Article Text

Cascade of decisions meet personal preferences in sciatica treatment decisions
  1. Wieteke van Dijk,
  2. Marit A C Tanke,
  3. Marjan J Meinders,
  4. Eva W Verkerk,
  5. Patrick P T Jeurissen,
  6. Gert P Westert
  1. Radboud Institute for Health Sciences, Scientific Institute for Quality of Healthcare (IQ healthcare), Radboudumc, Nijmegen, The Netherlands
  1. Correspondence to Ms Wieteke van Dijk; wieteke.vandijk{at}


Study design An in-depth interview study including patients, general practitioners, neurologists and neurosurgeons.

Objective To gain insight in decision-making in sciatica care, by identifying patients’ and physicians’ preferences for treatment options, and the differences between and within both groups.

Summary of background data Sciatica is a self-limiting condition, which can be treated both conservatively and surgically. The value of both options has been disputed, and the care pathway is known for a substantial amount of practice variation. Most Dutch patients are taken care of by general practitioners before they are referred to hospital-based neurologists, who might refer to a neurosurgeon, who can perform a surgical intervention. Dutch sciatica care thus follows the principles of stepped care, and a cascade of decisions precedes surgery. Better understanding of the decision-making within this cascade might reveal opportunities to improve shared decision-making and to reduce unwarranted practice variation.

Methods Interviews with 10 patients and 22 physicians were analysed thematically.

Results While physicians were confident of their clinical diagnosis, patients preferred confirmation trough imaging to exclude other possible explanations. Furthermore, many patients showed reluctance towards the use of (strong) opioids, while all physicians favoured this and underlined the benefits of opioids in the management of sciatica complaints, to buy time and to allow patients to recover naturally. Finally, individual physicians differed strongly in their opinion on benefits and optimal timing of surgical treatment and epidural injections.

Conclusions Dutch sciatica care is characterised by a cascade of decisions preceding surgery. Preferences differ within and between patients and physicians, which adds to the practice variation. To improve decision-making, physicians and patients should invest not necessarily more in the exchange of options or preferences, but in making sure the other understands the rationale behind them.

  • back pain
  • decision making
  • qualitative research

Data availability statement

No data are available.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Sciatica can be treated conservatively and surgically and is known to have a large practice variation.


  • The decision-making process in sciatica treatment includes two critical steps: (1) diagnosing of the problem and (2) deciding on the type of treatment. For patients certainty of diagnosis was most important, while for physicians timing and type of intervention was most important.


  • This study reveals that individual preferences guide decision in sciatica care, possibly resulting in high practice variation. Implementing shared decision-making in this care pathway as a strategy to reduces unwarranted practice variation can be improved.


Sciatica can be managed both conservatively and surgically. In the Netherlands, sciatica care is organised following the stepped care principle, involving a general practitioner (GP) as gatekeeper for hospital care, and a hospital-based neurologist, before a neurosurgeon is consulted. Therefore, a chain of decisions and referrals precede surgery. Both the diagnostic process and timing of surgical treatment and the (societal) value of surgery over conservative treatment are topics of ongoing discussion, and practice variation has been widely reported.1–4 This practice variation may be driven by both differences between and within patient and provider preferences.5 This makes the management of sciatica of interest. Which preferences of patients and physician(s) shape these decisions?

In the Netherlands, the GP is the point of entry for non-acute health complaints. For sciatica, the GP guideline advises conservative treatment for 6–8 weeks, combining a strategy of watchful waiting, informing the patient and providing painkillers. If the patients’ complaints last longer, the GP can refer to a neurologist.6 Neurologists can confirm the diagnosis and might order imaging. However, the neurologists’ guideline advises against imaging, unless surgery is considered or there is doubt about underlying cause or pathology.7 Recent research showed that only 11% of Dutch neurologists routinely order imaging.8 Apart from conservative options, the neurologist can also prescribe more invasive pain treatments, usually performed by a pain team. In case of persisting complaints, the patient might be referred to a neurosurgeon or orthopaedic surgeon to consider surgery. The Dutch neurosurgery guideline advices not to perform surgery on patients with less than 8 weeks of complaints, and proposes surgery after more than 6 months of complaints without any proper improvement.7 In the intermediary period natural recovery could occur. Therefore, a shared decision on treatment should be pursued with the patient. A multidisciplinary guideline exists, of which the implementation faces difficulties.9

