DNA in 4 μL of water.

Samples were analysed on an Illumina PsychArray-24 BeadChip and scanned using the Illumina iScan systems. Subsequently, quality control was conducted using the GenomeStudio v2011.1 (Illumina, San Diego, CA, USA) with Genotyping Module 1.9.4 software using the InfiniumPsychArray-24v1-1_A1 systems which creates and clusters files for calling variants. Genotype call rate across variants was 99.59%.

Imputation

Not all genotypes were available, necessitating imputation of genetic variants to ensure PGx was properly classified. Prior to imputation, quality control was performed to remove variants with a minor allele frequency <1%, a call rate <95% and a Hardy–Weinberg equilibrium threshold <0.000001. Genotypes were imputed using the Sanger Imputation Server (https://imputation.sanger.ac.uk/). Chromosomes were phased against the UK10K + 1000 Genomes Phase 3 reference panel using EAGLE2 before imputation with the Positional Burrows Wheeler Transform. The Sanger server was used for its superior coverage of variants of interest. Imputed variants with an info score >0.6 were kept for further analysis, consistent with best practices to ensure accuracy and minimise phenotype misclassification.

PGx calling

CYP2C19 PGx typing was conducted by examining the five key genetic variants. For each variant, the number of alleles of interest present was used to determine the metaboliser phenotype (Table 2). The genetic variants considered were rs3758581 (*1 wild-type), rs12769205 (*2 loss-of-function), rs4244285 (*2 loss-of-function), rs4986893 (*3 loss-of-function) and rs12248560 (*17 gain-of-function). While rs4244285 is the primary CYP2C19*2 variant, rs12769205 is in linkage disequilibrium with rs4244285, and therefore considered when calling CYP2C19 PGx.

Table 2.

Coding used to determine the CYP2C19 PGx for each participant.

CYP2C19 PGx	Diploid	CYP2C19 *1 rs3758581	CYP2C19 *2 rs12769205	CYP2C19 *2 rs4244285	CYP2C19 *17 rs12248560
UM	17/17	0	0	0	2
RM	1/17	0	0	0	1
RM	1/17	1	0	0	1
NM	1/1	0	0	0	0
NM	1/1	1	0	0	0
NM	1/1	2	0	0	0
IM	1/2	0	0	1	0
IM	1/2	0	1	1	0
IM	2/17	0	1	1	1
IM	1/2	1	1	1	0
PM	2/2	0	2	2	0

Genotypes were coded 0, 1 or 2 based on the count of the allele of interest.

Descriptive analysis

Descriptive analyses were performed using R (v4.5.0) and Microsoft Excel.

Results

Demographics

The study cohort included 283 veterans with CYP2C19 genotypes. The mean age was 68.8 (SD, 4.1) years. A notable proportion were university graduates (N = 87, 31%) and of European descent (N = 270, 98.9%) (Table 3).

Table 3.

Patient characteristics.

