Abstract
Background
Point-of-care testing (POCT) devices are a potential solution to improving access to faster and more accurate diagnosis of non-communicable diseases (NCDs), particularly in resource-limited settings. However, evidence about their implementation in low- and middle- income countries is limited.
Aim
To determine the potential utility of a co-created strategy for implementing multiparameter POCT devices to improve care for three NCDs at public primary healthcare centres in urban and rural areas in Tumbes, Peru.
Methods
This quasi-experimental cluster trial with a concurrent control was conducted in Tumbes, Peru. This study compared the current standard of care (conventional testing) to POCT in the screening and monitoring of certain NCDs (type 2 diabetes, kidney disease, and hypercholesterolemia) among patients at the primary healthcare level. The diagnosis was based on Peruvian clinical practice guidelines for NCDs. The intervention consisted of training for health professionals, implementing a POCT device in the laboratory for three months, advertising the population about the availability of this new device, and three months of follow-up care for participants. Relative risk (RR) was estimated.
Results
1,234 participants were recruited in eight facilities. 37.8% of participants received a new diagnosis. Using POCT alone significantly increased the diagnosis of new diabetes cases (adjusted relative risk [aRR]: 1.84) and high cholesterol detection (aRR: 1.24) compared to conventional testing. The combination of POCT plus conventional testing showed even greater increases in diabetes (aRR: 2.42) and high cholesterol diagnoses (aRR: 1.56) compared to control sites.
Conclusions
The use of POCT significantly increased the diagnosis of diabetes and high cholesterol compared to conventional laboratory testing. This suggests that POCT may be an effective tool for increased early detection of NCDs.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12913-025-13554-3.
Keywords: Point-of-care testing, Implementation science, Noncommunicable diseases, Peru
Background
Non-communicable diseases (NCDs) are a major public health challenge globally, accounting for 71% of all deaths worldwide, with low- and middle-income countries (LMICs) carrying a significant burden [1]. The World Health Organization estimates that NCDs will cause 52 million deaths per year by 2030, representing a significant threat to sustainable development [2]. In LMICs, the burden of NCDs includes challenges in detecting and managing these conditions due to poor infrastructure, limited resources, and lack of access to laboratory services [3].
Point-of-care testing (POCT) devices are a potential solution to improving access to faster and more accurate diagnosis, control and treatment of NCDs, particularly in resource-limited settings [4, 5]. Multiparameter POCT can detect multiple conditions simultaneously, such as diabetes and cardiovascular and kidney diseases, offering several advantages over single-parameter devices, including their portability, ease of use, reducing length of stay, and potential cost savings [6, 7].
Despite the potential benefits of multiparameter POCT for NCD detection and management, their implementation at the primary healthcare level faces several challenges, such as lack of regulatory approvals, high costs, and limited capacity for maintenance and quality control [8, 9]. Moreover, the evidence on the feasibility and sustainability of integrating these devices into public primary healthcare centres (PHCs) is limited in LMICs [10], as previous research was more focused on the validation process of these devices [10, 11].
Identifying barriers and facilitators in the use of a POCT devices can improve the implementation process [11]. Moreover, the co-creation of a process involving users, service providers and decision-makers offers the potential to develop efficient strategies for implementing the devices [12]. Therefore, the present study aimed to determine the potential utility of a co-created strategy for implementing multiparameter POCT devices to improve care (i.e., new diagnosis and control) for specific NCDs (type 2 diabetes, hypertension, hypercholesterolemia, and kidney disease) at public PHCs in urban and rural areas in the North of Peru.
Methods
Context
The current study was based on two previous phases before implementation. In the first phase, a scoping review was carried out to assess the literature related to implementation facilitators and barriers of POCT for the diagnosis or monitoring of cardiometabolic diseases [11]. Key findings were as follows: the most recognized facilitator was the quick turnaround of results and patients appreciated that obtaining the sample was simple and more comfortable if venipuncture was not used. However, implementation of POCT generated concerns in PHC professionals in terms of new training burdens, responsibility of being in charge of the device, technical abilities required to operate it, and staff turnover, all of which may could hamper the adoption or sustainability of the use of POCT devices [13].
In the second phase, a co-creation process was conducted with different stakeholders (patients, healthcare workers, and decision-makers) to develop an acceptable strategy to implement POCT devices at the PHC level in Tumbes, a semi-urban region in the North of Peru [14]. The following implementation strategies were used in this study: local consensus discussions; educational meetings; changes to the physical structure and equipment; ongoing training; distribution of educational materials; and reminders to physicians [15]. In the first workshop (October 2022), the participants created a “journey map” of the patient’s routine in the PHC and talked about barriers and facilitators to implement POCT device in the facilities. Their ideas, comments and topics were summarized and organized by the research team for the next meeting. In the second workshop (November 2022), participants developed an implementation strategy, selecting the most important topics (training modality, priority parameters, best location for POCT, modality advertising). The selected location for the POCT was the PHC laboratory, followed by the triage area and examination rooms. The most desired laboratory parameters were glucose, hemoglobin A1c (HbA1c), lipid profile, and urea/creatinine. The modalities of advertising to make the availability of the new POCT devices known to the community proposed were loudspeakers, flyers, and billboards located in the PHC, the municipality, and the town square, among others.
Study design and settings
This study aimed to compare the current standard of care to POCT in the screening and monitoring of certain NCDs in patients who received care at the PHC level. This was a quasi-experimental cluster trial with a concurrent control conducted in Tumbes, Peru (see Fig. 1). Tumbes is a tropical area with temperatures between 21 and 32° C [16]. The most recent national census (2017) showed that Tumbes had approximately 250,000 inhabitants. Tumbes had a prevalence greater than the national average of type 2 diabetes (10.3% vs. 7.0%), hypertension (26.9% vs. 19.7%), and obesity (31.6% vs. 26.9%) [17]; these differences may be due to the geographical and climatic diversity, in addition to the accelerated urbanization process in this Peruvian region [18–20]. The fieldwork was carried out between December 2023 and June 2024 in eight (of 43) public PHCs (four sites for the implementation of multiparameter POCT and four as control sites) in the region, divided by settings (rural/urban).
Fig. 1.
Map of Peru indicating the location of Tumbes
As part of the original protocol, a process and economic evaluation were also conducted; these results will be published separately later.
