Abstract
Introduction
Rational prescribing is essential for children under 5 years of age due to their vulnerability to dosing errors and adverse drug effects. Poor adherence to WHO prescribing guidelines may contribute to medication errors and antimicrobial resistance. This study evaluated paediatric prescribing patterns and adherence to WHO prescribing indicators across diverse healthcare settings in Sri Lanka.
Methods
A descriptive, cross-sectional, multicentre study was conducted across six provinces in Sri Lanka. A total of 1400 outpatient prescriptions issued for children under 5 years were systematically sampled from public and private healthcare facilities. Prescriptions were assessed against WHO prescribing criteria, including patient information, drug-related details, prescriber identifiers and antibiotic use. Comparisons across healthcare sectors (government vs private) and prescription formats (handwritten vs typed) were performed using χ2 tests.
Results
Of the 1400 prescriptions analysed, 94.6% were handwritten and 53.1% originated from the private sector. Critical patient details were incompletely documented, with weight recorded in only 43.0% and diagnosis in 36.0% of prescriptions. Antibiotics were prescribed in 57.2% of prescriptions. Government sector prescriptions showed significantly better documentation of patient identifiers and diagnoses (p<0.00001) and greater use of generic drug names (58.6% vs 26.6%, p<0.00001). In contrast, private sector prescriptions more frequently documented accurate dosing (92.9% vs 76.4%, p<0.00001), treatment duration (98.1% vs 86.3%, p<0.00001) and prescriber credentials (p<0.01). Typed prescriptions demonstrated superior documentation of patient identifiers, treatment duration, route of administration and prescriber identifiers compared with handwritten prescriptions (all p≤0.005).
Conclusion
Prescribing practices across the studied facilities showed notable gaps in prescription documentation and variability between healthcare sectors and prescription formats. These findings highlight areas that may benefit from improved prescribing practices, strengthened regulatory oversight and wider adoption of digital prescribing systems to enhance prescription quality and patient safety.
Keywords: Children, Child Health, Cross-Sectional Studies, Health Policy, Health services research
WHAT IS ALREADY KNOWN ON THIS TOPIC
Young children are at high risk of medication errors; prescriptions are often incomplete in low- and middle-income countries.
WHAT THIS STUDY ADDS
Paediatric prescriptions in Sri Lanka show poor documentation of weight (43%) and diagnosis (36%), with 57% containing antibiotics.
Government prescriptions had better patient details, while private prescriptions had more accurate dosing and prescriber information.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Findings support improved adherence to WHO prescribing standards and wider use of digital/typed prescriptions.
Introduction
Prescription practices play a critical role in ensuring the safe and effective treatment of illnesses, particularly in vulnerable populations such as children under 5 years of age.1 This age group is at heightened risk of adverse health outcomes due to developmental and physiological differences that influence drug absorption, metabolism and excretion.2 Inappropriate prescription practices, including errors in drug selection, dosing and documentation, may lead to serious consequences, including ineffective treatment, drug toxicity or antimicrobial resistance.3
Globally, studies have highlighted significant gaps in paediatric prescription practices.4 Research from low- and middle-income countries (LMICs) has reported high rates of inappropriate prescribing, including overuse of antibiotics and failure to specify critical drug information such as dosage and duration.5 6 Similarly, a review of prescription practices in European settings revealed common errors, including incomplete patient information and improper documentation of drug regimens, despite more advanced healthcare infrastructure.7 These findings underscore the universality of challenges in paediatric prescribing and the urgent need for standardisation and adherence to guidelines.
In Sri Lanka, healthcare services are delivered through a combination of public and private sector systems, each catering to diverse geographical and demographic populations. Despite substantial advancements in healthcare infrastructure, concerns persist about the quality and safety of prescriptions issued for paediatric patients. A national review reported that irrational drug use, including overprescription and use of brand names over generics, remains a prevalent issue in both outpatient and inpatient settings in Sri Lanka, particularly in the paediatric population.8 Studies have indicated that only 17.5–62.4% of prescriptions in public facilities adhered to essential drug list recommendations, with significant regional and institutional variability.8 This highlights the urgent need for systematic monitoring and interventions to standardise paediatric prescribing practices across the country.
The WHO provides guidance on good prescribing practices, which emphasise clear documentation of patient identifiers, use of generic medicine names, specification of dose, route, frequency and duration of treatment, and inclusion of prescriber identifiers.9 These elements are particularly important in paediatric care where weight-based dosing is essential to avoid underdosing or overdosing. While the WHO framework includes broader core drug use indicators, such as the average number of medicines per encounter and the proportion of medicines from essential drug lists, the more fundamental prescribing elements themselves have not been systematically examined for paediatric outpatient prescribing in Sri Lanka. In addition, few studies have evaluated prescription documentation quality across different healthcare sectors or prescription formats.10 Therefore, the present study aimed to assess adherence to WHO core prescribing elements in outpatient prescriptions issued for children under 5 years of age in Sri Lanka, and to compare prescribing documentation across public and private healthcare sectors, multiple provinces and handwritten versus typed prescriptions.