Resuming, Dutch sciatica care is characterised by involvement of multiple stakeholders, uncertainty in the aetiology of the disease and uncertainty about the value of diverse treatment options in each step of the care pathway. As a result, practice variation in sciatica care persists, despite efforts to harmonise care processes. Shared decision-making (SDM) is warranted in this situation, because given the uncertainties, patients preferences are extra important to take into account. However, research on the actual preferences that guide the different stakeholders in their decision-making process is not available. This qualitative study aims to: (1) identify key moments in this care pathway and (2) map the drivers and arguments in the decisions among the involved stakeholders.

Materials and methods

Qualitative methods were appropriate for this research question, because the goal was to discover underlying arguments and considerations. region. The COREQ checklist (COnsolidated criteria for REporting Qualitative research) was followed and is included in the online supplemental file 1.10

Supplemental material

Data collection

WD conducted in-depth semistructured interviews with sciatica patients (10), physiotherapists (3), GPs (7), neurologists (6) and neurosurgeons (6). Respondents were purposively recruited. Respondents were approached in several ways. The physicians mostly with snowballing, after first contacts were made through the personal networks of the project team. For the GPs, the working group ‘movement disorders’ of the Dutch General Practitioners’ Society was contacted. Physicians were asked to distribute a leaflet among eligible patients in their practice. In addition, an online call on the website of the Dutch association for sciatica was placed. One patient was reached through the personal network of one of the project group members. We aimed to reach a diverse sample of representatives from all stakeholders involved in sciatica care.

The semistructured interview guide for physicians contained prompts and questions about experiences with care for sciatica patients and the deliberations regarding treatment alternatives. The interview guide for patients focused on the personal patient journey and deliberations patients recalled regarding their treatment decisions. Table 1 provides an overview of the composition of the sample of physicians, table 2 provides an overview of the composition of the sample of patients.

Table 1

Basic characteristics of the sample of physicians

Table 2

Basic characteristics of the interviewed patients

The interviews with physiotherapist quickly revealed that, although they treated patients with back complaints, their involvement in patients with severe sciatica complaints (with surgery as a treatment option) was limited. As the decisional trajectory towards surgery was our main topic, we decided not to pursue further interviews with physiotherapists.

During the interview, patient 2 appeared to have misunderstood the inclusion criteria. He suffered not from sciatica. Therefore, this respondent was excluded from the analysis.

Data analysis

All interviews were transcribed verbatim and anonymised. Thematic analysis was conducted with three coders (WD1, EV2 and RS3), using Atlas-ti.11 WD1 and EV2 coded and analysed all physicians, WD1 and RS3 coded and analysed all patients. First, a sample of the same two interviews was coded separately. Next, the coders compared their findings and discussed shared themes. Then each coded two more interviews, further developing the code scheme. The previously coded interviews were reread using the improved code scheme, to check if coding could be more accurate. The team selected and translated (from Dutch) the key quotes that illustrated the themes. Literal translation was pursued, with respect to the natural ‘flow’ or ‘stammer’ in a quote.


Four steps in the care process proved relevant for the clinical pathway of patients. These were uncertainty about the diagnosis, analgesia, epidural injections and surgery. Patients and physicians differed in opinion on the subjects of a certain diagnosis and analgesia. They had relatively similar opinions with respect to the different treatment options. Different options existed on the topic of surgical treatment, within both the group of patients as well as the group of physicians. A summary of the positions of patients and physicians on the four steps is presented in table 3.

