In addition to medical practices, PHSs include other services such as dentistry, veterinary medicine, optometry, cosmetic surgery, and physiotherapy. PHSs are provided mainly by SMEs and micro-firms (Henry et al., 2016). These activities are essential in societies: their expenditures represent between one and three percent of European countries' GDPs. According to Chung and Tseng (2019) , PHSs are among the most representative KIBS activities due to the high level of knowledge required to ensure their functioning and performance and the high investment required in a specific equipment-qualified workforce.

Beyond the small size of PHSs firms, there are two aspects that characterize them. First, they conduct their activities in a traditionally inflexible institutional and legal environment. These professions adapt to the needs detected by their members in the territory within their scope of application. Professional colleges and associations (PAs) are institutions legitimized to organize these professions (Greenwoodet al., 2002). However, the deregulation policies in the last years have led to other institutions with greater bargaining power to enter the market. Examples include networks of clinics and insurance firms, which have incorporated a multitude of innovations into the sector (Brandt and Znotak, 2021). Consequently, these actors have also emerged as legitimized referents (DiMaggio and Powell, 1983), as they influence the organization and professional activities carried out by professional clinics and associations. Second, PHSs require a high and continuous investment in education and training to practice their profession with guarantees, "also requiring high qualified employees as the unique way to produce profitable results for clients” (Huggins and Weir, 2012: 93 ).

All these features represent clear entry barriers. Moreover, as a consequence of the fragmentation of the environment, in recent years, other types of knowledge that are more closely linked to business management-such as the implementation of market research and new strategies for the promotion and management of innovation-have become increasingly important (Trathen and Gallagher, 2009). These changes in leadership and management seek to improve clinics' competitiveness and efficiency. Despite their relevance, they have also introduced a certain level of conflict and stress among professionals, mainly due to the dilemma between focusing on the profession or improving the clinic's performance.

Following Meyer and Rowan (1977: 340-41) , institutionalized organizations are defined as those "systems that arise to coordinate and control specific activities that are based on technical actions in complex environments and where the limits of the activity are explored (...). The ultimate goal of these institutions is to maintain the ceremonial conformity of the profession in the face of environmental uncertainties." Zucker (1987) distinguishes two types: 1) the environment as an institution, which refers to the agents created by the governments to coordinate the actions of these professions (e.g., PAs); 2) organizations as institutions, which are those companies that, through the market, develop routines or incorporate innovations that are subsequently imitated by others (e.g., the market and leading firms).

In Spain, PAs are among the key institutions that regulate PHSs by establishing the necessary conditions to practice these professions. Some of the benefits of such interventions include the formulation of rules that favor more predictable behaviors, which may thereby increase the survival rate (DiMaggio and Powell, 1983), safeguard quality standards (Trathen and Gallagher, 2009), and/or encourage better performance within the profession (Perron et al., 2014). On the contrary, some researchers criticize the use of these agents because they standardize or homogenize the competitive strategy (e.g., Chadee and Roxas, 2013; Johnston, 2013; Thackeray et al., 2005).

Moreover, the market creates institutions when organizations engaged in the same activity are identified as inefficient (Davis and McMaster, 2015). The most influential institutional referents are the most prominent firms within a particular activity (Lieberman and Asaba, 2006). These firms include those considered when conducting a competitor analysis. PHSs have usually adopted a highly reluctant and critical attitude toward such organizations because they were often bigger and had implemented new business models based on the deregulation of activities regarded as "owned" by the qualified professionals engaged in them (Nash, 1994). Traditional professionals accept the actions of public agents aimed at maintaining control over the profession (centralization), and they reject disruptive strategic innovations that emerged from market failures and deregulation. However, despite the protectionism still in force, some large companies have penetrated powerfully into the PHSs. These companies have become a reference model for some entrepreneurs, while others believe their survival is threatened.

Intellectual Capital (IC) is defined as the intellectual assets that can create value, including knowledge, information, intellectual property, and experiences (Stewart, 1997). There is broad agreement in the literature on the close relationship between IC and PHSs (e.g., Cohen and Kaimenaki, 2007), mainly because high levels of investment in knowledge (e.g., training, experience, patient management) represent the most visible barrier to entry for those who wish to pursue this type of business. Given the competitive environment and similarities in structural capital due to the small size of PHS micro-clinics, it is possible to identify two critical dimensions of the IC of these firms: 1) human capital (HC), and 2) relational capital (RC) (Allameh, 2018).

