Abstract
Background:
Many governments and insurers are driving down the cost of medical devices, including glucose meters, by the central management of purchasing decisions. We report patients’ responses to an “enforced” change in brand of glucose meter, one year after the introduction of a national sole supplier arrangement for funded glucose meters and strips.
Method:
Specialist diabetes clinic attendees from two geographical locations completed a questionnaire one year after the final meter changeover date. In the first location, consecutive patients were asked to complete a glucose meter satisfaction questionnaire during their clinic visit. In the second location, this questionnaire was mailed to clinic attendees. Responses to open questions were analyzed thematically.
Results:
Response rates were 85% and 31% from the first and second locations, respectively and 378 questionnaires were suitable for analysis, 309 from the first and 69 from the second location. Insulin users composed 90% of participants. Results from the two locations were broadly similar. Most participants adapted well to the changeover, however 36% reported ongoing dissatisfaction with their “new” meter. The commonest concern, expressed by 23% of participants, related to meter accuracy and precision.
Conclusions:
One year after glucose meter changeover, a third of participants expressed dissatisfaction with their meter, with many participants describing a failure to adapt to the sole supplier arrangement. Providing a choice of meters and strips, ideally from two or more brands that have demonstrable differences in technical and ergonomic features, is likely to produce higher overall patient satisfaction than is a sole supplier arrangement.
Keywords: blood glucose self-monitoring, diabetes mellitus, health care economics and organizations, human factors and ergonomics, medical devices, patient satisfaction
Internationally, many governments and health agencies are managing procurement of medical devices, including glucose meters, in part to manage their associated costs.1-3 New Zealand took the unusual step in 2013 to 2014, of procuring publicly funded glucose meters and strips through what was effectively a sole supply arrangement. Only the CareSens® brand of meter (subsequently referred to as the “new” meter), was funded for most patients, thereby reducing patient choice of glucose testing device.4 Although these “new” meters had previously been funded together with several other meter brands, only a small number of patients had used this brand prior to the introduction of a sole supply arrangement.5
In general, change management in health is difficult and it is even more difficult if ownership of and commitment to the change process among stakeholders is low.4-6 At the time of glucose meter changeover, there was minimal evidence of patient ownership of the change process, despite the availability of free educational sessions and other up-skilling opportunities for both patients and clinicians.4,5 Much dissatisfaction was expressed publicly by patients and other clinical stakeholders at changeover about the impact of limiting glucose meter choice.4,7,8 It was however difficult at the time, to discriminate between dissatisfaction that related to factors such as the meter change process itself or brand loyalty,5 and negative technical and ergonomic aspects of the “new” meter/strip system that might represent long term barriers to optimal self-care. This study aimed to explore patient satisfaction and dissatisfaction with their “new” meter through a patient questionnaire, one year after the deadline for “enforced” glucose meter changeover.
Methods
Background: Glucose Meters Available to Patients
Prior to meter changeover, the majority of New Zealand patients chose the Roche Accu-Chek® meter as their preferred funded meter.4,5 In this article, the patient’s previous brand of meter is referred to as the “old” meter, irrespective of brand. After meter changeover, most patients chose to transition to one of the three government funded CareSens meters and strips: CareSens II, CareSens N and CareSens N POP (their “new” meter), rather than continue using their “old” meter and self-fund their glucose testing. While there are technical and ergonomic (human factors) differences between the three CareSens meters, these within-brand differences are less marked that are differences between brands.
Questionnaire Design
The study’s paper-based questionnaire was designed by a multidisciplinary team of local researchers, consumers, and diabetes health professionals using a collaborative, iterative process. The questionnaire focused on patient satisfaction and dissatisfaction with their “new” meter, within the context of national glucose meter changeover. After providing background demographic and clinical information, participants were then asked about their overall satisfaction with their “new” meter, using a five-point rating scale: “very dissatisfied” (1), “slightly dissatisfied” (2), “neither satisfied nor dissatisfied” (3), “slightly satisfied” (4), or “very satisfied” (5). The questionnaire also asked participants if they perceived any advantages of their “new” meter over their “old” meter and whether they would continue to use their “new” meter if they were given the opportunity to change brands. Open text boxes were provided for descriptive answers. A copy of the questionnaire is included in the Supplementary Materials.