Table 3

A summary of the perspectives of patients and physicians on the four main topics

Certainty and diagnosis

After the onset of sciatica, two prominent decisions or conflicts appear, concerning (1) diagnosing sciatica and the need for MRI-imaging to confirm the diagnosis and (2) the appropriate timing for referral to the hospital. Physicians felt confident about diagnosing sciatica, although the GPs mentioned that they sometimes doubted their diagnosis. They would discuss such doubts with their patients and generally would wait to see how symptoms developed. GPs acknowledged that their first objective in patients with subjected sciatica is to help the patient through the first period of pain and restricted mobility, rather than to determine the ‘true’ diagnosis rapidly. When patients did not improve, or pushed for referral, GPs would refer to a neurologist earlier than the 6–8 weeks suggested in the guideline. However, because of waiting lists, these patients often waited a few weeks before a specialist was available, and the total duration of complaints generally exceeded 6 weeks.

Patients understood the 6–8 weeks of conservative treatment by the GP as mandatory ‘waiting time’, after which their diagnosis would be confirmed by a specialist and with an MRI scan. When a neurologist deemed the scan to be unnecessary, patients were disappointed. Physicians were conscious of such patient expectations. Some kept refusing, others gave in:

Sometimes you notice within a minute that whatever you say, it won't matter.(…)And in such cases you can say ‘according to the guideline you are not entitled to a scan’. But I am not treating the guideline, I am treating the patient. In my experience, if you do not refer such a patient for a scan, if you do not manage to get the patient to understand, and you kind of quarrel with the patient, next month you will receive information from another hospital and they'll have made the scan. (N3)

Actually, I never have angry patients because they want surgery per se. I do have patients who are angry because they are refused to get a MRI. But I consider that to be something different. Why is that different? Because it is not related to the decision to pursue surgery or no surgery. When deciding to pursue surgery or not you always have the patients best interest at hart. (N4)

Sometimes I notice that patients first of all need more certainty about their diagnosis. And when they do have that certainty, they can accept that. This also depends on what they can expect from surgery. How long will the recovery take? How will surgery influence their quality of life, afterwards? In my experience, people sometimes say: ‘well, let’s wait a little bit more’. They are scared of surgery and have lots of fears about it. For example, the fear that for surgery they need to receive general anaesthesia, and they don’t want to. This can all play a role in the decision making. (GP6)

Both neurologists and neurosurgeons stated that they generally did not need an MRI scan to confirm the diagnosis. They trusted their clinical judgement and only required imaging for a-typical cases.For patients, this was difficult to grasp. All patients whose diagnosis was not confirmed by a scan kept ‘doubting’ their diagnosis.

So I say, dear neurologist, that’s not my spine. That’s a model on a table, my spine is in my back. And you think you can see on that model what’s wrong with my back? I thought I came here for a scan and to find out what’s in my back. No, he says, that MRI is only needed for the specialist pain team. (P10).


When it comes to analgesia, patients were hesitant towards using them, especially opioids, while physicians advocated their benefits. Patients were critical towards potential side effects, but accepted using them after either the GP, neurologist or pain specialist explained their benefits. They feared to become dependent on opioids and were afraid of using strong painkillers.

It is a mixed feeling, I would prefer to quit all medication. I would prefer to, but I also notice that the medication is necessary to be able to move. That’s a very strange balance. It don't think it’s right, I struggle with it. Because I feel as if I do not function as the real me. (P10)

Don’t reduce your medication too soon, because you need it, take it slow. I thought, well okay, if he says so… because I need someone to slow me down. (P11)

Did it help? The pain medication? Well, not at first. Then we got Tramadol as well and that didn’t work either, and then in combination with paracetamol, that knocked me of my feet. Then I just, eh, it was like I was totally drunk. (P4)

Physicians were aware of this hesitance, and stimulated patients to use painkillers to give natural recovery more time. Physicians also stated that they usually followed the WHO analgesic ladder and prescribed stronger medications when necessary, which is supported by the guidelines. They tried to explain why more aggressive painkillers, such as opioids, could be designated for these complaints. Not all patients were equally reluctant to use these medications under the circumstances, but all distrusted opiates.