HC is constituted by the “skills, experiences, attitudes, ideas, values and competencies of the people who are integrated into a company” (Watson and Stanworth 2006: 339 ). Some indicators related to HC include educational attainment, experience, personal motivations, employee knowledge, innovative capacity, satisfaction, and employee turnover (Tai and Chen, 2009; Zarandi et al., 2012). RC is the knowledge acquired from markets, suppliers, and customers, as well as from governments, representatives of economic activities, and the most relevant associations (Bianchi-Martini et al., 2016). Therefore, this dimension is not merely evaluated from an internal point of view but also from the perspective of customers and suppliers. There are many indicators to measure the RC. For example, Hosseini and Owlia (2016) detected more than 50 variables, such as brand value, licensing agreements, or subsidiary firms, among others.

Organization configurations are commonly occurring clusters of attributes related to structures, processes, and strategies (Miller, 1996). They are explained because groups of companies have similar organizational characteristics (Meyer et al., 1993). It can also be understood "by identifying distinct, internally consistent sets of firms and their relationships to their environments and performance outcomes over time rather than by seeking to uncover one universal set of relationships that hold across all organizations" (Ketchen et al., 1997: 224).

From an equifinality perspective, it is feasible to predict that different organizational factor (archetypes) combinations are associated with such strategic positioning (Fiss, 2007). Compared to other well-known options (e.g., Miller and Friesen, 1978; Mintzberg, 1979), in line with Helmig et al. (2014) , and considering the cri-tical dimensions under study, the typology of Miles and Snow (1978) may provide better guidance.

Summarizing, their four main archetypes can be defined as1) the prospective posture, which aims to identify new market opportunities and invest intensively in individuals while relying, organizationally, on the decentralization of institutions and improving IC; 2) the defensive posture, which is concerned with maintaining a stable environment (centralization), even if some innovations have to be abandoned while paying moderated attention to RC; 3) the analytical posture, which seeks to identify new opportunities (moderated decentralization), though only by developing upon core skills (HC), while also providing high-quality services; and 4) the reactive posture, which demonstrates inconsistent and unstable behaviors (e.g., sometimes moderated decentralization and sometimes moderated IC), and does not have a clear competitive strategy.

In short, given the current institutional deregulatory context of PHSs and the unequal IC endowment among these firms, different strategic positions can be envisaged. In line with Torgovicky et al. (2005) or Marlin et al. (2002) and opinions recorded in previous steps of our research, we suggest that, among companies providing PHSs, there would be three main positionings:

Focused on providing excellent service (i.e., "the quality of the service we provide is much better than that of our competitors").

In summary, we propose that differences in organizational configurations (degree of attention to the institutional environment and the market, as well as the emphasis placed on the development of IC) will influence the strategic response of companies (positioning) and will impact their abilities and potentialities to detect and capture opportunities and threats. That combination of configuration-positioning is evidence of the business model implemented (Zott et al., 2011) and can predict the success or failure of a company under certain competitive conditions (Miller, 1996).

A qualitative study was conducted on previous broader research to explore professionals' perceptions and re-fine the questionnaire to carry out the empirical analysis. It is based on data from companies providing PHSs in dentistry, optometry, and physiotherapy. The current study selected these professional activities because of their representativeness, research opportunity, and methodological reasons. Due to the normative and regulatory characteristics of the PHSs of each country, only companies located in Spain were selected. Given that most of the firms in these sectors are independent micro and small-sized companies, the population of firms analyzed had a maximum of nine employees.

PHS professionals were contacted through their respective business associations. The questionnaire was distributed and collected from June to September 2017 through the Limesurveycomputer platform. It was based on previous studies that examined the strategic positioning of companies (Robinson and Pearce, 1988), the role of institutions (Ang and Cummings, 1997), and the valuation of IC (Sveiby, 1997; Dewhurst and Cegarra-Navarro, 2004).

A total of 78 valid questionnaires were retrieved. The average sample size consisted of five employees. Regarding the distribution, 56% of the companies were engaged in dentistry, 26% in optometry, and 18% in physiotherapy. Given the sample size and the research purpose, the technique used was qualitative comparative analysis - QCA (Lee et al., 2019; Fainshmidt et al., 2019); a data analysis based on theory to examine the relationship of conditions to an outcome.

In this work, we deployed the QCA analysis technique. Although it has been criticized (Fiss, 2007), it has some advantages. According to Miller (2018) , QCA makes it possible to identify the configurations of causally re-levant characteristics and antecedent conditions most associated with strategic positioning of each company. Following this author, we can assume that QCA has become one of the most used techniques in organizational configurations.

Compared to other techniques, it is advantageous as it is possible to determine the conditions separately or in combination with others to identify the best fit for a given outcome (Rihoux and Ragin, 2008). For example, for those micro-clinics most aligned with a differentiation position, this technique identifies combinations of dimensions of the institutional environment and the organizational context they create.