Participant Recruitment
The study was conducted in two geographically distinct specialist diabetes centers. Center 1 (Christchurch, New Zealand) has a catchment population of around half a million and a temperate climate. Center 2 (Invercargill, New Zealand) has a smaller catchment population of around 120 000 and a slightly cooler climate. Both centers see youth and adults aged 15 years and older but antenatal diabetes care is provided at a different clinic.
In Center 1, consecutive patients attending diabetes outpatient clinics for physician, nurse, dietitian or podiatry review over a four-week period (April 29 to May 27, 2014) were considered for the study. Exclusions included patients not using a “new” glucose meter, either because they were using a brand other than CareSens or because they did not use any meter. Patients who were considered by attending staff to be vulnerable were also excluded. Center 1 participants provided written consent for participation. Center 2 had a much smaller volume of outpatient attendees, therefore a different recruitment method was used; patients due to be seen at the diabetes outpatient clinic within the period September 2014 to September 2015 were mailed the questionnaire. At Center 2, return of a completed questionnaire represented consent. Investigators at both centers were blinded to the participants’ identities.
Data Analysis
The primary questions were a rating frequency for different levels of satisfaction and also whether patients would choose to continue to use their “new” meter if they had an opportunity to change back to their “old” meter. Categorical variables were compared using chi-square test or Fisher’s Exact test (when half of the cell sizes were smaller than 5). Differences were deemed statistically significant if P < .05. In addition, qualitative analysis was undertaken of free text comments. The design of the questionnaire dictated the major qualitative themes, but not subthemes. Participant comments were therefore independently grouped into subthemes by two researchers (CM, DK) and a consensus reached on any variations in grouping.
Approval was obtained to conduct this study from the New Zealand Health and Disability Ethics Committee (HDEC number 14/STH/6).
Results
Complete or near complete questionnaires were returned by 85% of patients at Center 1 and 31% of patients at Center 2. Figure 1 shows participant (patient) inclusions/exclusions for Center 1, with 423 patient visits to the center yielding 309 questionnaires suitable for analysis.
Figure 1.
Open in a new tab
At Center 1, there were no statistically significant differences in demographic features, between participants agreeing to complete the questionnaire (n = 343) and nonparticipants (n = 59). The different method of recruiting participants at Center 2 (mailed questionnaire), precluded the researchers from constructing a Flow diagram or undertaking responder/nonresponder comparisons, at this center.
Participant Characteristics by Diabetes Center
Table 1 shows demographic and clinical results, including type of glucose meter currently used and also satisfaction ratings, from 378 participants by center.
Table 1.
Participant Characteristics, Including Satisfaction Ratings, by Diabetes Center.