Because, many patients resist using pain medication. Because they are scared of its side effects, or because they fear that they will be unable to feel the sciatica get worse. You have to discuss these fears, because people might push for surgery to avoid pain medication. (N3)

I almost never meet people who do not want pain medication. Because if it is really a radicular syndrome, well… then you want something… yes, then you really want something. So that is not really a problem. But they do experience difficulty from the side effects. Mostly with the morphine, the opioids. Yes, that makes people drowsy, it makes it difficult to go to the toilet, they are really bothered with it. So, that are problems you meet, but well, they have no other option. (GP1)

Lots of the people I treat need a revision of their pain medication. That is sort of related to who is prescribing. Of course, there a lot of literature about this topic and there also is some fright at the side of the professional, but when people are in a lot of pain you need to prescribe a lot of painkilling quickly. You better start with opioids and then reduce to paracetamol, compared to starting with paracetamol and a bit of this and a bit of that. You see? That’s following the WHO pain ladder from bottom to top. But with acute pain you’d better follow it from top to bottom. Yet I do notice, although I cannot support this notion with literature, but I do notice that GP’s are careful to be too aggressive with painkilling, so I often meet people who are still in quite some pain. (N5)

Epidural injections

For epidural injections differences were found among physicians, more so than among patients. Some physicians prescribed them regularly for sciatica, although in this small sample there were more opponents than proponents. One neurologist was a strong believer in epidural injections and was setting up a randomised controlled trial to prove their effectiveness.

Why do I prefer epidurals? Because, in my experience it works well, although there are always patients for whom it does not work. Those patients are quick to call and ask for another, the following step. (N5)

And treatment by epidural injection, do you prescribe that as well? Yes, sure. Especially with elderly people, with lots of other morbidities or when you think: ‘in this case surgery is really unpleasant, but she has so much pain’. Then we use them sometimes. Or people of whom you think ‘well, I have doubts about the amount of compression and the test injection worked very well’. And when the neurosurgeon says: ‘I think this compression is to limited for surgery to be of use’. (N6)

However, apart from two proponents, physicians were sceptical and had had more negative than positive experiences. Patients were also moderately inclined against epidural injections, but were willing to give it a try for the doctors sake.

So they discussed it: surgery or an epidural injection? On the one hand I did not want an injection, because I did not believe it would work. On the other hand, if they I advise it you have to be open to it. (P1)

So I had a pain blocker [epidural injection), but that didn’t work at all. No, it didn’t help? No, those never help, those pain blockers. I’ve spoken with so many people who had one, but it never helps. (P8)


Proponents and opponents of surgery were found among both physicians and patients. Opposing patients expressed fear and doubted the effectiveness of the procedure. Patients that proposed for surgery underlined the difficulty of their personal situation and hoped for quicker recovery than with conservative treatment.

Yes, well, choice, they let me choose, surgery or rest. But it was so troublesome that I said if surgery will help, then why not? (P3)

I am scared of surgery, because I am a nurse myself and I know how to judge the evidence. And people tell me that surgery is not scientifically proven to be effective. (P7)

Yes, well, it is quite an operation. I know I’ll probably be home within a night, but still, I found it a bit scary, to do. Also because there is another herniated disk underneath. When the one is removed, what will the other do? How will the scare tissue develop? Yeah, I did read a lot about it. So yes… In a way, I think, I’m… I’m a bit scared to get the surgery done. (P9)

Opposing physicians underlined that they wanted to maximise the chance of natural recovery, that they did not want to misuse public money, and that they feared the chance of complications or irreparable damage to a surgical treatment. On the other hand, proposing physicians focused on (the possibility of) quicker recovery with surgery, fewer residual complaints and earlier work resumption. Notably, neurologists held strong views, either pro or con, which influenced their referring behaviour.