Participants and sampling
Eligible subjects were those aged ≥ 18 years who were habitual residents and users of healthcare public services. Participants were able to understand procedures and give informed consent.
A sample size of 1344 participants (672 per arm) was estimated, divided into 168 for each of the 8 cluster (4 intervention sites and 4 control sites) pursuant to a cluster-randomized trial design [21]. Assuming a proportion of 78.9% of patients screened in the past year for NCDs [22], with an expected increase of 10% post-intervention, a coefficient of variation of 0.03 based on unpublished data from a previous trial in Tumbes [23], a significance level of 5% and a power of 80%, a minimum of 168 participants per cluster was required.
The selection process of the PHCs used a non-randomized approach with the help of regional health decision makers, selecting those that had similar characteristics such as population size, degree of urbanization (urban or rural), number of healthcare providers, PHC subtype (I-2, I-3 or I-4, per Peruvian Ministry of Health’s classification of healthcare centres) with laboratory facilities, and access to healthcare services. Two of the four PHCs assigned to the intervention had previously participated in the co-creation phase [14]. Some laboratories had equipment not enabled for one (i.e. HbA1c or creatinine) of the parameters evaluated, which reduced the number of results from some tests evaluated in several facilities.
Outcomes and covariates
The variables of interest were the number of diagnoses and the number of participants who achieved treatment goals. Diagnoses were based on Peruvian clinical practice guidelines for NCDs and other international guidelines when the former did not have specific information. Diagnosis of type 2 diabetes was based on the Peruvian guideline and on the American Diabetes Association criteria (fasting plasma glucose ≥ 126 mg/dL (7.0 mmol/L); HbA1c ≥ 6.5% (48 mmol/mol)) [24, 25]; hypercholesterolemia was defined as total cholesterol ≥ 200 mg/dL (5.18 mmol/L) [26, 27]; and kidney disease as creatinine > 1.1 mg/dL (97.24 µmol/L) and > 1.3 mg/dL (114.92 µmol/L) for women and men, respectively [28, 29]. The “any assessed NCDs” category was defined as the presence of none (No), or one or more (Yes) newly diagnosed NCDs. Treatment goals (controlled parameters) in NCDs were defined as: fasting plasma glucose < 130 mg/dL; HbA1c < 7% (53 mmol/mol) [30]; total cholesterol < 200 mg/dL [14] and, creatinine ≤ 1.1 mg/dL and ≤ 1.3 mg/dL for women and men, respectively [15].
Sociodemographic, economic, and health variables included were gender (male or female); age; partnership status (with [married or cohabiting] or without [single, divorced or widowed] current partner); educational level (primary or less [< 7 years], secondary [7–11 years] or higher [12 + years]); current employment status (having or not having a job); having public or private health insurance (yes vs. no); and prior diagnosis of certain NCDs (type 2 diabetes, arterial hypertension, hypercholesterolemia, and/or chronic kidney disease).
Intervention
The intervention consisted of:
A full-day training of PHC professionals (physicians, nurses, midwives, and dentists) in the prevention, diagnosis, and treatment of NCDs, took place in the intervention and control sites in November 2023. A half-day training for technicians, other PHC workers, and community health workers, focussed on identifying risk factors for NCDs also occurred. See Procedures section below.
A three-hour training of laboratory workers from the intervention sites in the use of POCT and preparation for the use of the POCT took place in each PHC laboratory in December 2023.
Billboards were placed in strategic locations outside and inside of the PHCs with information about the POCT’s availability.
The POCT SimplexTAS™ 101 (TASCOM, Korea) was selected for this study for its ability to collect the chosen parameters/indicators (glucose, HbA1c, total cholesterol and creatinine), as well as the devices’s ability to operate at a temperature range of 18 to 32 °C, with a function to operate at temperatures up to 40 °C; and for its use of one drop of venous blood or less (between 2 and 20 µL) by fingerstick for each parameter. It delivers results in 13 min for any parameter(s) analysed up to 4 simultaneously [31]. This device is self-calibrating and underwent quality control checks according to the manufacturer’s recommendations during the study. This device could electronically record participants’ results, which could be compared with those collected manually by fieldworkers. This multiparameter POCT device was used when the treating physician ordered any of the chosen tests for participants. Conventional laboratory tests were also applied to all participants (both intervention and control sites) because the selected POCT was not registered for use in the Sanitary Registration of Medical Devices of the General Directorate of Medicines, Supplies and Drugs (DIGEMID) for use in Peru at the start of the study; therefore, any patient result generated with the POCT needed to be accompanied with a test result from a laboratory device approved for medical use. POCT results (available in 13 to 15 min) were compared with conventional test results (available eight to 24 h after sampling). POCT results were used for clinical diagnosis and treatment decisions during the medical consultation. Patients were informed that they may be recontacted if there was a discrepancy greater than 10% between the POCT and the conventional tests as management readjustment could have been needed.
Procedures
Fieldworkers were selected from health technicians or professionals with experience working in PHC and/or conducting field research. Eight fieldworkers (one per site) were trained in participant recruitment and data collection. The latter was performed using tablets with a mobile application based on Research Electronic Data Capture (REDCap).
Healthcare workers (36 health professionals [physicians, nurses, dentists and obstetricians]; 49 technicians [laboratory or nursing] and other workers [administrators or nursing interns]; and 78 community health workers) in the intervention and control sites were trained to follow guidelines for basic NCD management for type 2 diabetes, arterial hypertension, hypercholesterolemia, and chronic kidney disease. Community health workers were also trained on the most common risk factors for NCDs. NCD training was provided by clinical physicians specializing in internal medicine. Laboratory workers in the intervention PHCs were trained to appropriately use the multiparameter POCT device. The training about the multiparameter POCT included theoretical and practical lessons and was conducted in a single day by staff trained by the manufacturer. The manufacturer’s user manual was translated into Spanish and provided to laboratory workers.
All the PHCs in the intervention arm received a multiparameter POCT device and supplies to carry out tests (glucose, HbA1c, total cholesterol and creatinine) over a course of two months. A three-month guided intervention was conducted, followed by a three-month period for patient follow-up. Control facilities followed the default standard of care. During the three-month follow-up period, the POCT continued to be used by the PHCs with a minimal intervention of the research team, providing consumables when required. During the last month of follow-up, calls to participants and the review of their medical records was performed to confirm the reported diagnoses and possible unrecorded follow-ups of participants.