| Center 1, n (%) | Center 2, n (%) | Total, n (%) | P value comparing centers | ||
|---|---|---|---|---|---|
| Total number of participants | 309 | 69 | 378 | ||
| Gender | Male | 172 (56) | 43 (62) | 215 (57) | .35 |
| Female | 134 (44) | 26 (38) | 160 (43) | ||
| Ethnicity | NZ European | 243 (79) | 57 (84) | 300 (80) | .31 |
| Maori | 28 (9) | 7 (10) | 35 (9) | ||
| Other | 38 (12) | 4 (6) | 42 (11) | ||
| Age group (years) | 0-49 | 111 (36) | 18 (26) | 129 (34) | .09 |
| 50-64 | 122 (39) | 28 (41) | 150 (40) | ||
| 65-74 | 51 (17) | 11 (16) | 62 (16) | ||
| 75+ | 25 (8) | 12 (17) | 37 (10) | ||
| Insulin use | Yes | 275 (90) | 64 (93) | 339 (90) | .42 |
| No | 32 (10) | 5 (7) | 37 (10) | ||
| Meter type | CareSens POP | 101 (33) | 29 (42) | 130 (35) | .35 |
| CareSens N | 187 (61) | 37 (54) | 224 (60) | ||
| CareSens II | 18 (6) | 3 (4) | 21 (5) | ||
| Previous usea | Yes | 253 (84) | 60 (92) | 313 (85) | .08 |
| Satisfaction | Very satisfied | 118 (39) | 16 (25) | 134 (36) | .01b |
| Slightly satisfied | 47 (15) | 6 (9) | 53 (14) | ||
| Neutral | 49 (16) | 16 (25) | 65 (18) | ||
| Slightly dissatisfied | 56 (18) | 11 (17) | 67 (18) | ||
| Very dissatisfied | 36 (12) | 16 (25) | 52 (14) |
Open in a new tab
a
“Previous use” denotes participants who had previously used a different brand of meter. bPost hoc tests (ie, follow-up multiple testing on each pair) suggest that Center 2 had significantly fewer “very satisfied” patients and more “very dissatisfied” patients than Center 1 (25% vs 39%, and 25% vs 12%, respectively).
Figure 2 provides a visual summary of satisfaction-dissatisfaction ratings, from all 378 participants.
Figure 2.
Open in a new tab
No between-center differences were observed in demographic features, insulin use or meter type, however post hoc analysis of satisfaction ratings showed that Center 2 had fewer satisfied and more dissatisfied participants compared to Center 1. This most likely reflects the difference in methodologies and also response rates between centers, with dissatisfied patients more likely to spend time responding to a postal (mailed) questionnaire, than satisfied patients.
When participants were asked, if given the opportunity to change their meter, would they do so, 47% responded that they would continue to use their current meter but 27% reported they would prefer to change. The remaining 26% were unsure whether they would elect to change meter. The most commonly used “new” meter type was the CareSens N, however younger participants preferred the smaller CareSens N POP meter. This age-related finding was statistically significant at both centers (P < .01).
Qualitative Analyses
Two macro themes, perceived advantages and perceived disadvantages, emerged in relation to the predefined questions asked of the participants. A third theme was also identified, which related to usability and ergonomic features of the “new” glucose meters. Subthemes identified from participants’ responses are detailed in the relevant subsections, described below.
Advantages of the “New” Meter
The four most commonly reported advantages (reported by greater than 5% of respondents) were (1) faster time for result, such as “faster,” “short test time” (18% of respondents); (2) small meter size, such as “smaller,” “lighter,” “easier to carry” (10%); (3) ease of use, such as “easy to use,” “easy to navigate,” “simple to use” (5%); and finally (4) requirement of a small blood volume, such as “less blood needed,” “needs little blood” (5%). The comments about the advantages of the small meter size was largely confined to younger patients who preferentially used the small version of the “new” meter.
In response to negative comments about perceived accuracy of the “new” meter (see below), we sought to provide balance by searching specifically for comments about improvement in accuracy. This was reported by 2% respondents. Typical responses were “Gives right readings,” “Results seem more accurate.”
Disadvantages of the “New” Meter
The most commonly reported disadvantage related to concerns about meter “accuracy” or analytical performance (23% of participants), with some participants commenting on more than one subtheme. Subthemes included (1) general comments about accuracy such as “results inaccurate,” “readings not accurate,” “I don’t think it gives the right results” (12%); (2) more specific comments related to lack of precision (imprecision) such as “different results in short time frame,” “different readings straight after each other” (5%); and (3) relative differences (bias) in results when compared with the previous meter, such as “reads higher than previous meter” (9%). In addition, three respondents reported concerns regarding the safety of the meter based on their perceptions of inaccuracy: “Results inaccurate, caused a bad turn . . . with low blood sugar levels, reads high . . . a safety issue! Someone could die,” “Tests do not reflect how I feel . . . show higher when I feel hypo,” “I feel the [“new”] glucose meters are too dangerous.”