Variable	N	Mean ± SD (min–max) or frequency in each group (% participants)
Age (years)	282	68.8 ± 4.1 (60–88)
Highest education level	281
University		87 (31.0%)
Year 11–12		64 (22.8%)
Year 10		50 (17.8%)
<Year 10		32 (11.4%)
Vocational		48 (17.1%)
Ethnic background	273
European		270 (98.9%)
Indigenous Australian		2 (0.7%)
Other		1 (0.4%)
Comorbidities
Generalised anxiety disorder	280	14 (5.0%)
Major depression	280	21 (7.5%)
Epilepsy	281	6 (2.1%)
Parkinson’s disease	281	0 (0.0%)
Obstructive sleep apnoea	283	89 (31.4%)
Regular pain	282	136 (48.2%)
Sleep disturbance
Nightmares	283	169 (59.7%)
DASS-21 scores
Depression	281	4.4 ± 4.5 (0–21)
Anxiety	281	3.9 ± 4.2 (0–20)
Stress	281	7.6 ± 5.2 (0–21)
Depression >4		108 (38.4%)
Anxiety >3		114 (40.6%)
Stress >7		130 (46.3%)
Current suicide risk	280	30 (10.7%)
Alcohol use
Abuse	257	4 (1.6%)
Dependence	279	24 (8.6%)
For sleep	241	35 (14.5%)
AUDIT score category	281	7.32 ± 5.8 (0–30)
Low (0–7)		170 (60.5%)
Risky (8–15)		86 (30.6%)
High (16–19)		17 (6.1%)
High, likely dependent (≥20)		8 (2.9%)
PTSD	201	102 (50.7%)
Substance use
Abuse	274	0 (0.0%)
Dependence	280	1 (0.4%)
CAPS-5 total symptom score	270	9.7 ± 10.1 (0–56)
MoCA	281	26.3 ± 2.5 (19–30)
ESS
Total score	283	8.4 ± 4.8 (0–24)
Score ≥10 (excessively sleepy)		107 (37.8%)
Total number of any medications	256	5.4 ± 3.5 (1–16)
Veterans taking ≥5 medications	134 (52.3%)
Number of any medications	256
0		136 (53.1%)
1		62 (24.2%)
2		35 (13.7%)
3		12 (4.7%)
4		8 (3.1%)
5		3 (1.2%)
CYP2C19 phenotype	283
Poor metaboliser (PM)		11 (3.9%)
Intermediate metaboliser (IM)		67 (23.7%)
Normal metaboliser (NM)		118 (41.7%)
Rapid metaboliser (RM)		78 (27.6%)
Ultra-rapid metaboliser (UM)		9 (3.2%)

Psychological and clinical assessments

For 201 veterans with data available (Table 3), 102 (50.7%) met the DSM-5 criteria for trauma exposure and were diagnosed with PTSD, by both a psychiatrist and CAPS-5. The average CAPS-5 total symptom severity score was 9.7 (10.1).

On average, veterans experienced mild symptoms of depression, anxiety and stress with DASS-21 sub-scores of 4.4 (SD, 4.5), 3.9 (SD, 4.2) and 7.6 (SD, 5.2), respectively (Table 3). Among the 281 veterans with data available (Table 3), the number (%) of veterans with mild symptom severity or greater for depression, anxiety and stress were 108 (38.4%), 114 (40.6%) and 130 (46.3%), respectively.

When assessing alcohol use, cognitive function and daytime sleepiness, the average scores for AUDIT, MoCA and ESS were 7.32 (SD, 5.8), 26.3 (SD, 2.5) and 8.4 (SD, 4.8), respectively (Table 3). Among the 281 veterans with data available, most veterans had a low-risk AUDIT score (N = 170, 60.5%), with relatively fewer individuals at high risk (N = 17, 6.1%) or with possible alcohol dependence (N = 8, 2.9%).

Among the 280 veterans with information available, 30 veterans (10.7%) were found to be at risk of suicide. Fewer veterans experienced generalised anxiety disorder (N = 14, 5.0%) and major depression (N = 21, 7.5%), according to MINI. In terms of sleep disturbances for the 283 veterans with data available, more than half of the participants experienced nightmares (N = 169, 59.7%). Obstructive sleep apnoea was also reported by this cohort (N = 89, 31.4%). A notable proportion of participants (N = 282 with available data) reported regular pain (N = 136, 48.2%).

Polypharmacy for all participants

Among veterans reporting current medication use (N = 256), the average number of medications taken was 5.4 (SD, 3.5). A significant proportion of participants reported general polypharmacy (N = 134, 52.3%) (Table 3). Psychotropic medication use was common, with 46.9% of veterans (N = 120) reporting at least one such medication. Psychotropic polypharmacy was observed in 22.7% (N = 58).

Medications with established recommendations

A total of 267 medications were used by veterans in this study. Of these, acetylsalicylic acid (aspirin), a non-steroidal anti-inflammatory drug (NSAID), was the most commonly prescribed, taken by 87 of 256 participants (34.0%), followed by atorvastatin (N = 49, 19.1%) and paracetamol (N = 48, 18.8%). For drugs metabolised by CYP2C19, omeprazole was the most frequently prescribed, taken by 22 of 256 participants (8.5%), followed by clopidogrel (N = 20, 7.8%) and escitalopram (N = 16, 6.3%). Forty-seven medications (17.6%) were classified as psychotropics (N class drugs). Nine (3.4%) medications were metabolised by CYP2C19 based on CPIC and DPWG recommendations. Among the 47 psychotropic medications, five antidepressants – escitalopram, doxepin, sertraline, citalopram and amitriptyline – were metabolised by CYP2C19.