Methods
Study design
This study employed a descriptive, cross-sectional, multicentre design to evaluate prescription practices for children aged 5 years and across diverse healthcare settings in Sri Lanka. The study was conducted over a 1-year period from November 2024 to October 2025, allowing for data collection across both public and private sector institutions situated in six geographically and demographically diverse provinces.
Study population
Data were collected from outpatient prescriptions issued at multiple sites including outpatient department (OPD) pharmacies in tertiary care hospitals, state pharmaceutical corporations (SPCs) and selected private hospitals and dispensaries across the country. A total of 23 study settings across six provinces were included: Western Province (Ragama), North Central Province (Anuradhapura), Northern Province (Jaffna), Central Province (Kandy), Southern Province (Galle) and Eastern Province (Trincomalee and Batticaloa) (figure 1).
Figure 1. Distribution of the study settings across Sri Lanka.
Study settings were selected using a purposive sampling approach to capture variation in paediatric prescribing practices across different geographical regions and healthcare sectors in Sri Lanka. Six provinces were included to ensure representation from multiple administrative regions with differing population densities, healthcare infrastructure and prescribing environments. Within each province, healthcare facilities were selected to represent the major types of outpatient prescribing settings commonly used by caregivers of young children. Tertiary care hospitals were included because they serve as major referral centres with high outpatient volumes and structured prescribing environments within the public healthcare system. SPC pharmacies were included as they dispense prescriptions from both public and private practitioners and provide access to essential medicines in community settings. Private hospitals and dispensaries were selected to reflect prescribing practices in the private healthcare sector, where a substantial proportion of paediatric outpatient care occurs in Sri Lanka. While purposive selection of facilities may introduce some risk of selection bias, inclusion of multiple facility types across several provinces was intended to capture a broad range of prescribing practices in Sri Lanka.
Inclusion and exclusion criteria
This study included outpatient prescriptions issued for children, specifically from birth up to 5 years of age. Prescriptions were eligible for inclusion if they were issued for children within this age group and originated from one of the six selected provinces during the study period. Prescriptions were excluded if essential information could not be extracted, or if they lacked sufficient patient demographic data to confirm eligibility.
Prescriptions were eligible regardless of whether they were issued at a first consultation or during a follow-up visit. In routine outpatient prescribing records, this distinction was not consistently documented and could not be reliably determined from the prescription sheets alone. Therefore, all eligible prescriptions issued for children under 5 years during the study period were included. Because the primary aim of the study was to evaluate prescription documentation and adherence to WHO prescribing elements rather than clinical decision-making across visit types, inclusion of both initial and follow-up prescriptions was considered appropriate.
During the data collection period, a total of 1436 prescriptions for children under 5 years of age were initially screened across the selected study settings. After applying the predefined eligibility criteria, 36 prescriptions were excluded from the analysis. The main reasons for exclusion included incomplete or illegible prescriptions that prevented extraction of essential study variables (n=21), prescriptions lacking sufficient demographic information to confirm the patient’s age eligibility (n=9) and duplicate prescriptions identified during data entry (n=6). Following these exclusions, a final sample of 1400 prescriptions was included in the analysis.
Sample size and sampling strategy
The minimum required sample size for estimating prescription documentation characteristics was calculated using a standard single-proportion formula assuming a prevalence of 50%, a 95% confidence level (Z=1.96) and a precision of 5%, resulting in a minimum sample size of 384 prescriptions. A prevalence of 50% was assumed in the absence of prior data, as this provides the most conservative estimate and yields the maximum required sample size. Allowing for incomplete records and regional comparisons, the minimum target sample was increased to 424. To improve representation across multiple provinces, healthcare sectors and facility types, the final study sample was expanded to 1400 prescriptions. This larger sample enabled more stable estimates and meaningful comparisons across study settings.
Healthcare facilities were selected purposively to represent different geographical regions and types of outpatient prescribing environments within Sri Lanka. Within each facility, prescriptions were selected systematically, with every third eligible prescription included until the required numbers were obtained.