Another subtheme related to technical features. Lack of utility at low temperature was reported by 14% of respondents and this was often linked to the need to test when physically active in cold environments, such as when tramping or playing soccer in winter. These comments came mainly from Center 2 participants, who lived in a cooler climate than Center 1 participants. Concerns regarding difficulty in recalling past results in the meter’s memory (6%) and dislike of the finger lancing device supplied with the meter (5%) were also mentioned.
Usability/Ergonomic Features
Specific comments about usability and ergonomic features were provided by 4% of participants. A typical comment from participants was that they found the “new” meter/strip system to be “fiddly.” More detailed descriptions of ergonomic features related to the spatial orientation of strips in the strip vial (which when compared to the “old” meter, were “upside down”) and how this related to placing the strip into the meter’s test strip portal. A typical comment was “Glucose sticks in the container are the wrong way round and this means you have to pull out by the blood test end and turn [the strip] around in your fingers.” Some patients used a slightly different approach, aimed at minimizing strip handling, which introduced another problem: “I’m right handed but every time I test, it [the meter] is upside down.” Another usability theme related to testing gear lacking a contrast of colors: “It’s all black—black meter, black strips, and black [glucose meter] case and at night blood looks black, so it is hard to work out what you are doing.” There was a single positive comment about ergonomic features, but this participant had developed a testing technique that was not in accordance with manufacturer’s instructions for meter use.
Discussion
The majority of study participants expressed satisfaction or felt neutral, when comparing their “new” meter with their “old” meter. The commonest reported positive experience using the “new” meter related to improvements in technical features. Also, one of the “new” meters was smaller than the “old” meter and this “new” meter was popular with younger patients. Dissatisfaction was however reported by 36% of participants. This related predominantly to concerns about meter accuracy but also related to negative technical features. Since undertaking this study, several changes have been made to the “new” meters. This includes an expansion of temperature range and also changes to finger lancing device. These changes are likely to mitigate some of the dissatisfaction expressed by patients at the time the current study was undertaken.
Although many participants perceived the “new” meter to be inaccurate, previous independent assessment of the “new” glucose meter’s performance showed acceptable accuracy and precision,9-13 including one assessment to the international standard ISO 15197:2013.9 Why the discrepancy between these technical reports of good analytical performance within research settings and the patient dissatisfaction with meter performance reported in this study, when meters are used by patients in real world settings? There are subtle but detectable differences (analytical biases) between glucose readings from the commonest brand of “old” meter and the “new” brand of meter.14 This may explain in part, initial concerns about accuracy, expressed at the time of meter changeover.7,8 We would nevertheless have anticipated that a year after changeover, nearly all patients would have adapted to how their meter “read,” in relation to the onset of symptoms, especially symptomatic hypoglycemia. The thematic comments about usability and ergonomics from the current study and also comments within a previous report,5 showed that some patients struggled with dexterity issues with the “new” meter system. This may have caused them to over-handle their test strips, which may have had an adverse effect on strip function.
Diabetes educators acknowledge the importance of human factors when matching patient characteristics with the ergonomic characteristics of different meter-strip systems.15 There are subtle ergonomic differences between the “old” and “new” meter systems, such as orientation of strips within the strip container, strip dimensions (size), strip texture (“slippery-ness”), and design of the strip container. For some patients, the fit between the “new” meter system and their physical and cognitive needs appears to have been suboptimal. This in turn is likely to have influenced the ability of some patients, to perform accurate tests on a consistent basis. Conversely, it is likely that those patients expressing satisfaction with their “new” meter found the “new” meter to be a good ergonomic fit with their own needs. Matching the ergonomic features of a glucose meter with the patient’s needs and preferences, may improve both safety and adherence, and also reduce psychological stress.3,16,17
This study has several limitations. The current questionnaire represents a single point in time and serial questioning would have allowed estimation of the time it took to reach a ceiling level of satisfaction/dissatisfaction. Participants were recruited from specialist diabetes clinics, thus insulin treated patients were over-represented in the study. Although study findings may not reflect the general diabetes population’s response to changes in glucose meters, the similarities between findings from the two participating centers did however offer reassurance that study results (other than comments about temperature range from the cooler climate of Center 2) were likely to be representative of specialist outpatient settings within New Zealand.