Of the 267 medications, three medications (1.1%) had an RCPA recommendation for CYP2C19. The RCPA website categorises 35 medications into three groups: recommended, consider and non-consensus which relate to the strength of the indication in regard to PGx testing. Among veterans, only clopidogrel (B01AC04) had a ‘recommended’ indication for PGx testing for CYP2C19. The remaining two medications had a ‘consider’ indication for PGx testing for CYP2C19: amitriptyline (N06AA09) and citalopram (N06AB04). Notably, two drugs metabolised by CYP2C19 – clopidogrel (B01AC04) and citalopram (N06AB04) – have aligned recommendations across CPIC, DPWG and RCPA. Amitriptyline (N06AA09) is supported by recommendation from CPIC and RCPA only. Overall, 89 participants (34.8%) were taking at least one medication with a recommendation (Table 4).

Table 4.

Number of participants using drugs with either a CYP2C19 according to CPIC/DPWG or RCPA PGx recommendation.

PGx target	Total number of participants (%)
Drugs metabolised by CYP2C19	89
CPIC/DPWG recommendations for drugs metabolised by CYP2C19	89
1 drug	72/89 (80.9%)
2 drugs	17/89 (19.1%)
RCPA indications for drugs metabolised by CYP2C19	33
1 drug	32/33 (97.0%)
2 drugs	1/33 (3.0%)

CYP2C19 PGx frequencies among participants

Genotype information was available for 283 veterans in this study. Of these, NM (N = 118, 41.7%) made up a major proportion of the cohort. Following NM, IM (N = 67, 23.7%) and RM (N = 78, 27.6%) made up a notable proportion of the sample (Table 5). Fewer participants were classified as PM (N = 11, 3.9%) and UM (N = 9, 3.2%) (Table 5). The frequencies observed among veterans align with frequencies expected in a Europeans as shown in Table 5.¹⁹ While the percentages are not identical, they are reasonably similar. Overall, a significant proportion veterans (N = 165, 58.3%) exhibited CYP2C19 phenotypes associated with altered drug metabolism (IM, RM, PM or UM) which may result in altered drug responses for medications metabolised by CYP2C19. As noted above, 89 veterans were taking at least one medication with a recommendation (Table 4), and of these, 52 (58.4%) had altered CYP2C19 metabolism (IM, RM, PM or UM).

Table 5.

Observed and expected CYP2C19 phenotype distribution.

	European frequencies (%)	Sample (N = 283)
Poor metaboliser (PM)	4.6	11 (3.9%)
Intermediate metaboliser (IM)	23.9	67 (23.7%)
Normal metaboliser (NM)	39.1	118 (41.7%)
Rapid metaboliser (RM)	26.9	78 (27.6%)
Ultra-rapid metaboliser (UM)	4.6	9 (3.2%)

CYP2C19 medication interactions

Overall, 15 medications were identified as potential interactors with CYP2C19 (Table 6). Of these, nine medications were substrates metabolised by CYP2C19, six medications were inhibitors, (reduce metabolism of CYP2C19 substrates) and one was an inducer (increase metabolism of CYP2C19 substrates). Notably, omeprazole was classified as both substrate and inhibitor, reflecting its complex pharmacokinetic profile. Among substrates, omeprazole was most frequently used by veterans, whilst esomeprazole was the most common CYP2C19 inhibitor. Other notable inhibitors include fluoxetine (moderate inhibitor) and fluvoxamine (strong inhibitor). Prednisone was the only potential inducer identified. Three veterans (IM, PM and RM phenotypes) were prescribed both omeprazole and clopidogrel, a known gene–drug–drug interaction.

Table 6.

Summary of medication frequency and CYP2C19 interaction potential.