Potential participants were invited by fieldworkers to join the study when they were in the PHC waiting rooms and if the treating physician had previously requested at least one of the four laboratory parameters evaluated (for screening or follow-up). Participants were explained the study aims and procedures before signing an informed consent. In the intervention site, participants were informed that, in addition to their regular blood test, a capillary blood sample (one drop per parameter by fingerstick) would be taken for processing with the POCT device and that these results would be available in 20 min. Demographic data was collected from those who agreed to participate. The study fieldworkers conducted follow-ups when participants returned to the PHCs for a new medical consultation related to NCDs.
Statistical analysis
Participant characteristics were described using relative and absolute proportions for categorical variables and mean and standard deviation for continuous variables. The intervention group was evaluated in two ways: POCT alone and POCT combined with conventional (laboratory) tests. The performance of the POCT device results was assessed via comparison with conventional equipment; a difference of less than 10% was considered satisfactory from an analytical perspective [32, 33].
Rates of new diagnoses for the intervention group was compared with the control group for total participants and for degree of urbanization using Pearson’s Chi-squared test. Crude and adjusted risk ratios (cRR and aRR) were performed using Poisson regression models with robust variance. Test results at baseline and follow-up were presented using means and standard deviations. Means difference was evaluated using Student’s t test for paired data in participants who had both results (baseline and follow-up tests). Analyses were performed using STATA 17.0 for Windows (StataCorp, College Station, TX, US).
Results
A total of 1,241 participants were initially enrolled, though seven of these were excluded for not having any laboratory test results. Of the remaining participants (1,234), 605 were recruited in the four facilities in the intervention arm, and 629 in the four facilities in the control arm. The mean age of all participants was 52.6 (SD: 16.7) years and the majority were women (74.6%, 921). 69.9% (863) of participants were unemployed, and 91.0% (1123) reported having health insurance (public or private). Around a quarter of participants had a history of diabetes (25.3%, 312) and one fifth, a history of hypertension (19.9%, 245). Characteristics by arm are shown in Table 1.
Table 1.
Socio-demographic characteristics and non-communicable disease history of participants, split by intervention and control arm at baseline
| Characteristics | Total | Control | Intervention | |
|---|---|---|---|---|
| (N = 1,234) | (N = 629) | (N = 605) | ||
| Gender | ||||
| Female | 921 (74.6%) | 478 (76.0%) | 443 (73.2%) | 0.26 |
| Age (mean ± SD) | 52.6 ± 16.7 | 53.5 ± 16.6 | 51.5 ± 16.7 | 0.04 |
| Partnership status | ||||
| Without partner | 446 (36.1%) | 242 (38.5%) | 204 (33.7%) | 0.08 |
| With partner | 788 (63.9%) | 387 (61.5%) | 401 (66.3%) | |
| Educational level | ||||
| Primary or less | 483 (39.1%) | 251 (39.9%) | 232 (38.3%) | 0.13 |
| Secondary | 562 (45.5%) | 288 (45.8%) | 274 (45.3%) | |
| Higher | 184 (14.9%) | 90 (14.3%) | 94 (15.5%) | |
| DK / NR | 5 (0.4%) | 0 (0.0%) | 5 (0.8%) | |
| Currently employed | ||||
| Yes | 362 (29.3%) | 181 (28.8%) | 181 (29.9%) | 0.87 |
| No | 863 (69.9%) | 443 (70.4%) | 420 (69.4%) | |
| DK / NR | 9 (0.7%) | 5 (0.8%) | 4 (0.7%) | |
| Monthly income | ||||
| Don’t receive income | 754 (61.1%) | 381 (60.6%) | 373 (61.7%) | 0.79 |
| < 1000 PEN (~ 270USD) | 330 (26.7%) | 175 (27.8%) | 155 (25.6%) | |
| ≥ 1000 PEN | 103 (8.3%) | 51 (8.1%) | 52 (8.6%) | |
| DK / NR | 47 (3.8%) | 22 (3.5%) | 25 (4.1%) | |
| Health insurance | ||||
| Yes | 1123 (91.0%) | 581 (92.4%) | 542 (89.6%) | 0.20 |
| No | 31 (2.5%) | 12 (1.9%) | 19 (3.1%) | |
| DK / NR | 80 (6.5%) | 36 (5.7%) | 44 (7.3%) | |
| Type 2 diabetes | ||||
| Yes | 312 (25.3%) | 178 (28.3%) | 134 (22.1%) | < 0.001 |
| No | 803 (65.1%) | 427 (67.9%) | 376 (62.1%) | |
| DK / NR | 119 (9.6%) | 24 (3.8%) | 95 (15.7%) | |
| Arterial hypertension | ||||
| Yes | 245 (19.9%) | 148 (23.5%) | 97 (16.0%) | < 0.001 |
| No | 890 (72.1%) | 469 (74.6%) | 421 (69.6%) | |
| DK / NR | 99 (8.0%) | 12 (1.9%) | 87 (14.4%) | |
| Hypercholesterolemia | ||||
| Yes | 250 (20.3%) | 142 (22.6%) | 108 (17.9%) | < 0.001 |
| No | 794 (64.3%) | 445 (70.7%) | 349 (57.7%) | |
| DK / NR | 190 (15.4%) | 42 (6.7%) | 148 (24.5%) | |
| Chronic kidney disease | ||||
| Yes | 6 (0.5%) | 5 (0.8%) | 1 (0.2%) | < 0.001 |
| No | 1178 (95.5%) | 613 (97.5%) | 565 (93.4%) | |
| DK / NR | 50 (4.1%) | 11 (1.7%) | 39 (6.4%) |
DK / NR: Don’t know / No response. SD: Standard deviation. PEN: Peruvian currency (soles; 1USD ≈ 3.75 PEN)
Overall, during the study period, 470 (37.8%) participants received a new diagnosis of a relevant condition of interest. Using POCT alone or POCT combined with conventional testing, the intervention facilities identified statistically significant more new cases for diabetes (p = 0.01 and p < 0.001, respectively) and high cholesterol (p = 0.02 and p < 0.001, respectively) as well as any assessed NCD (p = 0.02 and p < 0.001, respectively) than control facilities. However, there was no significant difference in the diagnosis of kidney disease (p = 0.11 and p = 0.82) between the study arms. In rural settings, POCT in combination with conventional tests detected more new cases of diabetes (p = 0.03), high cholesterol (p = 0.01), as well as the total of any assessed NCD (p < 0.001), compared with the control. In urban settings, POCT combined with a conventional test detected more cases of diabetes (p < 0.001), high cholesterol (p < 0.001) and any assessed NCD (p = 0.01). More details in Table 2.