Prescription collection process
Prescription data were collected from outpatient prescriptions presented for dispensing at pharmacy points within the selected study settings. In government hospitals, prescriptions were obtained from OPD pharmacy counters after they had been issued by clinicians and presented by caregivers for medication dispensing. These prescriptions typically originated from paediatric clinics, general OPDs and primary care consultation units within the hospitals. In SPC outlets and private sector facilities, prescriptions were collected at the point of dispensing after patients presented prescriptions issued by registered medical practitioners. Data collectors reviewed prescriptions after dispensing and recorded relevant information using a standardised electronic data extraction form. No patient identifiers were retained, and prescriptions were not removed from the pharmacy workflow to avoid disruption of routine clinical services. The use of an electronic data collection system helped minimise transcription errors, ensured uniform data entry across study sites and facilitated secure storage of anonymised data for subsequent analysis.
Study variables
The primary outcome was prescription documentation quality, defined as the completeness and accuracy of key elements recommended in WHO good prescribing guidance.11 These included documentation of patient information (name, age, gender, address and weight), drug-related information (generic drug name, dose, dose unit, route, frequency and duration) and prescriber identifiers (name, qualifications, signature, Sri Lanka Medical Council (SLMC) registration number and contact details).
Prescription assessment and inter-rater reliability
The primary outcome of the study was prescription documentation quality, assessed through the completeness and accuracy of individual elements recommended in the WHO Guide to Good Prescribing.11 Prescription evaluation was carried out by a team of four assessors, comprising two senior pharmacists and two consultant paediatricians, each with a minimum of 10 years of clinical experience in their respective fields. All assessors had prior involvement in clinical audit or quality assurance initiatives and were familiar with the WHO guidelines for good prescribing practices.11
Because the study focused on evaluating specific components of prescription quality rather than generating a composite index, results are presented as the proportion of prescriptions with satisfactory documentation for each element rather than a single overall quality score. Independent variables included healthcare sector (public vs private), geographical region (administrative district), prescription format (handwritten vs typed), prescriber category (consultant paediatrician, general practitioner, family physician or unspecified government prescriber), inclusion of a diagnosis and whether antibiotics were prescribed. These variables were analysed to explore differences in prescription documentation quality across healthcare settings. Each element was assessed individually as either present and appropriate (‘satisfactory documentation’) or absent/incomplete.
Prior to data collection, all assessors participated in a calibration session to align their interpretations of WHO criteria and to reduce inter-rater variability. In cases where discrepancies arose between the two assessors’ evaluations, the prescription was discussed in a consensus meeting involving both reviewers. If disagreement persisted, a third senior assessor (either a pharmacist or paediatrician not involved in the original evaluation) was consulted to provide an independent judgement. This process ensured the reliability and consistency of the data used for analysis. Agreement between the two assessors was assessed during the pilot review of prescriptions using percentage agreement for key variables including dose accuracy, documentation of weight, drug name and dosing frequency. Overall agreement exceeded 90% for these variables, indicating high consistency between reviewers.
Definition of prescription accuracy and errors
Drug dose accuracy was assessed by comparing the prescribed dose with recommended paediatric dosing ranges in a standard reference source, the British National Formulary for Children. When patient weight was available, weight-based dosing recommendations were used to determine whether the prescribed dose fell within the acceptable therapeutic range. When weight was not documented, age-based dosing guidance was used as the reference standard. Doses falling outside the recommended range were categorised as either lower than recommended (underdose) or higher than recommended (overdose).
Documentation of dosing frequency was considered satisfactory when the frequency was explicitly stated using standard terms (eg, once daily, two times per day, three times per day or equivalent hourly intervals). Missing frequency or ambiguous expressions were categorised as errors. Route of administration was considered adequately documented when the route (eg, oral, intravenous, topical) was explicitly specified. If the route was omitted or unclear, it was classified as not satisfactorily documented.
Drug names were evaluated based on the completeness of documentation. Generic names were recorded as satisfactory when written in full or with widely recognised abbreviations. Trade names were recorded separately. Partial drug names or unclear abbreviations that could lead to ambiguity were classified as incomplete documentation.
Data analysis
Comparative analyses were conducted to examine differences in prescription documentation across healthcare sectors (public vs private), geographical regions and prescription formats (handwritten vs typed). The χ² test was used to assess associations between categorical variables. As multiple comparisons were performed across several prescription indicators, findings were interpreted cautiously and considered exploratory rather than confirmatory.
Ethical considerations
To ensure patient confidentiality, all prescription data were anonymised prior to analysis. Identifiable information such as patient or prescriber names, addresses and contact details was either removed or coded during data extraction, and no personal identifiers were retained in the final dataset. Access to raw data was restricted to authorised members of the research team, and all data were stored on password-protected devices. These procedures were in accordance with ethical standards outlined in the Declaration of Helsinki and adhered to institutional data protection guidelines to safeguard participant privacy.