Medication	N	2C19 substrate	Flockhart substrate	2C19 inhibitor	Flockhart inhibitor	2C19 inducer
Amitriptyline	6	x	Weak evidence
Citalopram	8	x	Strong evidence
Doxepin	4	x	Moderate evidence
Escitalopram	16	x	Strong evidence
Fluoxetine	2			x	Moderate inhibitor
Fluvoxamine	5			x	Strong inhibitor
Sertraline	10	x
Chloramphenicol	1			x	Weak inhibitor
Clopidogrel	20	x	Strong evidence
Esomeprazole	44			x	Moderate inhibitor
Lansoprazole	6	x	Strong evidence
Modafinil	1			x	Weak inhibitor
Omeprazole	22	x		x	Weak inhibitor
Pantoprazole	14	x
Prednisone	9					x

Discussion

PGx can improve medication safety by guiding drug selection, but its use in Australia is limited. This study examined CYP2C19 variation and drug use in veterans.

Polypharmacy was prevalent among veterans, with over half taking multiple medications and more than one in five prescribed multiple psychotropics. While polypharmacy is common and often necessary in older adults,²⁰ it heightens the risk of drug–drug interactions, and adverse outcomes highlighting the need for PGx testing to minimising unnecessary polypharmacy.^2,20

Genetic variations in CYP2C19 clinically important due to its role in drug metabolism, particularly for medications used in cardiovascular and mental health disorders.¹² As such, CYP2C19 is routinely included in PGx panels to identify drug–gene interactions and guide medication dosing. The high proportion of veterans (89/256, 34.7%) prescribed medications with CYP2C19 pharmacogenomic recommendations (CPIC/DPWG or RCPA) highlights the potential clinical utility of implementing PGx testing in this population. Notably, many veterans carry genotypes associated with altered metabolism (165/283, 58.3%), which may place them at increased risk of therapeutic failure or adverse effects when prescribed certain medications. In fact, a number of these individuals are already taking CYP2C19-metabolised medications (52/89, 58.4%), meaning they may currently be at risk. These findings suggest that routine consideration of CYP2C19 phenotype could inform safer, more effective prescribing.

Panel-based PGx testing and programs such as PREDICT (The Pharmacogenomic Resource for Enhanced Decisions in Care and Treatment) or the Mayo-Baylor Right 10K study have been developed to support implementation of PGx into clinical practice.^21–23 These programs identify actionable drug–gene interactions and verify and incorporate this information into medical records. Through this pre-emptive approach, the Mayo-Baylor RIGHT 10K study found that most participants could benefit from the pre-emptive testing of 13/77 genes assessed, including CYP2C19.²² A recent review affirmed the value of such testing in improving prescribing but cautioned that existing PGx guidelines are largely derived from Caucasian ancestry, limiting their applicability to diverse populations.²⁴

Notably, the average age of veterans is 68.8 years, which highlights the potential value of PGx testing in older adults. A study conducted in Denmark consisting of community-dwelling participants in a similar age group (65–98 years) found 37 out of 87 individuals taking multiple drugs with CYP2C19 recommendations according to CPIC and DPWG (42.5%),²⁵ further supporting the relevance of PGx testing in this age group.

While drug–gene interactions are important in prescribing, phenoconversion can alter genotype-predicted metabolism. Phenoconversion occurs when non-genetic factors, such as concomitant CYP inhibitors or inducers, modify actual drug metabolism, making polypharmacy a key risk factor. CYP2C19 NM may phenoconvert to PM or UM phenotypes with concurrent inhibitors or inducers, respectively.²⁶ For instance, the concomitant use of omeprazole and clopidogrel in three veterans in our study may place them at risk of adverse cardiovascular outcomes due to these interactions.²⁷ The use of omeprazole can convert an NM into PM, reducing clopidogrel’s antiplatelet effect and may increase risk of cardiovascular events.²⁷ Drugs metabolised by CYP enzymes to varying extents and active metabolites – as well as lifestyle and other factors – can compound this complexity by contributing to drug–drug interaction risk.²⁸

Strengths and limitations

The study has several strengths, including its focus on a homogenous group of Australian Vietnam veterans which provides valuable insights into a population often underrepresented in research. The use of PGx guidelines enhanced the interpretation of dosing recommendations, while the availability of comprehensive data, including psychological and physical health measures, medication use and PGx profiles, enabled a nuanced and multi-dimensional analysis. This study was particularly relevant for older veterans who are at increased risk of polypharmacy and its associated complications.