Table 2.
New diagnoses of non-communicable diseases
| New Diagnosis of: | Total* | Control (CT) |
Intervention (POCT) |
P † | Intervention (CT + POCT) |
P ‡ |
|---|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | n (%) | |||
| Total ( N = 1,234) | ||||||
| Diabetes ( n = 878) | ||||||
| No | 769 (87.59%) | 401 (92.40%) | 385 (86.71%) | 0.01 | 368 (82.88%) | < 0.001 |
| Yes (Glu + HbA1c) | 109 (12.41%) | 33 (7.60%) | 59 (13.29%) | 76 (17.12%) | ||
| High cholesterol (n = 813) | ||||||
| No | 457 (56.21%) | 273 (66.10%) | 232 (58.00%) | 0.02 | 184 (46.00%) | < 0.001 |
| Yes | 356 (43.79%) | 140 (33.90%) | 168 (42.00%) | 216 (54.00%) | ||
| Kidney disease (n = 326) | ||||||
| No | 263 (80.67%) | 167 (80.29%) | 103 (87.29%) | 0.11 | 96 (81.36%) | 0.82 |
| Yes | 63 (19.33%) | 41 (19.71%) | 15 (12.71%) | 22 (18.64%) | ||
| Any assessed NCD§ (n = 1,233) | ||||||
| No | 763 (61.88%) | 438 (69.75%) | 383 (63.31%) | 0.02 | 325 (53.72%) | < 0.001 |
| Yes | 470 (38.12%) | 190 (30.25%) | 222 (36.69%) | 280 (46.28%) | ||
| Rural ( n = 408) | ||||||
| Diabetes ( n = 307) | ||||||
| No | 272 (88.6%) | 96 (94.12%) | 180 (87.80%) | 0.08 | 176 (85.85%) | 0.03 |
| Yes (Glu + HbA1c) | 35 (11.4%) | 6 (5.88%) | 25 (12.20%) | 29 (14.15%) | ||
| High cholesterol ( n = 283) | ||||||
| No | 132 (46.64%) | 44 (59.46%) | 118 (56.46%) | 0.65 | 88 (42.11%) | 0.01 |
| Yes | 151 (53.36%) | 30 (40.54%) | 91 (43.54%) | 121 (57.89%) | ||
| Kidney disease ( n = 47) | ||||||
| No | 38 (80.85%) | 19 (76.00%) | 19 (86.36%) | 0.37 | 19 (86.36%) | 0.37 |
| Yes | 9 (19.15%) | 6 (24.00%) | 3 (13.64%) | 3 (13.64%) | ||
| Any assessed NCD § ( n = 406) | ||||||
| No | 228 (56.16%) | 112 (73.68%) | 144 (56.69%) | 0.001 | 116 (45.67%) | < 0.001 |
| Yes | 178 (43.84%) | 40 (26.32%) | 110 (43.31%) | 138 (54.33%) | ||
| Urban ( n = 835) | ||||||
| Diabetes ( n = 571) | ||||||
| No | 497 (87.04%) | 305 (91.87%) | 205 (85.77%) | 0.02 | 192 (80.33%) | < 0.001 |
| Yes (Glu + HbA1c) | 74 (12.96%) | 27 (8.13%) | 34 (14.23%) | 47 (19.67%) | ||
| High cholesterol ( n = 530) | ||||||
| No | 325 (61.32%) | 229 (67.55%) | 114 (59.69%) | 0.07 | 96 (50.26%) | < 0.001 |
| Yes | 205 (38.68%) | 110 (32.45%) | 77 (40.31%) | 95 (49.74%) | ||
| Kidney disease ( n = 279) | ||||||
| No | 225 (80.65%) | 148 (80.87%) | 84 (87.50%) | 0.16 | 77 (80.21%) | 0.89 |
| Yes | 54 (19.35%) | 35 (19.13%) | 12 (12.50%) | 19 (19.79%) | ||
| Any assessed NCD § ( n = 827) | ||||||
| No | 535 (64.69%) | 326 (68.49%) | 239 (68.09%) | 0.90 | 209 (59.54%) | 0.01 |
| Yes | 292 (35.31%) | 150 (31.51%) | 112 (31.91%) | 142 (40.46%) | ||
CT: Conventional laboratory test, POCT: Point-of-care testing devices, Glu: Glucose
HbA1c: Hemoglobin A1c
* Considering control and Intervention of CT + POCT
† p-value of Intervention (POCT) vs. Control (CT)
‡ p-value of Intrvention (CT+POCT) vs. Control (CT)
§ Any assessed NCD: Total number of patients with none (“No”) or one or more (“Yes”) evaluated NCDs newly diagnosed. One participant was not considered because he/she had history of all three diagnoses evaluated
In the intervention sites that used POCT results alone, there was a significant increase in new diabetes cases (aRR: 1.84, 95% confidence interval [CI]: 1.22–2.78) and the detection of high cholesterol (aRR: 1.24, 95% CI: 1.02–1.50) compared to conventional testing. When the combination of POCT plus conventional testing was assessed, the intervention sites showed even greater increases in new diabetes cases (aRR: 2.42, 95% CI: 1.64–3.58) and high cholesterol diagnoses (aRR: 1.56, 95% CI: 1.31–1.85) compared to control sites. Additionally, the diagnosis of any relevant NCD was significantly higher using POCT alone (aRR: 1.19, 95% CI: 1.02–1.41) or in combination with conventional testing (aRR: 1.51, 95% CI: 1.30–1.75). However, no significant differences were observed for kidney disease diagnoses across the different strategies. Details in Table 3.
Table 3.
