International Journal of Statistics in Medical Research, 2013, 2, 67-75 67 E-ISSN: 1929-6029/13 © 2013 Lifescience Global Heterogeneity in Preferences for Primary Care Consultations: Results from a Discrete Choice Experiment Alessandro Mengoni1, Chiara Seghieri2,* and Sabina Nuti2 1 Azienda Sanitaria Unica Regionale, Ancona, Italy 2 Laboratorio Management e Sanità, Istituto di Management, Scuola Superiore Sant'Anna, Pisa, Italy Abstract: Purpose: The increasing importance of flexibility in the general practitioner (GP) -patient consultation approach in primary care requires healthcare managers and physicians to find a balance among all the potentially important characteristics of consultation. This study used a discrete choice experiment (DCE) to assess patients’ preferences for different attributes of GP consultation and how the rate at which they traded between different attributes is affected by socio-demographic characteristics and past experiences with primary care services . Methods: A survey was conducted to a sample of 6970 residents in Tuscany region, Italy. Besides socio-demographic characteristics the survey collected information about participants’ past experience with GP consultation in the last 12 months. Moreover, participants were asked to select their preferred option in a series of pairwise choices, defined by the following attributes: level of involvement in decision making, amount of information received from the GP and waiting time for the visit. Results: Results revealed that receiving information from the GP was more important than being involved in the decisions and that, approximately, a complete involvement had the same importance as a partial involvement. Participants' past experience with GP’s consultation appeared to have the greatest influence on the involvement level. The amount of information required by the respondents was also influenced by a complex interplay of personal and contextual factors. Conclusions: This large-scale study extends the body of literature on DCE applications for different GP consultation approaches, providing new information about the influence that patients’ socio-demographic characteristics and past experiences could have on consultation preferences. Keywords: Physician-patient relations, primary care, public preference elicitation, patient-centred care, discrete choice experiment. INTRODUCTION In Western countries, the physician-patient interaction has evolved from being a paternalistic one to a patient-centred one. This trend is particularly advocated in the primary care, considering the lack of complex diagnostic and therapeutic technologies and the long term nature of the doctor-patient relationship in this setting. Although existing research emphasizes support for increasing patient-centeredness in consultations, empirical evidence for the role of patient-centred care in patient outcomes is mixed [1]. While such interventions are generally successful in modifying styles of communication and increasing rates of patient satisfaction, it is much less clear as to whether they result in positive health outcomes [2]. Also in health economics, where the doctor-patient relationship has been modelled within the economic theory of “agency” [3], some of the theoretical arguments implicitly assume a patient-centred approach, while others imply *Address correspondence to this author at the Laboratorio Management e Sanità, Istituto di Management, Scuola Superiore Sant'Anna Piazza Martiri della libertà, 33, 56127 – Pisa, Italy; Tel: +39-050-883992; Fax: +39-050-883936; E-mail: c.seghieri@sssup.it a more paternalistic consultation style [4-6]. All the aforementioned indications suggest that patient preferences should be a more central element in determining the type of consultation approach [7]. However, addressing patients expectations could be arduous, as there are various consultation characteristics potentially important for patients that often generate conflicting results in the doctor-patient interaction. Patients, for instance, want fast access to good care, as well as to be helped to help themselves [8]. On the other hand, time constraints was the most often reported barrier to implementing shared decision making in clinical practice, according to the perceptions of health professionals [9]. This implied an increasing interest in priority setting analyses, where patients are required to trade off between some important attributes. Discrete choice experiments in primary care have presented a quite comprehensive evaluation of patients’ priorities for characteristics of primary care consultations [4, 10-12]. DCEs are a commonly used technique in health economics [13] in which individuals are presented with alternative