Too easy access to surgery can be deadly for patients. Deadly, really? Literally, because of the risk of complications, but also because some patients are operated on who would have recovered without surgery. And surgery always does harm as well. (N1)

Those large studies of Peul, the Sciatica trials, conclude that you should not operate. All neurologists have read that. All policy makers have read that. (…) But that interpretation is completely wrong. What they really say is that íf you do want to perform surgery, do it quickly. (NC6)

Interviewer: “And of 5 assessments [of MRI scans), overall, how many proceed to surgery? Respondent: I estimate about 60%. Three out of five. Why do those other two do not receive surgery? Yes, that depends. Partly, I think, it is patient related, like when complaints have decreased, that the situation is improving. Sometimes patients refuse surgery. Or patients want to wait, when they hear about other treatment options. Sometimes they chose one of the other treatment options. And sometimes they just don’t want surgery. Or, but that is really rare, if I think the risks of surgery are too high, I’ll try to steer them towards an alternative option. That are the patient related aspects. And then there are, well, the MRI related aspects, so to speak, when the MRI does show a herniated disk but it is an a-typical case. Or the MRI does not show deviations that can explain the complaints. Or the deviations do not correspond to the complaints. (NC2)

People do have strong preferences about surgery or not. They really have. When people are self-employed and cannot miss the earnings of 3 months of working, well yes, that is an argument for surgery. I listen to what people say. I do, I do have a certain advice in mind, but then you get into a dialogue, and then you migrate towards an advice that suits them. (N2)


The results of this study give insight into the critical decision-making steps in the care pathway of sciatica. The two key elements are: (1) diagnosis of the problem and (2) deciding on the type of treatment. The position that patients and their physicians took differed, also within the groups of patients and physicians.

For patients, certainty about their diagnosis was most important. They preferred the evidence of MRI. This is congruent with earlier research, which showed that patients placed much emphasis on the need for a confirmed diagnosis, as a starting point for further treatment decisions.12 13 Patients’ preference for imaging is well known, although not fully understood.14 A systematic review of qualitative studies about patients with low back pain and sciatica concluded that for patients, imaging gives more certainty and excludes other possible explanations.13 The patients we interviewed experienced the first weeks of GP-led conservative treatment as mandatory waiting time before they would receive a scan for confirmation.

Physicians seldom felt that they required imaging to confirm the diagnosis. This holds especially for neurologists and neurosurgeons. This is concurrent with research that showed that 89% of Dutch neurologists only order imaging under specific circumstances, such as after a long period of pain or with an abnormal neurological exam.8 For 27% of these neurologists, the patient requesting for an MRI is also sufficient reason to order imaging. For physicians, the presentation of sciatica is often so recognisable that confirmation by imaging is not required. Patients find this difficult to grasp.

Among physicians, the timing of intervention was the main subject of discussion. Physicians differed strongly in their opinions about the benefits of surgical or conservative treatment, especially with regard to the timing of surgery. In the Dutch context, the neurologist appears to be the most influential decision-maker here: he or she can accelerate or delay the decision to pursue surgical treatment, and decides about the timing of involvement of the neurosurgeon. Between approximately 6 weeks and 6 months after complaints have started, patients with persisting symptoms see a neurologist, who, with the patient, drives the decision about conservative treatment or surgical intervention. Without referral to a neurosurgeon, surgery is highly unlikely within the Dutch context. Of course, neurosurgeons can delay surgery further. Yet, they are unable to make it happen earlier.