Effect of the intervention in the diagnosis of non-communicable diseases
| Strategy | Diabetes | High cholesterol | Kidney disease | Any assessed non-communicable disease | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Crude model cRR (95%CI) |
Adjusted model aRR (95%CI) |
Crude model cRR (95%CI) |
Adjusted model aRR (95%CI) |
Crude model cRR (95%CI) |
Adjusted model aRR (95%CI) |
Crude model cRR (95%CI) |
Adjusted model aRR (95%CI) |
||||
| POCT vs. Conventional test | |||||||||||
| Intervention |
1.75 (1.17–2.62) |
1.84 (1.22–2.78) |
1.24 (1.04–1.48) |
1.24 (1.02–1.50) |
0.64 (0.37–1.11) |
0.60 (0.34–1.07) |
1.21 (1.04–1.42) |
1.19 (1.02–1.41) |
|||
| POCT + Conventional test vs. Control | |||||||||||
| Intervention |
2.25 (1.53–3.31) |
2.42 (1.64–3.58) |
1.59 (1.35–1.87) |
1.56 (1.31–1.85) |
0.95 (0.59–1.51) |
0.92 (0.56–1.49) |
1.53 (1.32–1.77) |
1.51 (1.30–1.75) |
|||
RR: Risk ratio; POCT: Point-of-care Testing; 95%CI: 95% confidence interval. Models adjusted by urbanism, age, gender, occupation, partnership status and education level
At baseline, 44.0% vs. 58.2% of glucose values, 59.5% vs. 29.0% of HbA1c values, and 57.9% vs. 37.3% of total cholesterol values were in non-pathological ranges in the control and intervention sites, respectively. The percentages of controlled creatinine did not differ between the arms (77.8% vs. 75.0%) (see Table S1. A total of 251 (20.2%) participants returned to the PHCs for follow-up visits, with 193 out of 629 in the control arm and 58 out of 605 in the intervention arm, showing a statistically significant difference (p < 0.001). Providers ordered laboratory test for only 210 (83.6%) of these 251. The mean results for the tests at baseline and in the follow-up visits can be found in Table S2 (see supplementary material). The mean of the baseline and follow-up test results in participants who returned to the PHCs for follow-up showed significant differences in total cholesterol (p = 0.02) in control sites, and in the intervention arm in glucose using conventional test (p = 0.04) and total cholesterol using POCT devices (p = 0.02) (Table S3).
Discussion
Main findings
Using POCT in comparison with conventional laboratory tests increased the detection of NCDs in urban and rural areas in the North of Peru. Our findings indicate that POCT implementation significantly increased the detection of new cases of diabetes and high cholesterol, but did not increase diagnoses of kidney disease. Doctors decided which tests to order for their patients; the creatinine test was probably only ordered for those with a longer illness duration and not for all those diagnosed with diabetes or hypertension. This small number of cases without much variability could explain the lack of statistical difference for kidney disease diagnosis. Furthermore, the population has little knowledge about chronic diseases, and they may believe that their initial test results were sufficient [34]. Moreover, the increase in the detection of diabetes and high cholesterol was consistent in both urban and rural areas. Only around one fifth of participants returned for follow-up visits and few differences were observed between the mean of test results at baseline and follow-up. These results may reflect differences in prior access to health services and the importance of implementing POCT in resource-limited areas [4].
Comparison with previous studies
Multiparameter POCT devices have been previously used for cardiovascular risk assessment using lipid profile tests [35] and for diagnosis and control of diabetes by HbA1c [36], with good performance compared to the accuracy of conventional laboratory tests [35, 36]. Similarly, POCT for creatinine has already been used to estimate glomerular filtration rate in patients undergoing routine intravenous contrast radiological investigations [37]. On the other hand, POCT glucometers have shown poor performance in detecting gestational diabetes mellitus when compared with the oral glucose tolerance test [38]. The environment can, but does not always, modify the results. For example, in Nepal, with altitudes ranging from 1,000 to 4,000 m, two POCT devices (including the SimplexTAS™ 101) showed good performance accuracy with low biases for HbA1c and creatinine, and moderate positive biases for glucose and total cholesterol [39]. In the Peruvian Amazon, the temperature did not affect HbA1c POCT results, but barometric pressure did [10].
A 10% difference between results of POCT devices and the gold standard (i.e. conventional tests) can be a useful indicator to assess accuracy of some laboratory parameters [32, 33]. However, a minimal variation in creatinine values usually means a high percentage difference [40]. The difference in the number of new diagnoses between the intervention and control arms in this study can be explained by the fact that POCT results were provided immediately, avoiding loss to follow-up for result collection and interpretation, compared to the control facilities, which used conventional lab tests only. As our study was not randomized, differences in the studied population in the variables relating to prior diagnoses of chronic diseases were observed; thus, in the intervention arm, more participants were unsure about their previous diagnoses. Therefore, POCT could play an important role in identifying and confirming these cases, thanks to its ease of performance in remote areas [13].
We cannot explain the difference between POCT and conventional laboratory test results; perhaps this discrepancy was due to the poor calibration process and quality control of the conventional devices. The selected device (SimplexTAS™ 101) is self-calibrating and underwent quality control checks according to the manufacturer’s recommendations during the study. Therefore, ensuring calibration process and quality control could lead to fewer diagnostic errors and better control of NCDs in the population. Adequate implementation of calibration process and quality control systems in the PHCs is an area that requires research in developing countries [41, 42].
Relevance of POCT in public health
An optimal POCT device should be easy to operate with minimal training, be mobile and low-cost (along with its supplies), and be able to perform multiple simultaneous tests with rapid results even in extreme conditions (i.e., temperature and humidity environments) [5, 43, 44]. Furthermore, the ease of using POCT devices and the lack of need for specialized laboratory staff to conduct testing makes their implementation feasible in rural areas and PHCs, given the limited or non-existent access to laboratory tests in those areas [44]. This strategy can improve access to health services, early diagnosis, and fewer complications [13, 45, 46]. The previous co-creation process showed that patients prefer rapid results to reduce their time spent in PHCs [14]. Nevertheless, this fast turnaround of results did not improve participant rates in this study. It’s possible that community communication and engagement was not sufficiently impactful [11, 13], and that the implementation of new laboratory devices alone may not be sufficient to improve follow-up rates in PHC.