hypothetical services consisting of a number of attributes with different levels, and they are asked to choose between these attributes. However, the existing work has not 68 International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 Mengoni et al. accounted enough for the influence that patients’ socio- demographic characteristics and past experiences could have on primary care consultations’ preferences. Previous studies on primary care demonstrated that such characteristics had an impact on patients’ preference for information exchange and involvement in decision making [14] and that the fit between physicians’ style and patients’ orientation influenced patient satisfaction and trust [15]. In addition, Charles and Gafni [16], rather than advocating a particular approach to patient care, emphasized the importance of flexibility in the decision making process so that individual differences in patient preferences are respected. Trying to deepen the knowledge on this matter, in this study a DCE approach was applied to Tuscany region (Italy) data to answer to the following research questions: What characteristics of primary care models are more important to Tuscan citizens? Are population preferences for different GP consultations affected by socio-demographic characteristics and past experiences with primary care services? The relative importance of the different primary care model attributes and the rate at which individuals trade between attributes and the relative value of different service configurations was examined by means of a DCE. These results could be of support to healthcare managers to configure existing and new primary care services in order to better meet population needs. METHODS Questionnaire Design The attributes and levels describing the different consultation scenarios were identified through a review of the existing literature and semi-structured interviews to primary care managers and District managers of Tuscan Local Health Authorities and they were validated in a focus group. In order to avoid placing a significant cognitive burden on respondents that could alter the trade off between the attributes [17], the number of attributes selected was limited to the three most important factors emerged [17]. Considering also the results of previous DCEs [4, 10-12, 20], plausible levels to each of the attributes were assigned (Table 1) [18, 19]. A full factorial design has been adopted [21] and 3 3 (27) combinations were obtained. The 27 alternatives were paired into choice sets using systematic level changes [22] to maintain orthogonality, level balance and minimal overlap [23]. Two different sets of questions were tested, including 10 and 4 choice tasks respectively, using a blocked design. The 27 choice sets were therefore distributed across three blocks of nine and nine blocks of three respectively, creating an extra column with a number of levels equal to the number of blocks which is uncorrelated with every attribute of every alternative. Level balance was satisfied within each block. In each version the sequence of questions was randomized (as to avoid possible ordering effects) with the first choice set repeated as the last choice set (to provide a check of response consistency). This experiment was embedded in the patient satisfaction and experience survey on primary care services (SEPC) performed in the Tuscany Region in 2009. Full details of the survey are reported elsewhere [24]. The 2009 SEPC questionnaire comprised four sections. The first section regarded participants’ past experience with primary care services. In more detail, this section, included questions such as the frequency and reason to see the GP in the last year, the working organization of the GP and the time waited in the clinic in the last consultation in the 12 months before the survey. The second section presented the attributes selected for the DCE and required each respondent to rank them in order of importance, for the identification of apparent non-traders. In the third participants were asked to make their choices in the context of a consultation for a non-urgent problem, and to express their preference for each choice set presented selecting one of the unlabelled options A or B. The fourth section comprised questions on current health status, presence of any chronic condition and socio- demographics. The survey was carried out using a computer aided telephone interview approach, as it allowed a wide geographic coverage with higher response rates than postal or internet approaches [25] and it was considered a viable method if used with a small number of choice sets per respondent [26]. Recruitment and Data Collection The reference population of the study consisted of