There is no single agreed approach to the practical, cost effective monitoring of the intended and unintended consequences that might occur following the implementation of a change in medical supplies, including glucose meters.18 Reliance on a single methodologic approach may provide an answer with a narrow perspective and this is unlikely to be in the patient’s best interest.3 A variety of assessment methodologies, designed by a range of stakeholders, is likely to represent the best way of monitoring a complex health care technology implementation plan.18 (ie, assessment of what worked, what did not work, why some aspects did not work and how to make it work better, next time around). This will in turn provide information about how best to optimize future implementations, balancing the need to preserve the health dollar against the risks of increasing the potential for harms and disruptions to patients. The results of the current study, which collected anonymized information from individual patients in a systematic way a year after meter changeover, represents one (clinician/patient) perspective. PHARMAC commissioned two assessments after meter changeover.5,19 The first assessment evaluated the implementation process from multiple perspectives, including the consumer (patient) perspective.5 It concluded that most of the 120 000 consumers who acquired a new meter experienced no negative impact, however a minority were concerned about the accuracy of their new meter and this impacted on some patient’s self-monitoring, thus these findings were congruent with that of the current study. The second assessment explored the impact of changeover on patient hospitalizations and concluded there was a possible impact on hospitalizations.19 The introduction of additional assessment methodologies, such as PHARMAC’s plan to assess meter usability during future competitive tender processes,20 is also welcomed.
In conclusion, a nationwide glucose meter changeover provided the opportunity to undertake a questionnaire based study which explored patient perceptions of a sole supplier arrangement, with patients acting as their own historical controls. One year after “enforced” glucose meter changeover, most study participants expressed satisfaction with their new meter but a minority expressed ongoing dissatisfaction. Levels of satisfaction/dissatisfaction may in part reflect an interaction between technical and ergonomic features of a particular brand of meter and strip, and patient specific factors. We recommend that medical device procurement agencies give patients sufficient choice of glucose meters, to allow them to choose a good fit between their personal needs and preferences and this essential medical device.
Supplementary Material
Supplementary material
Lunt_Supplemental_data_016.pdf (139.1KB, pdf)
Acknowledgments
We thank the staff and patients at the Christchurch Diabetes and also our local lay society, Diabetes Christchurch, for helping to develop the questionnaire used in this study. We also thank Ma Yi, biostatistician, Canterbury District Health Board, for her statistical advice. Several individuals and organizations kindly commented on interpretation of study findings. Finally, we are grateful to the study participants for sharing their reflections about meter changeover.
Footnotes
Abbreviation: PHARMAC, Pharmaceutical Management Agency.
Declaration of Conflicting Interests: The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: HL is on the speakers’ bureau for Sanofi (New Zealand) and has also undertaken glucose evaluation studies on behalf of PHARMAC (Pharmaceutical Management Agency), New Zealand. MD has received speaker’s honorarium and partial travel sponsorship to attend meetings from Novo Nordisk and Sanofi. CM and DK declare no competing interests.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material: The supplemental materials are available at http://journals.sagepub.com/doi/suppl/10.1177/1932296817693016.