The use of POCT for diabetes management has the potential to reduce HbA1c levels and increase the number of patients with controlled glycemia [47]. A study showed that obtaining immediate results could improve patient satisfaction and adherence to medical follow-up. Reducing waiting times can positively influence the management of chronic diseases [47–49]. In the present study, the follow-up of only three months was insufficient to evaluate the adherence to the treatment and medical controls. Patients with NCDs should visit their doctor every three months pursuant to clinical guidelines [25, 27, 50]; however, that did not occur for most patients in this study. This finding should be considered for subsequent interventions to improve follow-up rates.
The evaluation of the economic implications of the implementation of POCT devices have not been well explored, and the evidence shows contrary positions [46]. This may depend on the specific equipment and test being evaluated [51]. A Brazilian study demonstrated that the use of POCT devices for HbA1c in PHCs was a cost-effective alternative that increased early control of diabetes and reduced the cost of its complications [48]. A systematic review found that few studies advise against the use of POCT devices implementation [52]. Furthermore, the use of POCT devices can improve disease control, reduce consultation time, and improve adherence to treatment, all of which may offset the price differences with conventional testing [11].
Strengths and limitations
Several limitations are acknowledged in this study. Non-randomized PHC selection could have introduced selection bias, especially because two PHCs were part of the co-design phase of the study. We estimated limited gain from randomization due to the small number of clusters in each stratum, and because differences could be due to poor balance during randomization. According to the literature, at least six clusters are needed to reach an adequate randomization balance [53]. The lack of long-term follow-up prevented evaluation of the long-term impact of POCT on NCD treatment and control. Some parameters were not available in some laboratories; although we tried to have them maintained by a specialist, filters or other parts of the devices needed to be changed, which the study could not cover. This reduced the number of tests performed and/or compared. Another limitation is the use of different conventional laboratory devices in the control versus intervention sites for the diagnosis of the NCDs; variation in test performance may have contributed to the difference in detection rate. Despite these limitations, the study has strengths. The sample size is considerable and both urban and rural areas were included, which improves the representativeness of the sample. To maintain comparability, we provided laboratory consumables to ensure that all facilities could perform the evaluated tests.
Lessons learned
The development of the study has allowed for some reflections to consider in future research projects. First, prior to beginning an investigation, one should verify that the health system has implemented an adequate calibration and quality control process for its laboratory equipment to obtain comparable results between the POCT and its conventional equipment. Second, it is preferable to choose a POCT device that is authorized within the country of study to eliminate the need to compare the results with a test that fall within the current standard of care to avoid increasing the workload of the laboratory technician, and would have reduced doubts of the treating physicians by having two test results. Furthermore, this choice must consider the time it will take to obtain usage permits and the importation process in the country. Finally, resources must be available to cover possible deficits in the local health system and ensure the continuity of laboratory operations in general. During the study, researchers observed that the shortage of reagents and other consumables for conventional laboratory tests reduced the influx of patients to health centres, even though the study ensured reagents for the parameters evaluated. For subsequent studies, other factors such as health information systems, service delivery, financing, and greater involvement of the local and regional health authorities should be considered.
Conclusions
The use of POCT, both stand-alone and combined with conventional testing, significantly increased the diagnoses of diabetes and high cholesterol compared to conventional laboratory testing alone, but did not increase the diagnosis of kidney disease. This study suggests that POCT is an effective tool for early detection of diabetes and high cholesterol, potentially improving long-term health outcomes for people with these NCDs, despite generating low follow-up rates.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The research team would like to thank Ana Maria Sandoval for her collaboration during the study. We extend our gratitude to each of the fieldworkers who made data collection for this study possible.
Abbreviations
- LMICs
Low and middle-income countries
- POCT
Point-of-care testing
- PHC
Primary healthcare centre
- HbA1c
Hemoglobin A1c
- cRR and aRR
Crude and adjusted risk ratios
Author contributions
ML-P, AB-O, ES and BV conceived and design the research idea. ML-P, LA-F, CH-E and AB-O conducted the data collection. LA-F, ML-P and AB-O performed the data analysis. LA-F wrote the initial draft and prepared tables. All authors reviewed and edited critically the manuscript. All authors approved the final version of the manuscript.
Funding
The study was funded by FIND-Foundation for Innovative New Diagnostics. The study was funded by FIND through a grant from the Swiss Agency for Development and Cooperation. All reagents and materials were purchased with study funds, and the POCT devices were received on loan during the study. No donations were accepted. The study did not receive funding from any pharmaceutical or similar industry. Study design, analysis and interpretation of data, and writing of the manuscript were independent of the organization providing the grants.
Data availability
The data supporting the findings of this study are not publicly available in an open-access repository. However, study protocol and data may be made available upon reasonable request to the corresponding author, subject to approval from the appropriate ethical review board and in compliance with relevant data protection regulations.