The fuzzy-set QCA (fsQCA) variant of this technique was employed in our study because it admits dichotomous values typical of Boolean algebra and continuous values typical of Likert scales. This variant is suitable for constructing better causal theories and working with small and medium-sized samples (Fiss, 2011). This QCA modality requires three steps: 1) data calibration, 2) necessity analysis, and 3) sufficiency analysis.

Considering the objectives of this research, strategic positioning was taken as the dependent variable or outcome. In line with Robinson and Pearce (1988), three items were used to identify it, two of which were aligned with differentiation: ‘reaction capability’ (Murwatiningsihet al., 2019) and ‘quality service perceived’ (Hampel, 2021). The service price item (Auzair, 2011; Hilman and Kaliappen, 2014) was related to cost leadership. These items were assessed using a seven-point Likert scale. Based on this scale, the respondent evaluated its position compared to its main competitors for each item. For example, for the price of services, respondents were requested to rate their company's performance with respect toits competitors as follows: 1 = "much worse than its competitors"; 4 = "the same"; and 7 = "much better than its competitors."

We specified four independent variables or ante-cedent conditions of our research, two of which are based on the institutional context, while the other two relate to the organizational context. The institutional conditions include the PAs (professional colleges and associations) and the influence of the market (e.g., competitors, customers, and suppliers). The PA variable is made up of a single item which allowed us to assess how respondents perceived the help provided by these institutions to coordinate and achieve the objectives of their clinics. The market variable is a construct of three items based on Ang and Cummings (1997) . They reflect the level of vigilance and attention to consumer trends and the strategies implemented by competitors. The higher values recorded are associated with more favorable perceptions in all cases.

The antecedent conditions based on the organizational context are derived from the relevant IC literature (Sveiby, 1997; Dewhurst and Cegarra-Navarro, 2004). In this article, they are formed by the HC and the RC dimensions. The HC construct captures respondents' perceptions of the professionals' skills and experience required to perform their professional activities. Regarding RC, it measures its loyalty and reputation among its customers.

Table 1 presents the means, standard deviations, and the calibration made. Out of the causal conditions, HC stands out with the highest mean value (5.5), while market attention is the condition with the lowest score (4.5), together with the role of PAs (4.4) and RC (4.8). The differentiation positioning outcome has a mean of 5.4, which indicates that, in general, most clinics have a high perception of following this competitive strategy. The cost positioning shows that the companies are less alienated with this competitive strategy or out of it (4.6).

Table 1

Conditions and outcomes	M	SD	Threshold of non-membership (% of cases)	Crossover point (median)	Threshold of membership (% of cases)
Differentiation	5,4	1,1	4,0 (17)	5,5	6,5 (21)
Cost Leadership	4,6	1,7	3,0 (13)	4,0	6,0 (35)
PA influence	4,4	2,0	2,0 (24)	5,0	7,0 (19)
Marketinfluence	4,5	1,4	2,0 (10)	4,7	6,0 (14)
Human capital	5,5	1,0	3,3 (5)	5,3	6,7 (17)
Relational capital	4,8	1,3	2,5 (9)	5,0	6,5 (11)

Note: **p<0.05.

In the QCA analysis, the calibration of the data involves the establishment of membership thresholds. In line with Ragin (2006) , outcomes were set at 25%, while antecedent conditions were at 10%. For maximum ambiguity, we decided to use the median. We excluded those combinations from the truth table represented by only a single case (Schneider and Wagemann, 2010). As a result, the membership threshold was established at 6.5 (21% of the cases included) for the differentiation outcome, while the non-membership was at 4.0 (17%). For cost leadership positioning, the thresholds were 6.0 (35%) and 3.0 (13%). The membership thresholds show-ed that 56% of companies follow a precise positioning strategy based on differentiation or cost leadership. Therefore, the remaining 44% of the cases were classified as companies with a stuck-in-the-middle strategic position, i.e., those without a clear strategic orientation (Torgovicky et al., 2005).

The analysis of necessary conditions is the second step of QCA. Causal conditions will be excluded from the analysis if they show a consistency higher than 0.90 (Glaesser, 2008). Table 2 shows all the results. Given that all causal conditions are not consistently necessary, all of them are included in the sufficiency analysis.