Tuscan residents over 18 years of age. Taking into account the number of interviews required to return statistically significant results at health district level in SEPC survey, on the basis of previous experience and Heterogeneity in Preferences for Primary Care Consultations International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 69 the expected response rates, a sample of 6970 individuals was generated from the telephone directory of the Tuscany Region using a random sampling approach stratified by health districts. The questionnaires were pre-piloted to a sample of 34 individuals of different age and geographical location. In the final questionnaire configuration each respondent was randomly assigned to one of the nine blocks and was presented with four discrete choices. Figure 1 shows an example of a choice task. The DCE Models Each choice of the participants between pairs was included in the model as the binary dependent variable. Choice data were modeled using a random utility maximization framework [27] and, in order to account for multiple observations from a single respondent, a random effect probit model was used for modelling [28]. The baseline empirical model (model 1) was initially specified considering as the only independent variables the differences between the levels of each attribute in each pair of scenarios. Independent variables were effects-coded [29]. As second step, in addition to analysing the main attributes specified in the baseline model, it was hypothesised that respondents’ characteristics, such as socio-demographic, health condition and their past experience with the GP, would also influence preferences for consultation. Given that these characteristics do not differ between each choice, they were entered into the model analysis through interactions with the main effects (model 2). To create a more parsimonious model, model 2 was then Table 1: Attributes and Levels Selected Attributes Levels Names 0 Minutes Waiting time 90 Minutes Waiting time for the visit (WAIT) 180 Minutes Complete (you choose considering the doctor's opinion) Completely involv Partial (you and the doctor make a joint decision) Partially involv Involvement in decision making (INVOLV) No (the doctor chooses for you) * No involv A lot of information A lot of info Some information Some info Amount of information (INFO) A little information * No info *Denotes the base category. ALTERNATIVE 1 ALTERNATIVE 2 Wait …. Wait …. Nothing 1,5 Hours to be visited by a GP that …. to be visited by a GP that …. does not involve you in the decisions (the doctor chooses the treatment for you) partially involves you in the decisions (you and the doctor make a joint decision about your treatment) and that …. and that …. gives you some information on your problem / treatment (e.g. the most important information about drugs utilization) gives you a lot of information on your problem / treatment (e.g. how to benefit from specific healthcare services that could offer a better treatment or how to improve the quality of you life) Would you prefer Imagine that you need a visit by a general practicioner for a non-urgent problem and that you can choose between two alternatives Figure 1: Illustration of a choice task. 70 International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 Mengoni et al. Table 2: Respondents’ Description Attributes Levels Names Freq. % 18-49 years Age 18-49 938 29.5 50-69 years Age 50-69 1377 43.3 Age group > 69 years * Age > 69 868 27.3 Female Female 2419 75.0 Gender Male * Male 805 25.0 None / Primary level Educ No 1066 34.0 Secondary level Educ Sec 1759 56.1 Education University degree or higher * Educ Uni 308 9.8 Not working / Retired Empl No 1444 46.2 Working (High-skilled jobs) Empl High 293 9.4 Employment status Working (Medium / low-skilled jobs) + Students * Empl Low 1391 44.5 High Inc High 1214 42.1 Medium Inc Med 1199 41.6 Income Low * Inc Low 472 16.4 Yes Alone 364 11.8 Living alone No * Alone No 2733 88.2 Fair / Poor Health Low 414 13.3 Health status Excellent / Very good / Good * Health High 2697 86.7 Yes Chron 1132 36.1 Chronic disease No * Chron No 2008 63.9 Never / From 1 to 3 times Freq Low 903 31.7 Frequency to the GP clinic in the last year More than 3 times * Freq High 1950 68.3 General health check / Minor illness treatment Reas Min 434 16.0 Already existing illness check Reas Exist 579 21.4 Reason to see the GP Prescriptions / Certificates / Other * Reas Other 1692 62.6 Yes Assoc 757 28.0 The GP works with other GPs No * Assoc No 1948 72.0 Less than 1 hour Wait Less 1992 78.0 Time you waited in the clinic More than 1 hour * Wait More 563 22.0 Yes (Waited too much, GP unavailable, Clinic closed) Putoff 243 9.0 You have had to put off seeing the