Patient and public involvement
Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Results
The study included 23 healthcare facilities across six provinces, comprising both public (n=10) and private (n=13) sector institutions. These ranged from tertiary care hospitals to SPC and private dispensaries (table 1).
Table 1. Characteristics of sampled healthcare facilities (n=23).
| Province | Facility type | Sector | Facilities (n) | Estimated average of outpatient visits/day per facility |
|---|---|---|---|---|
| Western | TCH, SPC, PD | Public and private | 4 | 600 (TCH), 200 (SPC), 100 (PD) |
| North Central | TCH, PD | Public and private | 3 | 550 (TCH), 150 (PD) |
| Northern | TCH, SPC, PD | Public and private | 3 | 500 (TCH), 180 (SPC), 200 (PD) |
| Central | TCH, PD | Public and private | 3 | 700 (TCH), 120 (PD) |
| Southern | TCH, SPC, PD | Public and private | 3 | 600 (TCH), 160 (SPC), 110 (PD) |
| Eastern (Trincomalee) | SPC, PD | Public and private | 3 | 150 (SPC), 100 (PD) |
| Eastern (Batticaloa) | TCH, PD | Public and private | 4 | 500 (TCH), 130 (PD) |
PD, private dispensaries; SPC, state pharmaceutical corporation; TCH, tertiary care hospital.
All percentages reported in the results are calculated using the total number of prescriptions analysed (n=1400) as the denominator unless otherwise specified.
Out of the total number of prescriptions, 1325 (94.6%) were handwritten, while 75 (5.4%) were typed. In terms of sector distribution, 657 (46.92%) prescriptions were from government healthcare facilities, and 743 (53.08%) were from private sector establishments.
Patient information
The inclusion of patient details varied significantly across prescriptions. Name and initials were included in 703 (50.2%) prescriptions, while 666 (47.6%) included only the first name, and 30 (2.1%) had no name mentioned at all. Age was documented in 1331 (95.1%) prescriptions. Address details were recorded in only 428 (30.6%) cases. Gender was mentioned in only 713 (50.9%) prescriptions.
Diagnosis-related information
A diagnosis was documented in 504 (36.0%) prescriptions, consistent with the data presented in table 2.
Table 2. Patterns of prescription characteristics across the seven districts of Sri Lanka.
| Prescription characteristic | Satisfactory documentation | Total 1400 |
|||||
|---|---|---|---|---|---|---|---|
| Anuradhapura n=171 |
Galle n=206 |
Batticaloa and Trincomalee n=418 |
Ragama n=203 |
Kandy n=201 |
Jaffna n=201 |
||
| Patient information (%) | |||||||
| Name | 59 (34.5) | 31 (15.0) | 303 (72.5) | 70 (34.5) | 45 (22.4) | 196 (97.5) | 704 (50.3) |
| Age | 167 (97.7) | 188 (91.2) | 399 (95.45) | 191 (94.08) | 191 (95.0) | 195 (97.0) | 1331 (95.1) |
| Address | 3 (1.8) | 5 (2.4) | 256 (61.24) | 10 (4.92) | 3 (1.5) | 151 (75.1) | 428 (30.6) |
| Gender | 24 (14.0) | 73 (35.4) | 330 (78.94) | 134 (66.0) | 49 (24.37) | 103 (51.2) | 713 (50.9) |
| Weight | 23 (13.5) | 50 (24.3) | 131 (31.3) | 117 (57.6) | 145 (72.1) | 136 (67.7) | 602 (43.0) |
| Diagnosis-related information (%) | |||||||
| Diagnosis | 42 (24.5) | 51 (24.7) | 279 (66.7) | 46 (22.7) | 11 (5.5) | 75 (37.3) | 504 (36.0) |
| Drug-related information (%) | |||||||
| Documentation of the generic name | 83 (47.7) | 86 (41.7) | 250 (59.8) | 53 (26.1) | 23 (11.4) | 88 (43.78) | 583 (41.6) |
| Documentation of the complete name | 109 (63.7) | 139 (67.5) | 166 (39.7) | 150 (73.9) | 182 (90.5) | 142 (70.6) | 888 (63.4) |
| Documentation of dose units in milligrams (mg) | 85 (49.7) | 92 (44.7) | 91 (21.8) | 78 (38.4) | 71 (35.3) | 15 (7.5) | 432 (30.9) |
| Accurate dose | 169 (98.8) | 188 (91.3) | 281 (67.2) | 186 (91.6) | 196 (97.5) | 172 (85.6) | 1192 (85.1) |
| Documentation of the duration | 162 (94.7) | 194 (94.2) | 283 (67.7) | 196 (99.5) | 179 (89.1) | 175 (87.1) | 1189 (84.9) |
| Documentation of the frequency | 164 (95.9) | 197 (95.6) | 297 (71.1) | 184 (90.6) | 195 (97.0) | 184 (91.5) | 1221 (87.2) |
| Documentation of the route of administration | 71 (41.5) | 181 (87.9) | 279 (66.7) | 79 (38.9) | 35 (17.4) | 152 (75.6) | 797 (56.9) |
| Prescriber information (%) | |||||||
| Documentation of the academic qualifications | 56 (32.7) | 143 (69.4) | 224 (53.6) | 132 (65.0) | 182 (90.5) | 72 (35.8) | 809 (57.8) |
| Documentation of the SLMC registration number | 11 (6.4) | 146 (70.9) | 235 (56.22) | 78 (38.4) | 14 (7.0) | 71 (35.3) | 555 (39.6) |
| Signature | 166 (97.1) | 138 (67.0) | 365 (87.3) | 191 (94.1) | 179 (89.1) | 180 (89.6) | 1219 (87.1) |
SLMC, Sri Lanka Medical Council.