Despite these strengths, several limitations should be acknowledged. Although the homogeneous cohort allowed for a tighter genetic analysis, the sample size was small and may not represent all veterans. Since all participants are male, the clinical implications of these variants may be limited, given known sex differences in pharmacokinetics, drug response and prescribing patterns. As such, inclusion of female veterans in future studies will be important. Other factors – such as comorbidities, concurrent medications and lifestyle behaviours not captured in this study – can influence drug metabolism and remain clinically important. Another limitation is the lack of ethnic diversity with most participants being from European ancestry. Genetic variability in CYP2C19 differs substantially across ethnic or ancestral groups, potentially affecting drug metabolism and response. As such, PGx findings in a predominantly European population may not apply to genetically diverse groups, such as those of Asian or African descent.²⁹ Finally, this study focused on CYP2C19 due to data completeness and its relevance to commonly prescribed medications in this cohort. Other genes affecting drug metabolism, such as CYP2D6 and CYP3A4, are also important in the context of polypharmacy and represent priorities for future research.

Conclusion

This study suggests that PGx may benefit Australian veterans and the broader elderly population, given the prevalence of polypharmacy and CYP2C19-affected medications. Future research should expand to younger and more diverse veterans, include additional pharmacogenes and incorporate longitudinal outcomes to assess the real-world impact of PGx-guided therapy on effectiveness, polypharmacy and healthcare costs. Clinical pathways for PGx testing in veterans should be informed by international models, such as the US Veterans PHASER program and large-scale civilian initiatives, including PREDICT and Mayo Right 10K, which provide valuable frameworks for integrating pharmacogenomics into routine care.²³

Footnotes

Acknowledgements

Sullivan Nicolaides Pathology and Queensland X-Ray provided in-kind support, and the Australian Government Department of Veterans’ Affairs provided transport for eligible participants. We also gratefully acknowledge the dedicated efforts of the participants and their families, and the clinical and support staff involved in data collection.

Declaration of conflicting interests

The authors received no financial support for the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The original study from which this data was obtained (the PTSD Initiative at the Gallipoli Medical Research Institute) was funded by the Queensland Branch of the Returned and Services League of Australia (RSL).

Ethical considerations

The study was approved by the Departments of Defence and Veteran Affairs Human Research Ethics Committee (EO14-002) and the Queensland University of Technology Human Research Ethics Committee (LR 2024-6457-20350).

Consent to participate

Written informed consent was obtained from participants in advance.

ORCID iD

Wei Liang Andre Tan

Data Availability Statement

The de-identified data we analysed are not publicly available, but requests to the corresponding author for the data will be considered on a case-by-case basis.*

References

Marmar

Schlenger

Henn-Haase

, et al. Course of posttraumatic stress disorder 40 years after the Vietnam war: findings from the national Vietnam veterans longitudinal study. JAMA Psychiatry 2015; 72: 875–881.

Christopher

Shrestha

, et al. Medication use problems among older adults at a primary care: a narrative of literature review. Aging Med 2022; 5: 126–137. 20220315.

Morizio

Mistry

McKnight

, et al. Polypharmacy and high-risk medications in older veterans presenting for emergency care. J Geriatric Emerg Med 2021; 2: 5.

Raut

Mellor

Meurk

, et al. Prevalence and factors associated with polypharmacy in military and veteran populations: a systematic review and meta-analysis. J Affect Disord 2025; 369: 411–420.

Gardner

Brennan

Scott

, et al. The potential utility of pharmacogenetic testing in psychiatry. Psychiatry J 2014; 2014: 730956.

Al Jowf

Ahmed

Reijnders

, et al. To predict, prevent, and manage post-traumatic stress disorder (PTSD): a review of pathophysiology, treatment, and biomarkers. Int J Mol Sci 2023; 24: 5238.

Vandenberg

Echt

Kemp

, et al. Academic detailing with provider audit and feedback improve prescribing quality for older veterans. J Am Geriatr Soc 2018; 66: 621–627.

Weinshilboum

Wang

. Pharmacogenomics: precision medicine and drug response. Mayo Clin Proc 2017; 92: 1711–1722.