GP No * Putoff No 2462 91.0 Yes Listen 2651 98.2 The GP listened to you carefully No * Listen No 49 1.8 Yes Entime 2649 98.1 The GP gave you enough time to discuss No * Entime No 52 1.9 Yes Involv 2627 97.3 The GP involved you in the decisions No * Involv No 73 2.7 Yes Clear 2649 98.1 The GP gave you clear explanations No * Clear No 52 1.9 Yes Advice 1881 69.9 The GP gave you advices No * Advice No 810 30.1 Yes Trust 2661 98.4 You trust in your GP No * Trust No 44 1.6 *Denotes the base category. Heterogeneity in Preferences for Primary Care Consultations International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 71 reduced stepwise by excluding insignificant interaction effects one at a time with a p-value > 0.10 (model 3). The 95% confidence intervals (95% CIs) for the predicted utilities were calculated using non-parametric bootstrapping [30] with 2000 iterations. All statistical analyses were performed using Stata 10 (StataCorp, College Station, Texas). Internal Validity Investigation To measure internal validity the following three approaches were used: (a) consistency of preferences, which was tested with a stability (or completeness) test (b) willingness to trade which was tested by identifying respondents with dominant preferences following the approach in Scott et al. [31], and (c) consistency with theoretical predictions which was explored by examining the sign and significance of parameter estimates. RESULTS Response Patterns and Characteristics of Respondents Of the 6970 persons contacted, 3367 participated to the SEPC survey. Of these participants, 3225 completed the DCE, with a response rate of 46% - comparable to other DCEs in this setting (between 18% and 76%) [10, 24, 32, 33]. Details on responders’ characteristics are presented in Table 2. The respondents were equally distributed and without any significant differences in socio-demographic characteristics and past experience with the GP across the nine versions of the questionnaire used. Validity Issues The stability test showed that overall, 10% of respondents were inconsistent, which was considered to be acceptable. These levels are similar to those found in other studies [34-36]. While, for what concerns the willingness to trade, the level of dominant preferences was similar to other studies [31], as 17% of respondents chose always the scenario with the best level of a given attribute that they ranked as the most important attribute in the simple ranking. DCE Models The serial correlation obtained from running the random effect models were closed to zero and not statistically significant. This suggested that respondents treated the decision made in each pair- wise comparison as a separate hypothetical situation, and not in association with the decision made in each of the remaining pair-wise comparisons. Thus, all models were re-fitted to the data using the standard probit estimator. Results from model 1 showed that a large and a moderate amount of information from the GP is more important than being completely involved in the decisions and that participants would be willing to wait up to 92 minutes to be partially involved in the decisions (Table 3). Table 3: Results from DCE: Basic Mode (Model 1) Attribute Coefficient Std. Err. MRS (Min.) 95% CI (Lower) 95% CI (Upper) Constant 0.193 *** 0.016 - - - Waiting time -0.003 *** 0.000 - - - Completely involv 0.235 *** 0.014 93.4 79.4 109.6 Partially involv 0.233 *** 0.015 92.5 78.8 108.5 A lot of info 0.735 *** 0.015 292.3 263.9 328.5 Some info 0.289 *** 0.014 114.7 99.7 133.0 N 19212 Log Likelihood -10071.73 Likelihood ratio test (c 2 , d.f.) a 6490.03 (5) *** Pseudo R 2 McFadden a 0.244 ***p < 0.001. a Compared to a only constant model. 72 International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 Mengoni et al. Main findings from model 3 (Table 4) showed that individuals who trust their own GP, who in the last visit have been listened to carefully by the GP and received advice on eating or physical activity, as well as chronic patients did not prefer a consultation with a GP that totally involves them in the decisions about their treatment. Moreover, a GP consultation with a partial involvement was likely to be preferred by respondents living alone, who probably completely trust in their doctor and give him/her more decision power than those living with their family. Respondents who have been involved in the decisions in their last consultation with the GP were instead more likely to value being totally involved in the decisions about their treatment. With reference to the impact of respondents’ characteristics on preferences for the amount of information received, a consultation with a GP who gives a lot of information on the problem and