Drug-related information
In terms of medication details, the name of the drug was documented using generic names in 583 (41.6%) prescriptions and trade names in 817 (58.4%). The quality of documentation varied: abbreviations were used in 209 (14.9%), full names in 888 (63.4%) and partial names in 303 (21.6%). The dose unit was not specified in 282 (20.1%) prescriptions and was mentioned in milligrams (mg) in 432 (30.9%) and in millilitres (mL) in 686 (49%). The weight of the patient, which is often critical for accurate dosing, was omitted in 798 (57%) prescriptions. The dose was accurate in 1192 (85.1%) cases, but 85 (6.1%) prescriptions recorded doses that were too high, and 123 (8.8%) had doses that were too low. The duration of medication was specified in only 1289 (92.1%) prescriptions. The frequency of medication was correctly documented in 1221 (87.2%) cases, omitted in 98 (7%) and mentioned with errors in 81 (5.8%). The route of administration was not specifically mentioned in 603 (43.1%) prescriptions. An antibiotic was prescribed in 801 (57.2%) prescriptions.
Prescriber information
Prescriber details revealed that 297 (21.2%) prescriptions were from consultant paediatricians, 736 (52.6%) from general practitioners, 30 (2.1%) from consultant family physicians and 337 (24.1%) from government hospital facilities. Academic qualifications were mentioned in only 809 (57.8%) cases. The SLMC registration number was included in only 555 (39.6%) prescriptions. Prescriber signatures were present in only 1219 (87.1%) prescriptions. Table 2 shows the patterns of prescription characteristics across the seven districts of the country.
Government sector prescriptions demonstrated significantly better documentation of patient identifiers (name, address, gender) and diagnosis, as well as greater use of generic drug names (all p<0.00001) (table 3). In contrast, private sector prescriptions more frequently documented weight, complete drug names, accurate dosing, duration and frequency of administration (all p<0.00001). Antibiotics were prescribed more commonly in the government sector than the private sector (66.8% vs 48.7%, p<0.00001). Prescriber credentials, including academic qualifications and SLMC registration numbers, were substantially better documented in the private sector, while signatures did not differ significantly between sectors.
Table 3. Prescription practices in government and private sector healthcare settings (p<0.05 is considered statistically significant).
| Prescription characteristic | Satisfactory documentation | χ² | P value | |
|---|---|---|---|---|
| Government sector (n=657) | Private sector (n=743) | |||
| Patient information (%) | ||||
| Name with initials | 428 (65.1) | 276 (37.1) | 118.0 | <0.00001 |
| Age | 629 (95.7) | 702 (94.4) | 1.2 | 0.278 |
| Address | 314 (47.8) | 114 (15.3) | 172.9 | <0.00001 |
| Gender | 377 (57.4) | 336 (45.2) | 20.6 | <0.00001 |
| Weight | 223 (33.9) | 379 (51.0) | 41.4 | <0.00001 |
| Diagnosis-related information (%) | ||||
| Diagnosis | 329 (50.1) | 175 (23.6) | 106.5 | <0.00001 |
| Drug-related information (%) | ||||
| Documentation of the generic name | 385 (58.6) | 198 (26.6) | 146.5 | <0.00001 |
| Documentation of the complete name | 319 (48.6) | 569 (76.6) | 132.4 | <0.00001 |
| Documentation of dose units in milligrams (mg) | 177 (26.9) | 255 (34.3) | 102.2 | <0.00001 |
| Accurate dose | 502 (76.4) | 690 (92.9) | 84.9 | <0.00001 |
| Documentation of the duration | 567 (86.3) | 729 (98.1) | 41.8 | <0.00001 |
| Documentation of the frequency | 524 (79.7) | 697 (93.8) | 63.3 | <0.00001 |
| Documentation of the route of administration | 414 (63.0) | 383 (51.5) | 18.7 | 0.000015 |
| Antibiotic prescription | 439 (66.8) | 362 (48.7) | 46.5 | <0.00001 |
| Prescriber information (%) | ||||
| Documentation of the academic qualifications | 248 (37.7) | 561 (75.5) | 203.8 | <0.00001 |
| Documentation of the SLMC registration number | 232 (35.3) | 323 (43.5) | 9.7 | 0.001839 |
| Signature | 583 (88.7) | 636 (85.6) | 3.1 | 0.080737 |
SLMC, Sri Lanka Medical Council.