Polasek

Mina

Suthers

. Pharmacogenomics in general practice: the time has come. Aust J Gen Pract 2019; 48: 100–105.

10.

White

Scott

Paul

, et al. Pharmacogenomics in the era of personalised medicine. Med J Aust 2022; 217: 510–513.

11.

de Leon

. Personalizing dosing of risperidone, paliperidone and clozapine using therapeutic drug monitoring and pharmacogenetics. Neuropharmacology 2020; 168: 107656.

12.

Shubbar

Alchakee

Issa

, et al. From genes to drugs: CYP2C19 and pharmacogenetics in clinical practice. Front Pharmacol 2024; 15: 1326776. Review.

13.

Dean

Kane

. Clopidogrel therapy and CYP2C19 genotype. National Center for Biotechnology Information, 2012.

14.

McLeay

Harvey

Romaniuk

, et al. Physical comorbidities of post-traumatic stress disorder in Australian Vietnam war veterans. Med J Aust 2017; 206: 251–257.

15.

Lovibond

. Manual for the depression anxiety stress scales. Psychology Foundation of Australia, 1996.

16.

Theal

McLeay

Gibson

, et al. Psychotropic polypharmacy in Australian Vietnam war veterans with post-traumatic stress disorder: a descriptive cohort study. J Military Vet Health 2020; 28: 34–45.

17.

Flockhart

Thacker

McDonald

, et al. The flockhart cytochrome P450 drug-drug interaction table. https://drug-interactions.medicine.iu.edu (accessed 30 April 2025).

18.

Bruenig

Mehta

Morris

, et al. Genetic and serum biomarker evidence for a relationship between TNFα and PTSD in Vietnam war combat veterans. Compr Psychiatry 2017; 74: 125–133. 20170124.

19.

Huddart

Fohner

Whirl-Carrillo

, et al. Standardized biogeographic grouping system for annotating populations in pharmacogenetic research. Clin Pharmacol Ther 2019; 105: 1256–1262. 20190121.

20.

Sharp

Linder

Valdes

Jr . Polypharmacy: a healthcare conundrum with a pharmacogenetic solution. Crit Rev Clin Lab Sci 2019; 57: 161–180. 20191102.

21.

Peterson

Bowton

Field

, et al. Electronic health record design and implementation for pharmacogenomics: a local perspective. Genet Med 2013; 15: 833–841.

22.

Wang

Scherer

Bielinski

, et al. Implementation of preemptive DNA sequence–based pharmacogenomics testing across a large academic medical center: the Mayo-Baylor RIGHT 10K study. Genet Med 2022; 24: 1062–1072.

23.

Dong

Bates

Chanfreau-Coffinier

, et al. Veterans affairs pharmacogenomic testing for veterans (PHASER) clinical program. Pharmacogenomics 2021; 22: 137–144.

24.

Mosch

van der Lee

Guchelaar

, et al. Pharmacogenetic panel testing: a review of current practice and potential for clinical implementation. Annu Rev Pharmacol Toxicol 2025; 65: 91–109.

25.

Vermehren

Søgaard Nielsen

Jørgensen

, et al. Drug use among nursing home residents in Denmark for drugs having pharmacogenomics based (PGx) dosing guidelines: potential for preemptive PGx testing. J Personalized Med 2020; 10: 78, 20200731.

26.

Klomp

Manson

Guchelaar

H-J

, et al. Phenoconversion of cytochrome P450 metabolism: a systematic review. J Clin Med 2020; 9: 2890.

27.

Catapano

Srinivasan

Wakim

, et al. Omeprazole-clopidogrel interaction and neurovascular complications after flow-diverter device placement. J Neurointerventional Surg 2022; 14: 380–383.

28.

Zhao

, et al. Cytochrome P450 enzymes and drug metabolism in humans. Int J Mol Sci 2021; 22: 12808.

29.

Zhang

. Frequencies of clinically important CYP2C19 and CYP2D6 alleles across East Asian populations. In: Proceedings of the 2021 10th international conference on bioinformatics and biomedical science, Xiamen, China, 2022, pp. 63–69. Association for Computing Machinery.