Declarations
Ethics approval and consent to participate
This study was reviewed and approved by the Institutional Review Board (IRB) at Universidad Peruana Cayetano Heredia, Lima, Peru (IRB00001014, date of approval: September 4th, 2023). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. The study was further modified by an IRB-approved amendment adding a review of participants’ medical records to confirm reported diagnoses and follow-ups visits. All participants signed an informed consent after the aims and procedures of the study were explained. Personal data of the participants were available exclusively to the personnel assigned to each study site and to the research team. Deidentified databases were analysed using codes assigned to each participant. No personal data has been published.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Noncommunicable Diseases -. NCDs | Knowledge for policy [Internet]. [cited 2023 Feb 11]. Available from: https://knowledge4policy.ec.europa.eu/foresight/topic/shifting-health-challenges/non-communicable-diseases-ncds_en
- 2.Global action plan for the prevention and control of. NCDs 2013–2020 [Internet]. [cited 2023 Feb 18]. Available from: https://www.who.int/publications-detail-redirect/9789241506236
- 3.Mendis S, Al Bashir I, Dissanayake L, Varghese C, Fadhil I, Marhe E, et al. Gaps in capacity in primary care in low-resource settings for implementation of essential noncommunicable disease interventions. Int J Hypertens. 2012;584041. [DOI] [PMC free article] [PubMed]
- 4.Peeling RW, Mabey D. Point-of-care tests for diagnosing infections in the developing world. Clin Microbiol Infect Off Publ Eur Soc Clin Microbiol Infect Dis. 2010;16(8):1062–9. [DOI] [PubMed] [Google Scholar]
- 5.Vetter B, Beran D, Boulle P, Chua A, de la Tour R, Hattingh L, et al. Development of a target product profile for a point-of-care cardiometabolic device. BMC Cardiovasc Disord. 2021;21(1):486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dincer C, Bruch R, Kling A, Dittrich PS, Urban GA. Multiplexed Point-of-Care Testing – xPOCT. Trends Biotechnol. 2017;35(8):728–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chaisirin W, Wongkrajang P, Thoesam T, Praphruetkit N, Nakornchai T, Riyapan S, et al. Role of Point-of-Care testing in reducing time to treatment Decision-Making in urgency patients: A randomized controlled trial. West J Emerg Med. 2020;21(2):404–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kegler MC, Rigler J, Honeycutt S. The role of community context in planning and implementing community-based health promotion projects. Eval Program Plann. 2011;34(3):246–53. [DOI] [PubMed] [Google Scholar]
- 9.Lenters-Westra E, English E, Analysis. Investigating the quality of POCT devices for HbA1c, what are our next steps? J Diabetes Sci Technol. 2019;13(6):1154–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Saxton AT, Miranda JJ, Ortiz EJ, Pan W. Assessment of two diabetes Point-of-care analyzers measuring hemoglobin A1c in the Peruvian Amazon. Ann Glob Health. 2018;84(4):618–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Tenorio-Mucha J, Busta-Flores P, Lazo-Porras M, Vetter B, Safary E, Moran AE, et al. Facilitators and barriers of the implementation of point-of-care devices for cardiometabolic diseases: a scoping review. BMC Health Serv Res. 2023;23(1):412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lazo-Porras M, Perez-Leon S, Cardenas MK, Pesantes MA, Miranda JJ, Suggs LS, et al. Lessons learned about co-creation: developing a complex intervention in rural Peru. Glob Health Action. 2020;13(1):1754016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shephard M, Shephard A, Matthews S, Andrewartha K. The benefits and challenges of Point-of-Care testing in rural and remote primary care settings in Australia. Arch Pathol Lab Med. 2020;144(11):1372–80. [DOI] [PubMed] [Google Scholar]
- 14.Albitres-Flores L, Perez-Leon S, Bernabe-Ortiz A, Tenorio-Mucha J, Cardenas MK, Vetter B, et al. Co-creation process of an intervention to implement a multiparameter point-of-care testing device in a primary healthcare setting for non-communicable diseases in Peru. BMC Health Serv Res. 2024;24(1):401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Powell BJ, Waltz TJ, Chinman MJ, Damschroder LJ, Smith JL, Matthieu MM, et al. A refined compilation of implementation strategies: results from the expert recommendations for implementing change (ERIC) project. Implement Sci. 2015;10(1):21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.SENAMHI Tumbes. - Pronostico Detalle [Internet]. [cited 2024 Nov 4]. Available from: https://www.senamhi.gob.pe/main.php?dp=tumbes&p=pronostico-detalle
- 17.Bernabé-Ortiz A, Carrillo-Larco RM, Gilman RH, Checkley W, Smeeth L, Miranda JJ, et al. Contribution of modifiable risk factors for hypertension and type-2 diabetes in Peruvian resource-limited settings. J Epidemiol Community Health. 2016;70(1):49–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Popkin BM, Adair LS, Ng SW. Global nutrition transition and the pandemic of obesity in developing countries. Nutr Rev. 2012;70(1):3–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Allender S, Wickramasinghe K, Goldacre M, Matthews D, Katulanda P. Quantifying urbanization as a risk factor for noncommunicable disease. J Urban Health. 2011;88(5):906–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Goldstein J, Jacoby E, del Aguila R, Lopez A. Poverty is a predictor of non-communicable disease among adults in Peruvian cities. Prev Med. 2005;41(3–4):800–6. [DOI] [PubMed] [Google Scholar]
- 21.Hayes RJ, Bennett S. Simple sample size calculation for cluster-randomized trials. Int J Epidemiol. 1999;28(2):319–26. [DOI] [PubMed] [Google Scholar]
- 22.Chen Y, Sloan FA, Yashkin AP. Adherence to diabetes guidelines for screening, physical activity and medication and onset of complications and death. J Diabetes Complications. 2015;29(8):1228–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Miranda JJ, Taype-Rondan A, Bazalar-Palacios J, Bernabe-Ortiz A, Ariely D. The effect of a Priest-Led intervention on the choice and preference of soda beverages: A Cluster-Randomized controlled trial in Catholic parishes. Ann Behav Med Publ Soc Behav Med. 2019;54(6):436–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: standards of care in diabetes—2024. Diabetes Care. 2023;47(Supplement_1):S20–42. 10.2337/dc24-S002. [DOI] [PMC free article] [PubMed]
- 25.Ministerio de Salud – Dirección General de Intervenciones Estratégicas en Salud Pública. Guía de Práctica Clínica para el Diagnóstico, Tratamiento y Control de la Diabetes Mellitus Tipo 2 en el Primer Nivel de Atención [Internet]. Lima, Peru: Ministerio de Salud del Perú; 2016 [cited 2023 May 8]. Report No.: R.M. N° 719–2015/MINSA. Available from: http://bvs.minsa.gob.pe/local/MINSA/3466.pdf
- 26.Overview. High cholesterol. In: InformedHealth.org [Internet] [Internet]. Institute for Quality and Efficiency in Health Care (IQWiG); 2022 [cited 2024 Nov 4]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK279318/
- 27.Guía de Práctica Clínica para diagnóstico, manejo y control de dislipidemia, complicaciones renales y oculares en personas con diabetes mellitus tipo 2 [Internet]. Lima, Peru: Ministerio de Salud del Perú; 2017 [cited 2024 Nov 25]. Available from: https://bvs.minsa.gob.pe/local/MINSA/4186.pdf.