treatment was the option preferred by the chronics, by respondents whose GP works in association with other GPs, by those who received in their last consultation advice from the GP on eating or physical activity and by those who trust their GP. Significant interactions between the amount of information and education level, and between information and frequency of visits to the GP clinic in the last year were reported in a similar analysis [4]. Table 4: Results from DCE: Reduced Model with Interaction Term (Model 3) Attribute Coefficient Std. Err. Constant 0.1766 *** 0.019 Waiting time -0.0024 *** 0.000 Completely involv 0.2726 *** 0.076 Partially involv 0.2614 *** 0.075 A lot of info 0.7297 *** 0.061 Some info 0.2878 *** 0.019 Waiting time * Advice 0.0003 ** 0.000 Completely involv * Chron -0.0409 ** 0.015 Completely involv * Listen -0.1384 † 0.076 Completely involv * Involv 0.2140 ** 0.069 Completely involv * Advice -0.0357 * 0.016 Completely involv * Trust -0.1212 † 0.073 Partially involv * Female -0.0297 † 0.017 Partially involv * Alone 0.0921 *** 0.024 Partially involv * Involv -0.1540 ** 0.059 Partially involv *Trust 0.2024 ** 0.071 A lot of info * Chron 0.0565 *** 0.016 A lot of info * Assoc 0.0924 *** 0.017 A lot of info * Putoff 0.1225 *** 0.028 A lot of info * Advice 0.1941 *** 0.019 A lot of info * Trust 0.1441 * 0.058 Some info * Inc Med 0.0635 ** 0.020 Some info * Advice -0.1708 *** 0.018 N 14344 Log Likelihood -7396.96 Likelihood ratio test (c 2 , d.f.) a 5091.08 (22) *** Pseudo R 2 McFadden a 0.256 *** p < 0.001, ** 0.01 > p  0.001, * 0.05 > p  0.01, † 0.1 > p  0.05. a Compared to a only constant model. Heterogeneity in Preferences for Primary Care Consultations International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 73 DISCUSSION This large-scale study extends the body of literature on DCE applications for different GP consultation approaches, providing new information about the influence that selected patients’ characteristics and past experiences could have on consultation preferences. While the results confirmed the value of all the attributes that were identified, in line with other researches [4, 10, 38-41], receiving information from the GP was more important than being involved in the decisions and that, approximately, a complete involvement had the same importance as a partial involvement. Apart from the important impact of some socio- demographic variables and the health status, the characteristics connected to the participants' past experiences seem to have the greatest influence on the involvement level. Thus, our data suggest that preferences for a different involvement level could be relatively controllable by the caregivers, considering that, to a large extent, they seem to depend on the attitude of the GP in the previous visits and to the long- term ‘personal’ relationship between a patient and his/her GP, that is different from the occasional contacts with a specialist. Patients, in Italy, are in fact free to choose their GP at the age of 18 and once they make their final choice they hardly change it [37]. Consequently, this long term relationship, built over the years and most of the times based on reciprocal trust should not require patients’ total involvement during each consultation. The presence of heterogeneity was also confirmed when taking into account information preferences. Implications for Practice Given this diversity of patient preferences, according to some authors [38, 42, 43] a potentially beneficial strategy to patients would be that of matching the physician’s actual behaviour to the patients’ desired levels of information and involvement. Ideally, physicians should adapt their participatory style, impersonating easily the entire range of egalitarian to paternalistic role behaviours [44]. However, this solution is not easily viable, because physicians sometimes have difficulties in judging accurately patient expectations [45, 46] and attempting to change provider consultation behaviours is hard to sustain [47]. Thus, further strategies should be used, such as pre-consultation approaches (routine pre-visit assessments of patients preferences and behaviours [7, 48, 49], patient activation programs that train patients to disclose their preferences and to be actively involved in their consultations [50, 51] or tailored interventions with interactive computer based presentations providing detailed descriptions about the disease, the treatment alternatives and the potential outcomes [52]. While there are several ways to achieve successful preference-match interventions, it is also important to note that patients’ preferences could change as the patient–provider relationship proceeds [40, 53]. In any case, the physicians