Typed prescriptions demonstrated significantly better documentation of key patient identifiers, including name, address, gender and weight, compared with handwritten prescriptions (all p≤0.003), while age and diagnosis documentation did not differ significantly between formats (table 4). Documentation of complete drug names, treatment duration and route of administration was also higher in typed prescriptions (p≤0.01). In contrast, handwritten prescriptions more frequently documented generic drug names (43.1% vs 16.0%, p<0.001), with no significant differences observed in accurate dosing or dosing frequency. Prescriber-related details, including academic qualifications and SLMC registration numbers, were substantially better documented in typed prescriptions, while signature documentation was comparable between formats.
Table 4. Comparison of prescription characteristics of handwritten and typed prescriptions (p<0.05 is considered statistically significant).
| Prescription characteristic | Satisfactory documentation | χ² | P value | |
|---|---|---|---|---|
| Handwritten prescriptions (n=1325) | Typed prescriptions (n=75) | |||
| Patient information (%) | ||||
| Name | 640 (48.3) | 64 (85.3) | 39.1 | <0.001 |
| Age | 1258 (94.9) | 73 (97.3) | 0.9 | 0.352 |
| Address | 411 (31.0) | 67 (89.3) | 12.3 | 0.001 |
| Gender | 647 (48.8) | 66 (88.0) | 43.6 | <0.001 |
| Weight | 574 (43.3) | 58 (77.3) | 1.0 | 0.003 |
| Diagnosis-related information (%) | ||||
| Diagnosis | 475 (35.8) | 29 (38.7) | 0.2 | 0.620 |
| Drug-related information (%) | ||||
| Documentation of the generic name | 571 (43.1) | 12 (16.0) | 21.4 | <0.001 |
| Documentation of the complete name | 832 (62.8) | 56 (74.7) | 4.4 | 0.010 |
| Documentation of dose units in milligrams (mg) | 401 (30.3) | 31 (41.3) | 5.8 | 0.054 |
| Accurate dose | 1134 (85.6) | 68 (89.3) | 4.2 | 0.119 |
| Documentation of the duration | 1137 (85.8) | 72 (95.3) | 15.1 | <0.001 |
| Documentation of the frequency | 1165 (87.9) | 65 (87.0) | 1.1 | 0.524 |
| Documentation of the route of administration | 757 (57.1) | 60 (80.0) | 14.4 | 0.005 |
| Prescriber information (%) | ||||
| Documentation of the academic qualifications | 764 (57.7) | 65 (87.0) | 13.2 | 0.006 |
| Documentation of the SLMC registration number | 523 (39.5) | 62 (82.7) | 19.3 | <0.001 |
| Signature | 1155 (87.1) | 64 (85.3) | 0.2 | 0.644 |
SLMC, Sri Lanka Medical Council.
Discussion
This study highlights significant patterns and gaps in paediatric prescription practices across diverse healthcare settings in Sri Lanka. Key findings include inadequate adherence to WHO prescription guidelines, suboptimal documentation of critical patient and drug information, and a notable reliance on antibiotics. These trends mirror findings in other LMICs such as India and Nigeria, where similar deficiencies in paediatric prescription practices have been documented, including poor documentation of drug regimens, inappropriate dosing and excessive use of antibiotics.5 6
Regional variations in prescription documentation observed in this study may reflect several contextual factors. Differences in local regulatory practices, variations in clinical workload and the availability of typed or electronic prescription systems could potentially influence documentation practice.12 13 However, these explanations remain speculative, as the present study did not directly measure these factors. Future research should explore how organisational, technological and workforce factors influence prescription quality across regions.
The inclusion of patient demographics such as age and gender was relatively high, but documentation of essential details like addresses and weights, critical for accurate dosing, was inconsistent. These findings are consistent with studies from countries such as Saudi Arabia, Ethiopia and Nepal, where the omission of such information was a common issue in outpatient settings.14 15 In Sri Lanka, the documentation of weight in only 57% of prescriptions is particularly concerning, as weight-based dosing is essential in paediatric care to avoid underdosing or overdosing.