- 28.Creatinine blood test. MedlinePlus Medical Encyclopedia [Internet]. [cited 2024 Nov 4]. Available from: https://medlineplus.gov/ency/article/003475.htm
- 29.Malta DC, Machado ÍE, Pereira CA, Figueiredo AW, de Aguiar LK, de Almeida W et al. Evaluation of renal function in the Brazilian adult population, according to laboratory criteria from the National Health Survey. Rev Bras Epidemiol. 2019;22:E190010.SUPL.2. [DOI] [PubMed]
- 30.American Diabetes Association Professional Practice Committee. 6. Glycemic Goals and Hypoglycemia: Standards of Care in Diabetes—2024. Diabetes Care. 2023;47(Supplement_1):S111–25. [DOI] [PMC free article] [PubMed]
- 31.TASCOM [Internet]. [cited 2024 Nov 5]. Available from: http://tascom.org/
- 32.Jendrike N, Baumstark A, Kamecke U, Haug C, Freckmann G. ISO 15197: 2013 evaluation of a blood glucose monitoring system’s measurement accuracy. J Diabetes Sci Technol. 2017;11(6):1275–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.42 CFR 493.931. -- Routine chemistry. [Internet]. [cited 2025 Aug 4]. Available from: https://www.ecfr.gov/current/title-42/part-493/section-493.931
- 34.Alsaidan AA, Alotaibi SF, Thirunavukkarasu A, ALruwaili BF, Alharbi RH, Arnous MM, et al. Medication adherence and its associated factors among patients with type 2 diabetes mellitus attending primary health centers of Eastern Province, Saudi Arabia. Med (Mex). 2023;59(5):989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Jain A, Rao N, Sharifi M, Bhatt N, Patel P, Nirmal D, et al. Evaluation of the point of care Afinion AS100 analyser in a community setting. Ann Clin Biochem. 2017;54(3):331–41. [DOI] [PubMed] [Google Scholar]
- 36.Sølvik UØ, Røraas T, Christensen NG, Sandberg S. Diagnosing diabetes mellitus: performance of hemoglobin A1c Point-of-Care instruments in general practice offices. Clin Chem. 2013;59(12):1790–801. [DOI] [PubMed] [Google Scholar]
- 37.Mathur N, Lu ZX, MacKay L, Lau T, Kuganesan A, Lau KK. Is point of care renal function testing reliable screening pre-IV contrast administration? Emerg Radiol. 2021;28(1):77–82. [DOI] [PubMed] [Google Scholar]
- 38.Khambule L, Chikomba C, Adam Y, Khan L, Haldane C, Vetter B, et al. Performance of point-of-care glucose testing for the diagnosis of gestational diabetes in South Africa. Int J Gynaecol Obstet Off Organ Int Fed Gynaecol Obstet. 2024;00:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Giachino M, Vetter B, Perone SA, Correia JC, Erkosar B, Heller O, et al. Performance and usability of cardiometabolic point of care devices in nepal: A prospective, quantitative, accuracy study. PLOS Glob Public Health. 2024;4(10):e0003760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Griffin BR, Butler-Dawson J, Dally M, Krisher L, Cruz A, Weitzenkamp D, et al. Unadjusted point of care creatinine results overestimate acute kidney injury incidence during field testing in Guatemala. PLoS ONE. 2018;13(9):e0204614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Price CP, Smith I, Van den Bruel A. Improving the quality of point-of-care testing. Fam Pract. 2018;35(4):358–64. [DOI] [PubMed] [Google Scholar]
- 42.McGuire H, Weigl BH. Medical devices and diagnostics for cardiovascular diseases in Low-Resource settings. J Cardiovasc Transl Res. 2014;7(8):737–48. [DOI] [PubMed] [Google Scholar]
- 43.St John A, Price CP. Existing and emerging technologies for Point-of-Care testing. Clin Biochem Rev. 2014;35(3):155–67. [PMC free article] [PubMed] [Google Scholar]
- 44.Weigl BH, Drake JK. Developing an adaptable set of Point-of-Care diabetes screening technologies for Low-Resource settings. Point Care J -Patient Test Technol. 2013;12(1):33–40. [Google Scholar]
- 45.Spaeth B, Shephard M, Schatz S. Point-of-care testing for haemoglobin A1c in remote Australian indigenous communities improves timeliness of diabetes care. Rural Remote Health. 2014;14(4). [PubMed]
- 46.St John A, Price CP. Economic evidence and Point-of-Care testing. Clin Biochem Rev. 2013;34(2):61–74. [PMC free article] [PubMed] [Google Scholar]
- 47.Shephard M, Mazzachi B, Shephard A, McLaughlin K, Bj BJD, Barnes G. The impact of point of care testing on diabetes services along Victoria’s Mallee track: results of a community-based diabetes risk assessment and management program. Rural Remote Health. 2005;5(3). [PubMed]
- 48.Rosa LdeS, Mistro S, Oliveira MG, Kochergin CN, Cortes ML, de Medeiros DS, et al. Cost-Effectiveness of Point-of-Care A1C tests in a primary care setting. Front Pharmacol. 2021;11:588309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Price CP, Kricka LJ. Improving healthcare accessibility through Point-of-Care technologies. Clin Chem. 2007;53(9):1665–75. [DOI] [PubMed] [Google Scholar]
- 50.Instituto de Evaluación de Tecnologías en Salud e Investigación. Guía de Práctica Clínica para el tamizaje, diagnóstico y manejo de la Enfermedad Renal Crónica En estadios 1 al 3: Guía en Versión Extensa. Lima, Peru: EsSalud; 2020. [Google Scholar]
- 51.Nichols JH, Alter D, Chen Y, Isbell TS, Moore N, Shajani-Yi Z. AACC guidance document onmanagement of point of care testing [Internet]. 2020 [cited 2024 Oct 11]. Available from: https://www.myadlm.org/science-and-research/academy-guidance/management-of-point-of-care-testing [DOI] [PubMed]
- 52.Lingervelder D, Koffijberg H, Kusters R, IJzerman MJ. Health economic evidence of Point-of-Care testing: A systematic review. PharmacoEconomics Open. 2021;5(2):157–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Hayes RJ, Moulton LH. Cluster randomised trials [Internet]. 2nd edn. Chapman & Hall; 2022. Available from: https://www.routledge.com/Cluster-Randomised-Trials/Hayes-Moulton/p/book/9781032339580.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are not publicly available in an open-access repository. However, study protocol and data may be made available upon reasonable request to the corresponding author, subject to approval from the appropriate ethical review board and in compliance with relevant data protection regulations.