References
- 1.Leigh S, Idris I, Collins B, et al. Promoting health and reducing costs: a role for reform of self-monitoring of blood glucose provision within the National Health Service. Diabet Med. 2016;33:681-690. [DOI] [PubMed] [Google Scholar]
- 2.Graves K.Global best practices in medical device procurement—a road map to system success. J Med Marketing. 2011;11:1101-1108. [Google Scholar]
- 3.Puckrein GA, Nunlee-Bland G, Zangeneh F, et al. Impact of CMS competitive bidding program on Medicare beneficiary safety and access to diabetes testing supplies: a retrospective, longitudinal analysis. Diabetes Care. 2016;39:563-571. [DOI] [PubMed] [Google Scholar]
- 4.Lunt H, Florkowski CM.Transitioning to a national (New Zealand) sole supply scheme for glucose meters: lessons learnt and problems yet to be solved. J Diabetes Sci Technol. 2014;8:615-618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Carter M, Hardie-Boys N, Marwick J, Wilson J.Evaluation of the implementation of a decision to change the funding and supply of blood glucose meters and test strips. September2014. Available at: https://www.pharmac.govt.nz/assets/blood-glucose-meter-report-evaluation-of-implemention.pdf. Accessed December. 2016.
- 6.Cornescu V, Adam C-R.The consumer resistance behaviour towards innovation. Procedia Economics and Finance. 2013;6:457-465. [Google Scholar]
- 7.Penman M.Diabetes meters under fire. Available at: http://www.stuff.co.nz/national/health/8921771/Diabetes-meters-under-fire. Accessed December. 2016.
- 8.Weekes J.Bitter deal for diabetics. Available at: http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=10900924. Accessed December. 2016.
- 9.Link M, Pleus S, Schmid C, et al. Accuracy evaluation of three systems for self-monitoring of blood glucose with three different test strip lots following ISO 15197. J Diabetes Sci Technol. 2014;8:422-424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Huang TY, Chang HW, Tsao MF, et al. Evaluation of accuracy of FAD-GDH- and mutant Q-GDH-based blood glucose monitors in multi-patient populations. Clin Chim Acta. 2014;433:28-33. [DOI] [PubMed] [Google Scholar]
- 11.Huh HJ, Park HD, Lee SY, Kim JW.Evaluation of CareSens® N glucometer glucose monitoring system. J Lab Med Qual Assur. 2010;32:229-236. [Google Scholar]
- 12.Kong GWS, Tam WH, Chan MHM, et al. Comparison in the performance of glucose meters in blood glucose monitoring during pregnancy. Gynecol Obstet Invest. 2010;69:264-269. [DOI] [PubMed] [Google Scholar]
- 13.Lunt H, Logan F, Florkowski C.Evaluation of capillary glucose meter: CareSens N™. 2011. Available at: http://www.pharmac.govt.nz/2012/02/28/CareSensN%20evaluation.pdf. Accessed December 2016.
- 14.Thompson H, Chan H, Logan FJ, et al. A glucose meter evaluation co-designed with both health professional and consumer input. N Z Med J. 2013;126:90-97. [PubMed] [Google Scholar]
- 15.Tsai A.Meters: does your device measure up?Diabetes Forecast. 2015. Available at: http://www.diabetesforecast.org/2015/mar-apr/meters-does-your-device-measure-up.html?loc=ymal. Accessed December 2016. [PubMed]
- 16.Schaeffer NE.The role of human factors in the design and development of an insulin pump. J Diabetes Sci Technol. 2012;6:260-264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Liberman A, Buckingham B, Philip M.Diabetes technology and the human factor. Int J Clin Pract. 2012;66:79-84. [DOI] [PubMed] [Google Scholar]
- 18.Varela-Lema L, Ruano-Ravina A, Mota TC, et al. Post-introduction observation of healthcare technologies after coverage: the Spanish proposal. Int J Technol Assess Health Care. 2012;28:285-293. [DOI] [PubMed] [Google Scholar]
- 19.Jackson G, Thornley S.Hospitalisations for diabetes in New Zealand: measuring the impact of the change in subsidy to glucose monitoring test strips. August2015. Available at: www.pharmac.govt.nz/medicines/your-health/diabetes/blood-glucose-meters/evaluation/#impacts. Accessed December 2016.
- 20.PHARMAC. Future funding approach for blood glucose meters and test strips. March2015. Available at: www.pharmac.govt.nz/news/consultation-2015-03-06-blood-glucose-meters/. Accessed December 2016.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material
Lunt_Supplemental_data_016.pdf (139.1KB, pdf)