Antibiotics were included in 57.2% of prescriptions analysed. While this proportion is relatively high compared with some reports from high-income countries where antimicrobial stewardship programmes are well established, the present study did not assess clinical indications, disease severity or appropriateness of prescribing.16 17 Therefore, these findings should be interpreted cautiously. Rather than demonstrating inappropriate prescribing, the observed level of antibiotic use highlights an area that warrants further investigation, particularly regarding prescribing indications and adherence to clinical guidelines.
The comparison between public and private sector prescriptions revealed a multidimensional pattern rather than a consistent advantage for one sector. Government sector prescriptions more frequently documented patient identifiers, addresses, diagnoses and generic drug names, reflecting stronger adherence to certain documentation standards. In contrast, private sector prescriptions more often included accurate dosing, clearer documentation of treatment duration and frequency and more complete prescriber credentials. These findings suggest that different strengths and weaknesses exist across sectors, highlighting opportunities for targeted improvements in both settings rather than indicating overall superiority of one sector.
Regarding handwritten versus typed prescriptions, typed prescriptions demonstrated superior legibility and comprehensive documentation, as highlighted by this study and corroborated by international findings.18 Studies from other regions have consistently shown that typed prescriptions reduce medication errors and improve clarity, facilitating better patient outcomes.19 20 However, the limited adoption of typed prescriptions (5.4% in this study) reflects infrastructural and training constraints in Sri Lanka’s healthcare system, particularly in resource-limited settings. The differences observed suggest that interventions tailored to sector-specific needs—such as mandating typed prescriptions in the private sector and enhancing regulatory enforcement in the public sector—could bridge these gaps. Future policies should promote digital prescription systems, ensuring uniformity and compliance across both healthcare domains. These measures are imperative for improving paediatric care quality and minimising medication-related errors in Sri Lanka.
This study highlights critical gaps in paediatric prescription practices in Sri Lanka, including incomplete documentation, inconsistent adherence to WHO guidelines and potential overprescription of antibiotics. These findings align with international literature on prescription practices in LMICs, emphasising the universality of these challenges. However, these findings also highlight specific areas for improvement, including strengthening documentation standards and promoting the adoption of typed prescribing practices through enhanced health information technology across healthcare systems. Policy interventions should prioritise training and sensitisation of healthcare providers across both public and private sectors, incorporating international best practices in prescription writing. Future research should explore the clinical appropriateness of prescriptions and assess the impact of targeted interventions on prescribing quality. Addressing these challenges is essential for improving the safety and effectiveness of paediatric care in Sri Lanka.
The study has several limitations. First, the cross-sectional design provides a snapshot of prescription practices but does not capture longitudinal changes or seasonal variations. Second, data collection relied on prescriptions issued at healthcare facilities, potentially excluding informal or over-the-counter prescriptions, which are common in some rural settings. Third, while systematic sampling minimised selection bias, variations in documentation quality across regions might reflect regional differences in healthcare infrastructure rather than broader national trends. Although prescriptions were evaluated against WHO guidelines, this study did not assess clinical appropriateness—such as the validity of drug indications (including antibiotics) or therapeutic outcomes—which warrants further investigation. In addition, the study focused primarily on prescription documentation characteristics rather than the full WHO prescribing indicators. Therefore, variables such as the mean number of medicines per prescription and detailed classification of antibiotic types were not systematically analysed. Finally, multiple statistical comparisons were conducted across several prescription indicators without adjustment for multiple testing or multivariable modelling. Therefore, some statistically significant associations may reflect type I error, and the findings should be interpreted with caution.
Conclusion
This multiprovincial study suggests the presence of important gaps in paediatric prescription practices across selected healthcare facilities in Sri Lanka. Marked variations were observed across regions, healthcare sectors and prescription formats, with typed prescriptions showing superior documentation quality. These findings highlight areas that may benefit from strengthened regulatory oversight, prescriber training and wider adoption of digital prescribing systems to improve the safety and quality of paediatric prescribing practices in Sri Lanka.
Footnotes
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Data availability free text: All data were deidentified for this research report. The datasets used or analysed during the current study are available from the corresponding author on reasonable request.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by the Ethics Review Committee of the Postgraduate Institute of Medicine, University of Colombo (ERC/PGIM/2024/104). This study involved studying only the prescriptions made to children by health professionals. No direct or indirect contact with human participants was involved.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Map disclaimer: The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Data availability statement
Data are available upon reasonable request.