could assess patient preferences during the visit by directly asking patients [54], although patients are often unable or unwilling to express their desired roles and needs, and time restrictions may hamper the ability of the physician to elicit valid preferences [55]. Further research therefore is needed to reveal principally how patients’ preferences and needs change over time. Clarifying these issues could facilitate the implementation of potentially more appropriate preference match strategies, enabling physicians to deliver a more flexible care with respect to the patient’s varying requests for information- receiving and involvement and to consequently improve patient outcomes. Limitations The study does have a number of limitations, reflecting the explorative nature of the research. First, although all the characteristics in the experiment were considered important and statistically significant, the relatively small number of attributes used in the design to ensure that the task was manageable for respondents, may have led to the omission of other features probably captured within the constant term. However, there is little discussion on this in the health economics literature, and where a significant constant is identified, the problem tends to be ignored [56]. Second, although our response rate was acceptable, the sample contained a slightly larger proportion of older persons and women than in the Tuscan population at large. This result may be expected in an “in-home” interview survey of this type [57]. In future studies, the potential for such biases needs to be addressed through more resources spent on recruitment of subjects. Third, the incorporation into the model of several aspects of respondents’ characteristics, interacted with 74 International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 Mengoni et al. the attributes, consented the investigation of important features of preference heterogeneity. However, some variations in tastes will probably remain unknown to the extent that it cannot be related to observed characteristics. Estimation models such as mixed logit and latent class model, that relax the assumption of taste homogeneity by allowing for random taste variation, should be explored in the future. ACKNOWLEDGEMENTS This study was supported by Tuscan Regional Healthcare Government. The authors would like to thank Anna Maria Murante and all the interviewers from Laboratorio Management e Sanità for their precious support. CONFLICT OF INTEREST All the authors have no conflicts of interest. REFERENCES [1] Mead N, Bower P. Patient-centred consultations and outcomes in primary care: a review of the literature. Patient Educ Couns 2002; 48(1): 51-61. http://dx.doi.org/10.1016/S0738-3991(02)00099-X [2] Lewin SA, Skea ZC, Entwistle V, Zwarenstein M, Dick J. Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database of Systematic Reviews 2001; 4: CD003267. [3] Ryan M. Agency in Health Care: lessons for economists from sociologists. Am J Econ Sociol 1994, 53(2): 207-17. http://dx.doi.org/10.1111/j.1536-7150.1994.tb02588.x [4] Scott A, Vick S. Patients, doctor and contracts: an application of principal-agent theory of the doctor-patient relationship. Scott J Polit Econ 1999; 46: 111-34. http://dx.doi.org/10.1111/1467-9485.00124 [5] Williams A. Priority setting in public and private health care. A guide through the methodological jungle. J Health Econ 1988; 7: 173-83. http://dx.doi.org/10.1016/0167-6296(88)90015-X [6] Evans RG. Strained Mercy. The economics of Canadian health care. Butterworths, Toronto 1984. [7] Krupat E, Yeager CM, Putnam S. Patient role orientations, doctor-patient fit, and visit satisfaction. Psychol Health 2000; 15(5): 707-19. http://dx.doi.org/10.1080/08870440008405481 [8] Coulter A. What do patients and the public want from primary care? BMJ 2005; 331:1199-201. http://dx.doi.org/10.1136/bmj.331.7526.1199 [9] Légaré F, Ratté S, Gravel K, Graham ID. Barriers and facilitators to implementing shared decision - making in clinical practice: Update of a systematic review of health professionals’ perceptions. Patient Educ Couns 2008; 73(3): 526-35. http://dx.doi.org/10.1016/j.pec.2008.07.018 [10] Vick S, Scott A. Agency in health care. Examining patients’ preferences for attributes of the doctor-patient relationship. J Health Econ 1998; 17: 587-605. http://dx.doi.org/10.1016/S0167-6296(97)00035-0 [11] Longo MF, Cohen DR, Hood K, Edwards A, Robling M, Elwin G, et al. Involving patients in primary care consultation: assessing preferences using discrete choice experiments. Br J Gen Pract 2006; 56: 35-42. [12] Cheraghi-Sohi S, Hole AR, Mead N, McDonald R, Whalley D, Bower P, et al. What Patients Want From Primary care Consultation: A Discrete