Table 3 Qualitative checklist of eight clinical studies on chiropractic primary or early secondary prevention

Kessinger

(1997)

JVSR

?

USA

Prospective outcome study of lung function after 2 weeks of chiropractic care

No placebo or control group

NA because no control group

FEV-1 reported to be most reproducible of the two measurements with ref. provided but level of reproducibility not reported

NA

There was only one group

NA

There was only one treatment group

To test the effect on pulmonary function after chiropractic care of the neck, one could a) either compare it to a sham treatment, or to b) another type of treatment known to be effective or (possibly) to c) a treatment elsewhere in the spine, if the purpose is to see if the ‘neck’ is important.

Kessinger

(1998)

JVSR

?

USA

Prospective outcome study of visual acuity after six weeks of chiropractic.

Also the dose of treatment was studied.

No placebo or control group

NA because no control group

Not reported but used standard eye chart

NA

There was only one group

NA

There was only one treatment group

To test if the ‘dose’ of adjustments matter, patients should at baseline be randomly allocated into one of several groups each receiving different numbers of treatments/adjustment.

Morter

(1998)

JVSR

Morter Health System, Inc. and developer of Bio-Energetic Synchronization Technique

USA

Prospective outcome study of salivary pH in two groups defined as predominantly sympathetic or parasympathetic after 4-days of chiropractic treatment

No placebo or control group but patients were all treated in the same way and outcomes were compared in relation to whether they predominantly were sympathetic or parasympathetic

NA because study sample stratified on predetermined criteria

Not reported but used standard pH paper

NA

There was only one group

NA

There was only one treatment group

To establish if different subgroups react differently to the chiropractic treatment, then the groups could either a) be tested for outcome in a randomized controlled clinical trial design or (possibly) b) be tested for outcome in a sufficiently large non-controlled prospective outcome study that allows for subgroup analyses. The diagnosis of predominantly sympathetic and parasympathetic subjects must be valid and/or reproducible. The assessment should be done with valid/reproducible methods by assessors that are blinded to classification group.

Campbell

(2005)

JVSR

Camgen, Inc.

USA

A retrospective study comparing serum thiol levels in patients with active disease (? Abstract)/apparently disease free (? Materials and Methods) for two groups (? Materials and Methods) or perhaps three groups (? Table 1).

These groups had been treated with chiropractic care for a) less than one year or b) at least one year. Perhaps there was also a third non-symptomatic apparently healthy control group (? Table 1).

Perhaps, not clear

No

Serum thiols claimed to be valid as indicators for mortality and active disease

NA

There was only one group

Yes

To test if chiropractic care and dose of care can affect DNA repair then a study sample should have been randomly divided into treated and untreated, and this could have been done for different study populations, the sick and the healthy. The dose-response should be tested in a similar way, i.e. a group of patients receiving short-term and one long-term treatment in a random fashion.

The results in this study relate only to association and not effect. The claim that “results clearly support” etc. are unfounded.

Boone

(2006)

JVSR

Sherman College of Straight Chiropractic

USA

Prospective outcome study of immune status and health after three and nine months of chiropractic care

NA because no control group

NA because no control group

Not reported

NA

There was only one group

NA

There was only one treatment group

Just because a study sample is small, does not justify to call it a ‘pilot study’.

A pilot study should be used to test study procedures, ability to obtain patients, etc.

To draw any (even preliminary) conclusions on effect/benefits of treatment, a sham/control treatment is needed.

McMasters

(2013)

Journal of Chiropractic Medicine

Private practice

USA

Prospective outcome study of blood pressure after 21 to 23 chiropractic consultations

NA because no control group

NA because no control group

Probably valid

NA

There was only one group

NA

There was only one treatment group

As this is a feasibility study, it is not really appropriate to concentrate the discussion on ‘improvement’ but should concentrate more on reasons for/against the possibility to perform a proper randomized controlled trial (RCT).

To test the effect of spinal manipulation on blood pressure an RCT with a sham group would be necessary.

Jones

(2014)

Disability and Rehabilitation

School of Health Science and Social Care

UK

2-arm randomized controlled trial of dysfunctional breathing after 2, 4, 8, 12 and 26 weeks of either a) respiratory management (RM) or b) RM plus manual therapy

Yes, with a control group

Yes

Yes, for the questionnaire. The other variables are frequently used so probably valid.

Yes

Yes

The design is appropriate for testing difference in outcome between treatment groups, in this case to see if manual therapy can provide added benefit to another treatment.

Goertz

(2016)

JMPT

Palmer college of Chiropractic

USA

2-arm randomized controlled trial of blood pressure after 1, 6 and 12 visits of spinal manipulation

Yes, with a sham group

Yes

Probably valid

Yes

Yes

The design is appropriate for testing whether spinal manipulation has an effect on blood pressure. However, the absence of effect should be discussed more clearly.