References
- 1.D’Errico S, Zanon M, Radaelli D, et al. Medication Errors in Pediatrics: Proposals to Improve the Quality and Safety of Care Through Clinical Risk Management. Front Med (Lausanne) 2021;8:814100. doi: 10.3389/fmed.2021.814100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lu H, Rosenbaum S. Developmental pharmacokinetics in pediatric populations. J Pediatr Pharmacol Ther. 2014;19:262–76. doi: 10.5863/1551-6776-19.4.262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Llor C, Bjerrum L. Antimicrobial resistance: risk associated with antibiotic overuse and initiatives to reduce the problem. Ther Adv Drug Saf. 2014;5:229–41. doi: 10.1177/2042098614554919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barbieri E, Minotti C, Cavagnis S, et al. Paediatric medicine issues and gaps from healthcare workers point of view: survey results and a narrative review from the global accelerator for paediatric formulations project. Front Pharmacol. 2023;14:1200848. doi: 10.3389/fphar.2023.1200848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Otaigbe II, Elikwu CJ. Drivers of inappropriate antibiotic use in low- and middle-income countries. JAC-Antimicrobial Resistance. 2023;5:dlad062. doi: 10.1093/jacamr/dlad062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sulis G, Daniels B, Kwan A, et al. Antibiotic overuse in the primary health care setting: a secondary data analysis of standardised patient studies from India, China and Kenya. BMJ Glob Health. 2020;5:e003393. doi: 10.1136/bmjgh-2020-003393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Velo GP, Minuz P. Medication errors: prescribing faults and prescription errors. Br J Clin Pharmacol. 2009;67:624–8. doi: 10.1111/j.1365-2125.2009.03425.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Galappatthy P, Ranasinghe P, Liyanage CK, et al. WHO/INRUD Core drug use indicators and commonly prescribed medicines: a National Survey from Sri Lanka. BMC Pharmacol Toxicol. 2021;22:67. doi: 10.1186/s40360-021-00535-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.World Health Organization Action program essential drugs. guide to good prescribing: a practical manual, World Health Organization, Geneva. 1994
- 10.Khalid Abdullahi A, Senire Fatima I, Abdurrahaman U, et al. Prescription Writing Guidelines in Ahmadu Bello University Teaching Hospital Zaria - Kaduna State, Nigeria. Innov Pharm. 2023 doi: 10.24926/iip.v14i1.5164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.World Health Organization . Guide to Good Prescribing: A Practical Manual. Geneva: World Health Organization; 1994. [Google Scholar]
- 12.Guan X, Ni B, Zhang J, et al. Association Between Physicians’ Workload and Prescribing Quality in One Tertiary Hospital in China. J Patient Saf. 2021;17:e1860–5. doi: 10.1097/PTS.0000000000000753. [DOI] [PubMed] [Google Scholar]
- 13.Erku D, Khatri R, Endalamaw A, et al. Digital Health Interventions to Improve Access to and Quality of Primary Health Care Services: A Scoping Review. Int J Environ Res Public Health. 2023;20:6854. doi: 10.3390/ijerph20196854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shrestha R, Prajapati S. Assessment of prescription pattern and prescription error in outpatient Department at Tertiary Care District Hospital, Central Nepal. J Pharm Policy Pract . 2019;12:16. doi: 10.1186/s40545-019-0177-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Idris SA, Hussien TMA, Al-Shammari FF, et al. An Evaluation of Drug Prescribing Patterns and Prescription Completeness. Healthcare (Basel) 2024;12:2221. doi: 10.3390/healthcare12222221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gres E, Diallo IS, Besnier C, et al. Antibiotic prescribing practices according to the AWaRe classification among children under 5 of age attending public primary care centres in four West African countries: a cross-sectional study (AIRE project, 2021-2022) BMJ Paediatr Open . 2024;8:e002833. doi: 10.1136/bmjpo-2024-002833. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Klatte JM. Pediatric Antimicrobial Stewardship Programs: Current Perspectives. Pediatric Health Med Ther. 2020;11:245–55. doi: 10.2147/PHMT.S224774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ariaga A, Balzan D, Falzon S, et al. A scoping review of legibility of hand-written prescriptions and drug-orders: the writing on the wall. Expert Rev Clin Pharmacol. 2023;16:617–21. doi: 10.1080/17512433.2023.2223972. [DOI] [PubMed] [Google Scholar]
- 19.Albarrak AI, Al Rashidi EA, Fatani RK, et al. Assessment of legibility and completeness of handwritten and electronic prescriptions. Saudi Pharm J. 2014;22:522–7. doi: 10.1016/j.jsps.2014.02.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ciapponi A, Fernandez Nievas SE, Seijo M, et al. Reducing medication errors for adults in hospital settings. Cochrane Database Syst Rev. 2021;11:CD009985. doi: 10.1002/14651858.CD009985.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]