The interviewed sciatica patients were reluctant to use strong painkillers. This is congruent with earlier research that reports relatively reserved use of opioids in the Netherlands.15–17 Nevertheless, the use of analgesia has increased dramatically over the last decades.15 Broad concern has risen about opioid use and epidemics of opioid deaths, especially in the USA.18 In the Netherlands, this does not seem an immediate problem, as problematic use of opioids is relatively rare and at the fourth-lowest level in the EU.19 This study suggests that physicians were less scared of opioids than their patients.

Implications for practice

This study gives some indications for why practice variation still prevails in sciatica treatment. Personal preferences of patients and physicians guide decisions, within the boundaries set by guidelines and evidence. Based on this observation, SDM, as a strategy to reduce unwarranted practice variation, can be improved. A basic model to reach SDM is by following three steps: choice talk, option talk and decision talk.20 While Dutch patients are aware that there is a choice in this situation and do form individual preferences, this study suggests that the underlying, supporting process of deliberation is not fully developed. As a result, individual preferences of either the patient or the physician guide the care pathway more strongly than acknowledged or preferrable. To improve SDM Dutch physicians and patients should invest not necessarily more in the exchange of options or preferences, but in making sure the other understands the rationale behind them, as well as the applicability of the situation to the patient’s individual context. This means not only presenting the treatment options, but an in depth discussion of which option is most fitting for what situation, and why both parties in the conversation think so. The revised model of the three steps of SDM places active listening and deliberation at the centre.21 Implementation of this improved model would possibly bridge the gap between patients and physicians in this stage of decision-making. Implementing SDM in clinical practice takes training, practice and requires adjustment in the way physicians were used to work.22 For the implementation of SDM in this care pathway, explicating the benefits and risks, adjusted to the situation of the patient, could be included more prominently in the clinical guidelines.

Strengths and limitations

One major strength of our study is that we have added the perspective of all important physicians involved. We learnt that a range of physicians, notably neurologists, have a profound influence on patients’ chances to receive surgery. The neurologist times the moment of referral to the neurosurgeon, and thereby strongly influences the timing of surgery. Patients’ expectations or preferences appeared to be influential, be it moderated through the guidance of the physicians they met. Since our sample covered neurosurgeons working both in public and private clinics and patients that underwent surgery in public hospitals or private clinics, our interviews reflect a broad sample of physicians and patients in this decision-making process in the Dutch context. Though we included all important physicians involved in the care pathway, the inclusion of pain specialists might have contributed to the further understanding of the use of painkillers and opiates.

An important limitation of this study is that we were unable to reach patients who had complaints for less than 6 weeks. Furthermore, we may have missed typical patients who bypass classical care pathways. However, bypassing the GP is uncommon in the Netherlands, because a referral by a GP is needed to get hospital care. Further, GP consultations are fully covered by health insurance without any copayment.23


In conclusion, this study shows how the stepped care pathway of sciatica patients in the Netherlands is influenced by individual patient and physician preferences. The neurologist appears to be of central influence. To improve SDM, investments should be made in the deliberation about options and preferences and in a better understanding of the underlying rationale for an individual patient and the physician.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants, but as ethical approval is not required for this type of study under Dutch law, an exemption was obtained by the Medical Ethics Committee CMO Regio Arnhem-Nijmegen (dossier nr. 2015-1760; dossier code: YwNur). Participants gave informed consent to participate in the study before taking part.


We thank R. Sanders, MSc. for her participation in the analysis of this research.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors WvD (corresponding author, also author acting as guarantor) designed the study, conducted the interviews, transcribed all recorded interviews verbatim, conducted the qualitative analysis, drafted and revised the manuscript. MACT designed the study, participated in the interpretation of the analysis, revised the manuscript. MJM designed the study, participated in the interpretation of the analysis, revised the manuscript, reviewed the final manuscript. EWV contributed to the qualitative analysis, participated in the interpretation of the analysis. PPTJ designed the study, reviewed the final manuscript. GPW designed the study, reviewed the final manuscript. RS contributed to the qualitative analysis.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.