Table 2

	Differentiation		~Differentiation		Cost Leadership		~CostLeadership
	1	2	1	2	1	2	1	2
PA influences	0.59	0.67	0.52	0.54	0.52	0.70	0.61	0.50
~PA influences	0.59	0.58	0.68	0.61	0.63	0.73	0.63	0.44
Market influences	0.67	0.67	0.59	0.53	0.61	0.72	0.65	0.46
~Market influences	0.53	0.58	0.64	0.64	0.55	0.72	0.61	0.48
HC	0.76	0.68	0.66	0.53	0.64	0.69	0.80	0.51
~HC	0.48	0.60	0.61	0.70	0.55	0.82	0.51	0.46
RC	0.70	0.70	0.54	0.50	0.61	0.74	0.69	0.51
~RC	0.52	0.55	0.70	0.68	0.60	0.76	0.65	0.50

Note: (1) Consistency; (2) Coverage.

The U Mann-Whitney’s non-parametric test completes our descriptive results (Table 3). Here, we compare those cases included in the membership threshold with other cases not included. In terms of differentiation, it shows the significant and positive relationship between Human and Relational capital. According to the mean rank (48.63 vs. 37.15, and 48.94 vs. 37.06) and the Wilcoxon test (p value< 0.1 in both cases), this could be interpreted as firms with a strategic positioning in differentiation have high perception or are associated with a high value in their intellectual capital. Regarding the cost leadership positioning, no antecedent condition was remarkably different between cases included in the threshold and those excluded.

Table 3

		Differentiation				Cost leadership
		N	U	Mean Rank	Sig.	N	U	Mean Rank	Sig.
PA influence	Strong	16	483	40,31	0.87	27	638	41,37	0.59
	Not strong	62		39,29		51		38,51
Market influences	Strong	16	407	45,06	0.27	27	650	38,09	0.69
	Not strong	62		38,06		51		40,25
Human capital	Strong	16	350	48,63	0.07*	27	585	35,67	0.27
	Not strong	62		37,15		51		41,53
Relational capital	Strong	16	345	48,94	0.06*	27	673	38,93	0.87
	Not strong	62		37,06		51		39,80

Source: Own elaboration.

The analysis of sufficiency is the last of the QCA steps. It was conducted by the algorithm of Quine-McCluskey. This analysis shows those conditions or their combinations or factors associated with each outcome. Authors like Ragin (2006) or Fiss (2011) proposed that the model must have a consistency coefficient greater than 0.80 to be accepted.

Table 4 represents the relation of antecedent conditions with a strategic position in differentiation. Black circles indicate the presence of antecedent conditions, while white circles indicate their negation. The blank cells represent irrelevant conditions. Moreover, large circles indicate core conditions, and small circles refer to peripheral conditions (Fiss, 2011; Wang et al., 2019). The Goodness-of-fit is acceptable, with coverage of 0.58 and consistency of 0.82 (Glaesser, 2008; Rihoux and Ragin, 2008; Greckhamer et al., 2018).

Table 4

Outcome: Strong strategy positioning in differentiation
Path	PA	Market	HC	RC	Coverage		Consistency
					Raw	Unique
1					0.45	0.07	0.82
2					0.51	0.13	0.85
Solution coverage: 0.58 Solution consistency: 0.82
a) Algorithm: Quine-McCluskey

Source: Own elaboration.

Paths 1 and 2 are associated with strategic positioning in differentiation. In both, the dimensions associated with IC are fundamental pieces (HC and RC). The divergence between the two paths lies in the attention paid to environmental institutions. The first path pays more attention to the recommendations of the Pas (Coverage = 0.07; Consistency = 0.82). The second path pays greater attention to the trends set by the market (0.13; 0.85).

Table 5 summarizes the antecedent conditions associated with strategic positioning in cost leadership strategy. It describes the existence of four paths. Remarkably, its goodness-of-fit is acceptable, coverage of 0.82, and consistency of 0.71 (Glaesser, 2008; Rihoux and Ragin, 2008; Greckhamer et al., 2018).

Table 5

Outcome: Strong strategy positioning in cost leadership
Path	PA	Market	HC	RC	Coverage		Consist.
					Raw	Unique
1					0.43	0.12	0.76
2					0.41	0.04	0.78
3					0.50	0.16	0.78
4					0.32	0.06	0.82
Solution coverage: 0.82 Solution consistency: 0.71
a) Algorithm: Quine-McCluskey.

Source: Own elaboration.

The first configuration is characterized by the low influence of market trends and RC (Coverage = 0.12; Consistency = 0.76). The second stands out as it is deeply concerned with improving its HC but shows low attention to PAs in the profession (0.04; 0.78). The third is intensive in its IC; however, it does not pay attention to the environment (0.16; 0.78). The last configuration stands out as it is highly geared towards improving its RC: It studies the market's trends and denies the PAs' active role (0.06; 0.82).