Choice Experiment to Identify Patients’ Priorities. Ann Fam Med 2008; 6(2): 107-15. http://dx.doi.org/10.1370/afm.816 [13] De Bekker-Grob EW, Ryan M, Gerard K. Discrete choice experiments in health economics: a review of the literature. Health Econ 2010; 21(2): 145-72. http://dx.doi.org/10.1002/hec.1697 [14] Edwards M, Davies M, Edwards A. What are the external influences on information exchange and shared decision making in healthcare consultations: a meta-synthesis of the literature. Patient Educ Couns 2009; 75(1): 37-52. http://dx.doi.org/10.1016/j.pec.2008.09.025 [15] Krupat E, Bell RA, Kravitz RL, Thom D, Azari R. When physicians and patients think alike: patient-centered beliefs and their impact on satisfaction and trust. J Fam Pract 2001; 50(12): 1057-62. [16] Charles C, Gafni A, Whelan T. Decision making in the physician-patient encounter: revisiting the shared treatment. Soc Sci Med 1999; 49(5): 651-61. http://dx.doi.org/10.1016/S0277-9536(99)00145-8 [17] Ryan M, Gerard K. Using discrete choice experiments to value health care programmes: current practice and future research reflections. Applied Health Econ Health Policy 2003; 2(1): 55-64. [18] Torbica A, Fattore G. Understanding the impact of economic evidence on clinical decision making: A discrete choice experiment in cardiology. Soc Sci Med 2010; 70(10): 1536- 43. http://dx.doi.org/10.1016/j.socscimed.2009.12.030 [19] Salkeld G, Solomon M, Short L, Ward J. Measuring the importance of attributes that influence consumer attitudes to colorectal cancer screening. ANZ J Surgery 2003; 73(3), 128-32. http://dx.doi.org/10.1046/j.1445-2197.2003.02650.x [20] Gerard K, Lattimer V, Surridge H, George S, Turnbull J, Burgess A, et al. The introduction of integrated out-of-hours arrangements in England: a discrete choice experiment of public preference for alternative models of care. Health Expect 2006; 9: 60-69. http://dx.doi.org/10.1111/j.1369-7625.2006.00365.x [21] Viney R, Lancsar E, Louviere J. Discrete choice experiments to measure consumer preferences for health and healthcare. Expert Rev Pharmacoecon Outcomes Res 2002; 2(4): 319- 26. http://dx.doi.org/10.1586/14737167.2.4.319 [22] Street DJ, Burgess L, Louviere JJ. Quick and easy choice sets: Constructing optimal and nearly optimal stated choice experiments. IJRM 2005; 22: 459-70 [23] Huber J, Zwerina K. The Importance of Utility Balance in Efficient Choice Designs. J Mark Res 1996; 33: 307-17. http://dx.doi.org/10.2307/3152127 [24] Murante AM. Valutazione esterna. In: MeS Lab. (ed). Il Sistema di valutazione della performance della sanità Toscana - Report 2009. ETS, Pisa 2010; pp. 268-272. [25] Burge P, Devlin N, Appleby J, Gallo F, Nason E, Ling T. Understanding Patients’ Choices at the Point of Referral. RAND Corporation 2006. [26] Perkins WS, Roundy J. Discrete choice surveys by telephone. JAMS 1993; 21(1): 33-38. http://dx.doi.org/10.1177/0092070393211004 [27] McFadden D. Conditional logit analysis of qualitative choice behaviour. In: Zarembka P. (ed). Frontiers in Econometrics. Academic Press, New York 1974; pp. 105-142. [28] Propper C. The Disutility of Time Spent on the United Kingdom’s National Health Services Waiting Lists. J Hum Resour 1995; 30(4): 677-700. http://dx.doi.org/10.2307/146227 Heterogeneity in Preferences for Primary Care Consultations International Journal of Statistics in Medical Research, 2013 Vol. 2, No. 1 75 [29] Bech M, Gyrd-Hansen D. Effects coding in discrete choice experiments. Health Econ 2005; 14(10): 1079-83. http://dx.doi.org/10.1002/hec.984 [30] Efron B. Bootstrap methods another look at the jackknife. Ann Stat 1979; 7: 1-26. http://dx.doi.org/10.1214/aos/1176344552 [31] Scott A. Identifying and analysing dominant preferences in discrete choice experiments: An application in health care. J Econ Psychol 2002; 23: 383-98. http://dx.doi.org/10.1016/S0167-4870(02)00082-X [32] Morgan A, Shackley P, Pickin M, Brazier J. Quantifying patient preferences for out-of-hours primary care. J Health Serv Res Policy 2000; 5(4): 214-18. [33] Scott A, Watson M, Ross S. Eliciting preferences of the community for out of hours care provided by general practitioners: a stated preference discrete choice experiment. Soc Sci Med 2003; 56: 803-14. http://dx.doi.org/10.1016/S0277-9536(02)00079-5 [34] Howard K, Salkeld G. Does attribute framing in discrete choice experiments influence willingness to pay? Results from a discrete choice experiment in screening for colorectal cancer. Value Health 2009; 12(2). 354-63. http://dx.doi.org/10.1111/j.1524-4733.2008.00417.x [35] Phillips KA, Maddala T, Johnson F. Measuring Preferences for Health Care Interventions Using Conjoint Analysis: An Application to HIV Testing. Health Serv Res 2002; 37(6) 1681-705. http://dx.doi.org/10.1111/1475-6773.01115 [36] Maddala T, Philips KA, Johnson FR. An experiment of simplifying conjoint analysis designs for measuring preferences. Health Econ 2003; 12(12): 1035-47. http://dx.doi.org/10.1002/hec.798 [37] Buja A, Cavinato M, Perissinotto E, Rausa G, Mastrangelo G, Toffanin R. Why do patients change their general practitioner? Suggestions on corrective actions. Ir J Med Sci 2011; 180: 149-54. http://dx.doi.org/10.1007/s11845-010-0571-6 [38] Guadagnoli E, Ward P. Patient participation in decision- making. Soc Sci Med1998; 47(3): 329-39. http://dx.doi.org/10.1016/S0277-9536(98)00059-8 [39] Benbassat J, Pilpel D, Tidhar M. Patients’ preferences for participation in clinical decision making: a review of published surveys. Behav Med 1998; 24(2): 81-88. http://dx.doi.org/10.1080/08964289809596384 [40] Robinson A, Thomson R. Variability in patient preferences for participating in medical decision making: implication for the use of decision support tools. Qual Health Care 2001; 10(Suppl 1): i34-i38. http://dx.doi.org/10.1136/qhc.0100034 [41] Levinson W, Kao A, Kuby A, Thisted R. Not all patients want to participate in decision making. A national study of public preferences. J Gen Intern Med 2005; 20(6): 531-35. http://dx.doi.org/10.1111/j.1525-1497.2005.04101.x [42] Cecil DW, Killeen I. Control, compliance, and satisfaction in the family practice encounter. Fam Med 1997; 29(9): 653-7. [43] Auerbach SM. Should patients have control over their own health care? Empirical evidence and research issues. Ann Behav Med 2000; 22(3): 246-59. http://dx.doi.org/10.1007/BF02895120 [44] Makoul G. Perpetuating passivity: reliance and reciprocal determinism in physician–patient interaction. J Health Commun 1998; 3(3): 233-59. http://dx.doi.org/10.1080/108107398127355 [45] Coulter A, Peto V, Doll H. Patients’ preferences and general practitioners’ decisions in the treatment of menstrual disorders. Fam Pract1994; 11(1): 67-74. http://dx.doi.org/10.1093/fampra/11.1.67 [46] Hall JA, Stein TS, Roter DL, Rieser N. Inaccuracies in physicians’ perceptions of their patients. Med Care 1999; 37(11): 1164-68. http://dx.doi.org/10.1097/00005650-199911000-00008 [47] Hulsman RL, Ros WJ, Winnubst JA, Bensing JM. Teaching clinically experienced physicians communication skills. A review of evaluation studies. Med Educ 1999; 33(9): 655-68. http://dx.doi.org/10.1046/j.1365-2923.1999.00519.x [48] Krantz DS, Baum A, Wideman M. Assessment of preferences for self-treatment and information in health care. J Pers Soc Psychol 1980; 39(5): 977-90. http://dx.doi.org/10.1037/0022-3514.39.5.977 [49] Degner LF, Kristjanson LJ, Bowman D, Sloan, JA, Carriere KC, O’Neil J, et al. Information needs and decisional preferences in women with breast cancer. JAMA 1997; 277: 1485-92. http://dx.doi.org/10.1001/jama.1997.03540420081039 [50] Post DM, Cegala DJ, Miser WF. The other half of the whole: teaching patients to communicate with physicians. Fam Med 2002; 34(5): 344-52. [51] Harrington J, Noble LM, Newman SP. Improving patients’ communication with doctors: a systematic review of intervention studies. Patient Educ Couns 2004; 52(1): 7-16. http://dx.doi.org/10.1016/S0738-3991(03)00017-X [52] Lauver DR, Ward SE, Heidrich SM, Keller ML, Bowers BJ, Brennan PF, Kirchhoff KT, Wells TJ. Patient-centered interventions. Res Nurs Health 2002; 25(4): 246-55. http://dx.doi.org/10.1002/nur.10044 [53] Say R, Murtagh M, Thomson R. Patients’ preference for involvement in medical decision making: a narrative review. Patient Educ Couns 2006; 60(2): 102-14. http://dx.doi.org/10.1016/j.pec.2005.02.003 [54] Swenson SL, Buell S, Zettler P, White M, Ruston DC, Lo B. Patient-centered communication: do patients really prefer it? J Gen Intern Med 2004; 19(11): 1069-79. http://dx.doi.org/10.1111/j.1525-1497.2004.30384.x [55] Keating NL, Guadagnoli E, Landrum MB, Borbas C, Weeks JC. Treatment decision making in early-stage breast cancer: should surgeons match patients’ desired level of involvement? J Clin Oncol 2002; 20(6): 1473-79. http://dx.doi.org/10.1200/JCO.20.6.1473 [56] Richardson G, Bojke C, Kennedy A, Reeves D, Bower P, Lee V, et al. What Outcomes Are Important to Patients with Long Term Conditions? A Discrete Choice Experiment. Value Health 2009; 12(2): 331-39. http://dx.doi.org/10.1111/j.1524-4733.2008.00419.x [57] Green C, Gerard K. Exploring the social value of health-care interventions: a stated preference discrete choice experiment. Health Econ 2009; 18(8): 951-76. http://dx.doi.org/10.1002/hec.1414 Received on 29-11-2012 Accepted on 01-01-2013 Published on 31-01-2013 http://dx.doi.org/10.6000/1929-6029.2013.02.01.08 © 2013 Mengoni et al.; Licensee Lifescience Global. This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.